Dr. Krishnan has received consulting fees, speaking fees, and/or honoraria from URL Pharma, Metabolex, and Ardea Biosciences (less than $10,000 each) and Takeda (more than $10,000).
Chronic Kidney Disease and the Risk of Incident Gout Among Middle-Aged Men: A Seven-Year Prospective Observational Study
Article first published online: 27 NOV 2013
Copyright © 2013 by the American College of Rheumatology
Arthritis & Rheumatism
Volume 65, Issue 12, pages 3271–3278, December 2013
How to Cite
Krishnan, E. (2013), Chronic Kidney Disease and the Risk of Incident Gout Among Middle-Aged Men: A Seven-Year Prospective Observational Study. Arthritis & Rheumatism, 65: 3271–3278. doi: 10.1002/art.38171
ClinicalTrials.gov identifier: NCT 00000487.
This study was performed using a limited-access data set obtained from the National Heart, Lung, and Blood Institute (NHLBI) and does not necessarily reflect the opinions or views of the Multiple Risk Factor Intervention Trial (MRFIT) or the NHLBI.
- Issue published online: 27 NOV 2013
- Article first published online: 27 NOV 2013
- Accepted manuscript online: 27 AUG 2013 03:33PM EST
- Manuscript Accepted: 20 AUG 2013
- Manuscript Received: 29 SEP 2012
- National Heart, Lung, and Blood Institute in collaboration with the MRFIT investigators
The kidney is the major organ that facilitates excretion of urate in humans. Surprisingly, few studies have assessed whether a reduced glomerular filtration rate (GFR) and/or kidney damage is associated with a higher incidence of gout, and this study was undertaken to address this question.
Data from a 7-year followup of patients enrolled in the Multiple Risk Factor Intervention Trial, a primary prevention trial for cardiovascular disease among 12,866 men ages 35–57 years, were used for the present investigation. Presence of gout was determined by the study physicians from the original trial. Chronic kidney disease was defined using criteria similar to those proposed by the National Kidney Foundation. The Cox proportional hazards regression model was used to assess the association between gout and chronic kidney disease, after accounting for the effects of potential confounders.
Overall, there were 722 cases of physician- diagnosed incident gout over 76,602 person-years of followup. The standardized incidence ratio of gout among those with chronic kidney disease was 1,217 (95% confidence interval [95% CI] 1,191–1,244). The adjusted hazard ratio (HR) among those with chronic kidney disease was 1.61 (95% CI 1.60–1.61). Each standard deviation decline in the estimated GFR was associated with an HR of 1.43 (95% CI 1.35–1.51). Including the serum urate level, as well as the urate–chronic kidney disease interaction term, as variables in the second analysis did not attenuate the HR. Proteinuria and hematuria, two markers of kidney damage, were associated with an elevated risk of gout independent of the estimated GFR.
Chronic kidney disease manifesting as reduced glomerular function or as presence of blood or protein in the urine increases the risk of incident gout.
The kidneys are a major route of excretion of urate, a byproduct of purine catabolism (), and the role of urate in the kidney disease–gout link is likely to be complex. In the 19th century, Alfred Garrod wrote that “Gout … depends on a loss of power in the uric acid excretion function of the kidneys”—a view that is still popular today ([2, 3]). However, the experimental and epidemiologic data supporting this paradigm are scant. For instance, one study did not demonstrate differences in renal urate clearance between those with and those without gout (). A decrease in the glomerular filtration rate (GFR) results in an adaptive increase in fractional excretion of urate and urate excretion per nephron ([5, 6]). In severe chronic kidney disease (inulin clearance <15 ml/minute), glomerular filtration becomes the rate-limiting step, and up to 45% of the filtered urate is excreted (). Additionally, there is an increase in extrarenal clearance, although this mechanism may be overwhelmed in severe chronic kidney disease (). Complicating the picture, data from several prospective studies show that hyperuricemia is associated with a higher risk of incident renal disease (). More than 27 million Americans have chronic kidney disease, and this number is rising ([9, 10]). Thus, establishing whether chronic kidney disease is an independent risk factor for gout has public health implications.
