Re: 통풍환자, 만성신장질환(신부전) 관계 2015년 메타분석!!

[문형철]

Arthritis Res Ther

. 2015 Apr 1;17(1):90. doi: 10.1186/s13075-015-0610-9

Gout and risk of chronic kidney disease and nephrolithiasis: meta-analysis of observational studies

Matthew J Roughley 1, John Belcher 2, Christian D Mallen 3, Edward Roddy 3,✉

• Author information
• Article notes
• Copyright and License information

PMCID: PMC4404569  PMID: 25889144

Abstract

Introduction

To determine the preval‎ence of chronic kidney disease and nephrolithiasis in people with gout, and the association between gout and preval‎ent or incident chronic kidney disease and nephrolithiasis.

Methods

Systematic review and meta-analysis of epidemiological studies. Data sources; MEDLINE, EMBASE and CINAHL databases, hand-searched reference lists, citation history and contact with authors. Eligibility criteria: cohort, case–control or cross-sectional studies which examined the occurrence of chronic kidney disease or nephrolithiasis amongst adults with gout (with or without a non-gout comparator group) in primary care or general population samples. Preval‎ence and risk estimate meta-analyses were performed using a random-effects model.

Results

Seventeen studies were included in the meta-analysis (chronic kidney disease n = 7, nephrolithiasis n = 8, both n = 2). Pooled preval‎ence estimates of chronic kidney disease stage ≥3 and self-reported lifetime nephrolithiasis in people with gout were 24% (95% confidence interval 19% to 28%) and 14% (95% CI 12% to 17%) respectively. Gout was associated with both chronic kidney disease (pooled adjusted odds ratio 2.41, 95% confidence interval 1.86 to 3.11) and self-reported lifetime nephrolithiasis (1.77, 1.43 to 2.19).

Conclusions

Chronic kidney disease and nephrolithiasis are commonly found amongst patients with gout. Gout is independently associated with both chronic kidney disease and nephrolithiasis. Patients with gout should be actively screened for chronic kidney disease and its consequences.

서론

통풍 환자의 만성 신장 질환 및 신장 결석 유병률을 확인하고,

통풍과 만성 신장 질환 및 신장 결석의 유병률 또는 발생률 간의 연관성을 규명하기 위함.

방법

역학 연구에 대한 체계적 문헌고찰 및 메타분석.

데이터 출처: MEDLINE, EMBASE 및 CINAHL 데이터베이스, 수동 검색된 참고문헌 목록, 인용 이력 및 저자 연락. 적격 기준: 일차 진료 또는 일반 인구 표본에서 통풍 환자(통풍이 없는 대조군 포함 또는 제외)의 만성 신장 질환 또는 신장 결석 발생을 조사한 코호트, 사례-대조 또는 횡단면 연구. 유병률 및 위험도 추정치 메타분석은 랜덤 효과 모델을 사용하여 수행.

결과

메타분석에 포함된 연구는

총 17건이었다(만성 신장병 n=7, 신장결석증 n=8, 양측 모두 n=2).

통풍 환자의 만성 신장병 3기 이상 및 평생 신장결석증 유병률 통합 추정치는

각각 24%(95% 신뢰구간 19%~28%) ,

평생 신장결석증은 14%(95% 신뢰구간 12%~17%)였다.

통풍은

만성 신장질환(통합 조정 오즈비 2.41, 95% 신뢰구간 1.86~3.11) 및

평생 신장결석증(1.77, 1.43~2.19)과 모두 연관성이 있었다.

결론

만성 신장 질환과 신장 결석증은

통풍 환자들 사이에서 흔히 발견됩니다.

통풍은

만성 신장 질환과 신장 결석증 모두와

독립적으로 연관되어 있습니다.

통풍 환자들은

만성 신장 질환 및 그 결과에 대해 적극적으로 선별 검사를 받아야 합니다.

Introduction

Gout is the most preval‎ent inflammatory arthritis, affecting 2.4% of adults in the UK [1]. Gout is associated with considerable co-morbidity including hypertension, diabetes mellitus, obesity, metabolic syndrome, and vascular disease [2]. Associations between gout and renal disease and nephrolithiasis have long been recognised, yet early studies undertaken in specialist secondary care populations are likely to be unrepresentative of most patients with gout who are managed exclusively in primary care settings [3-6]. To the best of our knowledge, no previous systematic reviews examining this association have been performed.

Chronic kidney disease (CKD) stages 3 to 5 (glomerular filtration rate <60 ml/minute/1.73 m2)

affects 8.5% of the UK population, is more common in females than males (10.6% vs. 5.8%),

and is associated with both all-cause mortality and cardiovascular disease [7,8]. If left untreated, CKD may progress to end-stage renal disease (ESRD) requiring expensive renal replacement therapy (RRT) [9]. CKD is an independent risk factor for the development of gout [10,11] yet there are several reasons to explain why gout may predispose to renal disease, including hyperuricaemia, chronic inflammation, co-morbid hypertension and diabetes mellitus, and use of nonsteroidal anti-inflammatory drugs. The management of gout in patients with renal disease is challenging because nonsteroidal anti-inflammatory drugs and colchicine should be used with caution and lower doses of allopurinol are required [12].

The lifetime preval‎ence of nephrolithiasis is approximately 8.8% [13]. Kidney stones most commonly present as renal colic but can be complicated by infection, renal tract obstruction requiring surgical intervention, and renal failure [14]. Gout is associated with lower urinary pH theoretically predisposing to the formation of both uric acid and other stones [5,15]. Diabetes mellitus, obesity, and hypertension are also known to be independent risk factors for nephrolithiasis [16-18].

The objectives of this systematic review and meta-analysis of epidemiological studies were to quantify the preval‎ence of CKD and nephrolithiasis in gout, and to determine whether gout is associated with preval‎ent or incident CKD and nephrolithiasis in the general population.

서론

통풍은 가장 흔한 염증성 관절염으로,

영국 성인의 2.4%가 앓고 있습니다 [1].

통풍은

고혈압, 당뇨병, 비만, 대사증후군, 혈관 질환 등

상당한 동반 질환과 연관되어 있다[2].

통풍과 신장 질환 및 신장결석증 간의 연관성은

오래전부터 인식되어 왔으나,

전문 2차 진료 환자군을 대상으로 수행된 초기 연구들은

일차 진료 환경에서만 관리되는 대부분의 통풍 환자를 대표하지 못할 가능성이 높다[3-6].

우리의 지식 범위 내에서,

이 연관성을 검토한 이전의 체계적 문헌고찰은 수행된 바 없다.

만성 신장 질환(CKD) 3~5기(사구체 여과율 <60 ml/분/1.73 m2)는

영국 인구의 8.5%에게 영향을 미치며,

남성보다 여성에서 더 흔하다(10.6% vs. 5.8%).

Chronic kidney disease (CKD) stages 3 to 5 (glomerular filtration rate <60 ml/minute/1.73 m2)

affects 8.5% of the UK population, is more common in females than males (10.6% vs. 5.8%),

체표면적 1.73㎡는
성인 남성의 평균 체표면적을 나타내는 수치.

