Re:임중혁 교수의 지병에 관하여...

[김기모]

우선 의과대학교수로 있는 친구에게서 받은 이멜을 올린다.

의약계 있는 친구들이 읽어보면 나보다 무슨 말인지 잘 알끼라.. 영욱아, 동식아.. 도와줘..

금요일경 미국에 돌아가는대로 미시간의대에 있는 몇친구들 직접 만나서 이야기해보기로 했다. 이번 주말경에 어느정도 정리된 소식을 다시 전할께.. 

-------------------------

Kimo,

Below is a topic review taken from the UpToDate service, I thought you might find it of interest for your friend in Seoul, Korea.
UpToDate is a clinical information service that has thousands of topic reviews such as this one, designed to give instant answers to the questions that arise in practice.

---

Official topic review from UpToDate^??, the clinical information service on the web, CD-ROM and Pocket PC. To subscribe to UpToDate^??, visit us online at https://store.utdol.com/app/index.asp.

ATS guidelines: Diagnosis and treatment of disease caused by non-TB mycobacteria

ACKNOWLEDGMENT — This official statement of the American Thoracic Society was approved by the Board of Directors, March 1997. Published by The American Thoracic Society, Am J Respir Crit Care Med 1997; 156:S1.

This statement was prepared by an ad hoc committee of the Scientific Assembly on Microbiology, Tuberculosis, and Pulmonary Infections. Members of the committee were: Richard J Wallace, Jr, MD, Jeffrey Glassroth, MD, David E Griffith, MD, Kenneth N Oliver, MD, James L Cook, MD, Fred Gordin, MD. Outside consultants: Drs. Timothy Kiehn, Laboratory; Clark Inderlied, Laboratory; Mark Goldberger, HIV Issues; and Barbara B Brown, Laboratory.

SUMMARY
Diagnostic criteria in HIV seropositive and seronegative hosts — The following criteria apply to symptomatic patients with infiltrate, nodular or cavitary disease, or a high resolution computed tomography scan that shows multifocal bronchiectasis and/or multiple small nodules.

A. If three sputum/bronchial wash results are available from the previous 12 months:
  1. Three positive cultures with negative acid-fast bacilli (AFB) smear results or
  2. Two positive cultures and one positive AFB smear

B. If only one bronchial wash is available:
  1. Positive culture with a 2+, 3+, or 4+ AFB smear or 2+, 3+, or 4+ growth on solid media

C. If sputum/bronchial wash eval‎uations are nondiagnostic or another disease cannot be excluded:
  1. Transbronchial or lung biopsy yielding nontuberculous mycobacteria (NTM) or
  2. Biopsy showing mycobacterial histopathologic features (granulomatous inflammation and/or AFB) and one or more sputums or bronchial washings are positive for an NTM, even in low numbers

  Comments — These criteria fit best with M avium complex, M abscessus, and M kansasii. Too little is known of other NTM to be certain how applicable these criteria will be.

At least three respiratory samples should be eval‎uated from each patient. Other reasonable causes for the disease should be excluded. Expert consultation should be sought when diagnostic difficulties are encountered.

Key laboratory features of the nontuberculous mycobacteria
  Staining and culture — Current methods of specimen staining and culture used for M tuberculosis are acceptable for most NTM species. The preferred methodology includes fluorochrome staining and culture in a liquid medium as well as on Middlebrook 7H10 or 7H11 agar. Species for which special growth conditions are needed include those responsible for cutaneous disease, which need lower incubation temperatures, and the relatively fastidious species M haemophilum, M genavense, and M conspicuum.

  Species identification — Methods of rapid species identification including commercial DNA probes (M avium complex, M kansasii, M gordonae) and high-pressure liquid chromatography are preferred over the slower traditional biochemical methods.

  Susceptibility testing of M avium complex — Susceptibility testing of clinically significant rapidly growing mycobacteria (M fortuitum, M abscessus, M chelonae) should not be performed with the antituberculosis agents. They should be tested against antibacterial drugs including amikacin, doxycycline, imipenem, the fluorinated quinolones, a sulfonamide, cefoxitin, and clarithromycin.

  Susceptibility testing of M Kansasii — Routine susceptibility testing of M Kansasii should include only rifampin, because currently used resistance breakpoints for isoniazid and streptomycin often give misleading results and methods for the other drugs have not been established.

  Susceptibility testing of the rapid growers — Susceptibility testing of clinically significant rapidly growing mycobacteria (M fortuitum, M abscessus, M chelonae) should not be performed with the antituberculosis agents. They should be tested against antibacterial drugs including amikacin, doxycycline, imipenem, the fluorinated quinolones, a sulfonamide, cefoxitin, and clarithromycin.

Prophylaxis and treatment of nontuberculous mycobacteria disease
  Treatment of M kansasii pulmonary disease — A regimen of daily isoniazid (300 mg), rifampin (600 mg), and ethambutol (25 mg/kg for 2 months, then 15 mg/kg) for 18 months with a minimum of 12 months' culture negativity is recommended for pulmonary disease in adults caused by M kansasii. Clarithromycin or rifabutin will need to be substituted for rifampin in HIV-positive patients who take protease inhibitors.

  Treatment of M avium complex pulmonary disease — A regimen of daily clarithromycin (500 mg twice a day) or azithromycin (250 mg), rifampin (600 mg) or rifabutin (300 mg), and ethambutol (25 mg/kg for two months, then 15 mg/kg) is recommended for therapy of adults not infected with the HIV virus. Streptomycin two to three times per week should be considered for the first eight weeks as tolerated. Patients should be treated until culture-negative on therapy for one year.

  Treatment of disseminated M avium complex disease — Therapy in adults should include daily clarithromycin (500 mg twice a day) or azithromycin (250 to 500 mg), plus ethambutol 15 mg/kg per day. Consideration should be given to the addition of a third drug (preferably rifabutin at a dose of 300 mg/d). Therapy should be continued for life until more data becomes available.

  Prophylaxis of disseminated M avium complex disease — Prophylaxis should be given to adults with the acquired immune deficiency syndrome (AIDS) with CD4 counts < 50 cells, especially with a history of a prior opportunistic infection. Rifabutin 300 mg/d, clarithromycin 500 mg twice daily, azithromycin 1,200 mg once weekly and azithromycin 1,200 mg once weekly plus rifabutin 300 mg daily are all proven effective regimens.

  Treatment of nontuberculous mycobacteria cervical Iymphadenitis — Nontuberculous mycobacteria cervical Iymphadenitis is still treated primarily by surgical excision alone, with a 95 percent cure rate. A clarithromycin- containing regimen should be considered for patients with extensive disease or a poor response to surgery.

  Treatment of nonpulmonary rapidly growing mycobacteria — Therapy of nonpulmonary disease caused by M fortuitum, M abscessus, and M chelonae should include drugs such as amikacin and clarithromycin, based on in vitro susceptibility tests.

INTRODUCTION — The continued growth in the number and preval‎ence of mycobacteria species other than the Mycobacterium tuberculosis complex, and recent advances in diagnostic methods and drug therapies for disease caused by these agents, has prompted us to put forth this second, updated diagnostic and therapeutic standard that deals exclusively with the nontuberculous mycobacteria. As in the first statement [1], we refer to these mycobacterial species collectively as the nontuberculous mycobacteria (NTM). The principles of therapy and diagnosis of disease caused by M tuberculosis have been dealt with separately and appear in two ATS statements published most recently in 1990 and 1994 [2,3]. Like the previous NTM statement published in 1990 [1], this statement is designed as a basic guide for professionals involved in the diagnosis and management of disease caused by NTM Although not all-inclusive, the areas of discussion are referenced in enough detail to allow the reader to assess the scientific basis for ideas and recommendations that are put forth. Included within this statement are revised recommendations for diagnostic criteria that apply primarily to the NTM and updated recommendations of specific therapeutic drug regimens for disease caused by M avium complex and other species of NTM, recognizing the major impact of the newer macrolides and rifabutin, which have become available since 1990. Unless otherwise stated, these drug dosages are for adults. Pediatric doses are described where available.

EPIDEMIOLOGY AND PATHOGENESIS
Sources of infection — Most NTM organisms have been isolated from water and soil [4,5]. The best studied of these has been M avium complex. Extensive environmental studies in the United States have shown that M avium complex grows well in natural waters, particularly in the Southeast [6]. Mycobacterium avium complex strains with plasmids, possibly associated with virulence, have been shown to be preferentially aerosolized, providing a possible mechanism for airborne acquisition of these organisms [7]. Although M avium is an important cause of disease in poultry and swine, serologic studies have suggested that animal-to-human transmission is not important in human infection [8], and recent molecular studies involving IS901-IS902 and IS1245 have shown that strains infecting humans and animals (especially swine) are different [9,10]. It is now generally accepted that environmental sources, especially natural waters, are the reservoir for most human infections caused by M avium complex. The reservoir of M avium for most patients with disseminated disease has not been identified, but it is assumed to be the same as or similar to that for patients with non-HIV-related M avium complex lung disease. Mycobacterium avium complex is present in tap water, and one study of disseminated M avium complex disease in AIDS demonstrated that some cases are likely acquired from hospital tap water [11]. Interestingly, there does not appear to be a geographic predilection with disseminated disease in the United States, as there is with skin test reactivity and with chronic lung disease.

Water is also the likely source of infection for numerous other NTM species including M marinum, M kansasii, nosocomial outbreaks or pseudo-outbreaks due to rapidly growing mycobacteria, M xenopi, and M simiae. Mycobacterium marinum has been commonly associated with salt water, fresh water, fish tanks, and swimming pools [12]. Mycobacterium kansasii has not been recovered from soil or natural water supplies [13]. It has been isolated repeatedly, however, from tap water [14,15] in the same communities where M kansasii disease exists. Interestingly, it has been shown to survive up to 12 months in tap water but not in soil.

Rapidly growing mycobacteria such as M fortuitum, M chelonae, and M abscessus can be recovered from soil and natural water supplies, and are the most common NTM associated with nosocomial disease [16-24]. Investigations of nosocomial outbreaks or pseudo-outbreaks caused by these species including the use of DNA fingerprinting with pulsed-field gel electrophoresis [25,26] have demonstrated that tap water [18,19], ice prepared from tap water [20,21], processed tap water used for dialysis [22], and distilled water used for preparing solutions such as gentian violet [23,24] are the usual nosocomial sources of the organisms.

Mycobacterium xenopi is an obligate thermophile that requires temperatures of 28º C or above to grow [4]. It has been recovered almost exclusively from hot water and hot water taps within hospitals [15,27-29], where it has been associated with multiple positive [ie, probably contaminated] clinical samples and a few cases of clinical pulmonary and soft tissue disease [28-30]. These clusters of hospital isolates have been reported from the United States, the United Kingdom, and other areas in Europe. In two studies, the clinical isolates and hospital water isolates have been shown to be identical by DNA fingerprinting [28,29]. It has been speculated that the organism enters the hospital from municipal water mains, then multiplies in the hospital heating tanks where the temperature is 43-45º C, the optimal temperature for growth of this organism [30].

Reports of recovery of M simiae from clinical specimens have been clustered in three geographic areas: Israel [31], Cuba, and the southwestern United States—Texas, Arizona, and New Mexico [32-34]. Most recoveries have been single positive specimens that are smear-negative [32,33] and not associated with clinical disease [33], suggesting environmental contamination as a likely source. For several clusters of isolates, organisms were also recovered from the local tap water [34,35, M Yakrus, personal communication], suggesting it as the likely organism source.

Mycobacterium malmoense, which has emerged as a major NTM pathogen in northern Europe, has been recovered from natural waters in Finland [35] and soils in Zaire [36] and Japan [37]. The recently recognized pathogen M genavense has not been recovered from soil or water, but it has been recovered from a dog and a variety of pet birds including psittacine birds [38,39]. Mycobacterium ulcerans disease occurs in discrete but widely dispersed geographic areas in the watersheds of tropical rain forests, primarily in Africa, Southeast Asia, Australia, and South and Central America [40].

Much less is known about the environmental epidemiology and sources of infection for the other NTM A number of NTM species have yet to be recovered from the environment, including M ulcerans, M haemophilum, M szulgai, M celatum, M genavense, and M conspicuum. Despite this, environmental sources of infection are highly likely. Good reviews of environmental studies have been provided by Wolinsky and Rynearson [13], Portaels [40], and Falkinham [41].

Much remains to be understood about the pathogenesis of NTM infection and disease in humans. Epidemiologic studies, skin test surveys, and more recently DNA fingerprinting studies suggest that person-to-person transmission of infection is rare. It is assumed that most persons are infected by environmental NTM Of the likely sources of infection, airborne NTM may play an important role in respiratory disease, whereas ingestion may be the source of infection for children with NTM cervical Iymphadenitis and for most patients with AIDS whose disseminated M avium or M genavense begins as gastrointestinal colonization. Bacteremic spread of the organism in patients with AIDS then involves multiple organ systems, including bone marrow, Iymph nodes, liver, and spleen. Direct inoculation with NTM organisms from water or other material is likely the source of infection for patients with soft tissue infections. It is not known whether NTM disease (especially pulmonary disease) develops soon after infection or, like tuberculosis, develops after a period of latency.

Preval‎ence in humans — Although first observed soon after Koch's discovery of the tubercle bacillus, NTM were not widely recognized as human pathogens until the 1950s, when several large series of patients with NTM lung disease were reported [42-44]. These patients were epidemiologically distinct from patients with tuberculosis, being older, more commonly white, and quite often having underlying chronic lung disease such as bronchiectasis, silicosis, and healed tuberculosis. Positive reactions of 10 mm or more to purified protein derivative (PPD) tuberculin were less common than among tuberculous patients, and family contacts tended to be tuberculin-negative.

As reports of patients with NTM disease increased, it became apparent there was marked geographic variability both in the preval‎ence of disease and in the mycobacterial species responsible for disease. Most patients in the southeastern United States with NTM lung disease were from rural areas and had isolates of M avium complex, whereas those in the central United States more commonly had disease caused by M kansasii [45].

In addition, patients with NTM disease tended to react more strongly to skin test antigens prepared from the infecting mycobacterial species than to standard PPD-S or PPD-T, antigens prepared from M tuberculosis [46]. Skin test surveys using NTM antigens suggested that infection by NTM was common, especially in rural areas and in the Southeast [47].

NTM disease is not reportable in the United States, and reliable estimates of its incidence or preval‎ence have been limited. Two national surveys in the early 1980s were the first to try to define the extent of NTM infection in the United States. The initial study, based on state laboratory reports from 1979 to 1980, indicated that NTM comprised approximately one-third of the 32,000 mycobacterial isolates [48]. Of these, 61 percent were M avium complex, 19 percent were M fortuitum complex, and 10 percent were M kansasii.

A second surveillance study based on reports from tuberculosis control officers on isolates of NTM recovered between 1981 and 1983 showed higher rates of NTM disease among nonwhites, women, and patients residing in urban areas when compared with the initial study. White males, however, continued to serve as the major diseased population [49]. Using combinations of national surveillance data, the preval‎ence of NTM (pulmonary) disease at that time was estimated to be 1.8 cases per 100,000 population for the entire United States. Of this, M avium complex represented 1.1/100,000.

A more recent Centers for Disease Control (CDC) study from 1991 to 1992 [50] that included results from 33 state laboratories demonstrated a dramatic change in the preval‎ence of NTM Despite the increases in isolates of M tuberculosis noted in the United States since 1985, there were now more isolates of M avium complex than M tuberculosis, with the latter representing only 26 percent of the total mycobacterial isolates. The reasons for this dramatic increase in numbers for NTM is unknown, but better clinical recognition and more culturing for both pulmonary and disseminated disease are felt to play important roles.

One category of people with NTM disease not represented in the two earliest studies but almost certainly represented in the 1993 study were patients with AIDS and disseminated NTM disease. HIV-infected patients were at especially high risk of disease due to NTM The majority of disease in this population (> 95 percent) is due to M avium [51]. Disseminated M avium infection is the most common bacterial infection in patients with AIDS, occurring in 20 to 40 percent of all patients in several reported series [52-55]. Disease in these patients is highly correlated with severe immunosuppression, with the average CD4 cell count at the time of dissemination in the 25 to 30 range [53-55]. Patients with < 100 CD4 cells, not receiving prophylaxis, develop disseminated M avium at the rate of approximately 20 percent per year [53]. The overall incidence of M avium as an initial diagnosis has increased among AIDS patients while other complications of AIDS, such as Pneumocystis pneumonia, have decreased [56]. Disseminated M avium occurs in similar rates in all geographic regions and various HIV risk groups [57]. Localized pulmonary disease in AIDS due to M avium occurs in less than 5 percent of patients [58].

Other NTM species, including M kansasii [51,59-61]. M scrofulaceum [51], M gordonae [51], M haemophilum [60,61], M genavense [38], M celatum [62], M conspicuum [63] M xenopi [64], M fortuitum [51,65], M marinum [66], M malmoense [67], and M simiae [68] have also been described as a cause of pulmonary and/or disseminated NTM disease in AIDS. Some of these, especially M haemophilum, M kansasii, and M genavense, have occurred in localized geographic areas. More than 95 percent of cases of disseminated disease, however, are due to isolates of M avium [51].

CLINICAL PRESENTATION AND DIAGNOSTIC CRITERIA

Pulmonary disease — Chronic pulmonary disease is the most common localized clinical manifestation of NTM [41,69]. Mycobacterium avium complex, followed by M kansasii, is the most frequent pathogen causing lung disease in the United States. Other pathogens occasionally causing pulmonary disease include M abscessus, M fortuitum, M szulgai, M simiae, M xenopi, M malmoense, M celatum, M asiaticum, and M shimodii. Mycobacterium xenopi is second to M avium complex as a cause of NTM lung disease in areas of Canada, the United Kingdom, and other areas of Europe, while M malmoense is second to M avium complex in Scandinavia and areas of northern Europe [70]. The patients with chronic lung disease due to NTM are generally older adults. Except for patients with cystic fibrosis, children rarely develop this form of NTM disease [69]. Although some NTM patients have a history of underlying chronic lung disease, not all do. The interpretation of NTM in the sputum of HIV-positive patients presents a particular problem, as these patients are frequently infected with NTM without evidence of pulmonary disease. Such infection may be transient, but it may also reflect disseminated NTM disease or subclinical NTM pulmonary disease. In addition, some NTM species that are generally considered nonpathogenic have been associated with pulmonary disease in the HIV-infected host.