In 2009–2010, an estimated 7.5 million people in the US were reported to have gout. Additionally, 1.25 million men and 0.78 million women were reported to have moderate or severe renal impairment as well as gout. The age-standardized prevalence of gout was 2.9% among those with a normal GFR compared to 24% among those with a GFR of <60 ml/minute/1.73 m2 (). Older epidemiologic studies have demonstrated that gout is an infrequent consequence of chronic kidney disease, with an incidence rate of ∼1–5% ([12-14]). More recently, 2 prospective studies and many cross-sectional studies have shown that chronic kidney disease is an independent risk factor for gout ([15-18]) and that the prevalence of moderate or severe chronic kidney disease among those with gout is 7–10 times higher than in the general population ([19-21]). However, these studies were limited by their cross-sectional design, narrow patient population (), and imprecise case definitions for chronic kidney disease ([15-17]). Further, none of these studies examined whether hyperuricemia related to renal dysfunction played a role in the observed association.
The aim of the present epidemiologic study was to address whether chronic kidney disease is associated with the risk of incident gout. Kidney disease manifests as a reduction in glomerular filtration and/or presence of excessive protein and blood in the urine. Accordingly, this study tested the hypotheses that a reduced GFR and/or proteinuria and hematuria are associated with an increased risk of incident gout.
PATIENTS AND METHODS
Data were available through the limited access program of the National Heart, Lung, and Blood Institute. Multiple Risk Factor Intervention Trial (MRFIT) participants provided informed consent for data collection and analysis. The present analyses were approved by the Stanford Institutional Review Board.
Participant enrollment and followup
From 1973 to 1975, a total of 356,222 men ages 35–57 years who were at high risk of coronary artery disease and who did not have a history of hospitalization for myocardial infarction were screened in 22 clinical centers in 18 cities in the US, through the MRFIT study. Exclusion criteria were as follows: serum creatinine level >2.0 mg/dl, diastolic blood pressure ≥115 mm Hg, and serum cholesterol level >9.05 mmoles/liter (350 mg/dl). Other exclusion criteria included current treatment for diabetes, previous hospitalization for ≥2 weeks for a heart attack, treatment with guanethidine, hydralazine, or lipid-lowering drugs, illness or disability likely to impair full participation in the trial, a medically prescribed diet that was incompatible with the MRFIT food program, such as a gluten-free diet, and intention to leave the clinic's geographic area in the near future. Patients receiving guanethidine and hydralazine who were identified prior to randomization were excluded, since the study intervention included these drugs and because there was no washout period. Of the 361,662 men invited to be screened for the study, 1,293 (0.4%) were excluded for dietary reasons.
Participants were randomly assigned to a special intervention group (n = 6,428) or a usual care group (n = 6,438). Although the study was designed to encompass 6 annual visits per patient, some participants who entered the study early in the enrollment phase were followed up for 7 visits. At each of these visits, participants were examined by study physicians, completed dietary and health questionnaires, and provided fasting blood and urine samples. Vital status was assessed throughout the trial for an average of 7 years. Additional information about the MRFIT design has been published elsewhere ([22-24]).
At the baseline and followup visits, participants completed questionnaires on medical history, alcohol consumption, and 24-hour food frequency. Total daily consumption of fructose and alcohol in grams was estimated based on responses to the questionnaires. The medical history questionnaire included a self-reported diagnosis of gouty arthritis, defined as an affirmative answer to the following question: “During the past 12 months has a doctor told you that you have gout?” No case validation for this definition is available through the MRFIT study, although it was found to have good reliability in other epidemiologic settings ([25-27]).
Information on the frequency of treatment with aspirin, diuretics, other blood pressure medications, and gout medications was collected at all visits. Therapy with allopurinol, probenecid, or colchicine was recorded under the “anti-gout medicines” category. Blood pressure medications were separated into 2 categories—diuretics and nondiuretics. The diuretics included triamterene, spironolactone, hydrochlorothiazide, and chlorthalidone. The nondiuretics included hydralazine, reserpine, guanethidine, methyldopa, and propranolol. Calcium-channel blockers and other modern antihypertensive therapies were not available/not used by subjects evaluated in the MRFIT. Information on individual gout medications taken and the doses prescribed was not available.
Tests to measure serum creatinine, lipids, and urate were performed in a central laboratory using autoanalyzers ([28, 29]). Since serum creatinine was measured over many years and the method of measurement was not standardized, we reduced the recorded serum creatinine values by 5% for purposes of analysis. Urine protein, blood, and hematocrit levels, white blood cell count, and pH were measured locally by urine dipstick (Ames Labstix) (). The GFR was estimated as the number of milliliters per minute per 1.73 m2, using equations from the Chronic Kidney Disease Epidemiology Collaboration ().