신장은 우리 몸의 노폐물을 걸러내는 중요한 기관인데,
사람마다 신장 크기가 다르기 때문에
체표면적을 기준으로 하여 사구체 여과율을 보정하면 좀 더 정확한 신장 기능을 평가

1. 주요 구조

구심성 소동맥 (Afferent arteriole): 혈액이 네프론으로 들어오는 입구. 산소와 영양분이 풍부한 혈액이 이곳을 통해 들어옵니다 (빨간색 화살표로 표시).
사구체 모세혈관 (Glomerular capillaries): 구심성 소동맥에서 갈라져 나오는 모세혈관 네트워크. 고압 환경에서 혈액 성분이 여과됩니다.
보우만 낭 (Bowman's capsule): 사구체를 둘러싼 주머니 모양 구조. 여과된 액체(원시 여과액, filtrate)가 이곳으로 모입니다 (노란색 부분).
세뇨관 (Tubule): 보우만 낭에서 이어지는 긴 관. 여기서 재흡수와 분비가 일어납니다 (노란색 관).
원심성 소동맥 (Efferent arteriole): 사구체에서 여과된 혈액이 나가는 출구. 여과 후 혈액이 이곳으로 이동합니다 (빨간색).
관주위 모세혈관 (Peritubular capillaries): 원심성 소동맥에서 갈라지는 모세혈관. 세뇨관에서 재흡수된 물질이 혈액으로 다시 흡수되는 장소 (파란색).
신정맥 (Renal vein): 최종적으로 여과와 재흡수 후 남은 혈액이 나가는 길 (파란색 화살표).

2. 네프론의 주요 과정 (다이어그램의 숫자 1~4 참조)
네프론은 혈액을 처리하는 4단계 과정을 거칩니다. 이는 다이어그램 하단의 방정식 **배설(Excretion) = 여과(Filtration) - 재흡수(Reabsorption) + 분비(Secretion)**로 요약됩니다. 이 방정식은 신장이 하루 약 180L의 여과액을 만들지만, 그중 99%를 재흡수하여 1~2L의 소변만 배설한다는 원리를 나타냅니다.

1. 여과 (Filtration):

위치: 사구체 모세혈관과 보우만 낭 사이.
과정: 구심성 소동맥의 고압으로 혈액의 물, 전해질, 포도당, 아미노산, 요소(urea) 등이 여과막을 통과해 원시 여과액으로 만듭니다. 적혈구, 단백질 등 큰 분자는 여과되지 않습니다.
목적: 혈액의 불순물을 분리.


2. 재흡수 (Reabsorption):

위치: 세뇨관의 근위 세뇨관(proximal tubule)과 원위 세뇨관(distal tubule)에서 주로 일어남.
과정: 여과액에서 포도당, 나트륨(Na+), 물, 칼륨(K+) 등 유용한 물질이 관주위 모세혈관으로 다시 흡수됩니다. (세뇨관 벽을 통해 능동/수동 수송).
목적: 체내 수분과 영양분 보존. 호르몬(알도스테론, ADH)에 의해 조절됨.


3. 분비 (Secretion):

위치: 세뇨관, 특히 원위 세뇨관과 집합관(collecting duct).
과정: 혈액에서 추가로 H+ 이온, 칼륨, 약물, 독소 등을 세뇨관으로 분비합니다. (관주위 모세혈관에서 세뇨관으로 이동).
목적: pH 균형 유지와 불필요한 물질 제거.


4. 배설 (Excretion):

위치: 세뇨관 말단에서 집합관으로 이어짐.
과정: 재흡수되지 않고 분비된 물질이 포함된 최종 여과액(소변)이 신우(renal pelvis)로 모여 방광으로 배출됩니다 (하단 화살표: Urinary excretion).
목적: 노폐물(요소, 크레아티닌) 제거.



3. 전체 과정의 흐름

혈액 → 구심성 소동맥 → 사구체 (여과) → 보우만 낭.
원시 여과액 → 세뇨관 (재흡수 + 분비) → 집합관.
여과 후 혈액 → 원심성 소동맥 → 관주위 모세혈관 (재흡수 흡수) → 신정맥.
소변 → 요관 → 방광 → 배설.

이 과정은 신장의 항상성 유지(혈압, pH, 전해질 균형)에 필수적입니다. 다이어그램에서 색상(빨강: 혈액 유입, 파랑: 혈액 유출, 노랑: 여과/세뇨관)은 흐름을 직관적으로 보여줍니다.

전원 사망률 및 심혈관 질환과 연관되어 있습니다[7,8]. 치료하지 않을 경우 CKD는 고가의 신대체요법(RRT)이 필요한 말기 신부전(ESRD)으로 진행될 수 있습니다[9].

CKD는

통풍 발병의 독립적 위험인자입니다[10,11]

그러나

통풍이 신장 질환을 유발할 수 있는 여러 이유가 존재하는데,

고요산혈증, 만성 염증, 동반 고혈압 및 당뇨병, 비스테로이드성 항염증제(NSAID) 사용 등이

포함됩니다.

신장 질환 환자의 통풍 관리는

비스테로이드성 항염증제와 콜히친 사용 시 주의가 필요하며,

알로퓨리놀의 저용량 투여가 요구되기 때문에 [12].

신장 결석증의 평생 유병률은 약 8.8%이다 [13].

신장 결석은

주로 신장 통증으로 나타나지만 감염,

수술적 개입이 필요한 요로 폐쇄, 신부전 등으로 합병될 수 있다 [14].

통풍은

요산 및 기타 결석 형성에 이론적으로 취약하게 만드는

낮은 요 pH와 연관되어 있다 [5,15].

당뇨병, 비만, 고혈압 역시

신장결석증의 독립적 위험인자로 알려져 있다[16-18].

본 역학 연구에 대한 체계적 문헌고찰 및 메타분석의 목적은

통풍 환자의 만성 신장질환(CKD) 및 신장결석증 유병률을 정량화하고,

일반 인구 집단에서 통풍이 기존 또는 신규 발생 CKD 및 신장결석증과 연관성이 있는지 규명하는 것이다.

Methods

Data sources and searches

A systematic literature search was undertaken by two co-investigators (MJR and ER) in the MEDLINE, EMBASE, and CINAHL databases using the National Health Service’s Healthcare Databases Advanced Search from inception date to November 2013 for epidemiological studies examining the association between gout and either CKD or nephrolithiasis. Search terms pertaining to gout were combined with terms describing the outcome (CKD or nephrolithiasis) and observational study design using the Boolean operator “AND”. No unpublished data sources were searched. The references and citation history of all relevant studies were checked for additional data sources.

Eligibility criteria

Eligible studies were required to include adults with gout (with or without a nongout comparator group), describe the occurrence of CKD or nephrolithiasis amongst adults with gout (and nongout comparator group where included), be of an epidemiological design (cross-sectional, cohort, case–control), and have been undertaken in a primary care or general population sample. No geographical or language restrictions were imposed. Conference abstracts were not included.

방법데이터 출처 및 검색

두 명의 공동 연구자(MJR 및 ER)가 MEDLINE, EMBASE, CINAHL 데이터베이스에서 영국 국민건강서비스(NHS)의 의료 데이터베이스 고급 검색 기능을 활용하여 2013년 11월까지의 시점부터 통풍과 CKD 또는 신장결석증 간의 연관성을 조사한 역학 연구를 체계적으로 문헌 검색하였다. 통풍 관련 검색어는 결과(CKD 또는 신장결석증)를 설명하는 용어 및 관찰 연구 설계와 부울 연산자 “AND”를 사용하여 결합되었다. 미발표 자료원은 검색 대상에서 제외되었다. 모든 관련 연구의 참고문헌 및 인용 이력을 추가 자료원 확인을 위해 검토하였다.

적격 기준

적격 연구는 통풍 성인(통풍이 없는 대조군 포함 여부 불문)을 대상으로 해야 하며, 통풍 성인(및 포함된 경우 통풍이 없는 대조군)에서의 CKD 또는 신장결석 발생률을 기술하고, 역학 연구 설계(횡단면, 코호트, 사례-대조)를 채택하며, 1차 진료 또는 일반 인구 표본에서 수행된 연구여야 했다. 지역적 또는 언어적 제한은 적용되지 않았다. 학회 초록은 포함되지 않았다.