Signs and symptoms of NTM pulmonary disease are variable and nonspecific. They include chronic cough, sputum production, and fatigue. Less commonly, malaise, dyspnea, fever, hemoptysis, and weight loss can also occur, usually with advanced NTM disease. Eval‎uation is often complicated by the symptoms caused by co-existing lung diseases. These conditions include chronic obstructive airway disease associated with smoking, bronchiectasis, previous mycobacterial diseases, cystic fibrosis, and pneumoconiosis.

There are some differences in the radiographic features of NTM lung disease compared with those produced by M tuberculosis with regard to conventional radiographic studies. Nontuberculous mycobacteria tend to cause thin-walled cavities with less surrounding parenchymal infiltrate, have less bronchogenic but more contiguous spread of disease, and produce more marked involvement of pleura over the involved areas of the lungs. Occasionally, they may produce dense pneumonic disease or a solitary pulmonary nodule without cavitation. Basal pleural disease is not often found, and pleural effusion is rare. Recent studies with high-resolution computed tomography (HRCT) of the chest have shown that up to 90 percent of patients with mid and lower lung field noncavitary disease with M avium complex have associated multifocal bronchiectasis, with many patients having clusters of small (< 5 mm) nodules in associated areas of the lung [71-74].

There has been a great deal of interest in the availability of species-specific skin test antigens. Unfortunately, many antigens are shared by different mycobacterial species for which there are previously tested NTM skin test antigen preparations, and extensive cross-reactions were observed with PPDs. However, recent studies provide hope for increased specificity of a preparation for M avium complex testing, although it is not yet FDA approved [75]. Specific skin test reagents for other NTM infections are not standardized and are neither available nor undergoing clinical trials at this time.

In the absence of specific diagnostic features in the history and physical examination, the chest roentgenogram, and differential skin testing, isolation of the NTM in a culture is essential for diagnosis. However, as these organisms are commonly found in nature, contamination of culture material or transient infection does occur. Thus, a single positive sputum culture, especially with small numbers of organisms, does not always suffice to diagnose NTM disease. Some previous authors suggested that the respiratory tract may be infected with the organism without disease, particularly in patients with chronic respiratory disease [48,49,69]. This condition was often referred to as "colonization," and was described most often with M avium complex. It was characterized by the presence of noncavitary, stable, and usually minimal radiographic disease in women, and was associated with sporadic excretion of organisms from the respiratory tract. No pathologic studies were done to demonstrate the absence of tissue invasion, and more recent studies with HRCT have shown that these patients often have a combination of multifocal bronchiectasis and nodular parenchymal disease [71-74], with the latter or both now felt to be due to mycobacterial disease. "Colonization" in the true sense [ie, no tissue invasion] is probably quite rare. In addition, not all patients with this disease have a benign prognosis, a point first emphasized by Prince and colleague in a landmark 1989 paper [76].

Given these observations, the diagnosis of lung disease caused by NTM is usually not difficult if a combination of clinical, radiographic, and bacteriologic criteria is used. AFB smear, culture results, and clinical status suggest a close correlation among the three. Minimal eval‎uation should include three or more sputums for AFB and efforts to exclude other confounding disorders such as tuberculosis and lung malignancy. In most patients, a diagnosis can be made without a lung biopsy. Although criteria are based on experience with M avium complex, there is no reason to believe these criteria would not be applicable to other species. The diagnostic criteria are presented in Table 1 (show table 1).

Clinical studies have established the validity of bronchial washings as a culture source for M tuberculosis. Although similar studies have not been done for NTM, bronchial washings are considered to be more sensitive than routine expectorated sputums; however, their relative specificity for clinical disease is unknown. Approximately 90 percent of patients with disease caused by M kansasii and most patients with disease caused by M avium complex have cavitary infiltrates [11] and can be readily identified. Among patients without cavities, the presence of clinical symptoms and HRCT abnormalities are important adjuncts to defining the presence of NTM disease. Bacteriologic criteria have been best analyzed with cavitary disease for M avium complex and M kansasii, and the HRCT abnormalities, with M avium complex. Although these are reasonable criteria for diagnosing other NTM, their use with other species has not been studied in detail.

In the patient with nondiagnostic cultures and radiographic studies, or concern about the presence of another disease producing radiographic abnormalities, a lung biopsy is often required for diagnosis. If a tissue sample from a transbronchial, percutaneous, or open-lung biopsy yields an NTM organism and shows mycobacterial histopathologic changes (ie, granulomatous inflammation with or without AFB), this by itself is sufficient to establish the diagnosis of NTM lung disease. If the lung biopsy has a negative culture (something that often happens when transbronchial biopsies are performed because of the small size of the tissue sample) but demonstrates mycobacterial histopathology features (without a history of other granulomatous or mycobacterial disease), NTM lung disease is considered to be present when one or more sputums or bronchial washes are culture-positive for NTM, even if they are negative for AFB on smear and result in light growth on culture.

Lymphadenitis — Infection of the submandibular, submaxillary, cervical, or preauricular Iymph nodes in children between one and five years old is the most common presentation of NTM Iymphadenitis [77-80]. It is the most common disease manifestation of NTM in children and, in the absence of HIV infection, rarely affects adults. The disease occurs insidiously, with only rare associated systemic symptoms. The involved Iymph nodes are generally unilateral (95 percent) and not tender. The nodes may enlarge rapidly, and even rupture, with formation of sinus tracts that result in prolonged local drainage. Other nodal groups outside of the head and neck may be involved occasionally [79]. There is typically no history of exposure to tuberculosis, screening PPD skin tests of family members are usually negative, and the chest radiograph is normal.

Most children with NTM Iymphadenitis will react to skin test antigens prepared from M avium complex, such as PPD-B [78,81,82]. A 1991 multicenter study of NTM antigens from the CDC that used PPD-B, however, was terminated early, due to a blistering reaction in several of the children [81]. More recent studies using a less potent, protein weight-standardized M avium skin-test material called "sensitin" and a dual skin test technique to determine M avium dominant versus PPD-dominant reactions have suggested improved specificity with this antigen preparation in study populations with known disease [75]. Although these antigens may prove beneficial for future eval‎uation of cervical Iymphadenitis, no commercial NTM skin-test material is currently available for clinical use in the United States, and this procedure is not recommended for diagnosis of NTM All children in this setting should be tested using PPD tuberculin. Most children tested with intermediate strength (5 tuberculin unit (TU)) PPD tuberculin will have a weakly reactive skin test (5-9 mm) due to cross-reactivity with NTM, but some children may be negative, and as many as one-third will have reactions with 10 mm or more induration [79].

Distinguishing tuberculous from nontuberculous Iymphadenitis is key, because the former requires drug therapy and public health tracking, whereas the latter does not. The presumptive diagnosis of NTM Iymphadenitis is based on the histopathologic appearance of the Iymph node showing caseating granulomata with or without AFB and a negative tuberculin skin test. Failure of the node to yield M tuberculosis provides stronger presumptive evidence for the diagnosis of NTM Iymphadenitis.

The utility of fine needle aspiration in obtaining diagnostic material is controversial [83-85]. However, granulomata or other compatible cytopathology such as a mixture of degenerating granulocytes, Iymphocytes, and epithelioid histiocytes are seen in most cases. A positive culture may be obtained in up to 50 percent of HIV-seronegative patients and in even higher proportions of HIV-positive patients with tuberculous adenitis.

A definite diagnosis of NTM Iymphadenitis is made by recovery of the causative organism from Iymph node cultures. A simple diagnostic biopsy or incision and drainage of the involved Iymph nodes should be avoided, since most of these procedures will be followed by fistulae formation with chronic drainage [78]. However, even with excised nodes with compatible histopathology, only about 50 percent will yield positive cultures [78], although the recovery rate may be as high as 82 percent in some centers [79]. Some of these smear-positive, culture-negative cases may be due to fastidious species such as M haemophilum [86] or M genavense [38]. Currently, approximately 80 percent of culture-proven cases of NTM Iymphadenitis are due to M avium complex [87]. In the United States and Australia the remaining cases are caused by M scrofulaceum [78,79,87], while in Scandinavia, the United Kingdom, and other areas of northern Europe, M malmoense has recently emerged as the major pathogen after M avium complex [70,88,89]. The predominance of M avium complex is a change from 20 years ago, when most geographic areas reported M scrofulaceum as the most common etiologic agent [77,79]. Now in the United States, only about 10 percent of the culture-proved mycobacterial cervical Iymphadenitis in children is due to M tuberculosis; the remainder is due to M avium complex and M scrofulaceum [87]. In contrast, in adults more than 90 percent of the culture-proven mycobacterial Iymphadenitis is due to M tuberculosis.

Localized skin, soft tissue, and skeletal infection — The NTM species that most commonly cause localized infections of the skin and subcutaneous tissue are M fortuitum, M abscessus, M marinum, and M ulcerans [13]. However, virtually all species of NTM have been described as a cause of cutaneous disease [41,69]. Localized drainage or abscess formation at the site of puncture wounds (such as occurs after stepping on a nail), or open traumatic injuries or fractures are most often due to the rapidly growing mycobacterial species M fortuitum, M abscessus, or M chelonae [90]. Nosocomial skin and soft-tissue disease caused by these three species is also seen [16-26]. These include infections of long-term intravenous or peritoneal catheters [90,91], post-injection abscesses, or surgical wound infections such as those occurring after augmentation mammoplasty [23,92] or cardiac-bypass surgery [16,17,21,93]. Diagnosis is made by culture of the specific pathogen from drainage material or tissue biopsy.

Mycobacterium marinum is the cause of "swimming pool granuloma" or "fish tank granuloma" [69]. The lesions usually appear as papules on an extremity, especially on the elbows, knees, and dorsum of feet and hands, progressing subsequently to shallow ulceration and scar formation. Most lesions are solitary, although occasional "ascending" lesions develop that resemble sporotrichosis. Clinical involvement of regional nodes is uncommon. The organisms may be introduced into the skin through previous abrasions contaminated while cleaning fresh-water fish tanks ("fish tank granuloma") or by scratches or puncture wounds from salt water fish, shrimp, fins, etc. Diagnosis is made from biopsy material, histologic examination, and culture.

Mycobacterium ulcerans causes indolent necrotic lesions of the skin and underlying tissue in Australia and tropical areas of the world [34,69,94]. It is not endemic in the United States. The lesions occur most commonly in children and young adults and often result in severe deformities of the extremities [94]. Drug treatment of the disease has been disappointing; surgical debridement combined with skin grafting is the usual treatment of choice.

Infection of bursae, joints, tendon sheaths, and bones — Chronic granulomatosis infection caused by NTM may develop in tendon sheaths, bursae, joints, and bones after direct inoculation of the organisms through accidental traumas, surgical incisions, puncture wounds, or injections. Mycobacterium marinum [41,69] and M avium complex [95] are particularly prone to causing tenosynovitis of the hand, although M fortuitum, M abscessus, M chelonae, and M kansasii have also been implicated [41,69]. Mycobacterium terrae complex (especially M nonchromogenicum) has also been isolated from synovial tissue of the hand or wrist, and it tends to be associated with a very indolent, chronic type of disease. Occasionally, axial bones and extremities have been infected without apparent trauma and are due presumably to hematogenous infection. After open-heart surgery, osteomyelitis of the sternum caused by M abscessus or M fortuitum has been described, with both epidemic and sporadic disease [16,17,21,93].

Disseminated disease in patients without AIDS — Dissemination of NTM in adult patients with immunosuppression but without AIDS (eg, those with renal or cardiac transplantation, chronic corticosteroid use, leukemia, etc) has been observed. Mycobacterium avium complex [69,96], M kansasii [97], M chelonae [90,98-100], M scrofulaceum [69], M abscessus [69], and M haemophilum [60] have all been reported to cause disease in this setting. In general, the disease caused by M avium complex presents as a fever of unknown origin [96], whereas disease caused by M kansasii, M chelonae, M abscessus, and M haemophilum generally presents as multiple subcutaneous nodules or abscesses that drain spontaneously [60,69,98-100]. The mortality relates directly to the type and severity of underlying disease [100]. Lincoln and Gilbert [77] reviewed 12 cases, all fatal, of disseminated NTM disease in children. Most of the children were infected with M avium complex, were less than three years old, and had no apparent underlying disease. A recent review of disseminated M avium complex disease in children noted its occurrence, rarely, in the setting of severe combined immunodeficiency syndrome or chemotherapy for malignancy [101]. The isolation of organisms from sterile, closed sites such as bone marrow or blood or from a skin biopsy (in the setting of multiple lesions) is diagnostic of the disease.

Disseminated disease in patients with AIDS — Disseminated disease due to NTM in patients with HIV infection usually occurs only in those with very advanced immunosuppression [51-57]. Because these patients frequently have other complications, the diagnosis of mycobacterial infection may be confused or delayed. The diagnosis is exceedingly rare in person with > 100 CD4 cells, and it should usually be suspected only in persons with < 50 CD4 cells [53-55]. Most patients (> 90 percent) have prolonged fevers, which may be as high as 103-104º F, frequently accompanied by night sweats. Weight loss is common, and some patients complain of abdominal pain and diarrhea. Physical findings may be only those of advanced HIV disease, although abdominal or retroperitoneal adenopathy and hepatosplenomegaly may be present. Anemia is the most striking laboratory abnormality, with many patients having a hematocrit of < 25 percent. Alkaline phosphatase is elevated in approximately one-third of patients and may be indicative of hepatic disease due to M avium. Thus, the diagnosis of disseminated M avium should be aggressively pursued in any person with < 50 CD4 cells who has a history of fever, weight loss, anemia, diarrhea, or elevated alkaline phosphatase, especially in one with a history of other opportunistic infections.

The diagnosis of disseminated M avium is most commonly confirmed by isolation of M avium in blood, using any of the culture techniques described in the laboratory section. The bacteremia in M avium is ongoing, and a single culture has a sensitivity of approximately 90 percent. It is recommended that a single culture be drawn, with repeat cultures only if the first is negative. Routine blood cultures of asymptomatic patients has a very low yield and is not recommended. In a prospective study of HIV-infected patients with < 50 CD4 cells, approximately 67 percent of patients with M avium in sputum or stool had disseminated disease within one year, although most did not develop pulmonary disease. However, only one-third of all patients with disseminated disease had a prior positive stool or sputum. Therefore, routine screening of stool or sputum is not indicated, but a positive culture of one of these sites needs to raise concern about future dissemination. Sputums that are smear-positive for AFB in the setting of HIV should always be regarded as tuberculosis until proven otherwise, since M tuberculosis is a common cause of pulmonary disease in HIV-infected patients and is more likely than M avium complex to produce positive AFB smears.

Nontuberculous mycobacteria other than M avium may present as disseminated disease. Disseminated M kansasii is usually associated with pulmonary disease [51,59]. Mycobacterium genavense has been isolated from the blood of patients with AIDS and requires extensive laboratory analyses for isolation and identification [38]. Mycobacterium haemophilum has been associated with infections of the skin, soft tissue, bones, and joints [60,61]. Disseminated disease has rarely been reported with other species, including M fortuitum [51], M marinum, M simiae, M scrofulaceum, M celatum, and M malmoense [41].

Copyright 1997 The American Thoracic Society, Am J Respir Crit Care Med 1997; 156:S1. Reprinted with permission from The American Thoracic Society. For more information about the ATS, please visit their website at www.thoracic.org.

LABORATORY METHODS

Digestion, decontamination, and staining procedures — Methods used for digestion and decontamination of clinical samples to recover M tuberculosis have also proved useful for the NTM In general, however, NTM are more susceptible to killing by NaOH, and for this reason, care must be taken not to exceed the recommended concentration and time guidelines. Because of the frequent presence of bronchiectasis in patients with M avium complex and M abscessus lung disease, Pseudomonas aeruginosa overgrowth in specimens from these patients is a more frequent problem than with tuberculosis. Growth of P. aeruginosa can be minimized by processing the specimens with the conventional N-acetyl-L-cysteine-sodium hydroxide (NALC-NaOH) solution followed by 5 percent oxalic acid, a procedure that should be considered when one or more specimens are contaminated in this setting [102].

Staining and microscopy of the NTM also follows the guidelines used for M tuberculosis. Both conventional basic fuchsin method (Kinyoun stain) and the fluorochrome method [auramine stain] are effective in recognizing NTM in clinical material, with the fluorochrome method being preferred [103]. The appearance of NTM by microscopy is generally indistinguishable from M tuberculosis.

Culture techniques for nontuberculous mycobacteria — The principles and practices of culturing M tuberculosis were updated in 1993 by the CDC [103], these methods proving very effective for NTM species. At least three respiratory (sputum) cultures should be used for the initial eval‎uation. Cultures should be inoculated onto one or more solid medias and into a liquid medium. Use of solid media as the primary or sole culture is no longer recommended by us or by the CDC [103], given the greater recovery rate and more rapid recovery of all mycobacteria, including M tuberculosis, in rapid broth systems [103-106]. Mycobacterial blood cultures may use a single medium, with the BACTEC 13A broth [104] or the lysis centrifugation method with plating on 7H10 or 7H11 (Isolator; Wampole Laboratories, Cranbury, New Jersey) being the recommended methods. Two general types of solid media are available: egg-potato-base media (commonly, Löwenstein-Jensen agar) and a clear agar-base media (commonly, Middlebrook 7H1O or 7H11 agar). Quantitation of growth on agar plates (generally 0 to 4+) is important to estimates of clinical significance and responses to therapy, and it is recommended for all samples other than blood cultures. Blood cultures using the Isolator System can also be quantitated, which may be useful for similar reasons. Because of its greater recovery rate for M avium complex and ease of quantitation, Middlebrook 7H10 or 7H11 agars are the preferred solid media. Löwenstein-Jensen is an excellent medium for recovery of M tuberculosis, but is generally inferior to Middlebrook agar as an all-purpose medium for both M tuberculosis and NTM [104,105].