Study selection

MJR and ER independently screened the title and abstract of all identified studies. Studies which could not be excluded on title/abstract review were retained for full-text review undertaken by the same reviewers. At both the title/abstract and full-text review stages, disagreement over study eligibility was resolved by consensus discussion between the reviewers. A third reviewer (CDM) was available where disagreement could not be resolved by consensus.

Data extraction and quality assessment

The following data were extracted independently by the same two reviewers from studies meeting the inclusion criteria: date and location of study, study population and size, study type, demographic characteristics (age and gender), method of ascertainment of gout and CKD/nephrolithiasis, preval‎ence and incidence of CKD/nephrolithiasis in gout and comparator populations, unadjusted and adjusted risk estimates (odds ratio, relative risk, hazards ratio) and corresponding 95% confidence interval (CI) or standard error, and details of covariates included in multivariate models. Studies utilising any method of gout ascertainment (for example, crystal identification, classification criteria, clinical diagnosis) were included. Staging of CKD was defined according to the National Kidney Foundation classification (stage 1, glomerular filtration rate ≥90 ml/minute/1.73 m2; stage 2, glomerular filtration rate 60 to 89 ml/minute/1.73 m2; stage 3, glomerular filtration rate 30 to 59 ml/minute/1.73 m2; stage 4, glomerular filtration rate 15 to 29 ml/minute/1.73 m2; stage 5 (ESRD), glomerular filtration rate <15 ml/minute/1.73 m2) [19]. Assessment of methodological quality was performed independently by the same two reviewers using the Newcastle–Ottawa quality appraisal tool with additional items regarding the method of diagnosis of gout, CKD, and nephrolithiasis [20]. Disagreement over data extraction and assessment of methodological quality was resolved by consensus discussion between the reviewers. Authors were contacted to request additional information and data where necessary.

연구 선정

MJR과 ER은 식별된 모든 연구의 제목과 초록을 독립적으로 선별했습니다. 제목/초록 검토로 배제할 수 없는 연구는 동일한 검토자가 수행한 전문 검토 대상으로 유지되었습니다. 제목/초록 검토 단계와 전문 검토 단계 모두에서 연구 적격성에 대한 의견 불일치는 검토자 간 합의 토론을 통해 해결되었습니다. 합의로 해결되지 않는 불일치 사항이 있을 경우 제3의 검토자(CDM)가 참여했습니다.

데이터 추출 및 질 평가

포함 기준을 충족한 연구에서 동일한 두 검토자가 독립적으로 다음 데이터를 추출했습니다: 연구 시기 및 장소, 연구 대상 및 규모, 연구 유형, 인구통계학적 특성(연령 및 성별), 통풍 및 만성 신장 질환/신장 결석증 확인 방법, 통풍 및 대조군 인구에서의 만성 신장 질환/신장 결석증 유병률 및 발생률, 조정되지 않은 및 조정된 위험 추정값(오즈비, 상대 위험도, 위험비) 및 해당 95% 신뢰구간(CI) 또는 표준오차, 다변량 모델에 포함된 공변량 세부사항. 통풍 확인 방법(예: 결정체 확인, 분류 기준, 임상 진단)을 활용한 모든 연구를 포함하였다. CKD 단계는 미국신장재단(National Kidney Foundation) 분류에 따라 정의하였다(1단계: 사구체여과율 ≥90 ml/분/1.73 m2; 2기, 사구체 여과율 60~89 ml/분/1.73 m2; 3기, 사구체 여과율 30~59 ml/분/1.73 m2; 4단계, 사구체 여과율 15~29 ml/분/1.73 m2; 5단계(말기 신부전), 사구체 여과율 <15 ml/분/1.73 m2) [19]. 방법론적 질 평가를 위해 동일한 두 명의 검토자가 뉴캐슬-오타와 질 평가 도구(Newcastle–Ottawa quality appraisal tool)를 사용해 독립적으로 수행하였으며, 통풍, 만성 신장 질환(CKD), 신장 결석증의 진단 방법에 관한 추가 항목을 포함하였다[20]. 데이터 추출 및 방법론적 질 평가에 대한 의견 불일치는 검토자들 간의 합의 토론을 통해 해결하였다. 필요한 경우 저자들에게 추가 정보 및 데이터 제공을 요청하기 위해 연락하였다.

Data analysis

Estimates of point/lifetime preval‎ence of CKD/nephrolithiasis in adults with gout and, where possible, risk estimates of the association between gout and CKD/nephrolithiasis were calculated. These estimates were derived from summary 2 × 2 tables, together with regression estimates arising from adjusted analyses using logistic and Cox regression.

To normalize the distribution, preval‎ence rates were transformed by means of the logit event rate:

L⁢p=L⁢n⁢(p/(1‐p))

with p being the preval‎ence rate, Ln the natural logarithm, and Lp the logit event rate. The sampling variance of each logit event rate, V(Lp), was calculated by means of:

V⁡(L⁢p)=1/(n⁢p)+1/(n⁢(1‐p))

with n being the sample size. Once the statistical analyses were carried out, the results were back transformed to preval‎ence rates to facilitate their interpretation.

Separate analyses were undertaken for CKD and nephrolithiasis. Where sufficient data were present in individual studies but risk estimates were not reported, these were calculated using the available data. Pooled estimates were calculated using a random effects model. The 95% CIs and forest plots were produced for all pooled estimates.

Heterogeneity was assessed visually with forest plots, and quantified numerically using the I2 index [21] and Cochran’s Q test. The I2 test describes the percentage of variation across studies due to heterogeneity rather than chance. Meta-analyses were performed using the metan command within STATA 12.1 (StataCorp, College Station, TX, USA).

결과연구 선정

검색을 통해 2,032건의 잠재적 관련 논문이 확인되었으며, 중복 제거 후 1,419건이 남았다(그림 1). 제목/초록 검토 후 1,333건이 제외되었다. 남은 86건 중 전문 검토 후 57건이 추가로 제외되었고, 29건이 적격 기준을 충족하였다. 17건의 연구가 통합에 적합한 데이터를 포함하였다. 제외 사유는 그림 1에 제시되어 있다.

Results

Study selection

The search identified 2,032 potentially relevant articles: after removal of duplicates, 1,419 remained (Figure 1). Following title/abstract review, 1,333 articles were excluded. Of the 86 articles remaining, 57 were excluded after full-text review and 29 met the eligibility criteria. Seventeen studies contained data suitable for pooling. Reasons for exclusion are shown in Figure 1.

Figure 1.

Open in a new tab

Selection of studies included in the review. CKD, chronic kidney disease.

Study characteristics

Twelve studies examined CKD [22-33], 15 studies examined nephrolithiasis [13,34-47], and two studies examined both CKD and nephrolithiasis [48,49] (Table 1). Nineteen cross-sectional studies [13,22,25,27,29,30,32,34-38,40-42,46-49], one case–control study [26], and nine cohort studies [23,24,28,31,33,39,43-45] were identified. Six of the nine cohort studies included relevant outcome data only at baseline and so were treated as cross-sectional studies [24,31,33,43-45].

Table 1.

Characteristics of included studies

First authorStudy periodLocationSource of data/study populationStudy typeTotal sampleMean age (SD) a% maleGout ascertainmentOutcomes reported (method of ascertainment)

Chen [22]	2006	China	Residents of Changning district of Shanghai	CS	2,554	58.4 (SE 15.3)	36	Self-report	Adjusted ORs for CKD any stage and CKD stage ≥3 (biochemical measurement)
Hsu [23]	1964 to 2000	USA	Kaiser Permanente Northern California linked to US Renal Data System	Cohort	177,570	ESRD 42.3 (10.3)	ESRD 59	Self-reported physician diagnosis	Univariate and multivariate HRs for ESRD requiring RRT (record linkage)
						No ESRD 40.7 (14.0)	No ESRD 46
Kuo [24]	2000 to 2006	Taiwan	Chang Gung Memorial Hospital health screening programme	CS data from cohortb	61,527	Males 49.1 (11.0)	56	Self-reported or ICD code or crystal identification	Preval‎ence of CKD stage ≥3 (biochemical measurement)
						Females 50.8 (10.8)
Fuldeore [25]	2002 to 2005	USA	Managed Healthcare Database	CS	3,929	CKD	CKD	ICD code on two occasions or ICD code plus gout medication	Preval‎ence of CKD stage 2, 3 and 4 (biochemical measurement)
						54.3 (NR)	86
						No CKD	No CKD
						47.8 (9.0)	89
Johnson [26]	1997 to 2004	USA	Kaiser Permanente North West	CC	5,335	64.4 (NR)	46	ICD code or SUA ≥7 mg/dl	Unadjusted and adjusted ORs for ESRD requiring RRT (medical record review)
Keenan [27]	2007 to 2008	USA	New York Veterans Affairs database	CS	575	71.8 (11.6)	99	ICD code	Preval‎ence of CKD stage ≥3 (biochemical measurement or ICD code)
Yu [28]	2000 to 2008	Taiwan	Taiwan National Health Insurance Database	Cohort	656,108	41.1 (15.6)	73	ICD code on two occasions or ICD code plus gout medication	Multivariate HR for ESRD requiring RRT (record linkage)
O’Sullivan [34]	1964	USA	Population of Sudbury, Massachusetts	CS	4,626	NR	NR	New York/Rome criteria	Preval‎ence of nephrolithiasis (self-report, not defined)
Schaffalitzky De Muckadell [35]	1973	Denmark	Male office workers aged 40 to 59 years in Copenhagen	CS	312	NR	100	Self-reported physician diagnosis	Preval‎ence of nephrolithiasis (self-report, not defined)
Currie [36]	1969 to 1975	UK	General practice records	CS	604	52.3 (NR)	77	GP diagnosis	Preval‎ence of nephrolithiasis (GP diagnosis)
Currie [37]	1976	UK	General practice records	CS	64,454	NR	NR	GP diagnosis	Incidence of nephrolithiasis (GP diagnosis)
Kramer [38]	1988 to 1994	USA	National Health and Nutrition Examination Survey (NHANES) III	CS	17,030	Stones 53.7 (0.7)	48	Self-reported physician diagnosis	Preval‎ence and adjusted ORs for nephrolithiasis (self-reported lifetime preval‎ence)
						No stones 44.2 (0.4)
Kramer [39]	1986 to 1998	USA	Health Professionals Follow-up Study	Cohort	51,529	Gout 59.9 (9.1)	100	Self-reported physician diagnosis, ARA criteria	Preval‎ence and adjusted RRs for nephrolithiasis (self-reported lifetime preval‎ence)
						No gout 54.4 (9.8)
Mikuls [40]	1990 to 1999	UK	General Practice Research Database (GPRD)	CS	207,350	Gout 60.5 (15.4)	46	GP diagnosis	Preval‎ence, unadjusted and adjusted ORs for nephrolithiasis (medical records)
						OA controls 66.8 (13.4)
Harrold [41]	1999 to 2003	USA	Health Maintenance Organisations (HMO) Research Network Centre for Education and Research on Therapeutics (CERT)	CS	6,133	Males 58 (14)	81	ICD code on two occasions	Preval‎ence of nephrolithiasis (ICD code)
						Females 70 (12)
Padang [42]	Not stated	Indonesia	Community Health Centres in Northern Sulawesi	CS	380	NR	NR	≥3 attacks of gout/year and/or tophi plus ARA criteria	Preval‎ence of nephrolithiasis (self-report, not defined)
Sarawate [43]	1999 to 2004	USA	Managed Care Database	CS data from cohortb	5,942	57.4 (14.1)	76	ICD code on two occasions or ICD code plus gout medication	Preval‎ence of nephrolithiasis (ICD code)
Solomon [44]	Not stated	USA	US Medicare system and Pharmacy Assistance Contract for the Elderly (PACE)	CS data from cohortb	9,823	Male 78 (7)	16	Use of ULT	Preval‎ence of nephrolithiasis (medical records)
						Female 80 (7)
Harrold [45]	2000 to 2006	USA	Health Maintenance Organisations (HMO) Research Network Centre for Education and Research on Therapeutics (CERT)	CS data from cohortb	4,166	62 (14)	75	ICD code plus use of ULT	Preval‎ence of nephrolithiasis (ICD code)
Roddy [49]	2005	UK	Registered population of two general practices in Nottingham	CS	3,082	Gout 63.8 (10.6)	Gout 81	Self-reported gout, validated by rheumatologist	Preval‎ence of CKD stage ≥3, ≥4 (biochemical measurement) and nephrolithiasis (self-reported lifetime preval‎ence)
						No gout 57.0 (14.4)	No gout 41
Zhu [48]	2007 to 2008	USA	National Health and Nutrition Examination Survey (NHANES)	CS	5,707	47 (NR)	48	Self-reported physician diagnosis	Preval‎ence and adjusted OR for CKD stage ≥2, ≥3c (biochemical measurement) and nephrolithiasis (self-reported lifetime preval‎ence)
Liote [29]	2008 to 2009	France	General Practitioner data from GOSPEL survey	CS	810	62.7 (11.3)	87.2	Physician diagnosis	Preval‎ence of CKD stage ≥3, ≥4 (biochemical measurement)
Lin [30]	2007	Taiwan	Community-based survey, sampling via Household Register Database	CS	3,352	47.5 (17.4)	48.6	Self-reported physician diagnosis of gout or hyperuricaemia	Preval‎ence of CKD stage ≥3 (biochemical measurement)
Kuo [31]	2000 to 2008	Taiwan	Longitudinal Health Insurance Database	CS data from cohortb	704,503	42.73 (16.6)	Gout 70.3, no gout 50.4	ICD code	Preval‎ence and unadjusted OR for ESRD requiring RRT (record linkage)
Krishnan [32]	2009 to 2010	USA	National Health and Nutrition Examination Survey	CS	5,589	44 (21)	49	Self-reported physician diagnosis	Preval‎ence of CKD stage 2, 3, ≥4 (biochemical measurement)
Kok [33]	2003 to 2007	Taiwan	Taiwan National Health Insurance Database	CS data from cohortb	3,858,840	NR	48.3	ICD code on three occasions	Preval‎ence of CKD, stages not specified (ICD code)
Trifiro [46]	2005 to 2009	Italy	Health Search/Cegedim Strategic Data Longitudinal Patient Database	CS	12,276	NR	74	ICD code, related keywords for free text search	Preval‎ence and unadjusted OR for nephrolithiasis (medical records)
Scales [13]	2007 to 2010	USA	National Health and Nutrition Examination Survey	CS	12,110	NR	NR	NR	Adjusted OR for nephrolithiasis (self-reported lifetime preval‎ence)
Ando [47]	1995 to 2001	Japan	Men undergoing medical examination at Gifu Prefectural Center for Health Check and Health Promotion	CS	13,418	Controls 48.5 (8.8), past stones 49.7 (8.4)	100	Receiving medical treatment for gout or SUA ≥7.0 mg/dl	Preval‎ence nephrolithiasis (self-reported lifetime preval‎ence)

Open in a new tab

ARA, American Rheumatism Association; CC, case–control study; CKD, chronic kidney disease; CS, cross-sectional; ESRD, end-stage renal disease; GP, general practitioner; HR, hazard ratio; ICD, International Classification of Disease; NR, not reported; OA, osteoarthritis; OR, odds ratio; RR, relative risk; RRT, renal replacement therapy; SD, standard deviation; SE, standard error; SUA, serum uric acid level; ULT, urate-lowering therapy. aAssumed to be SD unless otherwise stated. bCohort study reporting only cross-sectional data relevant to this review. cCKD stages clarified with authors; CSD, Cegedim Strategic Data.