The broth medium can involve one of several automated commercial systems, including radiolabeled BACTEC 12B broth used in the BACTEC TB 460 radiometric system (Becton Dickinson Instruments Systems, Sparks, Maryland), and the nonradiolabeled ESP Culture System II (Difco Laboratories, Detroit, Michigan) [106]. For lower volume laboratories, the recently introduced mycobacterial growth indicator tubes with a fluorescent detection system (MGIT; Becton Dickinson Microbiology Systems, Cockeysville, Maryland) or biphasic agar/ broth (Septi- Chek AFB System; Becton Dickinson Microbiology Systems) may prove more practical.

Most slowly growing NTM produce detectable growth in two to four weeks on the solid media and in one to two weeks with the BACTEC system. Cultures are generally incubated at 35-37º C for six weeks. All of the currently recognized NTM pathogens will grow on these media in this time except M haemophilum, M genavense. and M conspicuum. If M haemophilum is suspected, a commercial paper surface containing hemin (X factor) used to identify Haemophilus influenzae should be added to the surface of the 7H10 or 7H11 plate [107], or hemin or ferric ammonium citrate should be incorporated into the medium. Mycobacterium genavense often grows only from the blood in BACTEC 13A medium or comparable broth media, and requires incubation for at least eight weeks [108]. Some authors have identified better growth in the slightly acidic (pH 6) radiometric Middlebrook 7H12 broth (pyrazinamide test medium). Mycobacterium conspicuum will grow in BACTEC media at 35-37º C but will only grow on solid media at lower temperatures, 22-31º C [63]. The presence of an AFB smear-positive sample with no growth on solid media should immediately bring to mind M haemophilum, M conspicuum, or M genavense. Because of the relatively poor growth of M genavense (only approximately 50 percent of autopsy-proven cases are culture-positive [38]), molecular techniques such as polymerase chain reaction are the optimal method of identification [38].

The major modification of culture techniques for recovering NTM species is the need to incubate all skin or soft tissue samples at two temperatures: 35º C and 28-32º C. This is because a number of the common pathogens of these tissues, including M haemophilum, M ulcerans, M marinum, and M chelonae, may grow only at the lower temperatures, especially on primary isolation. Five to ten percent CO2 enhances the growth of some NTM species, such as M haemophilum, and should be used for primary isolation because of its definite growth enhancement for M tuberculosis.

Identification of nontuberculous mycobacteria — Traditional identification of NTM, as well as M Tuberculosis, has relied upon statistical probabilities of presenting characteristic reaction pattern in a battery of biochemical tests. The niacin test was the most useful for separating NTM and M tuberculosis because the former is usually negative, whereas isolates of M tuberculosis are positive. Runyon devised the first good scheme for grouping NTM based on growth rates and colony pigmentation [69]. Species identification has become more sophisticated and the number of potential pathogens has increased since this scheme was introduced.

A more appropriate grouping currently for these organisms is based on the type of clinical disease they produce: Iymphadenitis, cutaneous disease, disseminated disease, and pulmonary disease. Grouping of the NTM by this scheme is shown in table 2 (show table 2).

Because of the extremely slow nature of traditional biochemical tests, most clinical and public health laboratories now use one or more rapid diagnostic methods for species identification [109,110]. These rapid methods are recommended for identification of the NTM when possible; they include HPLC, the BACTEC NAP test, and commercial DNA probes. The HPLC examines the mycolic acid fingerprint patterns that differ among most species or complexes of mycobacteria [111]. Recent increased sensitivity of this technique using fluorescence detection has allowed identification directly from the sputum sample in approximately 50 percent of AFB smear-positive samples of M tuberculosis and 33 percent of M avium complex [112]. A small number of species (complexes) are not separable by HPLC, including most of the pathogenic rapidly growing mycobacterial species. Two additional techniques for rapid identification of NTM are currently in use. These are the species-specific DNA probes and the BACTEC NAP test. Commercial nonradiolabeled DNA probes complementary to ribosomal RNA are available for identifying isolates of M tuberculosis, M gordonae, M kansasii, M avium, and M intracellulare. They are highly sensitive and specific, providing species identification using a culture directly from BACTEC. broth within two to four hours. The presence or absence of serpentine cording on AFB smear of the BACTEC 12B bottle [113] and the time required for the bottle to turn positive [114] will help the laboratory decide which probe (M tuberculosis or M avium complex) should be used initially.

The BACTEC TB 460 system can be used to differentiate between M tuberculosis and the NTM using a selective growth inhibitor called NAP (p-nitro-a-acetylamino-b-hydroxypropiophenone) [115]. This compound, at a concentration of 5 mg/mL, inhibits the growth of M tuberculosis complex but not NTM (with the exception of M genavense). The average time for the NAP test is five days. However, a species identification schema other than the use of DNA probes for NTM has not been worked out for use with the BACTEC, and such identification must depend on one of the previously discussed methods.

Because of its generally poor growth, identification of M genavense can be difficult [43,108]. A presumptive identification can be made on the basis of organism morphology (small, coccobacillary forms on AFB smear), failure to grow on subculture to solid media, a negative nucleic acid probe for M avium complex, and a positive NAP test [38]. Mycobacterium genavense is one of the few mycobacteria, other than the M tuberculosis complex, which is inhibited by NAP.

Antimicrobial susceptibility testing — Although there are specific recommendations from the CDC, ATS, and the National Committee for Clinical Laboratory Standards (NCCLS) regarding which isolates of M tuberculosis should have antimicrobial susceptibility tests, which test methods to use, and which antimicrobial agents to test, the same is not true for the NTM There are however, sufficient data now available to make temporary recommendations regarding when, how, and to which agents the NTM should be tested. Recommendations will differ for different groups or species of the NTM (show table 3). Although routine testing of all NTM is discouraged, there are circumstances where susceptibility testing is warranted, including having baseline data available if the patient does not respond to therapy, or when relapses occur.

Slow-growing mycobacteria — Antimicrobial susceptibility testing of the slow-growing mycobacteria can be performed using either the agar proportion or the radiometric (BACTEC) methods used for testing M tuberculosis [116-124]. However, the two methods give varying results between some NTM and antimicrobial agents, with the radiometric broth method tending to give lower minimal inhibitory concentrations [MICs] than the agar method [117,118]. Too little experience is available with the antibiotic gradient strip method (E-test; AB Biodisk, Piscataway, NJ) to make any general recommendations on its use.

The agar proportion method uses Middlebrook 7H10 or 7H I L agar and the modified method of proportions, defining resistance as growth on the drug-containing medium of 1 percent or more of the number of colonies that grow on the drug-free control medium. Details of the agar proportion method are included in the 1990 ATS statement "Diagnostic Standards and Classification of Tuberculosis" [3], and a more detailed description is now available as a tentative standard (for M tuberculosis) by the NCCLS [125].

The radiometric BACTEC method is a more rapid method combining antimicrobial agents, the mycobacterium, and a C14-labeled substrate in a broth medium. Resistance is determined by the rate and amount of labeled CO2 produced, which is directly proportional to the rate and amount of growth that occurs in the broth medium. The BACTEC method has been widely used by some laboratories for testing all drugs for the NTM, but at present no universally agreed upon method has been developed.

Mycobacterium avium complex — When and how isolates of the M avium complex should be tested remains controversial [126]. Recent data suggest that the radiometric broth method is more reliable than the agar method for testing the M avium-complex. There is, however, considerable controversy regarding the size of the inoculum, the drug concentrations to use and interpretation of the BACTEC results. Strains of the complex are almost always resistant to the relatively low drug concentrations of isoniazid, rifampin, streptomycin, and ethambutol used for defining susceptibility of M tuberculosis. Using higher concentrations of the antituberculous agents with specific NTM breakpoints for susceptibility and resistance, determination of MICs, or determination of the activity of combined drugs are some of the newer approaches that may be helpful in predicting clinical response [117,119,120,127]. Such a benefit has not yet been shown by clinical trials, so susceptibility testing of M avium complex isolates to the antituberculosis drugs is not recommended.

Other drugs have also been tested, including amikacin, rifabutin [117,121,122], ciprofloxacin [123,124], clofazimine [128], azithromycin [126], and clarithromycin [126]. Changes in MICs of the antituberculosis drugs following unsuccessful therapy of M avium complex disease have been difficult to demonstrate. Such changes have been readily demonstrable with microbiologic relapses following monotherapy with clarithromycin, however [129]. Pretreatment isolates with this drug have MICs greater than or equal to 4.0 µg/mL when done in media with a pH of 7.4. Post-therapy or relapse isolates following macrolide therapy have MICs to clarithromycin of > 32 µg/mL [104,129-131] and a point mutation involving one of two base pairs in the 23S ribosomal macrolide binding site [130,131].

Susceptibility testing should not be performed on pretreatment or initial isolates against clarithromycin, but it should be performed with clarithromycin for all isolates from patients on prior macrolide therapy, including those on macrolide prophylaxis for disseminated disease. The recommended clarithromycin resistance breakpoint is > 32,µg/mL for broth or agar with pH corrected to 7.4 [104]. Until more data are available, patients on azithromycin should have their isolates characterized as susceptible or resistant based on clarithromycin susceptibility values.

Alterations or changes in MICs following treatment or prophylaxis failure have also been difficult to demonstrate with rifabutin. For this reason, testing of susceptibility to rifabutin even after therapy is not recommended. Because of lack of standardization, results of testing other nontuberculous drugs such as amikacin, clofazimine, and ciprofloxacin should be used with caution.

Mycobacterium kansasii — Although wild strains of M kansasii are initially susceptible to rifampin, acquired resistance does develop during therapy [132,133]. Since the correct history of therapy may not be known, all initial isolates of M kansasii should be tested against rifampin, using the agar proportion method and the interpretive criteria for M tuberculosis (resistance breakpoint of 1 µg/mL) [133]. Also, testing should be performed when the patient's sputum fails to convert from smear- and/or culture-positive or when a relapse occurs during therapy. Treatment for rifampin-susceptible isolates is empiric and is not influenced by susceptibility to drugs other than rifampin (ie, ethambutol and isoniazid); hence their routine testing is not recommended. A rifampin-resistant isolate could be tested against ciprofloxacin or ofloxacin, clarithromycin, ethambutol, streptomycin and a sulfonamide (eg, sulfamethoxazole) [132,133].

Other slow-growing nontuberculous mycobacteria — Susceptibility testing of infrequently isolated species of NTM may be helpful, since knowledge of susceptibility patterns is limited. Pulmonary infections caused by M malmoense, M xenopi, and M szulgai have been successfully treated with combinations of ethambutol, isoniazid, rifampin, and most recently clarithromycin [70,134-136], whereas ciprofloxacin, clarithromycin, and rifampin are suggested for treating M haemophilum infection [60,61]. Thus, susceptibility tests for these slow-growing NTM might include these five drugs. The agar proportion and broth methods have usually been used for testing. For M marinum, several methods have been used [137], with the desired test drugs being rifampin, ethambutol, doxycycline or minocycline, clarithromycin, and a sulfonamide.

Rapidly growing nontuberculous mycobacteria — Because of differences in susceptibilities among species of rapidly growing mycobacteria and even within species, susceptibility testing should be performed on all clinically significant isolates as well as isolates that have been recovered after treatment failure or relapse. Antimicrobial susceptibility testing of the rapidly growing mycobacteria differs from the other NTM Most drugs are different, although the methods are similar to those used to test other bacteria [137-141]. Most infections are caused by three species; M abscessus, M chelonae, and M fortuitum [90]. A primary panel of drugs for these species could include amikacin, cefoxitin, ciprofloxacin, clarithromycin, doxycycline, imipenem, and a sulfonamide. The most convenient method for susceptibility testing is to use microtiter MIC trays containing cation-supplemented Mueller-Hinton broth. Agar dilution, agar disk elusion, and disk diffusion methods have also been used. Details of these methods can be found in several laboratory handbooks, including the American Society for Microbiology's Manual of Clinical Microbiology [141].

TREATMENT OF MYCOBACTERIUM KANSASII DISEASE — Disease caused by M kansasii is the second most common NTM pulmonary disease in the United States. It occurs in geographic clusters and affects primarily adult white men, but it can affect patients of any sex, race, or age. Pulmonary disease is the most frequent clinical presentation. The organism frequently exhibits a beaded or cross-barred appearance on acid fast stain, and produces rough buff-colored colonies that develop a yellowish pigmentation because of the deposition of beta-carotene crystals after exposure to light. Disease-producing strains are usually highly catalase-positive. Recent DNA based studies suggest that up to five taxonomic groups or subspecies are present among both environmental and human isolates [142].

Untreated strains of M kansasii are inhibited by rifampin, isoniazid, ethambutol, ethionamide, streptomycin, and clarithromycin at concentrations readily achievable in the serum with usual therapeutic doses [133,143,144]. Because the concentrations of antituberculous drugs used in susceptibility testing were chosen for their usefulness with M tuberculosis, and because M kansasii is less susceptible to these drugs, some isolates of the latter species may be reported resistant to isoniazid at 0.2 or 1 µg/mL and to streptomycin at 2 µg/mL. These isolates are susceptible to slightly higher drug concentrations [133,143], and laboratory reports of resistance to the low concentrations of these two drugs have no clinical or therapeutic significance as long as a rifampin regimen is being used [145]. Thus, when clinically indicated, isoniazid and/or streptomycin should be used against M kansasii regardless of in vitro susceptibility results. Mycobacterium kansasii is also susceptible in vitro to clarithromycin [133], sulfamethoxazole [132], amikacin [132], the newer quinolones [124], and rifabutin [133], although there is limited information on the clinical usefulness of these drugs [132,133]. Isolates are usually resistant to achievable serum levels of p-aminosalicylic acid, capreomycin, and pyrazinamide. Acquired resistance to rifampin, ethambutol and isoniazid has been demonstrated in isolates from treatment failure cases [132,133,144], and resistance to the first two agents is reliably demonstrated by current M tuberculosis susceptibility test methods [133].

Treatment — The natural history of pulmonary disease caused by M kansasii in patients receiving no drug treatment has been assessed [146]. In general, the history has shown persistence of sputum positivity and progression of clinical and radiographic disease. On this basis, patients with pulmonary disease should receive drug therapy.

There have been no randomized comparative trials of treatment for disease caused by M kansasii, comparing one drug regimen with another or with no drug treatment at all. There have been, however, several retrospective and prospective studies of various treatment regimens [144-150] that have given us a good basis for drug therapy recommendations. Earlier reports of treatment with antimycobacterial drugs in the pre-rifampin period were disappointing when compared with the much higher success rates achieved in treating tuberculosis with these same drugs. The sputum conversion rates at 6 months ranged from 52 to 81 percent, and relapse rates of approximately 10 percent were seen in patients achieving an initial response [144,147]. Surgical resection was often recommended to achieve better initial control and prevent relapse. The advantage of adding surgery was never established, however [147].

With the advent of rifampin, the picture changed considerably for the better. Four-month sputum conversion rates with rifampin-containing regimens were 100 percent in 180 patients from three studies [144,145,148]. There were two treatment failure cases, however (an incidence of 1.1 percent). These patients converted their sputa but then became culture positive again while still receiving therapy. Both had been treated with isoniazid, rifampin, and ethambutol, and both failures were associated with the development of rifampin resistance [144]. Long-term relapse rates with rifampin-containing regimens also appear to be very low, with only one relapse recorded among 134 patients (0.8 percent) who received long-term follow-up in three studies [144,145,149]. Surgery is now considered to have no role in managing routine cases of pulmonary disease.

The current recommendation for treatment of pulmonary disease caused by M kansasii in adults is the regimen of isoniazid (300 mg), rifampin (600 mg), and ethambutol (25 mg/kg for the first two months, then 15 mg/kg) given daily for 18 months with at least 12 months of negative sputum cultures. In patients who are unable to tolerate one of these three drugs, clarithromycin would seem a reasonable alternative, but its effectiveness has not been established by clinical trials (see below). Pyrazinamide is unacceptable as an alternate or third drug for M kansasii because all isolates are resistant.

The use of intermittent drug regimens or short-course treatment for M kansasii has not been studied enough to recommend it. One study of 40 patients did demonstrate that adding intermittent streptomycin at 1 g twice weekly for the first three months to the previously recommended three-drug regimen given for 12 months resulted in apparent cure of all but one patient [148]. A trial of daily low-dose ethambutol (15 mg/kg) and daily rifampin given for nine months sponsored by the British Medical Research Council was completed in 155 adult patients [150]. Sputum conversion was achieved in 99.4 percent of patients, but with a relapse rate of 10 percent with a five year follow-up. This suggests that isoniazid does not contribute greatly to the treatment of M kansasii; however, nine months is not a long enough treatment period for the studied two-drug regimen.

In adult patients whose organisms have become resistant to rifampin as a result of previous therapy, a regimen consisting of high-dose daily isoniazid (900 mg), pyridoxine (50 mg daily), high-dose ethambutol (25 mg/kg per day), and sulfamethoxazole (1.0 gm three times per day) until the patient is culture-negative for 12 to 15 months has been under investigation [132,133]. The oral therapy has been combined with daily or five times per week streptomycin or amikacin for the initial two to three months, followed by intermittent streptomycin or amikacin for a total of 6 months. Results with this regimen described sputum conversion in 18 of 20 patients (90 percent) after a mean of 11 weeks, with only one relapse (8 percent) among patients who were culture-negative for at least 12 months on therapy [133]. The excellent in vitro activity of clarithromycin against M kansasii [133] suggests this agent will also be highly useful in retreatment regimens, perhaps allowing for omission of the aminoglycoside. The newer quinolones may also be potentially useful in this setting but have not been studied.

For treatment of extrapulmonary disease in adults, the regimen of antimycobacterial drugs should be the same as for pulmonary disease. Pulmonary disseminated disease has been described in patients with AIDS [59], and it is the second most common NTM that produces disease in this setting [51]. Of the cases detailed in the literature, most have been fatal [59]. In the treatment of Iymph node disease in children, excision of all accessible nodes at the time of the initial biopsy should be done, since the etiologic agent is probably an NTM other than M kansasii, for which excision is the indicated treatment.

The use of protease inhibitors for the treatment of HIV disease complicates the management of M kansasii disease because rifampin dramatically enhances the metabolism of these drugs and cannot be used with them concurrently. Options for treating HIV-infected patients who receive a protease inhibitor are to substitute clarithromycin for rifampin in the standard regimen, or to substitute rifabutin 150 mg/d for rifampin if the patient is receiving indinavir. None of these regimens has been studied clinically, however they appear likely to be successful.

Copyright 1997 The American Thoracic Society, Am J Respir Crit Care Med 1997; 156:S1. Reprinted with permission from The American Thoracic Society. For more information about the ATS, please visit their website at www.thoracic.org.

TREATMENT OF PULMONARY MYCOBACTERIUM AVIUM COMPLEX DISEASE (M AVIUM, M INTRACELLULARE) — Medical treatment of M avium complex pulmonary disease in HIV-negative patients has historically been frustrating and disappointing. In the few studies in which initial sputum conversion rates have been high (> 80 percent), long-term follow-up to establish continued sputum conversion is rarely documented [151-155]. Relapses after medical therapy with pre-macrolide treatment regimens are common, and the best outcomes have frequently been in those patients subjected to resectional surgery [156,157]. Recent significant advances in the drugs available for treatment of M avium complex have been made, however, and there is now greater expectation that pulmonary M avium complex disease can be effectively treated (defined as high rates of sputum conversion with long-term culture negativity) with medications alone.

The major limitations for effective therapy have been the absence of antimicrobial agents with low toxicity and good in vivo activity against the organism. Most first-line antituberculosis drugs have 10-100 times less in vitro activity against M avium complex isolates than against M tuberculosis. This diminished activity may be due to the lipophilic cell wall of M avium complex, which prevents drug penetration [158].

The major therapeutic advances in the treatment of pulmonary M avium complex disease have come as a result of progress in treating disseminated M avium complex in the setting of HIV disease. Recent studies have shown excellent in vivo and clinical activity against M avium complex by the newer macrolides, clarithromycin, and azithromycin, presumably due to the high phagocyte and tissue levels achieved by these agents. (The structure of azithromycin is technically an azalide; however, because of the close similarity of azalides to macrolides, the term macrolide will be used to refer to both in subsequent discussion.) Human trials of patients with disseminated M avium complex disease and AIDS have shown both newer macrolides to have clinical and microbiologic activity as monotherapy [129,159-161] and clarithromycin to have clinical and microbiologic activity in drug combinations [160]. Studies have also demonstrated significant sterilizing activity of both azithromycin and clarithromycin with short-term initial treatment as single agents in pulmonary M avium complex disease [162,163]. Although it is not appropriate to treat patients with disease (outside of clinical trials) with single agents, these are the first studies demonstrating significant in vivo activity of any single agent for pulmonary M avium complex disease.

Rifabutin, a derivative of rifamycin S, has been shown to be more active in vitro than rifampin against isolates of M avium complex [116]. Peak serum rifabutin concentrations are much lower than those of rifampin and exceed the MICs for only approximately 70 percent of M avium complex strains [122]. The relevance of this finding is uncertain, given higher levels of the drug in tissue than in serum. Rifabutin has demonstrated effectiveness as a prophylactic agent against disseminated M avium complex disease in AIDS [54], and there is some evidence of improved clinical response of pulmonary M avium complex disease in HIV-negative patients when rifabutin is added to multidrug regimens [164]. The newer macrolides and rifabutin form the basis of improved treatment regimens for pulmonary M avium complex disease.

Clinical presentations — The natural history of M avium complex lung disease is unpredictable in HIV-negative patients. Some patients maintain a stable clinical and radiographic picture for years, whereas others have a relatively rapid progression of their disease. The variability in the natural history of M avium complex lung disease appears to relate in part to the presence of two types of clinical disease and presentation. The traditional presentation of M avium complex lung disease has been as apical fibrocavitary lung disease, sometimes with huge cavities, in males in their late 40s and early 50s who have a history of heavy cigarette smoking and, frequently, alcohol abuse. This form of disease is generally progressive within one to two years if left untreated. More recently, it has become apparent that M avium complex lung disease also presents as bilateral nodular and interstitial/nodular disease or as isolated right middle lobe or lingular disease, predominantly in elderly nonsmoking females [71-74,165]. These syndromes were often referred to as M avium complex "colonization," with bronchiectasis considered to be the only real disease. This form of disease (nodular bronchiectasis) tended to have a much slower progression, such that long term follow-up (5-10 years) was often needed to show clinical or radiographic changes. For the nodular bronchiectatic forms of M avium complex lung disease, sputum submission for AFB analysis was often delayed months to years as the patient had radiographic and symptomatic progression.

In the past, many patients with nodular bronchiectatic M avium complex disease received a variety of nonspecific measures ("pulmonary toilet") rather than specific antimycobacterial therapy for their pulmonary disease. These nonspecific therapies included bronchodilators, postural drainage, smoking cessation, and broad spectrum antibiotics [166]. Clearly, these measures may be appropriate and can be associated with symptomatic improvement in bronchiectasis unrelated to any more active agents against M avium complex. Before the development of newer, more active agents against M avium complex, it had been proposed that patients in this group with a stable clinical picture (usually over three months) and minimal symptoms, and those whose sputum "cleared" with nonspecific therapy, should not be treated. However, based on recent studies using chest HRCT scans, these patients have specific radiographic features of parenchymal disease in addition to their multifocal bronchiectasis [71-74]. The term "colonization" is, therefore, incorrect. The appropriate distinction is not between colonization and invasive disease but between those patients with the disease of nodular bronchiectasis who require immediate therapy directed at M avium complex and those in whom such a decision can be delayed. If a decision is made to observe such a patient (eg, one with minimal symptoms and radiographic findings such that the treatment seems worse than the disease, or one with other major medial problems), it is incumbent upon the treating physician to continue collecting respiratory specimens for AFB analysis as well as follow-up chest radiographs and/or CTs over a relatively long period of time (perhaps the patient's entire lifetime), as the M avium complex disease will likely progress at some time and the patient's symptoms and chest radiographs will likely worsen. We recommend careful eval‎uation of each respiratory M avium complex isolate in the context of the patient's overall status. In general, almost any patient with two or three positive respiratory cultures for M avium complex has M avium complex lung disease. Although not all patients with an M avium complex respiratory isolate will require immediate therapy, the clinician must resist the temptation to make facile clinical decisions, particularly if that decision is to not treat a patient with a M avium complex respiratory isolate. As treatment regimens improve both in effectiveness and the ability of the patient to tolerate the medications, there should be less reluctance to treat M avium complex lung disease at an earlier stage. For patients with substantial symptoms and/or advanced or progressive radiographic abnormalities, an observation period is not needed to establish the need for therapy.

Drug treatment — Drug therapy for M avium complex disease involves multiple drugs; therefore, the risk of drug toxicities is relatively high. Additionally, the optimal therapeutic regimen has yet to be established. For these reasons, the treatment of M avium complex disease may best be served by physicians experienced in pulmonary or mycobacterial diseases.

With empiric combination regimens that include clarithromycin and usually ethambutol and a rifamycin (rifampin or rifabutin), sputum conversion rates for pulmonary M avium complex disease in adult patients able to tolerate the medications are about 90 percent [162,167,168]. Rifabutin is the preferred rifamycin because it is more active in vivo than rifampin against M avium complex, but it may also produce more problematic adverse events (uveitis, leukopenia). All untreated strains of M avium complex are macrolide susceptible (clarithromycin MICs of 0.25 of 4.0 ~g/mL), while microbiologic relapses associated with symptom recurrence reveal isolates with MICs of > 32 µg/mL [169]. Patients with either pulmonary or disseminated M avium complex do not respond to macrolide containing regimens in which the macrolide is the sole or principle agent if the patient's isolate is macrolide resistant in vitro [129,162,167,168]. Isolates of M avium complex resistant to clarithromycin are cross-resistant to azithromycin [169].

The newer macrolides are the cornerstone of contemporary therapy for pulmonary M avium complex disease, as they are for disseminated M avium complex disease. Initial therapy for adult HIV-negative patients with M avium complex disease needing treatment should consist of a minimum three-drug regimen of clarithromycin (500 mg twice a day) or azithromycin (250 mg/d or 500 mg three times a week), rifabutin (300 mg/d) or rifampin (600 mg/d), and ethambutol (25 mg/kg per day for 2 months followed by 15 mg/kg per day). For patients of small body mass and/or an age over 70, clarithromycin at 250 mg twice a day or azithromycin 250 mg three times a week may be better tolerated. Studies are currently ongoing to determine the feasibility and efficacy of both azithromycin- and clarithromycin-containing regimens with all drugs given intermittently (three times weekly) for pulmonary M avium complex disease.

The potential and method for treating pediatric patients (eg, those with underlying cystic fibrosis) with the above regimen has not been studied, nor have drug doses for the newer agents such as clarithromycin and rifabutin.

Intermittent streptomycin for the first two to three months of therapy may be considered, in addition to the above regimen, for extensive disease. The exact dose of streptomycin in this multidrug regimen will depend on the patient's age and weight. For extensive disease, we recommend at least two months of intermittent (twice or three times weekly) streptomycin, although longer therapy with streptomycin may be desirable in patients with very extensive disease or for those who do not tolerate other agents. There are no data, however, comparing clarithromycin-containing regimens with and without an aminoglycoside. The patient and physician should be alert to signs and symptoms of streptomycin toxicity, and these may prevent completion of the full course of therapy. Because ototoxicity due to streptomycin is often irreversible, patients receiving streptomycin should be instructed in the signs and symptoms of toxicity [unsteady gait, tinnitus, diminished hearing] at the start of therapy and again on subsequent visits, with discontinuation or decrease in dosage or frequency if suggestive signs of toxicity occur. Suggested doses of streptomycin based on patient age and weight are shown in table 4 (show table 4).

The optimal length of drug therapy for M avium complex lung disease has not been established. Recommendations in the pre-macrolide era were that patients be treated for 18 to 24 months without considering how long the sputum culture results were negative. This recommendation was based on empiric data, in part, drawing from the early experience in treating tuberculosis. With macrolides, a shorter length of therapy seems acceptable. Only two long-term studies of M avium complex lung disease have been reported to date, both investigating clarithromycin [167,168]. One study using 12 months of culture negativity as the treatment endpoint observed no pulmonary disease relapses with a mean follow-up of 18 months [167], while the second study, which used 7 to 9 months of culture negativity resulted in no early pulmonary disease relapses with a mean follow-up of 7 months [168]. Early relapses with less than 10 months of culture negativity were seen in the first study. These initial studies suggest that culture negativity of 10 to 12 months while on a clarithromycin-containing regimen is adequate for most patients.

Acid-fast bacilli smears and cultures of sputum should be obtained monthly during therapy for pulmonary M avium complex disease to assess response, then periodically after completion of therapy to eval‎uate possible relapse. The desired endpoint is negative sputum cultures; patients who respond to therapy should develop negative AFB smears and cultures. One or more cultures containing small numbers of M avium complex organisms (single colonies on solid media or positive liquid media cultures only) may occur after sputum conversion and should not necessarily be interpreted as indicative of treatment failure or relapse. Rather, these culture results should be interpreted in light of the patient's overall clinical status.

All patients should show clinical improvement within 3 to 6 months and should convert their sputum to negative within 12 months on macrolide-containing regimens [167]. Thus, patience is required in eval‎uating response to therapy. However, failure to respond in these time periods should prompt investigation for possible noncompliance or macrolide resistance.

For patients whose disease has failed to respond to a macrolide-containing regimen (usually as a consequence of in vitro macrolide resistance, noncompliance, or drug intolerance) and have progressive, symptomatic disease, an alternative drug regimen will be necessary. The treatment for pulmonary M avium complex disease in HIV-negative adult patients as recommended in the first ATS NTM statement published in 1990, consisted of the four-drug regimen of isoniazid (300 mg/d), rifampin (600 mg/d), ethambutol (25 mg/kg per day for the first two months, then 15 mg/kg per day) with streptomycin for the initial three to six months of therapy [1]. With the use of rifabutin instead of rifampin, this may be a reasonable regimen for patients who are macrolide resistant or intolerant. Which drugs are most useful in treating macrolide-resistant strains is a major issue to be addressed in future studies as resistant strains become more preval‎ent. Although there are no trials comparing regimens with and without a macrolide for treating pulmonary M avium complex, most experts consider the regimen without a macrolide to be inferior, and one such regimen (rifampin, ethambutol, ciprofloxacin, and clofazimine) has been shown to be inferior to a clarithromycin-containing regimen for disseminated M avium [170].

A number of other drugs have been used in multidrug regimens in the past, but they are limited by toxicity (eg, cycloserine and ethionamide) or little or no evidence of clinical efficacy (eg, clofazimine, newer quinolones, capreomycin). Some experts feel that clofazimine 100 mg/d and ciprofloxacin 750 mg twice daily or ofloxacin 400 mg twice daily are useful in the setting of M avium complex lung diseases, although there are no data corroborating their efficacy. Therapy with only two drugs, especially isoniazid and rifampin only, is strongly. Antituberculosis drugs are generally well tolerated, even by the elderly population with M avium complex lung disease.

If patients are unable to tolerate or fail first-line antituberculosis medications, an alternative "salvage" regimen should be considered with potential agents including ciprofloxacin 750 mg twice daily or ofloxacin 400 mg twice daily, clofazimine 100 mg daily, ethionamide (250 mg twice a day, then increased to three times a day as tolerated), and prolonged use of streptomycin or amikacin [three to five times per week]. A multiple drug regimen including these potentially toxic drugs can also be associated with at least short-term conversion of the sputum to AFB-negative. The long-term success rate for salvage regimens is unknown but is likely very low.

The role of immune therapy in patients who fail drug therapy has not been established. Interleukin and gamma interferon have been used in selected patients, and some investigation in this area continues.

Surgical treatment — Patients whose disease is localized to one lung and who can tolerate resectional surgery might also be considered for surgery, if there has been poor response to drug therapy or if the patient's isolate has become macrolide resistant. For some patients successfully treated by surgical resection, the prognosis has been better than for patients treated medically, although these results predated the use of macrolide-containing regimens [156,157]. Lung resectional surgery for mycobacterial disease is associated with significant morbidity and mortality [171,172]. In one recent series from a thoracic surgeon experienced in mycobacterial surgery, 8 of 38 (21 percent) of patients undergoing surgery and 8 of 17 (47 percent) of patients undergoing pneumonectomy developed postoperative bronchopleural fistulae, especially following a right pneumonectomy [171]. Whenever possible, this surgery should be performed at centers with thoracic surgeons who have considerable experience with this type of surgery. Overall, the bilateral nature of M avium complex lung disease, the advanced age of the patients, and the frequency of underlying chronic lung disease have limited the number of patients who are good candidates for surgery.

Toxicity monitoring — Monitoring of patients for toxicity, given the number of drugs and the older age of these patients, is essential. Monitoring should include visual acuity (ethambutol and rifabutin), red-green color discrimination (ethambutol), liver enzymes (clarithromycin, azithromycin, rifabutin, rifampin, isoniazid, ethionamide) [173], auditory and vestibular function (streptomycin, amikacin, clarithromycin, azithromycin), renal function (streptomycin and amikacin), leukocyte and platelet counts (rifabutin) [174,175], and the central nervous system (cycloserine). Patients who receive both a macrolide and rifabutin must be monitored for the development of toxicity related to the interaction of these drugs [174,175]. Clarithromycin enhances rifabutin toxicity (especially uveitis) while the rifamycins, rifampin more than rifabutin, lower clarithromycin serum drug levels. Details are provided in the section on monitoring for drug toxicity.

TREATMENT OF LOCALIZED EXTRAPULMONARY MYCOBACTERIUM AVIUM COMPLEX DISEASE

Lymphadenitis — Excisional surgery without chemotherapy is the recommended treatment for children with NTM cervical Iymphadenitis, including those with disease caused by M avium complex and M scrofulaceum [78,79,176,177]. The success rate with this procedure is about 95 percent [78]. Incisional biopsy or the use of antituberculosis drugs alone [without a macrolide] has frequently been followed by persistent clinical disease, including sinus tract formation and chronic drainage, and should be avoided [78,79,176,177]. For children with recurrent disease, a second surgical procedure is usually performed. An alternative for recurrent disease or for children in whom surgical risk is high (eg, risk of facial nerve involvement) may be the use of a clarithromycin multidrug regimen such as that used for pulmonary disease [79,178,179]. Experience with such an approach is limited [178-181], but the proven activity of clarithromycin against M avium complex in other clinical settings and preliminary reports makes this approach appear promising.

A special problem is created by the child who has granulomatous disease with or without AFB on examination of the excised Iymph nodes, and whose PPD tuberculin skin test is strongly positive (eg, more than 15 mm). A course of antituberculosis therapy while awaiting the results of the Iymph node culture is reasonable, especially when there are any risk factors for tuberculosis (positive family history, foreign-born child, etc). If the cultures fail to yield any mycobacteria, antituberculosis therapy should be discontinued unless there are significant risk factors for tuberculosis.