Fourteen studies were performed in the USA [13,23,25-27,32,34,38,39,41,43-45,48], four in the UK [36,37,40,49], five in Taiwan [24,28,30,31,33], and one study each in China [22], Denmark [35], France [29], Indonesia [42], Italy [46], and Japan [47]. One study restricted participation to people aged over 65 years [44] and three studies included only males [35,39,47]. Fourteen studies were undertaken in large healthcare databases [23-28,31,33,40,41,43-46] whereas 15 studies used empirically collected data [13,22,29,30,32,34-39,42,47-49].

Assessment of methodological quality

Only three of 29 studies based the diagnosis of gout on validated clinical classification criteria: two studies [39,42] used the American Rheumatism Association criteria [50], and one study [34] used the New York and Rome criteria [51,52]. No studies based the diagnosis of gout on crystal identification. Three studies reported combined outcomes for gout and hyperuricaemia [26,30,47]. Nine studies used biochemical measures to define CKD [22,24,25,27,29,30,32,48,49], four studies examined ESRD requiring RRT [23,26,28,31], whereas nine of 17 studies of nephrolithiasis defined outcome by self-report [13,34,35,38,39,42,47-49], six studies used medical record linkage [40,41,43-46], and two studies used a general practitioner completed questionnaire [36,37].

Participants were considered not representative of the typical community gout patient in eight studies due to including only patients prescribed urate-lowering therapy [25,44,45], only males [35,39,47], or only people with chronic gout [42] and one study being undertaken in a Veterans database which resulted in a predominantly male (99%) older population (mean age 72 years) [27]. All studies featuring a nongout comparator group had drawn this from the same community as the exposed cohort. Response and data usage rates were frequently not reported [13,26,38,41-45,48], Two surveys reported response rates under 60% [30,49] and two database studies based in a managed care setting used less than 70% of records available to them [24,31].

Of the three cohort studies, none reported rates of attrition and one did not state whether the outcome (ESRD) was absent at the start of the study [23]. Length of follow-up was deemed adequate to determine outcome in all three studies.

Selection for meta-analysis

Seventeen out of 29 studies were included in the meta-analysis [13,22-24,28-30,32,34,35,38,39,42,46-49]. Two studies examining CKD were deemed unsuitable for pooling due to use of diagnostic codes [27,33] rather than biochemical testing to ascertain CKD that has been shown to significantly underestimate preval‎ence [53,54]. One study demonstrated significant incompleteness of biochemical data and was not pooled [25]. Ten out of 17 studies examining self-reported nephrolithiasis were pooled [13,34,35,38,39,42,46-49], and six studies were deemed too methodologically different to pool due to using a general practitioner completed questionnaire [36] or diagnostic codes and/or a limited time period for ascertaining nephrolithiasis [40,41,43-45]. Three studies reported outcomes that were not reported by other studies, thus precluding meta-analysis [26,31,37].

Preval‎ence and risk of chronic kidney disease in gout

Six studies provided suitable data to allow the pooled preval‎ence of CKD stage ≥3 in people with gout to be calculated [24,29,30,32,48,49]. Further data [29,30] and outcome clarification [48] was requested from the authors of three of these studies, enabling their inclusion. No reply was received from the authors of a further paper from whom the preval‎ence of CKD in those with and without gout was requested. The pooled preval‎ence estimate of CKD stage ≥3 in people with gout was 24% (95% CI 19%, 28%) (Figure 2). Statistically significant heterogeneity was identified (I2 = 84.3%, P <0.001). Data from three studies [29,32,49] were pooled providing a preval‎ence estimate of CKD stage ≥4 in gout of 2% (95% CI 0%, 4%) (I2 = 82.5%, P = 0.003).

통풍 환자의 만성 신장 질환 유병률 및 위험도

통풍 환자에서 만성 신장 질환(CKD) 3기 이상

유병률의 통합 계산이 가능한 적절한 데이터를 제공한 연구는 여섯 건이었다 [24,29,30,32,48,49].

이 중 세 연구의 저자에게 추가 데이터 [29,30] 및 결과 명확화 [48]를 요청하여

해당 연구를 포함시킬 수 있었다.

통풍 유무에 따른 CKD 유병률 데이터를 요청한 추가 논문 저자로부터는 답변을 받지 못했다. 통풍 환자의 CKD 3기 이상 통합 유병률 추정치는 24%(95% CI 19%, 28%)였다 (그림 2). 통계적으로 유의한 이질성이 확인되었다(I2 = 84.3%, P < 0.001). 세 연구[29,32,49]의 데이터를 통합하여 통풍 환자의 만성 신장병 4기 이상 유병률 추정치는 2%(95% 신뢰구간 0%, 4%)로 나타났다(I2 = 82.5%, P = 0.003).

Figure 2.

Open in a new tab

Forest plot of individual and pooled (random effects) preval‎ences of chronic kidney disease stage ≥3 in gout. CI, confidence interval.

Comparing patients with gout to those without gout, the pooled unadjusted odds ratio (OR) between gout and CKD stage ≥3 was 4.32 (95% CI 3.82, 4.89) (Figure 3) based on the findings of three studies [24,30,48]. Pooling an age-adjusted and gender-adjusted OR and two multivariate ORs (adjustment included age, gender, obesity, hypertension, and diabetes mellitus) between gout and CKD stage ≥3 resulted in an adjusted OR of 2.41 (95% CI 1.86, 3.11) (Figure 3) [22,30,48]. There was no significant heterogeneity identified for either pooled unadjusted or adjusted ORs (unadjusted I2 = 0.0%, P = 0.803; adjusted I2 = 0.0%, P = 0.416).

Figure 3.

Open in a new tab

Forest plot of individual and pooled (random effects) unadjusted and adjusted odds ratios between gout and chronic kidney disease stage ≥3. CI, confidence interval.

One study reported preval‎ence of ESRD requiring RRT in gout (0.6%) and provided an unadjusted OR for ESRD in gout versus no gout (OR 2.65; 95% CI 2.25, 3.12) [31]. One retrospective matched case–control study compared the likelihood of a prior combined exposure of gout or hyperuricaemia between cases with ESRD requiring RRT and controls without ESRD matched for age, gender, and year of RRT (adjusted OR 2.51; 95% CI 1.78, 3.54) [26].

Following contact with the authors of one study to provide an unadjusted hazard ratio (HR) (95% CI) between gout and ESRD [28], there were two prospective cohort studies [23,28] that provided suitable data to allow a pooled unadjusted HR between gout and incident ESRD to be calculated (HR 2.69; 95% CI 1.00, 4.39). Mean follow-up periods were 24.5 years in one study [23] and 8.0 years overall in the other [28]. Significant heterogeneity was identified (I2 93.1%, P <0.001). We contacted the authors of one of these studies to request an adjusted HR (95% CI) between gout and incident ESRD [23]. The authors replied but were unable to provide this information. Only one study provided a multivariate adjusted HR between gout and incident ESRD (HR 1.57; 95% CI 1.38 to 1.79; adjusted for age, sex, diabetes and hypertension) [28].

통풍에서 만성 신장병 3기 이상 유병률의 개별 및 통합(무작위 효과) 포레스트 플롯. CI, 신뢰 구간.