Skin, tissue, and skeletal disease — For adult patients with extrapulmonary, localized M avium complex disease involving skin, soft tissue, tendons and joints, and occasionally bone, a combination of excisional surgery (or surgical debridement) and chemotherapy is usually performed. Whether a three-drug regimen alone in this setting would be adequate is not known. The optimal duration of treatment is also unknown, but drug treatment usually lasts 6 to 12 months.

TREATMENT OF DISSEMINATED MYCOBACTERIUM AVIUM DISEASE — Disseminated M avium is associated with an increased mortality in patients with AIDS. In one natural history study, the median survival was 134 days after the first positive blood culture, and only 13 percent of patients were alive at one year [53]. Initially, some clinicians questioned whether M avium was a direct cause of death or only present in persons who were dying of other reasons. Several controlled studies have shown shortened survival in patients with disseminated M avium when compared to cohorts of patients without M avium [182]. Based on the increased morbidity and mortality associated with disseminated M avium, prophylaxis should be strongly considered in high-risk patients and therapy should be offered to all patients with established disease.

Early (pre-macrolide) studies of the treatment of M avium in patients with AIDS demonstrated the ability of multidrug regimens to lower the burden of mycobacteria in the blood and improve symptoms [183,184]. The drugs used in these studies such as ethambutol, clofazimine, rifampin, and ciprofloxacin have been shown to have modest activity in vitro, and two of the agents (ethambutol and rifabutin) to have modest activity in single-drug therapy studies of patients with AIDS and M avium [185,186]. A major advance in therapy came with the recognition that clarithromycin and azithromycin were potent agents against M avium complex. Both clarithromycin and azithromycin were shown to markedly reduce the number of bacteria in the blood of patients in small pilot studies [159,161]. In a larger study of 154 adult patients with AIDS and M avium bacteremia, clarithromycin was given as single-drug therapy in doses of 500, 1,000 or 2,000 mg twice daily. All three groups had clearance of bacteremia and reduction in symptoms, although the groups receiving the higher doses had greater toxicity and a higher mortality [129]. It was also noted, however, that resistance was a problem, as clinical relapse and in vitro resistance developed in approximately 20 percent of individuals by 12 weeks.

Rifabutin has also been demonstrated to be effective in several small studies of patients with AIDS who had disseminated M avium [183,187]. As monotherapy, it was shown to reduce colony counts in the blood [186]. Clearance of bacteremia occurred in 7 of 11 patients receiving rifabutin, clofazimine, and ethambutol, compared to 0 of 13 patients with clofazimine alone in another study [187].

Due to problems with drug resistance, as well as the need to eradicate large numbers of organisms, multidrug therapy is considered essential in the treatment of patients with disseminated M avium. There are currently few well done comparative trials of the many possible multidrug regimens. A Canadian HIV Trials Network study [170] involving 229 patients did demonstrate the combination of clarithromycin, rifabutin, and ethambutol to be superior to rifampin, ethambutol, clofazimine, and ciprofloxacin at both reducing bacteremia (69 percent versus 29 percent, p < 0.001) and prolonging median survival (8.6 months versus 5.2 months, p = 0.001).

Based on currently available data, it would be advisable to always use a minimum of three drugs, one of which should be clarithromycin (500 mg twice daily) or azithromycin (250 mg or 500 mg daily). Most investigators would use ethambutol as the second agent at a dose of 15 mg/kg per day, although consideration should be given to an initial course of 25 mg/kg for the first two months. Rifabutin has the best potential as the third agent. Use of rifabutin will be problematic, however, in patients also on protease inhibitors, given its induction of the cytochrome P-450 system that metabolizes all currently approved members of this drug class. Clofazimine has also been used, as has a quinolone, but neither seems to contribute much to the regimen, and clofazimine has been associated with a higher mortality in two comparative treatment trials [188]. Amikacin [189] and streptomycin are both active, and one or the other should be considered for use in patients with severe symptoms due to M avium complex, especially as part of initial therapy.

It should be noted that drugs used to treat mycobacterial diseases in patients with AIDS are associated with frequent adverse effects, and changes to therapeutic regimens may often be required. Of particular note has been the frequent occurrence of uveitis when doses of clarithromycin higher than 500 mg twice daily have been used in combination with rifabutin doses of 600 mg daily [174]. This incidence fell to only 6 percent (3 of 53 patients) in the Canadian HIV trial when the rifabutin dose was reduced to 300 mg daily, the currently recommended dose [170].

Another problem is the interaction of rifamycins with the recently introduced protease inhibitors (saquinavir, ritonavir, and indinavir) for treatment of AIDS. Rifampin, and to a lesser degree rifabutin, enhances hepatic metabolism of the protease inhibitors, which may result in subtherapeutic levels of these agents and promote the emergence of resistant HIV strains. The protease inhibitors inhibit metabolism, and therefore promote dose-related adverse effects, of the rifamycins (especially rifabutin). Recent recommendations, made in the context of tuberculosis therapy, suggest that rifampin should not be used with the protease inhibitors, but that rifabutin can be used at modified doses with at least one of these agents, indinavir [190]. This recommendation would have little impact on the treatment or prophylaxis of disseminated M avium disease since rifabutin is the preferred rifamycin. The impact on the treatment of other NTM such as M kansasii, where rifampin has traditionally been used is less clear. Alternative strategies include treatment of NTM infections without a rifamycin or withholding the protease inhibitor until the mycobacterial infection has been treated. The improved immune function resulting from aggressive antiretroviral therapy, including a protease inhibitor, might ultimately be the most important factor for clearance of disseminated NTM infection in AIDS patients; therefore, continuing a protease inhibitor is a high priority. Optimal therapy for disseminated NTM disease, especially M avium, requires a multidrug treatment regimen including a rifamycin. Overall, in a patient on protease inhibitors with proven disseminated M avium, it still seems prudent to include rifabutin in the treatment regimen, even if the dose is attenuated. For other NTM, the importance of the rifamycin should be eval‎uated based on the specific organism being treated.

Copyright 1997 The American Thoracic Society, Am J Respir Crit Care Med 1997; 156:S1. Reprinted with permission from The American Thoracic Society. For more information about the ATS, please visit their website at www.thoracic.org.

PROPHYLAXIS OF DISSEMINATED DISEASE IN AIDS — The incidence of disseminated M avium can be reduced by prophylactic antimicrobials. Rifabutin was demonstrated to be effective in two placebo-controlled, double-blind studies. Mycobacterium avium bacteremia developed in 8 percent of adult patients receiving 300 mg of rifabutin daily and in 17 percent of patients on placebo [54]. Because rifabutin is highly active against M tuberculosis, it is probable that daily use of rifabutin would also provide prophylaxis against tuberculosis. Active tuberculosis must be ruled out before initiating rifabutin prophylaxis in order to prevent the development of drug-resistant tuberculosis. Clarithromycin in a dose of 500 mg twice daily was effective in a controlled trial of 667 adult patients in reducing the incidence of M avium complex bacteremia from 16 percent in the placebo group to 6 percent in the treatment group [191,192], while in a related trial it was shown to be more effective than rifabutin [192]. Azithromycin at a dose of 1,200 mg once weekly, either alone or in combination with rifabutin, has also been shown to be effective in a published clinical trial involving 693 adult patients [193]. The final selection of agents may depend on cost, tolerability, and potential drug interactions of the agents. Rifabutin should generally be avoided in patients on protease inhibitors because it markedly enhances their metabolism and reduces serum levels of the protease inhibitors. Some clinicians and the United States Public Health Service have advocated use of indinavir but not other currently available protease inhibitors (ritonavir, saquinavir) with reduced-dose rifabutin if both drugs are deemed essential.

The development of drug resistance during prophylaxis is a concern, and it has already been noted to occur with the use of clarithromycin [191,192] or azithromycin [193] as monotherapy, but not the rifabutin monotherapy [54] or azithromycin when combined with rifabutin [193]. Because of the very high risk of disseminated M avium in persons with advanced HIV infection, prophylaxis should be offered to all patients with < 50 CD4 cells, especially in patients with a history of opportunistic infection [194,195].

TREATMENT OF RAPIDLY GROWING MYCOBACTERIAL DISEASE — Disease caused by the rapidly growing mycobacteria, especially cutaneous disease, has come to be recognized as relatively common in selected areas of the United States. The southeastern United States from Georgia to Texas appears to be the major endemic area, although disease has been reported from all over the United States. Most clinical disease is sporadic and community acquired, although nosocomial outbreaks or clustered cases have been reported [16-26], and the association of rapidly growing mycobacteria wound infections with augmentation mammoplasty [23,92] and cardiac surgery [16,17,21,93] is well recognized.

Most clinical disease (more than 90 percent) is due to three species of rapidly growing mycobacteria: M fortuitum, M abscessus, and M chelonae [90]. Two taxa originally identified as biovariants within M fortuitum (M fortuitum third biovariant complex, sorbitol-positive and sorbitol-negative) are still undergoing taxonomic eval‎uation. Both groups may eventually attain species status. Mycobacterium smegmatis [196], M peregrinum [90], M mucogenicum (formerly known as the "M chelonae-like organisms" or MCLO) [197], and rarely, chromogenic rapidly growing mycobacteria [198-200] may also occasionally be responsible for human disease.

Mycobacterium fortuitum, M abscessus, and M chelonae are resistant to the antituberculous agents, but they are susceptible (especially M fortuitum) to a number of traditional antibacterial agents [99,137-140]. Isolates of M fortuitum treatment are susceptible to amikacin (100 percent), ciprofloxacin and ofloxacin (100 percent), sulfonamides (100 percent) cefoxitin (80 percent), imipenem (100 percent), clarithromycin [80 percent], and doxycycline (50 percent). Isolates of M abscessus are susceptible to clarithromycin (100 percent), clofazimine, amikacin (90 percent), and cefoxitin (70 percent) and imipenem (50 percent). Isolates of M chelonae are susceptible to amikacin (80 percent), tobramycin (100 percent), clarithromycin (100 percent), imipenem (60 percent), clofazimine, doxycycline (25 percent), and ciprofloxacin (20 percent).

Cutaneous diseases — Clinical disease caused by the rapidly growing mycobacteria usually follows accidental trauma or surgery in a variety of clinical settings [90]. Some minor infections will resolve spontaneously or after surgical debridement. However, several studies of post-injection abscesses in which no therapy was given revealed disease that persisted in most patients for 8 to 12 months before spontaneously resolving. In two outbreaks of sternal wound infections caused by M abscessus in the era when little was known of chemotherapy or surgery for these organisms, approximately one-third of the patients died of uncontrolled infection [16,17]. Drug therapy or combined surgical and medical therapy clearly produce better results than these historical controls.

No controlled clinical trials of treatment for disease caused by M fortuitum, M abscessus, or M chelonae, comparing one form of treatment with another or with no drug treatment at all, have been performed. However, susceptibility studies [138-140] have demonstrated excellent in vitro activity of drugs such as clarithromycin, imipenem, cefoxitin, cefmetazole, and amikacin. Several case studies [196,201,202] and one clinical trial [203] of patients with cutaneous disease treated on the basis of in vitro susceptibilities have shown good results.

On the basis of these studies, guidelines have been suggested for drug therapy of nonpulmonary disease caused by rapidly growing mycobacteria [204]. Because of variable drug susceptibility among species and even within species and subgroups, susceptibility testing of all clinically significant isolates is essential for good patient management. The first-line antituberculosis drugs (isoniazid, rifampin, pyrazinamide, etc) have no role in the therapy of rapidly growing mycobacterial disease, with the exception of ethambutol, to which M smegmatis is susceptible [196].

For serious disease caused by M fortuitum and M abscessus, intravenous amikacin is given at a dose of 10 to 15 mg/kg in two divided doses to adult patients with normal renal function (average 400 mg twice a day) to provide peak serum levels in the low 20 µg/mL range. The lower dose (10 mg/kg) should be used in patients over the age of 50, once-daily dosing is unproven clinically but appears reasonable. the amikacin combined with high-dose cefoxitin (12 g/d given intravenously) is recommended for initial therapy (minimum two weeks) until clinical improvement is evident. For M chelonae, tobramycin is more active in vitro than amikacin. Imipenem appears to be a reasonable alternative to cefoxitin for these two species, and it should be used with isolates of M smegmatis and M chelonae that are resistant to cefoxitin [99,138,196]. Monitoring of renal function, eighth nerve function, and white blood cell counts (for the beta lactams) should be done on patients receiving this regimen. If organisms are susceptible to oral agents, therapy can be switched to one or more of these agents. For M abscessus, the only oral agents available for therapy are clofazimine and clarithromycin [139]. For M chelonae, clarithromycin, clofazimine, and (for approximately 20 percent of strains) ciprofloxacin and doxycycline are the only oral drugs susceptible in vitro [99,139]. The only clinical trial for M chelonae skin disease was done with clarithromycin. Of patients (all adults) treated with monotherapy at 500 mg twice a day for 6 months, all were cured except one patient (8 percent) who relapsed with an isolate that developed resistance to clarithromycin [203]. For serious disease, a minimum of four months of therapy is necessary to provide a high likelihood of cure. For bone infections, six months of therapy is recommended [201].

Surgery is generally indicated with extensive disease, abscess formation, or where drug therapy is difficult. Removal of foreign bodies such as breast implants, percutaneous catheters, etc. is important, or even essential, to recovery.

Pulmonary disease — The preval‎ence of lung disease caused by rapidly growing mycobacteria is unknown; however, it is likely more common than early estimates as these organisms have gained increasing recognition as pathogens. The largest group of patients with this lung disease are elderly (older than 60), Caucasian, female nonsmokers with no predisposing conditions or known lung disease. Underlying disorders that are associated with the disease include lung damage produced by prior mycobacterial infection (usually tuberculosis or M avium complex), gastroesophageal disorders with chronic vomiting, lipoid pneumonia, cystic fibrosis, and bronchiectasis due to a prior respiratory infection [205]. The distinguishing feature of patients with a recognized underlying disease is that their rapidly growing mycobacteria lung disease occurs at a younger age, usually less than 50, and almost all patients under 40 have one of these disorders [205].

Although early studies identified most respiratory isolates of rapidly growing mycobacteria as M fortuitum, use of modern identification schemes have shown that M abscessus (formerly M chelonae subspecies abscessus) accounts for approximately 80 percent of rapidly growing mycobacterial respiratory disease isolates, while M fortuitum (formerly M fortuitum biovariant fortuitum) accounts for approximately 15 percent of these isolates [205]. An important exception is the small group of patients who have gastroesophageal disorders with chronic vomiting and rapidly growing mycobacterial lung disease, in whom M abscessus and M fortuitum occur with equal frequency. Overall, M abscessus appears to be a more virulent respiratory pathogen than M fortuitum. Obtaining a single respiratory isolate of M abscessus is more likely to indicate significant disease than a single isolate of M fortuitum, although careful clinical eval‎uation and follow-up is always necessary to determine the significance of an NTM respiratory isolate. In lung disease due to rapidly growing mycobacteria in patients with no apparent risk factors, the chest radiograph usually shows multilobar, patchy, reticulonodular or mixed interstitial-alveolar infiltrates with an upper lobe predominance. Cavitation occurs in only approximately 15 percent of cases [205]. The chest radiograph is usually not typical for or suggestive of reactivation pulmonary tuberculosis, which likely accounts for a delay in ordering sputum for AFB analysis and therefore a delay in diagnosis. High resolution computed tomography of the lung frequently shows associated cylindrical bronchiectasis and multiple small (< 5 mm) nodules, a pattern also seen in nonsmokers with M avium complex lung disease [65-68]. Interestingly, approximately 15 percent of patients with M abscessus will also have M avium complex, suggesting the close relationship of the disorders [205]. Some patients have positive sputum cultures for Pseudomonas aeruginosa, further evidence of bronchiectasis. The radiographic features of this disease in cystic fibrosis are still under investigation.

The usual presenting symptoms are cough and easy fatiguability, often attributed for months or years to bronchitis or bronchiectasis. Fever, night sweats, and weight loss occur, but they are much less common and less severe than with M tuberculosis. The constellation of typical presenting symptoms in an elderly nonsmoking patient with no underlying lung disease, a compatible chest radiograph, and multiple positive sputa is sufficient to make a diagnosis. The presence of other diseases or unusual features may necessitate obtaining a lung biopsy (bronchoscopy with transbronchial biopsy) to be certain of the diagnosis.

The natural history of this disease depends primarily on the presence or absence of underlying disorders. For most patients with M abscessus and no underlying disorder, the disease is indolent and slowly progressive. Some patients show little radiographic change over years. More fulminant, rapidly progressive disease can occur, particularly in association with gastroesophageal disorders. Death occurs as a consequence of M abscessus in 20 percent of cases [205].

Mycobacterium fortuitum isolates, when they do occur, are usually susceptible to multiple oral antimicrobial agents including the newer macrolides and quinolones, doxycycline and minocycline, and sulfonamides [138-140]. Drug susceptibilities for this species are essential for effective therapy. Six to twelve months of therapy with two oral agents to which the M fortuitum isolate is susceptible in vitro usually results in clinical cure. Unfortunately, the M fortuitum group produces less than 20 percent of lung disease due to rapidly growing mycobacteria.

Mycobacterium abscessus isolates are usually susceptible in vitro only to the parenteral agents amikacin, cefoxitin, and imipenem, and to the newer oral macrolides (clarithromycin and azithromycin). Preliminary studies suggest that monotherapy with the newer macrolides is not sufficient to produce microbiologic cure for M abscessus. Combination therapy of low-dose amikacin plus high-dose cefoxitin for two to fours weeks almost invariably produces clinical and microbiologic improvement, but cost and morbidity prohibit potentially curative courses of treatment (four to six months). Surgical resection for limited disease related to prior localized lung disease can also be curative [205]. Unfortunately, suppressive therapy, including periodic parenteral antibiotic or oral macrolide therapy, may be all that can be realistically administered to control the symptoms and progression of M abscessus lung disease.

TREATMENT OF MYCOBACTERIUM MARINUM DISEASE — A number of treatment modalities have been used for cutaneous disease caused by M marinum [12,69]. These include simple observation for minor lesions, surgical excision, the use of antituberculous agents, and the use of single antibiotic agents. By standard susceptibility testing, these isolates are susceptible to rifampin and ethambutol, intermediately susceptible to streptomycin, and resistant to isoniazid and pyrazinamide. Isolates are also susceptible to clarithromycin, sulfonamides or trimethoprim-sulfamethoxazole and susceptible or intermediately susceptible to doxycycline and minocycline.

Acceptable treatment regimens in adults include clarithromycin 500 mg twice a day, minocycline or doxycycline at 100 mg twice a day [206-208], trimethoprim-sulfamethoxazole at 160/800 mg twice a day [209], or rifampin (600 mg) plus ethambutol (15 mg/kg) daily [207,210], with each regimen being given for at least three months. Rifampin alone has also been recommended, but little experience with this regimen has been reported [211]. The rate of clinical response is quite variable, and a minimum of four to six weeks of therapy should be given before considering that the patient may not be responding. Surgical debridement may also be important, especially for disease involving the closed spaces of the hand or disease that responds poorly to drug therapy [12,210]. If a lesion is excised surgically, many clinicians provide drug coverage during the perioperative period. It is not clear if longer durations of drug treatment after surgery offer any additional advantage.

TREATMENT OF PULMONARY DISEASE DUE TO OTHER NONTUBERCULOUS MYCOBACTERIA — Although most species of NTM have been reported to cause pulmonary disease, there are only four additional species that merit consideration. Because of the small number of reported cases and the absence of therapeutic trials or treatment of disease caused by these species, only limited recommendations on drug therapy can be made now. In most cases, if there has been satisfactory clinical and bacteriologic response to chemotherapy, a treatment period of 18 to 24 months is recommended.

Mycobacterium malmoense is a slowly growing, non-pigmented NTM species that causes pulmonary disease. Although rare in the United States, it has been increasingly recognized in England, Wales, and northern Europe. Most isolates are susceptible to ethambutol, and many are susceptible to rifampin and streptomycin. The four-drug regimen recommended for treating M avium complex (before the availability of macrolides and rifabutin) has resulted in clinical and bacteriologic responses in most cases [70,212]. Potential improvements in therapeutic response with the macrolides and rifabutin has not been assessed.

Mycobacterium simiae is a slowly growing, non-pigmented Mycobacterium that may be confused with M tuberculosis, as it is the only NTM that is niacin-positive. It is an uncommon cause of pulmonary and disseminated infection, and many patients with respiratory isolates of this species do not have disease. Most isolates are resistant to all first-line antimycobacterial drugs, and response to chemotherapy has been variable [32,213], For patients with disseminated or progressive pulmonary disease in need of treatment, initial therapy may be started with the four-drug regimen recommended for M avium complex (clarithromycin, ethambutol, rifabutin, and streptomycin), modified as needed using results of susceptibility tests.

Mycobacterium szulgai is a slowly growing Mycobacterium that has been associated with skin, joint, Iymphatic, pulmonary, and disseminated disease [134]. When isolated from humans, it should be considered pathogenic. Through 1986, only 24 cases of disease had been reported in English literature. The organism is usually susceptible to rifampin and higher concentrations of isoniazid, streptomycin, and ethambutol. Enhanced activity of rifampin, ethambutol, and streptomycin when used in combination has been shown in vitro [214]. Most patients treated with these drugs respond to therapy.

Relatively uncommon in the United States, M xenopi has been reported as a common cause of slowly progressive NTM pulmonary disease in western Europe. In southeast England, it is the most common NTM recovered in the laboratory and has been since 1977 [215]. Disseminated disease and joint disease caused by this organism have also been reported. In vitro susceptibility to antituberculosis agents is variable, although enhanced drug activity has been shown with the combination of rifampin and streptomycin [214]. Although some investigators have reported success with surgical therapy similar to that used for selected patients with M avium complex disease, others have had disappointing results [135,172]. Results with drug therapy alone in the pre-macrolide era have also shown variable results. One recent study of clarithromycin-containing regimens [136] demonstrated an excellent sputum conversion rate compared to these older studies. For most patients, initial therapy should consist of a macrolide, rifampin or rifabutin, and ethambutol with or without initial streptomycin. Patients who fail therapy or who relapse after treatment might be considered for surgery [30,135,172].

MONITORING FOR DRUG TOXICITY — Monitoring for drug toxicity of patients who are being treated for NTM disease is important, given the number and type of drugs used and the older age of these patients. It should include monitoring of the visual system including visual acuity (ethambutol), the presence of eye pain and decreased visual acuity or uveitis (rifabutin) [174,175], and red-green color discrimination (ethambutol); the central nervous system (cycloserine. ciprofloxacin, ofloxacin, ethionamide); the liver (isoniazid, rifampin, ethionamide, clarithromycin, rifabutin) [173,175]; the kidney (streptomycin, amikacin); auditory and vestibular function (streptomycin, amikacin, azithromycin); and hematologic indices (sulfonamides, cefoxitin, rifabutin) [175]. Major side effects and monitoring procedures are listed in Tables 1A and 1B (show table 5).

For the aminoglycosides, this monitoring should include routine questioning about balance, ability to walk (especially in the dark), tinnitus, dizziness, and difficulty hearing. Baseline blood urea nitrogen and creatinine measurements should be obtained, with reduction of the streptomycin dose and/or frequency of administration if these are abnormal. Periodic monitoring of renal function is recommended for high-risk patients receiving drugs daily or five times a week, especially with patients older than 50 years or who have impairment of renal function. A baseline hearing test should be considered, especially in high-risk patients, and then repeated if signs or symptoms of seventh nerve toxicity appear.

For cycloserine, careful attention should be given to signs of central nervous system toxicity, which may include seizures, lethargy, depression, alterations in personality, and even suicidal ideations. Because toxicity with this drug relates primarily to excessive serum levels (>40 µg/mL), patients with central nervous system symptoms or abnormal baseline renal function should have serum level determinations. Unfortunately, such tests are not available in most clinics and are rarely available in a timely fashion to help with decisions on dosing. If serum levels are not available, physicians should remember that the drug is excreted by the kidney, and high serum levels tend to occur in the elderly or in the presence of renal failure. A lower dose (eg, 250 mg twice a day) is often given in such cases, and discontinuation of the drug may be necessary if central nervous system symptoms occur and one cannot monitor serum levels.

Given that isoniazid hepatotoxicity is higher in the older population, some physicians obtain a baseline measurement of aspartate aminotransferase (AST), then repeat this determination at two, four and eight weeks, and thereafter during therapy as clinically indicated. Other monitoring considerations are the same as those for patients with tuberculosis (see "ATS/CDC/IDSA statement: Treatment of tuberculosis 0. Summary" ) [2,216].

The incidence of gastrointestinal side effects such as nausea, vomiting, abdominal pain, and cramping is almost prohibitive with ethionamide. These symptoms can be minimized by starting with a 250 mg single dose, slowly increasing the dose, and administering it with food. In the older patient, limiting the total dose to 500 mg may help reduce toxicity. A good review of side effects and toxicities experienced at the National Jewish Hospital with ethionamide and cycloserine is provided in a study by Lester [217].

Some adverse events with rifabutin are comparable to those seen with rifampin. These include rash, fever, nausea, vomiting, and hepatitis. Several adverse events are unique to rifabutin, including anterior uveitis, skin hyperpigmentation or pseudojaundice, and a polymyalgia/polyarthralgia syndrome [174,175]. The last three adverse events are seen almost exclusively in patients concurrently receiving clarithromycin. These effects can be minimized by giving no more than 300 mg/d of rifabutin when combined with clarithromycin. Leukopenia is also seen with rifampin, but it is much more common with rifabutin. Periodic white blood cell counts should be performed on all patients taking rifabutin, perhaps at monthly intervals. Mild leukopenia is common (8 of 26 or 31 percent of patients in one series of HlV-negative patients on 600 mg/d developed WBC counts below 4,000 cells/mm3) [175], and the drug should be discontinued only if cell counts fall below a comfortable range (perhaps 2,500 cells/mm3 or an absolute granulocyte count of 1,500 or less in HIV-negative patients).

Copyright 1997 The American Thoracic Society, Am J Respir Crit Care Med 1997; 156:S1. Reprinted with permission from The American Thoracic Society. For more information about the ATS, please visit their website at www.thoracic.org.

Use of UpToDate is subject to the Subscription and License Agreement.

REFERENCES

1.  American Thoracic Society. Diagnosis and treatment of disease caused by nontuberculous mycobacteria. Am Rev Respir Dis 1990; 142:94.

2.  American Thoracic Society. Treatment of tuberculosis and tuberculosis infection in adults and children. Am J Respir Crit Care Med 1994; 149:1359.

3.  American Thoracic Society. Diagnostic standards and classification of tuberculosis. Am Rev Respir Dis 1990; 142:725.

4.  Collins, CH, Grange, JM, Yates, MD. Mycobacteria in water. J Appl Bacteriol 1984; 57:193.

5.  Reznikov, M, Leggo, JH, Dawson, DJ. Investigation by seroagglutination of strains of the Mycobacterium intracellulare-. scrofulaceum group from house dusts to sputum in southeastern Queensland. Am Rev Respir Dis 1974; 104:951.

6.  Gruft, H, Falkinham, JO, Parker, BC. Recent experience in the epidemiology of disease by atypical mycobacteria. Rev Infect Dis 1981; 3:990.

7.  Meissner, PS, Falkinham, JO. Plasmid DNA profiles as epidemiologic markers for clinical and environmental isolates of Mycobacterium avium, Mycobacterium intracellulare, and Mycobacterium scrofulaceum. J Infect Dis 1986; 153:325.

8.  Meissner, PS, Anz, W. Sources of Mycobacterium avium-complex infection resulting in human disease. Am Rev Respir Dis 1977; 116:1057.

9.  Ahrens, P, Giese, S, Klausen, J, Inglis, NF. Two markers, IS901-IS902 and p40 identified by PCR and by using monoclonal antibodies in Mycobacterium avium strains. J Clin Microbiol 1995; 33:1049.

10.  Guerrero, C, Bernasconi, C, Burki, D, et al. A novel insertion element from Mycobacterium avium, IS1245, is a specific target for analysis of strain relatedness. J Clin Microbiol 1995; 33:304.

11.  von Reyn, CF, Maslow, JN, Barber, TW, et al. Persistent colonization of potable water as a source of Mycobacterium avium infection in AIDS. Lancet 1994; 343:1137.

12.  Collins, CH, Grange, JM, Noble, WC, Yates, MD. Mycobacterium marinum infections in man. J Hyg (Camb) 1985; 94:135.

13.  Wolinsky, E, Rynearson, TK. Mycobacteria in soil and their relation to disease-associated strains. Am Rev Respir Dis 1968; 97:1032.

14.  Steadman, JE. High-catalase strains of Mycobacterium kansasii isolated from water in Texas. J Clin Microbiol 1980; 11:496.

15.  McSwiggan, DA, Collins, CH. The isolation of M. kansasii and M. xenopi from water systems. Tubercle 1974; 55:291.

16.  Hoffman, PC, Fraser, DW, Robicsek, F, et al. Two outbreaks of sternal wound infections due to organisms of the Mycobacterium fortuitum complex. J Infect Dis 1981; 143: 533.

17.  Szabo, I. Mycobacterium chelonei endemy after heart surgery with fatal consequences. Am Rev Respir Dis 1980; 121:607.

18.  Lowry, PW, Jarvis, WR, Oberle, WA, et al. Mycobacterium chelonae causing otitis media in an ear-nose-and-throat practice. N Engl J Med 1988; 319:978.

19.  Maloney, S, Welbel, S, Daves, B, et. Mycobacterium abscessus pseudoinfection traced to an automated endoscope washer: Utility of epidemiologic and laboratory investigation. J Infect Dis 1994; 169:1166.

20.  Laussucq, S, Baltch, AL, Smith, RP, wt. Nosocomial Mycobacterium fortuitum colonization from a contaminated ice machine. Am Rev Respir Dis 1988; 138:891.

21.  Kuritsky, JN, Bullen, M, Broome, CV, et. Sternal wound infections and endocarditis due to organisms of the Mycobacterium fortuitum complex: A potential environmental source. Ann Intern Med 1983; 98:938.

22.  Bolan, G, Reingold, AL, Carson, et al. Infections with Mycobacterium chelonei in patients receiving dialysis and using processed hemodialyzers J Infect Dis 1985; 152:1013.

23.  Safranek, TJ, Jarvis, WR, Carson, A, et. Mycobacterium chelonae wound infections after plastic surgery employing contaminated gentian violet skin-marking solution. N Engl J Med 1987; 317:197.

24.  Wenger, JD, Spika, JS, Smithwick, RW, et al. Outbreak of Mycobacterium chelonae infection associated with use of jet injectors. JAMA 1990; 264:373.

25.  Hector, JS, Pang, Y, Mazurek, GH, et al. Large restriction fragment patterns of genomic Mycobacterium fortuitum: DNA as strain-specific markers and their use in epidemiologic investigation of four nosocomial outbreaks. J Clin Microbiol 1992; 30:1250.

26.  Wallace, RJ, Jr, Zhang, Y, Brown, BA, et al. DNA large restriction fragment patterns of sporadic and epidemic nosocomial strains of Mycobacterium chelonae and Mycobacterium abscessus J Clin Microbiol 1993; 31:2697.

27.  Gross, WM, Hawkins, JE, Murphy, DB. Origin and significance of Mycobacterium xenopi in clinical specimens. Bull Int Union Tuberc Lung Dis 1976; 51:267.

28.  Desplaces, N, Picardeau, M, Dinh, V, et al. Spinal infections due to Mycobacterium xenopi after discectomies. 35th Interscience Conference on Antimicrobial Agents and Chemotherapy. San Francisco, CA (Abstract). 1995. p.162.

29.  Bennett, SN, Peterson, DE, Johnson, DR, et al. Bronchoscopy-associated Mycobacterium xenopi pseudoinfections. Am J Respir Crit Care Med 1994; 150:245.

30.  Costrini, AM, Mahler, DA, Gross, WM, et al. Clinical and roentgenographic features of nosocomial pulmonary disease due to Mycobacterium xenopi. Am Rev Respir Dis 1981; 123:104.

31.  Lavy, A, Yoshpe-Purer, Y. Isolation of Mycobacterium simiae from clinical specimens in Israel. Tubercle 1992; 63:279.

32.  Valero, G, Peters, J, Jorgensen, JH, Graybill, JR. Clinical isolates of Mycobacterium simiae in San Antonio, Texas. Am J Respir Crit Care Med 1995; 152:1555.

33.  Rynkiewicz, DL, Ampel, NM. Lack of clinical significance of Mycobacterium simiae. 34th Annual Meeting of the Infectious Disease Society of America, October 8, 1994, Orlando, FL. Abstract 305, p. 92.

34.  Crossey, MJ, Yakrus, MA, Cook, MB, et al. Isolation of Mycobacterium simiae in a southwestern hospital and typing by multilocus enzyme electrophoresis. 94th General Meeting, American Society for Microbiology, May 24,1994, Las Vegas, NV. Abstract U38, p. 179.

35.  Livanainen, EK, Martikainen, PJ, Vaananen, PK, Katila, M-L. Environmental factors affecting the occurrence of mycobacteria in brook waters. Appl Environ Microbiol 1993; 59:398.

36.  Portaels, F, Larsson, L, Jenkins, PA. Isolation of Mycobacterium malmoense from the environment in Zaire. Tuber Lung Dis 1994; 76:160.

37.  Saito, H, Tomioka, H, Sato, K, et al. Mycobacterium malmoense isolated from soil. Microbiol Immunol 1994; 38:313.

38.  Bottger, EC. Mycobacterium genavense: an emerging pathogen. Eur J Clin Microbiol Infect Dis 1994; 13:932.

39.  Kiehn, TE, Hoefer, H, Bottger, EC, et al. Mycobacterium genavense infections in pet animals. J Clin Microbiol 1996; 34:1840.

40.  Portaels, F. Epidemiology of mycobacteria! diseases. Clin Dermatol 1995; 13:207.

41.  Falkinham, JO. Epidemiology of infection by nontuberculous mycobacteria. Clin Microbiol Rev 1996; 9:177.

42.  Crow, HE, King, CT, Smith, E, et. A limited clinical, pathologic, and epidemiologic study of patients with pulmonary lesions associated with atypical acid-fast bacilli in the sputum. Am Rev Tuberc 1957; 75:199-222.

43.  Lewis, AG, Lasche, EM, Armstrong, AL, Dunbar, FP. A clinical study of the chronic lung disease due to nonphotochromogenic acid-fast bacilli. Ann Intern Med 1960; 53:273.

44.  Timpe, A, Runyon, EH. The relationship of "atypical" acid-fast bacteria to human disease: A preliminary report. J Lab Clin Med 1954; 44:202.

45.  Christianson, LC, Dewlett, HJ. Pulmonary disease in adults associated with unclassified mycobacteria. Am J Med 1960; 29:980.

46.  Edwards, LB, Palmer, CE. Epidemiologic studies of tuberculin sensitivity: preliminary results with purified protein derivatives prepared from atypical acid-fast organisms. Am J Hyg 1958; 68:213.

47.  Edwards, LB, Acquaviva, FA, Livesay, VT, et al. An atlas of sensitivity to tuberculin, PPD-B, and histoplasmin in the United States. Am Rev Respir Dis 1969; 99:1.

48.  Good, RC, Snider, DE. Isolation of nontuberculous mycobacteria in the United States, 1980. J Infect Dis 1982; 146:829.

49.  O'Brien, RJ, Geiter, LJ, Snider, DE. The epidemiology of nontuberculous mycobacteria! diseases in the United States: Results from a national survey. Am Rev Respir Dis 1987; 135:1007.