통풍 환자와 비통풍 환자를 비교했을 때, 세 연구[24,30,48] 결과를 바탕으로 통풍과 만성 신장병 3기 이상 사이의 통합 미조정 오즈비(OR)는 4.32(95% CI 3.82, 4.89)였다(그림 3). 연령 조정 및 성별 조정 OR과 두 개의 다변량 OR(조정 변수: 연령, 성별, 비만, 고혈압, 당뇨병)을 통합한 결과, 조정된 OR은 2.41(95% CI 1.86, 3.11)로 나타났다(그림 3) [22,30,48]. 통합된 미조정 또는 조정된 OR 모두에서 유의한 이질성은 확인되지 않았다(미조정 I2 = 0.0%, P = 0.803; 조정 I² = 0.0%, P = 0.416).

그림 3.

새 탭에서 열기

통풍과 만성 신장병 3기 이상 간의 개별 및 통합(랜덤 효과) 미조정 및 조정 오즈비에 대한 포레스트 플롯. CI, 신뢰 구간.

한 연구는 통풍 환자에서 RRT가 필요한 말기 신부전(ESRD)의 유병률(0.6%)을 보고했으며, 통풍 유무에 따른 ESRD 발생에 대한 미조정 OR을 제시하였다(OR 2.65; 95% CI 2.25, 3.12) [31]. 한 후향적 매칭 사례-대조 연구는 RRT가 필요한 ESRD 환자군과 연령, 성별, RRT 연도를 매칭한 비-ESRD 대조군 간에 통풍 또는 고요산혈증의 과거 복합 노출 가능성을 비교하였다(조정된 OR 2.51; 95% CI 1.78, 3.54) [26].

통풍과 ESRD 간 미조정 위험비(HR)(95% CI)를 제공하기 위해 한 연구의 저자와 연락한 결과[28], 통풍과 신규 발생 ESRD 간 통합 미조정 HR을 계산할 수 있는 적절한 데이터를 제공한 두 개의 전향적 코호트 연구[23,28]가 있었습니다(HR 2.69; 95% CI 1.00, 4.39). 평균 추적 관찰 기간은 한 연구[23]에서 24.5년, 다른 연구[28]에서 전체적으로 8.0년이었다. 상당한 이질성이 확인되었다(I2 93.1%, P <0.001). 이 중 한 연구의 저자에게 통풍과 신부전 발생 간 조정된 위험비(95% CI)를 요청하기 위해 연락했습니다[23]. 저자들은 답변했으나 해당 정보를 제공할 수 없었습니다. 단 한 연구만이 통풍과 신부전 발생 간 다변량 조정 위험비(HR 1.57; 95% CI 1.38~1.79; 연령, 성별, 당뇨병 및 고혈압 조정)를 제공했습니다[28].

Preval‎ence and risk of nephrolithiasis in gout

We were able to use additional unpublished data from one of our previous studies [49] concerning the preval‎ence of self-reported lifetime nephrolithiasis in people with gout and ORs (95% CI) between gout and self-reported lifetime nephrolithiasis, unadjusted and then adjusted, firstly, for age and gender and, secondly, for age, gender, diabetes, hypertension, and diuretic use. The pooled estimate of the preval‎ence of nephrolithiasis in people with gout was 14% (95% CI 12%, 17%) (Figure 4) based on the findings of eight studies [34,35,38,39,42,47-49]. Significant statistical heterogeneity was identified (I2 = 79.4%, P <0.001).

통풍 환자의 신장결석 유병률 및 위험도

우리는 통풍 환자의 평생 신장결석 유병률에 관한 이전 연구[49]의 미발표 추가 데이터를 활용할 수 있었으며, 통풍과 평생 신장결석 사이의 오즈비(95% 신뢰구간)를 미조정 및 조정된 형태로 확인할 수 있었습니다. 조정 변수는 첫째, 연령 및 성별을 조정하고, 두 번째로 연령, 성별, 당뇨병, 고혈압 및 이뇨제 사용을 조정하여 계산한 오즈비(OR)를 확인할 수 있었다. 통풍 환자의 신장결석 유병률에 대한 통합 추정치는 8개 연구[34,35,38,39,42,47-49]의 결과를 바탕으로 14%(95% CI 12%, 17%)였다(그림 4). 통계적으로 유의한 이질성이 확인되었다(I2 = 79.4%, P <0.001).

오즈비가 1보다 크다는 것은
흡연자가 비흡연자보다 폐암에 걸릴 확률이 높다는 것을 의미.

예를 들어, 오즈비가 2라면
흡연자가 비흡연자보다 폐암에 걸릴 확률이 2배 높다고 해석.

Figure 4.

Open in a new tab

Forest plot of individual and pooled (random effects) preval‎ences of nephrolithiasis in gout. CI, confidence interval.

Eight studies [34,35,38,39,42,46,48,49] provided suitable data to calculate a pooled unadjusted OR between gout and self-reported preval‎ence of nephrolithiasis (OR 2.66; 95% CI 2.16, 3.27) (Figure 5). The pooled age-adjusted and gender-adjusted OR was 1.87 (95% CI 1.25, 2.80), based on the findings of four studies [42,47-49]. Three studies provided suitable data to permit a pooled multivariate adjusted OR to be calculated (OR 1.77; 95% CI 1.43, 2.19) [13,38,49]. All three studies adjusted for age, gender, and body mass index; two studies additionally adjusted for hypertension [38,49] and the other study for diabetes mellitus [13]. Significant heterogeneity was identified for the pooled unadjusted and the age-adjusted and gender-adjusted ORs (unadjusted I2 = 69.1%, P = 0.002; age-adjusted and gender-adjusted I2 = 70.2%, P = 0.018) but not for the multivariate pooled estimate (I2 = 0.0%, P = 0.510). One prospective study found men with gout to have a relative risk of 2.12 (95% CI 1.22, 3.68) of developing incident nephrolithiasis over a 12-year period compared with men without gout, after adjustment for age, body mass index, diuretic use, hypertension, and dietary factors [39].

통풍과 자가 보고 신장결석증 유병률 간 통합 미조정 오즈비(OR) 계산에 적합한 데이터를 제공한 8개 연구[34,35,38,39,42,46,48,49]를 기반으로 한 통합 오즈비(OR 2.66; 95% CI 2.16, 3.27) (그림 5). 4건의 연구[42,47-49] 결과를 바탕으로 한 통합 연령 조정 및 성별 조정 OR은 1.87(95% CI 1.25, 2.80)이었다. 3건의 연구에서 다변량 조정 오즈비(OR 1.77; 95% CI 1.43, 2.19)를 계산할 수 있는 적절한 데이터를 제공하였다[13,38,49]. 세 연구 모두 연령, 성별, 체질량 지수를 조정하였으며, 두 연구는 고혈압[38,49]을 추가로 조정했고, 다른 한 연구는 당뇨병[13]을 추가로 조정하였다. 통합 미조정 및 연령·성별 조정 오즈비에서는 유의한 이질성이 확인되었으나(미조정 I2 = 69.1%, P = 0.002; 연령·성별 조정 I2 = 70.2%, P = 0.018), 다변량 통합 추정치에서는 이질성이 관찰되지 않았다(I2 = 0.0%, P = 0.510) 한 전향적 연구에서는 통풍이 있는 남성이 통풍이 없는 남성에 비해 12년 동안 신장결석 발생의 상대 위험도가 2.12(95% CI 1.22, 3.68)로 나타났으며, 이는 연령, 체질량지수, 이뇨제 사용, 고혈압 및 식이 요인을 조정한 후의 결과이다[39].

Figure 5.

Open in a new tab

Forest plot of individual and pooled (random effects) unadjusted and adjusted odds ratios between gout and nephrolithiasis. CI, confidence interval.

Assessment of publication bias

When pooling the six studies describing the preval‎ences of CKD stage ≥3, Egger’s test yielded a borderline result of P = 0.05. No evidence of small study effects was seen for nephrolithiasis (Egger’s test P = 0.548).