50.  Ostroff, S, Hutwagner, L, Collin, S. Mycobacterial species and drug resistance patterns reported by state laboratories1992. 93rd American Society for Microbiology General Meeting, May 16,1993, Atlanta, GA. Abstract U-9, p. 170.

51.  Horsburgh, CJ, Jr, Selik, RM. The epidemiology of disseminated nontuberculous mycobacteria! infection in the Acquired Immunodeficiency Syndrome (AIDS). Am Rev Respir Dis 1989; 139:47.

52.  Hawkins, CC, Gold, JWM, Whimbey, E, et al. Mycobacterium avium- complex infections in patients with the acquired immunodeficiency syndrome. Ann Intern Med 1986; 105:184.

53.  Nightingale, SD, Byrd, LT, Southern, PM, et al. Incidence of Mycobacterium avium-intracellulare complex in human immunodeficiency virus-positive patients. J Infect Dis 1992; 165:1082.

54.  Nightingale, SD, Cameron, DW, Gordin, FM, et al. Two controlled trials of rifabutin prophylaxis against Mycobacterium avium complex infections in AIDS. N Engl J Med 1993; 329:828.

55.  Hoover, DR, Graham, NMH, Bacellar, H, et al. An epidemiologic analysis of Mycobacterium avium complex disease in homosexual men infected with human immunodeficiency virus type 1. Clin Infect Dis 1995; 20:1250.

56.  Hoover, DR, Saah, AJ, Bacellar, et al. Clinical manifestations of AIDS in the era of pneumocystis prophylaxis. N Engl J Med 1993; 329:1922.

57.  Horsburgh, CR. Mycobacterium avium complex infection in the acquired immunodeficiency syndrome. N Engl J Med 1991; 324:1332.

58.  Kalayjian, RC, Tossi, Z, Tomashefski, JF, et al. Pulmonary disease due to infection by Mycobacterium avium complex in patients with AIDS. Clin Infect Dis 1995; 20:1186.

59.  Sherer, R, Sable, R, Sonnenberg, M, et al. Disseminated infection with Mycobacterium kansasii in the Acquired Immunodeficiency Syndrome. Ann Intern Med 1986; 105:710.

60.  Kiehn, TE, White, M. Mycobacterium haemophilum: an emerging pathogen. Eur Clin Microbiol Infect Dis 1994; 13:925.

61.  Lerner, CW, Safdar, A, Coppel, S. Mycobacterium haemophilum infection in AIDS. Infect Dis Clin Prac 1995; 4:233.

62.  Butler, WR, O'Connor, SP, Yakrus, MA, et al. Mycobacterium celatum sp.. nov. Int J Syst Bacteriol 1993; 43:539.

63.  Springer, B, Tortoli, E, Richter, I, et al. Mycobacterium conspicuum sp. nov., a new species isolated from patients with disseminated infections. J Clin Microbiol 1995; 33:2805.

64.  Ausina, V, Barrio, J, Luquin, M, et al. Mycobacterium xenopi infections in the acquired immunodeficiency syndrome. Ann Intern Med 1988; 109:927.

65.  Rodriguez-Barradas, MC, Clarridge, J, Darouiche, R. Disseminated Mycobacterium fortuitum disease in an AIDS patient. Am J Med 1992; 93:473.

66.  Ries, KM, White, GL, Jr, Murdock, RT. Atypical mycobacterial infection caused by Mycobacterium marinum. N Engl J Med 1990; 322:633.

67.  Jemni, L, Hmouda, H, Letaief, A. Disseminated infection due to Mycobacterium malmoense in a patient infected with Human Immunodeficiency Virus. Clin Infect Dis 1994; 19:203.

68.  Huminer, D, Dux, S, Samra, Z, et al. Mycobacterium simiae infection in Israeli patients with AIDS. Clin Infect Dis 1993; 17:508.

69.  Wolinsky, E. State of the Art: Nontuberculous mycobacteria and associated diseases. Am Rev Respir Dis 1979; 119:107.

70.  Henriques, B, Hoffner, SE, Petrini, B, et al. Infection with Mycobacterium malmoense in Sweden: Report of 221 cases. Clin Infect Dis 1994; 18:596.

71.  Hartman, TE, Swensen, SJ, Williams, DE. Mycobacterium avium-intracellulare complex: eval‎uation with CT. Radiology 1993; 187:2326.

72.  Primack, SL, Logan, PM, Hartman, TE, et al. Pulmonary tuberculosis and Mycobacterium avium-intracellulare: A comparison of CT findings. Radiology 1995; 194:413.

73.  Moore, EH. Atypical mycobacteria! infection in the lung: CT appearance. Radiology 1993; 187:777.

74.  Patz, EF, Swenson, SJ, Erasmus, J. Pulmonary manifestation of nontuberculous mycobacteria. Radiol Clin North Am 1995; 33:719.

75.  von Reyn, CF, Green, PA, McCormick, D, et al. Dual skin testing with Mycobacterium avium sensitin and purified protein derivative: An open study of patients with M. avium complex infection or tuberculosis. Clin Infect Dis 1994; 19:15.

76.  Prince, DS, Peterson, DD, Steinger, RM, et al. Infection with Mycobacterium avium complex in patients without predisposing conditions. New Engl J Med 1994; 321:863.

77.  Lincoln, EM, Gilbert, LA. Disease in children due to mycobacteria other than Mycobacterium tuberculosis. Am Rev Respir Dis 1972; 105:683.

78.  Schaad, UB, Votteler, TP, McCracken, GH, Nelson, JD. Management of atypical mycobacteria! Iymphadenitis in childhood: A review based on 380 cases. J Pediatr 1979; 95:356.

79.  Wolinsky, E. Mycobacterial Iymphadenitis in children: a prospective study of 105 nontuberculous cases with long-term follow-up. Clin Infect Dis 1995; 20:954.

80.  Margileth, AM, Chandra, R, Altman, P. Chronic Iymphadenopathy due to mycobacteria! infection. Am J Dis Child 1984; 138:917.

81.  Huebner, RE, Schein, MF, Cauthen, GM, et al. Usefulness of skin testing with mycobacteria! antigens in children with cervical Iymphadenopathy. Pediatr Infect Dis J 1992; 11:450.

82.  Del Beccaro, MA, Mendelman, PM, Nolan, C. Diagnostic usefulness of mycobacteria! skin test antigens in childhood Iymphadenitis. Pediatr Infect Dis J 1989; 8:206.

83.  Lau, SK, Wei, WI, Kwan, S, Yew, WW. Combined use of fine-needle aspiration cytologic examination and tuberculin skin test in the diagnosis of cervical tuberculous Iymphadenitis. Arch Otolaryngol Head Neck Surg 1991; 117:87.

84.  Baily, TM, Akhtar, M, Ali, MA. Fine needle aspiration biopsy in the diagnosis of tuberculosis. Acta Cytol 1985; 29:732.

85.  Gupta, SK, Chugh, TD, Sheikh, ZA, Al-Rubah, NAR. Cytodiagnosis of tuberculous Iymphadenitis. Acta Cytol 1993; 37:329.

86.  Armstrong, KL, James, RW, Dawson, DJ, et al. Mycobacterium haemophilum causing perihilar or cervical Iymphadenitis in healthy children. J Pediatr 1992; 121:202.

87.  Lai, KK, Stottmeier, KD, Sherman, IH, McCabe, WR. Mycobacterial cervical Iymphadenopathy. JAMA 1984; 251:1286.

88.  Zaugg, M, Salfinger, M, Opravil, M, Luthy, R. Extrapulmonary and disseminated infections due to Mycobacterium malmoense: Case report and review. Clin Infect Dis 1993; 16:540.

89.  Grange, JM, Yates, MD, Pozniak, A. Bacteriologically confirmed non-tuberculous mycobacteria! Iymphadenitis in southeast England: A recent increase in the number of cases. Arch Dis Child 1995; 72:516.

90.  Wallace, RJ Jr, Swenson, JM, Silcox, JA, et al. Spectrum of disease due to rapidly growing mycobacteria. Rev Infect Dis 1983; 5:657.

91.  Hoy, JF, Rolston, KVI, Hopfer, RL, Bodey, GP. Mycobacterium fortuitum bacteremia in patients with cancer and longterm venous catheters. Am J Med 1987; 83:213.

92.  Clegg, HW, Foster, MT, Sanders, WE Jr, Baine, WB. Infection due to organisms of the Mycobacterium fortuitum complex after augmentation mammoplasty: Clinical and epidemiologic features. J Infect Dis 1983; 147:427.

93.  Wallace, RJ Jr, Musser, JM, Hull, SI, et al. Diversity and sources of rapidly growing mycobacteria associated with infections following cardiac surgery. J Infect Dis 1989; 159:708.

94.  Marston, BJ, Diallo, MO, Horsburgh, CR Jr, et al. Emergence of Buruli ulcer in the Daloa region of Cote d'lvoire. Am J Trop Med Hyg 1995; 52:219.

95.  Hellinger, WC, Smilack, JL, Greider, JL Jr, et al. Localized soft tissue infections with Mycobacterium avium/Mycobacterium intracellulare complex in immunocompetent patients: Granulomatous tenosynovitis of the hand or wrist. Clin Infect Dis 1995; 21:65-69.

96.  Horsburgh, CR Jr, Mason, UG, Farhi, DC, Iseman, MD. Disseminated infection with Mycobacterium avium-intracellulare. Medicine 1985; 64:36.

97.  Lichtenstein, IH, MacGregor, RR. Mycobacterial infections in renal transplant recipients: Report of five cases and review of the literature. Rev Infect Dis 1983; 5:216.

98.  Cooper, JF, Lichtenstein, MJ, Graham, BS, Schaffner, W.

99.  Wallace, RJ Jr, Brown, BA, Onyi, GO. Skin, soft tissue, and bone infections due to Mycobacterium chelonae (chelonae): Importance of prior corticosteroid therapy, frequency of disseminated infections, and resistance to oral antimicrobials other than clarithromycin. J Infect Dis 1992; 166:405.

100.  Ingram, CW, Tanner, DC, Durack, DT, et al. Disseminated infection with rapidly growing mycobacteria. Clin Infect Dis 1993; 16:463.

101.  Stone, AB, Schelonka, RL, Drehner, DM, et al. Disseminated Mycobacterium avium complex in non-human immunodeficiency virus-infected pediatric patients. Pediatr Infect Dis J 1992; 11:960.

102.  Whittier, S, Hopfer, RL, Knowles, MR, Gilligan, PH. Improved recovery of mycobacteria from respiratory secretions of patients with cystic fibrosis. J Clin Microbiol 1993; 31:861.

103.  Tenover, FC, Crawford, JT, Huebner, RT, et al. The resurgence of tuberculosis: Is your laboratory ready? J Clin Microbiol 1993; 31:767.

104.  Heifets, LB. Quantitative cultures and drug susceptibility testing of Mycobacterium avium clinical isolates before and during the antimicrobial therapy. Res Microbiol (Paris) 1994; 145:188.

105.  Wilson, ML, Stone, BL, Hildred, MV, Reves, RR. Comparison of recovery rates for mycobacteria from BACTEC 12B vials, Middlebrook 7H11-Selective 7H11 Biplates, and Löwenstein-Jensen Slants in a public health mycobacteriology laboratory. J Clin Microbiol 1995; 33:2516.

106.  Woods, GL, Fish, G, Plaunt, M, Murphy, T. Clinical eval‎uation of Difco ESP Culture System II for growth and detection of mycobacteria J Clin Microbiol 1997; 35:121.

107.  Vadney, FS, Hawkins, JE. Eval‎uation of a simple method for growing Mycobacterium haemophilum. J Clin Microbiol 1985; 22:884.

108.  Coyle, MB, Carlson, LDC, Wallis, CK, et al. Laboratory aspects of "Mycobacterium genavense," a proposed species isolated from AIDS patients J Clin Microbiol 1992; 30:3206.

109.  Woods, GL, Witebsky, FG. Mycobacterial testing in clinical laboratories that participate in the College of American Pathologists' mycobacteriology E survey: Results of a 1993 questionnaire. J Clin Microbiol 1995; 33:407.

110.  Huebner, RE, Good, RC, Tokars, JI. Current practices in mycobacteriology: Results of a survey of state public health laboratories. J Clin Microbiol 1993; 31:771.

111.  Butler, WR, Ahearn, DG, Kilburn, JO. High-performance liquid chromatography of mycolic acids as a tool in the identification of Corynebacterium, Nocardia, Rhodococcus, and Mycobacterium species. J Clin Microbiol 1986; 23:182.

112.  Jost, KC, Dunbar, DF, Barth, SS, et al. Identification of Mycobacterium tuberculosis and M avium complex directly from smear-positive sputum specimens and BACTEC 12B cultures by high-performance liquid chromatography with fluorescence detection and computer-driven pattern recognition models. J Clin Microbiol 1995; 33:1270.

113.  Kaminski, DA, Hardy, DJ. Selective utilization of DNA probes for identification of Mycobacterium species on the basis of cord formation in primary BACTEC 12B cultures. J Clin Microbiol 1995; 33:1548.

114.  Metchock, B, Diem, L. Algorithm for use of nucleic acid probes for identifying Mycobacterium tuberculosis from BACTEC 12B bottles. J Clin Microbiol 1995; 33:1934.

115.  Lazlo, A, Siddiqi, SH. Eval‎uation of a tepid radiometric differentiation test for the Mycobacterium tuberculosis complex by selective inhibition with p-nitro-a-acetylamino-b-hydroxypropiophenone. J Clin Microbiol 1984; 19:694.

116.  Woodley, CL, Kilburn, JO. In vitro susceptibility of Mycobacterium avium complex and Mycobacterium tuberculosis strains to a spiro-piperidyl rifamycin. Am Rev Respir Dis 1982; 126:586.

117.  Heifets, L. MIC as a quantitative measurement of the susceptibility of Mycobacterium avium strains to seven antituberculosis drugs. Antimicrob Agents Chemother 1988; 32:1131.

118.  Steadham JE, Stall, SK, Simmank, JL. Use of the BACTEC system for drug susceptibility testing of Mycobacterium tuberculosis, M. kansasii, and M. avium complex. Diagn Microbiol Infect Dis 1985; 3:33.

119.  Heifets, LB. Synergistic effect of rifampin, streptomycin, ethionamide, and ethambutol on Mycobacterium intracellulare. Am Rev Respir Dis 1982; 132:710.

120.  Kuze, F, Kurasawa, T, Bando, K, et al. In vitro and in vivo susceptibility of atypical mycobacteria to various drugs. Rev Infect Dis 1981; 3:885.

121.  Heifets, LB, Iseman, MD. Determination of in vitro susceptibility of mycobacteria to ansamycin. Am Rev Respir Dis 1985; 132:710.

122.  Heifets, LB, Iseman MD, Linhold-Levy, PF, Kanes, W. Determination of ansamycin MlCs for Mycobacterium avium complex in liquid medium by radiometric and conventional methods. Antimicrob Agents Chemother 1985; 28:570.

123.  Leysen, DC, Haemers, A, Pattyn, SR. Mycobacteria and the new quinolones. Antimicrob Agents Chemother 1989; 33:1.

124.  Gay, JD, DeYoung, DR, Roberts, GD. In vitro activities of norfloxacin and ciprofloxacin against Mycobacterium tuberculosis, M. avium complex, M. chelonei, M. fortuitum, and M. kansasii. Antimicrob Agents Chemother 1984; 26:94.

125.  Kiehn, TE, Cynamon, MH, Inderlied, CB, et al. Antimycobacterial susceptibility testing for Mycobacterium tuberculosis: Tentative standards. National Committee for Clinical Laboratory Standards document M24-T, December 1995; 15:1.

126.  Heifets, L. Susceptibility testing of Mycobacterium avium complex isolates. Antimicrob Agents Chemother 1996; 40:1759.

127.  Horsburgh, CR Jr, Mason, UG III, Heifets, LB, et al. Response to therapy of pulmonary Mycobacterium avium-intracellulare infection correlates with results of in vitro susceptibility testing. Am Rev Respir Dis 1987; 135:418.

128.  Ausina, V, Condom, MJ, Mirelis, B, et al. In vitro activity of clofazimine against rapidly growing nonchromogenic mycobacteria. Antimicrob Agents Chemother 1986; 29:951.

129.  Chaisson, RE, Benson, CA, Dube, MP, et al. Clarithromycin therapy for bacteremic Mycobacterium avium complex disease. Ann Intern Med 1994; 121:905.

130.  Meier, A, Kirschner, P, Springer, B, et al. Identification of mutations in 23S rRNA gene of clarithromycin-resistant Mycobacterium intracellulare. Antimicrob Agents Chemother 1994; 38:381.

131.  Nash, KA, Inderlied, CB. Genetic basis of macrolide resistance in Mycobacterium avium isolated from patients with disseminated disease. Antimicrob Agents Chemother 1995; 39:2625.

132.  Ahn, CH, Wallace, RJ Jr, Steele, LC, Murphy, DT. Sulfonamide-containing regimens for disease caused by rifampin-resistant Mycobacterium kansasii. Am Rev Respir Dis 1987; 135:10.

133.  Wallace, RJ Jr, Dunbar, D, Brown, BA, et al. Rifampin-resistant Mycobacterium kansasii. Clin Infect Dis 1994; 18:736.

134.  Maloney, JM, Gregg, CR, Stephens, DS, et al. Infections caused by Mycobacterium szulgai in humans. Rev Infect Dis 1987; 9:1120.

135.  Banks, J, Hunter, AM, Campbell, IA, et al. Pulmonary infection with Mycobacterium xenopi: Review of treatment and response. Thorax 1984; 39:376.

136.  Dautzenberg, B, Papillon, F, Lepitre, M, et al. Mycobacterium xenopi infections treated with clarithromycin- containing regimens (Abstract). Annual Meeting, 33rd Interscience Conference on Antimicrobial Agents and Chemotherapy, 1993. p.1125.