Discussion

In this systematic review and meta-analysis of population-based epidemiological studies, renal disease and nephrolithiasis were common findings in people with gout. The pooled preval‎ence estimates of CKD stage ≥3 and self-reported nephrolithiasis in people with gout were 24% and 14% respectively. Furthermore, after adjustment for confounding variables, people with gout were more than twice as likely to have CKD stage ≥3 (OR 2.41) and over one and a half times (OR 1.77) as likely to have ever had nephrolithiasis as people who did not have gout. Most studies employed a cross-sectional design and there was a paucity of prospective studies. The pooled unadjusted HR for incident ESRD in people with gout was 2.69; however, the lower limit of the 95% CI bordered unity. Only one study provided an adjusted HR which suggested that gout is an independent risk factor for incident ESRD (HR 1.57). Similarly, the single prospective study of the risk of developing nephrolithiasis found people with gout to have twice the risk (adjusted relative risk 2.12) of people without gout.

To our knowledge, this is the first meta-analysis of associations between gout and CKD/nephrolithiasis. Strengths of our methodology include the comprehensive search strategy and literature review process undertaken by two independent reviewers, translation of non-English-language publications, contacting study authors to request additional data to maximise the number of included studies, and an outcome which required biochemically defined CKD rather than more heterogeneous and less precise clinical diagnostic labels.

As with any systematic review, our findings are dependent upon the size and quality of the published literature. There was only a small number of studies that had examined either CKD or nephrolithiasis. Studies of CKD were all published from 2007 onwards, which is not surprising as the National Kidney Foundation CKD classification was only introduced in 2002 [19]. Widening the outcome definition to include clinical diagnostic labels for kidney disease would have increased the number of available studies but would have increased heterogeneity. Furthermore, CKD classification is widely used and highly relevant to clinical practice. The small number of published studies also raises the possibility of publication bias. Although not evident for nephrolithiasis, there were too few studies of CKD to examine for publication bias. There were several methodological limitations of the included studies worthy of further discussion. The majority of included studies were not designed with the primary aim of assessing the association between gout and CKD/nephrolithiasis. Most studies based the diagnosis of gout on physician diagnosis (which has been shown to be reasonably accurate [49]) or patient self-report rather than more rigorous methods such as the American Rheumatism Association criteria [50] or crystal identification. This limitation risks misclassification bias but, as highlighted by this study, the current evidence base offers limited numbers of epidemiological studies using more robust methods. Several studies did not include control groups without gout and only reported the preval‎ence of CKD/nephrolithiasis in gout but not the strength of association between the two. Although our findings demonstrate a clear independent association between gout and both CKD and nephrolithiasis, there were few prospective studies so we could not draw firm conclusions about temporal aspects of these associations. Previous epidemiological studies have shown that chronic renal disease is an independent risk factor for gout [10,11], yet there are several plausible mechanisms by which gout might predispose to CKD. Renal damage can result from co-morbid hypertension and diabetes, hyperuricaemia-mediated endothelial dysfunction and renovascular disease [55], and use of nonsteroidal anti-inflammatory drugs. Although allopurinol is widely believed to have a deleterious effect on renal function based on early observations [56], a recent systematic review suggested that it may protect against progression of CKD [57]. Inflammation in gout is increasingly recognised to persist in the inter-critical period between acute attacks [58,59], raising the possibility that inflammatory mechanisms contribute to vascular risk, as has been proposed for other inflammatory arthropathies [60]. Statistical heterogeneity was demonstrated in the meta-analyses for CKD and nephrolithiasis preval‎ence, the unadjusted HR between gout and incident ESRD, and the unadjusted and age-adjusted and gender-adjusted ORs between gout and nephrolithiasis. Possible sources include differences between the demographic and co-morbid characteristics of the populations studied and the different follow-up periods in the prospective studies. It is noteworthy, however, that the pooled multivariate estimates adjusting for such demographic and co-morbid factors did not demonstrate significant heterogeneity.

토론

이 체계적 문헌고찰 및 메타분석에서,

인구 기반 역학 연구를 통해 통풍 환자에게서

신장 질환과 신장 결석이 흔히 발견됨을 확인하였다.

통풍 환자의 CKD 3기 이상 및 자가 보고 신장 결석의 통합 유병률 추정치는

각각 24%와 14%였다.

또한,

혼란 변수를 조정한 후에도 통풍 환자는

통풍이 없는 사람에 비해 CKD 3기 이상 발생 위험이 2배 이상 높았으며(오즈비 2.41),

신장결석증 이력 발생 위험도 1.5배 이상 높았다(오즈비 1.77).

대부분의 연구는 횡단면 설계를 사용했으며 전향적 연구는 부족했다.

통풍 환자의 신규 말기 신부전(ESRD) 발생에 대한

통합 미조정 위험비(HR)는 2.69였으나,

95% 신뢰구간 하한은 1에 근접했다.

단 한 연구만이 조정된 위험비를 제공했으며,

이는 통풍이 신규 ESRD 발생의 독립적 위험인자임을 시사했다 (위험비 1.57).

마찬가지로,

신장결석 발생 위험에 대한 단일 전향적 연구에서도

통풍 환자의 위험도가 통풍이 없는 사람에 비해

두 배 높다는 결과가 나왔다(조정 상대 위험도 2.12).

우리의 지식 범위 내에서, 이는 통풍과 만성 신장 질환(CKD)/신장 결석증 간의 연관성을 분석한 최초의 메타분석이다. 본 연구 방법론의 강점으로는 두 명의 독립적인 검토자가 수행한 포괄적인 검색 전략 및 문헌 검토 과정, 비영어권 출판물의 번역, 포함된 연구 수를 극대화하기 위한 추가 데이터 요청을 위한 연구 저자 연락, 그리고 더 이질적이고 덜 정확한 임상적 진단 라벨보다는 생화학적 정의에 따른 CKD를 요구한 결과 평가 등이 포함된다.

모든 체계적 문헌고찰과 마찬가지로, 본 연구 결과는 발표된 문헌의 규모와 질에 의존한다. CKD 또는 신장결석을 조사한 연구는 극소수에 불과했다. CKD 연구는 모두 2007년 이후 발표되었는데, 이는 미국신장재단(National Kidney Foundation)의 CKD 분류 체계가 2002년에야 도입되었기 때문에 놀랍지 않다 [19]. 결과 정의 범위를 신장 질환의 임상적 진단 라벨까지 확대했다면 이용 가능한 연구 수는 증가했을 것이나 이질성도 함께 증가했을 것이다. 게다가 CKD 분류는 임상 실무에서 널리 사용되며 매우 관련성이 높다. 발표된 연구 수가 적다는 점은 출판 편향 가능성도 제기한다. 신장결석증의 경우 뚜렷하지 않았으나, CKD 연구는 출판 편향을 검토하기에는 너무 적었다. 포함된 연구들에는 추가 논의가 필요한 몇 가지 방법론적 한계가 있었다. 포함된 연구 대부분은 통풍과 만성 신장 질환/신장 결석증 간의 연관성 평가를 주요 목적으로 설계되지 않았다. 대부분의 연구는 통풍 진단을 의사 진단(상당히 정확함이 입증됨 [49]) 또는 환자 자가 보고에 의존했으며, 미국 류마티스 학회 기준[50]이나 결정체 확인과 같은 보다 엄격한 방법은 사용하지 않았다. 이러한 한계는 오분류 편향을 초래할 위험이 있으나, 본 연구에서 강조된 바와 같이 현재의 증거 기반은 보다 견고한 방법을 사용한 역학 연구가 제한적이다. 여러 연구는 통풍이 없는 대조군을 포함하지 않았으며, 통풍에서의 CKD/신장결석증 유병률만 보고했을 뿐 양자 간 연관성의 강도는 보고하지 않았다. 본 연구 결과는 통풍과 CKD 및 신장결석증 모두 사이에 명확한 독립적 연관성이 있음을 보여주지만, 전향적 연구가 거의 없어 이러한 연관성의 시간적 측면에 대해 확고한 결론을 내릴 수 없었다. 기존 역학 연구들은 만성 신장질환이 통풍의 독립적 위험인자임을 보여주었으나 [10,11]을 입증했으나, 통풍이 CKD를 유발할 수 있는 몇 가지 타당한 기전이 존재한다. 신장 손상은 동반 고혈압 및 당뇨병, 고요산혈증에 의한 내피 기능 장애 및 신혈관 질환[55], 그리고 비스테로이드성 항염증제 사용으로 인해 발생할 수 있다. 알로퓨리놀은 초기 관찰 결과에 근거하여 신기능에 해로운 영향을 미친다고 널리 알려져 있지만[56], 최근의 체계적 문헌고찰에서는 CKD 진행을 보호할 수 있다는 제안이 제기되었다[57]. 통풍의 염증은 급성 발작 사이의 간발기 동안에도 지속된다는 점이 점차 인식되고 있으며[58,59], 이는 다른 염증성 관절병증에서 제안된 바와 같이 염증 기전이 혈관 위험에 기여할 가능성을 제기한다 [60]. CKD 및 신장결석 유병률에 대한 메타분석, 통풍과 신부전 발생 간 미조정 위험비(HR), 통풍과 신장결석 간 미조정 및 연령/성별 조정 오즈비(OR)에서 통계적 이질성이 확인되었다. 가능한 원인으로는 연구 대상 집단의 인구통계학적 및 동반질환 특성 차이, 전향적 연구 간 추적 관찰 기간 차이 등이 있다. 그러나 이러한 인구통계학적 및 동반 질환 요인을 조정하여 통합한 다변량 추정값에서는 유의한 이질성이 관찰되지 않았다는 점은 주목할 만하다.

Conclusions

The main clinical implications of our findings are that patients with gout should be screened for CKD and that clinicians should be made aware of the associations between gout and CKD/nephrolithiasis. Unless sought for, CKD usually progresses subclinically until reaching more advanced stages. In view of this, the significant morbidity and mortality associated with CKD [8] and the risk of CKD associated with gout, a presentation with gout in primary care should be viewed as a red flag for CKD and should prompt screening for and treatment of both CKD and its associated risk factors such as hypertension and diabetes mellitus, which are also risk factors for gout [10]. American College of Rheumatology guidelines advise considering CKD and nephrolithiasis as part of the management of gout [61] but current national and international guidelines regarding CKD [62] and nephrolithiasis [63] do not recognise gout as a risk factor for these conditions. Only one in five people presenting to primary care with acute gout are screened for CKD within a month of presentation [64].

In summary, gout is associated with both CKD and nephrolithiasis. However, there were insufficient prospective studies to determine the temporal nature of these associations or to determine the mechanisms underlying them. Whilst the associations seen suggest that clinicians should be made aware of them and that patients with gout should be screened for CKD, further studies are required to investigate these relationships further.

결론

본 연구 결과의 주요 임상적 함의는

통풍 환자에게 CKD 검진을 시행해야 하며,

의료진이 통풍과 CKD/신장결석증 간의 연관성을 인지해야 한다는 점이다.

의도적으로 검진하지 않는 한,

CKD는 일반적으로 더 진행된 단계에 도달할 때까지

무증상적으로 진행된다.

이러한 점과 CKD와 관련된 상당한 이환율 및 사망률[8],

그리고 통풍과 관련된 CKD 위험을 고려할 때,

일차 진료에서 통풍 증상이 나타나는 경우 CKD에 대한 경고 신호로 간주해야 하며,

CKD와 고혈압 및 당뇨병과 같은 관련 위험 인자

(이들 역시 통풍의 위험 인자임[10])에 대한 검진 및 치료를 즉시 시행해야 한다.

미국 류마티스학회 지침은 통풍 관리의 일환으로 CKD 및 신장결석증을 고려할 것을 권고하지만[61],

현재 CKD[62] 및 신장결석증[63]에 관한 국내외 지침은 통풍을 이들 질환의 위험 요인으로 인정하지 않습니다.

일차 진료에서

급성 통풍으로 내원한 환자 중 CKD 검사를 받는 비율은

5명 중 1명에 불과합니다[64].

요약하면, 통풍은 만성 신장 질환 및 신장 결석증과 연관되어 있다. 그러나 이러한 연관성의 시간적 특성이나 그 기전을 규명하기에는 전향적 연구가 불충분했다. 관찰된 연관성은 임상의들이 이를 인지하고 통풍 환자에게 만성 신장 질환 검진을 실시해야 함을 시사하지만, 이러한 관계를 더 깊이 조사하기 위해서는 추가 연구가 필요하다.

Acknowledgements

The authors would like to thank the authors of included studies who responded to queries enabling their studies to be included in this meta-analysis: Dr Chang-Fu Kuo, Professor Nan Chen, Dr Ming-Yen Lin, and Professor Frederic Liote for provision of additional data from their studies; Professor Hyon Choi, Professor Ryosuke Ando, and Dr Victor Kok for clarifying methodological aspects of their studies; and Professor Chi-Yuan Hsu for replying to our query and confirm‎ing that additional data were not available. The authors would also like to thank Professor Danielle van der Windt, Dr Nadia Corp, and Dr Olalekan Uthman for methodological advice and Professor Danielle van der Windt, Professor Victoria Barskova, Dr Kate Dunn, Simona Tchakarova, Dr Maryia Horava, and Kanchan Vohora for assistance with translation of non-English-language articles.

MJR received a bursary from the Jean Shanks Foundation to fund his intercalated MPhil. CDM is funded by the National Institute for Health Research (NIHR) Collaborations for Leadership in Applied Health Research and Care West Midlands, the NIHR School for Primary Care Research, and a NIHR Research Professorship in General Practice. This article presents independent research funded by the NIHR. The views expressed are those of the author(s) and not necessarily those of the National Health Service, the NIHR, or the Department of Health. The study funders had no role in the study design; the data collection, analysis, or interpretation; the writing of the paper; or the decision to submit the paper for publication.

AbbreviationsCI

confidence interval

CKD

chronic kidney disease

ESRD

end-stage renal disease

HR

hazard ratio

OR

odds ratio

RRT

renal replacement therapy

Footnotes

Competing interests

All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that: the authors do not have any support from any company for the submitted work; the authors do not have any relationships with any companies that might have an interest in the submitted work in the previous 3 years; their spouses, partners, or children do not have any financial relationships that may be relevant to the submitted work; and the authors do not have any non-financial interests that may be relevant to the submitted work.

Authors’ contributions

ER and CDM conceived the study. Analysis was undertaken by MJR and JB. All authors were involved in the design, interpretation of data, and drafting, revising and final approval of the manuscript. ER is guarantor and affirms that the manuscript is an honest, accurate and transparent account of the study being reported, and that no important aspects of the study have been omitted. There are no discrepancies from the study as planned. All authors had full access to all of the data (including statistical reports and tables) in the study and can take responsibility for the integrity of the data and the accuracy of the data analysis.

Contributor Information

Matthew J Roughley, Email: mattjroughley@gmail.com.

John Belcher, Email: j.belcher@keele.ac.uk.

Christian D Mallen, Email: c.d.mallen@keele.ac.uk.

Edward Roddy, Email: e.roddy@keele.ac.uk.

References

• 1.Kuo C, Grainge MJ, Mallen C, Zhang W, Doherty M. Rising burden of gout in the UK but continuing suboptimal management: a nationwide population study. Ann Rheum Dis. 2014;0:1–7. doi: 10.1136/annrheumdis-2013-204463. [DOI] [PMC free article] [PubMed] [Google Scholar]