137.  Stone, MS, Wallace, RJ Jr, Swenson, JM, et al. An agar disk elusion method for clinical susceptibility testing of Mycobacterium marinum and the Mycobacterium fortuitum-complex to sulfonamides and antibiotics. Antimicrob Agents Chemother 1983; 34:486.

138.  Swenson, JM, Wallace, RJ Jr, Silcox, VA, Thornsberry, C. Antimicrobial susceptibility testing of 5 subgroups of Mycobacterium fortuitum and Mycobacterium chelonae. Antimicrob Agents Chemother 1985; 28:807.

139.  Brown, BA, Wallace, RJ Jr, Onyi, GO, et al. Activities of four macrolides, including clarithromycin, against Mycobacterium fortuitum, Mycobacterium chelonae, and M. chelonae-like organisms. Antimicrob Agents Chemother 1992; 36:180.

140.  Wallace, RJ Jr, Brown, BA, Onyi, G. Susceptibilities of Mycobacterium fortuitum biovar. fortuitum and the two subgroups of Mycobacterium chelonae to imipenem, cefmetazole, cefoxitin, and amoxicillan-clavulanic acid. Antimicrob Agents Chemother 1991; 35:773.

141.  Hawkins , JE, Wallace, RJ Jr, Brown, BA. Antibacterial susceptibility tests: mycobacteria. In: Manual of clinical microbiology, 5th ed, Balows, A, (Ed). American Society for Microbiology, Washington, DC, 1991; 114:1138.

142.  Picardeau, M, Prod'Hom, G, Raskine, L, et al. Genotypic characterization of five subspecies of Mycobacterium kansasii. J Clin Microbiol 1997; 35:25.

143.  Hobby, GL, Redmond, WB, Runyon, EH, et al. A study of pulmonary disease associated with mycobacteria other than M. tuberculosis: Identification and characterization of the mycobacteria. Am Rev Respir Dis 1967; 95:954.

144.  Pezzia, W, Raleigh, JW, Bailey, MC, et al. Treatment of pulmonary disease due to Mycobacterium kansasii: Recent experience with rifampin. Rev Infect Dis 1981; 3:1035.

145.  Ahn, CH, Lowell, JR, Ahn, SA, et al. Chemotherapy for pulmonary disease due to Mycobacterium kansasii: Efficacies of some individual drugs. Rev Infect Dis 1981; 3:1028.

146.  Francis, PB, Jay, SJ, Johanson, WG Jr. Course of untreated M. kansasii disease. Am Rev Respir Dis 1975; 111:477.

147.  Jenkins, DE, Bahar, D, Chofuas, I. Pulmonary disease due to atypical mycobacteria: Current concepts. Transactions 19th Conference on Chemotherapy of Tuberculosis 1960. p. 224.

148.  Ahn, CH, Lowell, JR, Ahn, SS, et al. Short-course chemotherapy for pulmonary disease caused by Mycobacterium kansasii Am Rev Respir Dis 1983; 128:1048.

149.  Banks, J, Hunter, AM, Campbell, IA, et al. Pulmonary infection with Mycobacterium kansasii in Wales, 1970-9: Review of treatment and response. Thorax 1983; 38:271.

150.  Jenkins, PA, Banks, J, Campbell, IA, Smith, AP. Mycobacterium kansasii pulmonary infection: A prospective study of the results of nine months of treatment with rifampicin and ethambutol. Thorax 1994; 49:442.

151.  Dutt, AK, Stead, VW. Long-term results of medical treatment in Mycobacterium intracellulare infection. Am J Med 1979; 67:449.

152.  Ahn, CH, Ahn, SS, Anderson, RA, et al. A four-drug regimen for initial treatment of cavitary disease caused by Mycobacterium avium complex. Am Rev Respir Dis 1986; 134: 438.

153.  Seibert, AF, Bass, JB. Four drug therapy of pulmonary disease due to Mycobacterium avium complex (Abstract). 1989 Annual Meeting of American Thoracic Society, May 14-17. Cincinnati, OH. Am Rev Respir Dis 1989; 139(S):A399.

154.  Davidson, PT, Khanijo, V, Goble, M, Moulding, TS. Treatment of disease due to Mycobacterium intracellulare. Rev Infect Dis 1981; 3:1052.

155.  Reich, J, Johnson, R. Mycobacterium avium complex pulmonary disease. Am Rev Respir Dis 1991; 143:1381.

156.  Corpe, RF. Surgical management of pulmonary disease due to Mycobacterium avium-intracellulare. Rev Infect Dis 1981; 3:1064.

157.  Moran, JF, Alexander, LG, Staub, EW, et al. Long-term results of pulmonary resection for atypical mycobacteria! disease. Ann Thorac Surg 1983; 35:597.

158.  Rastogi, N, Frehel, C, Ryter, A, et al. Multiple drug resistance in Mycobacterium avium: Is the wall architecture responsible for the exclusion of antimicrobial agents? Antimicrob Agents Chemother 1981; 20: 666.

159.  Dautzenberg, B, Truffot, C, Legris, S, et al. Activity of clarithromycin against Mycobacterium avium infection in patients with acquired immune deficiency syndrome. Am Rev Respir Dis 1991; 144:564.

160.  Dautzenberg, B, Saint Marc, T, Meyohas, MC, et al. Clarithromycin and other antimicrobial agents in the treatment of disseminated Mycobacterium avium infections in patients with acquired immunodeficiency syndrome. Arch Intern Med 1993; 153:368.

161.  Young, LS, Wiviott, L, Wu, M, et al. Azithromycin for treatment of Mycobacterium avium-intracellulare complex infection in patients with AIDS. Lancet 1991; 338:1107.

162.  Wallace, RJ Jr, Brown, BA, Griffith, DE, et al. Initial clarithromycin monotherapy for Mycobacterium avium-intracellulare complex lung disease. Am J Respir Crit Care Med 1994; 149:1335.

163.  Griffith, DE, Wallace, RJ Jr, Brown, BA, et al. Azithromycin monotherapy for HIV (-) patients with Mycobacterium avium-intracellulare complex lung disease (Abstract). Am J Respir Crit Care Med 1994; April:151(4): PA477.

164.  O'Brien, RJ, Lyle, MA, Snider, DE Jr. Rifabutin (ansamycin LM 427): a new rifamycin-S derivative for the treatment of mycobacterial diseases. Rev Infect Dis 1987; 9:519.

165.  Reich, JM, Johnson, RE. Mycobacterium avium complex pulmonary disease presenting as an isolated lingular or middle lobe pattern: The Lady Windermere Syndrome. Chest 1992; 101:1605.

166.  Ahn, CH, McLarty, JW, Ahn, SS, et al. Diagnostic criteria for pulmonary disease caused by Mycobacterium kansasii and Mycobacterium intracellulare. Am Rev Respir Dis 1982; 125: 388.

167.  Wallace, RJ Jr, Brown, BA, Griffith, DE, et al. Clarithromycin regimens for pulmonary Mycobacterium avium complex: The first 50 patients. Am J Respir Crit Care Med 1996; 153:1766.

168.  Dautzenberg, B, Piperno, D, Diot, P. et al., and the Clarithromycin Study Group of France. Clarithromycin in the treatment of Mycobacterium avium lung infections in patients without AIDS. Chest 1995; 107:1035.

169.  Heifets, L, Mor, N, Vanderkolk, I. Mycobacterium avium strains resistant to clarithromycin and azithromycin. Antimicrob Agents Chemother 1993; 37:2364.

170.  Shafran, SD, Singer, J, Zarowny, DP, et al. A comparison of two regimens for the treatment of Mycobacterium avium complex bacteremia in AIDS: Rifabutin, ethambutol, and clarithromycin versus rifampin, ethambutol, clofazimine and ciprofloxacin. N Engl J Med 1996; 335:377.

171.  Pomerantz, M, Madsen, L, Goble, M, Iseman, M. Surgical management of resistant mycobacteria! tuberculosis and other mycobacterial pulmonary infections. Ann Thorac Surg 1991; 52:1108.

172.  Parrot, RG, Grosset, JH. Post-surgical outcome of 57 patients with Mycobacterium xenopi pulmonary infection. Tubercle 1988; 69:47.

173.  Brown, BA, Wallace, RJ Jr, Griffith, DE, Girard, W. Clarithromycin-induced hepatotoxicity. Clin Infect Dis 1995; 20:1073.

174.  Shafran, SD, Deschenes, J, Miller, M, et al. Uveitis and pseudojaundice during a regimen of clarithromycin, rifabutin, and ethambutol. N Engl J Med 1994; 330:438.

175.  Griffith, DE, Brown, BA, Girard, WM, Wallace, RJ Jr. Adverse events associated with high-dose rifabutin in macrolide-containing regimens for the treatment of Mycobacterium avium complex lung disease. Clin Infect Dis 1995; 21:594.

176.  Castro, DJ, Hoover, L, Castro, DJ, Zuckerbraun, L. Cervical mycobacteria! Iymphadenitis: Medical vs. surgical management. Arch Otolaryngol 1985; 111:816.

177.  Taha, AM, Davidson, PT, Bailey, WC. Surgical treatment of atypical mycobacteria! Iymphadenitis in children. Pediatr Infect Dis 1985; 4:664.

178.  Green, PA, von Reyn, CF, Smith, RP Jr. Mycobacterium avium complex parotid Iymphadenitis: Successful therapy with clarithromycin and ethambutol. Pediatr Infect Dis 1993; 12:615.

179.  Stewart, MG, Starke, JR, Coker, NJ. Nontuberculous mycobacteria! infections of the head and neck. Arch Otolaryngol Head Neck Surg 1994; 120:873.

180.  Taj Jadavji, MD, Wong, A. Atypical mycobacteria cervical adenitis in normal childrenis clarithromycin effective? The Third International Conference on the Macrolides, Azalides and Streptogramins, Lisbon, Portugal, 1996. Abstract 7.23.

181.  Berger, CH, Pfyffer, GE, Nadal, D. Drug therapy of Iymphadenitis with non-tuberculous mycobacteria (NTM). The Third International Conference on the Macrolides, Azalides and Streptogramins, Lisbon, Portugal, 1996. Abstract 7.22.

182.  Chin, DP, Reingold, AL, Stone, EN, et al. The impact of Mycobacterium avium complex bacteremia and its treatment on survival of AIDS patientsa prospective study. J Infect Dis 1994; 170:578.

183.  Hoy, J, Mijch, A, Sandland, M, et al. Quadruple-drug therapy for Mycobacterium avium-intracellulare bacteremia in AIDS patients. J Infect Dis 1990; 161:801.

184.  Chiu, J, .Nussbaum, J, Bozzette, S, et al. Treatment of disseminated Mycobacterium avium complex infection in AIDS with amikacin, ethambutol, rifampin, and ciprofloxacin. Ann Intern Med 1990; 113:358.

185.  Kemper, CA, Halvir, D, Haghighat, D, et al. The individual microbiologic effect of three mycobacterial agents, clofazimine, ethambutol, and rifampin, on Mycobacterium avium complex bacteremia in patients with AIDS. J Infect Dis 1994; 170:157.

186.  Dautzenberg, B, Castellani, P, Pellegrin, J-L, et al. Early bactericidal activity of rifabutin versus that of placebo in treatment of disseminated Mycobacterium avium complex bacteremia in AIDS patients. Antimicrob Agents Chemother 1996; 40:1722.

187.  Sullam, PM, Gordin, FM, Wynne, BA. Efficacy of rifabutin in the treatment of disseminated infection due to Mycobacterium avium complex. Clin Infect Dis 1994; 19:84.

188.  Chaisson, RE, Keiser, P, Pierce, M, et al. Controlled trial of clarithromycin/ethambutol with or without clofazimine for Mycobacterium avium complex bacteremia in AIDS. The Third International Conference on the Macrolides, Azalides and Streptogramins, Lisbon, Portugal, 1996. Abstract 7.16.

189.  Baron, EJ, Young, LS. Amikacin, ethambutol, and rifampin for treatment of disseminated Mycobacterium avium-intracellulare infections in patients with acquired immune deficiency syndrome. Diagn Microbiol Infect Dis 1986; 5:215.

190.  Centers for Disease Control. Impact of HIV protease inhibitors on the treatment of HlV-infected tuberculosis patients with rifampin. MMWR 1996; 45:921.

191.  Pierce, M, Crampton, S, Henry, D, et al. A randomized trial of clarithromycin as prophylaxis against disseminated Mycobacterium avium complex infection in patients with advanced acquired immunodeficiency syndrome. N Engl J Med 1996; 335:384.

192.  Benson, CA, Cohn, DL, Williams, P. A Phase 111 prospective, randomized, double-blind study of the safety and efficacy of clarithromycin vs. rifabutin vs. clarithromycin and rifabutin for prevention of Mycobacterium avium complex disease in HIV positive patients with CD4 cells greater than or less than 100 cells/mL. Program and abstracts of the 3rd Conference on Retroviruses and Opportunistic Infections, Washington, DC. Infectious Diseases Society of America, 1996. Abstract 205.

193.  Havlir, DV, Dube, MP, Sattler, FR, et al. Prophylaxis against disseminated Mycobacterium avium complex with weekly azithromycin, daily rifabutin, or both. N Engl J Med 1996; 335:392.

194.  Masur, H. Recommendations on prophylaxis and therapy for disseminated Mycobacterium avium complex disease in patients infected with the human immunodeficiency virus. N Engl J Med 1993; 329: 898.

195.  Centers for Disease Control. U.S. Public Health Service/Infectious Disease Society of America. Guidelines for the Prevention of Opportunistic Infections in Persons Infected with Human Immunodeficiency Virus: A Summary. MMWR 1995; 44 RR-8:1.

196.  Wallace, RJ Jr, Nash, DR, Tsukamura, M, et al. Human disease due to Mycobacterium smegmatis. J Infect Dis 1988; 158:52.

197.  Band, JD, Ward, JI, Fraser, DW, et al. Peritonitis due to a Mycobacterium chelonei-like organism associated with intermittent chronic peritoneal dialysis. J Infect Dis 1982; 145:9.

198.  Weitzman, I, Osadczyi, D, Corrado, ML, Karp, D. Mycobacterium thermoresistibile: A new pathogen for humans. J Clin Microbiol 1981; 14:593.

199.  Jadeja, L, Bolivar, R, Wallace, RJ Jr, et al. Bacteremia caused by a previously unidentified species for rapidly growing Mycobacterium successfully treated with vancomycin. Ann Intern Med 1983; 99:475.

200.  Davidson, MB, McCormack, JG, Blacklock, ZM, et al. Bacteremia caused by Mycobacterium neoaurum. J Clin Microbiol 1988; 26:762.

201.  Wallace, RJ Jr, Swenson, JM, Silcox, VA, Bullen, MG. Treatment of non-pulmonary infections due to Mycobacterium fortuitum and Mycobacterium chelonei on the basis of in vitro susceptibilities. J Infect Dis 1985; 152:500.

202.  Dalovisio, JR, Pankey, GA, Wallace, RJ Jr, Jones, DB. Clinical usefulness of amikacin and doxycycline in the treatment of infection due to Mycobacterium fortuitum and Mycobacterium chelonei. Rev Infect Dis 1981; 3:1068.

203.  Wallace, RJ Jr, Tanner, D, Brennan, PJ, Brown, BA. Clinical trial of clarithromycin for cutaneous (disseminated) infection due to Mycobacterium chelonae. Ann Intern Med 1993; 119:482.

204.  Wallace, RJ Jr. The clinical presentation, diagnosis, and therapy of cutaneous and pulmonary infections due to the rapidly growing mycobacteria M. fortuitum and M. chelonae. Clin Chest Med 1989; 10:419.

205.  Griffith, DE, Girard, WM, Wallace, RJ Jr. Clinical features of pulmonary disease caused by rapidly growing mycobacteria: An analysis of 154 patients. Am Rev Respir Dis 1993; 1271.

206.  Kim, R. Tetracycline therapy for atypical mycobacteria! granuloma. Arch Dermatol 1974; 110:299.

207.  Edelstein, H. Mycobacterium marinum skin infections. Arch Intern Med 1994; 154:1359.

208.  Loria, PR. Minocycline hydrochloride treatment for atypical acid-fast infection. Arch Dermatol 1976; 112:517.

209.  Black, MM, Eykyn, S. The successful treatment of tropical fish tank granuloma (Mycobacterium marinum) infections with cotrimoxazole. Br J Dermatol 1977; 97:689.

210.  Chow, SP, Ip, FK, Lau, JHK, et al. Mycobacterium marinum infection of the hand and wrist. J Bone Joint Surg Am 1987; 69-A:1161.

211.  Donta, ST, Smith, PW, Levitz, RE, Quintiliani, LR. Therapy of Mycobacterium marinum infections. Arch Intern Med 1986; 146:902.

212.  Banks, J, Jenkins, PA, Smith, AP. Pulmonary infection with Mycobacterium malmoense: A review of treatment and response. Tubercle 1985; 66:197.

213.  Bell, RC, Higuchi, JH, Donovan, WN, , et al. Mycobacterium simiae: clinical features and follow- up of twenty-four patients. Am Rev Respir Dis 1983; 127:35.

214.  Banks, J, Jenkins, PA. Combined versus single antituberculosis drugs on the in vitro sensitivity patterns of non-tuberculous mycobacteria. Thorax 1987; 42:838.

215.  Yates, MD, Grange, JM, Collins, CH. The nature of mycobacterial disease in southeast England, 1977-84. J Epidemiol Community Health 1986; 40:295.

216.  American Thoracic Society. Treatment of tuberculosis and tuberculosis infection in adults and children. Am Rev Respir Dis 1986; 134:355.

217.  Lester, W. Treatment of drug-resistant tuberculosis. Dis Mon 1971; April:3.

[김태영]

기모야... 이게 무슨 내용인지... 읽어볼 엄두가 안나는데...

[하필상]

태영아! 이글을 카피해서 이재일에게 보내서 자문을 받아라.