The study physician reviewed medical and medication histories with the participants. Physical examination included anthropometry, blood pressure measurement, and a clinical evaluation. The physician assessed all participants for gout and classified participants into 1 of 3 categories: gout present, no evidence of gout, and suspicion of gout. The MRFIT protocol did not specify any standard criteria for this clinical assessment, and the determination was left to the judgment of the individual study physician. Patients seldom presented with acute swollen joints for which aspiration and verification of presence of urate crystals would have been medically appropriate.
Key case definitions
A physician diagnosis of definite gout was considered the primary case definition for gout. Participants in whom gout was suspected were not included in this definition. Three alternate definitions were used for sensitivity analyses: self-reported gout, self-reported gout and physician diagnosis, and self-reported gout, physician diagnosis, and either a serum urate level of >7.0 mg/dl or treatment with any anti-gout medication.
With respect to renal function, terminology and case definitions in this study were similar to those that have been proposed by the National Kidney Foundation. The estimated GFR was classified as normal, mildly impaired, moderately to severely impaired, or severely impaired as per the following cutoff values: ≥90 ml/minute/1.73 m2, 60–89 ml/minute/1.73 m2, 30–59 ml/minute/1.73 m2, and <30 ml/minute/1.73 m2, respectively (). Due to the small sample size, the last 2 strata were combined for most analyses. For this study, a dipstick proteinuria measurement of 1+ or higher (corresponding to proteinuria ≥30 mg/dl) or “moderate or large” hematuria was deemed as signifying kidney damage (). Chronic kidney disease was defined as either the presence of kidney damage or an estimated GFR of <60 ml/minute/1.73 m2 regardless of the kidney damage status ().
Hyperuricemia was defined as a serum urate level of >7.0 mg/dl (>417 μmoles/dl), consistent with prior analyses ([32-35]). Systolic blood pressure of ≥135 mm Hg or diastolic blood pressure of ≥85 mm Hg or treatment with any antihypertensive medication was used to identify hypertension (). Diabetes was defined as a fasting glucose level of ≥126 mg/dl or treatment with any antidiabetic medications (). Metabolic syndrome was defined according to the US National Cholesterol Education Program Adult Treatment Panel III criteria ().
Initially, data from both the special intervention group and the usual care group were analyzed separately; data were then pooled, since there were no systematic differences. Survival model methods were used in the analysis (). For participants who never had kidney disease, collection of data started with the baseline visit and continued until new-onset gout was diagnosed, the patient died or was lost to followup, or the study ended. For participants with kidney disease, collection of data corresponded to the onset of chronic kidney disease and was discontinued as described above.
Incidence rates and ratios
Incidence rates were calculated as the number of new cases of gout per 1,000 person-years of followup. Standardized incidence ratios (SIRs) were computed using the age-specific incidence data for the year 1977–1978 from the Rochester Epidemiology Project program, a contemporaneous incidence study that used the American College of Rheumatology 1977 criteria for defining gout ([40-42]).
The primary dependent variable of interest was the physician diagnosis of gout. Sensitivity analyses were performed by repeating analyses using alternate case definitions. The main independent variable of interest, kidney disease, was analyzed in multiple formats: 1) based on the estimated GFR with continuous and categorical variables, 2) based on the National Kidney Foundation criteria for kidney damage with a binary variable, and 3) based on the National Kidney Foundation criteria for chronic kidney disease with a binary variable.
Cox proportional hazards regression models were used to calculate hazard ratios (HRs) for kidney disease status based on the risk of incident gout before and after adjustment for covariates (). In these models, all values of covariates were updated at every visit (i.e., all variables were treated as time-varying covariates).
The first Cox proportional hazards regression model assessed the impact of each of the 5 individual metrics of renal function after accounting for continuous measures (age, body mass index, diastolic blood pressure, alcohol and fructose consumption, and fasting serum triglyceride levels), as well as categorical measures (ethnicity [African American or not], use of aspirin, baseline or incident diabetes, and use of diuretics). Systolic blood pressure was not used as a variable, since it was not significantly associated with gout. The second Cox model included serum urate level as a continuous variable, in addition to all of the covariates that were used in the first Cox model. Clustering within the 2 randomized arms of the study was accounted for by using appropriate methods for calculating 95% confidence intervals (95% CIs): the jackknife method for rates and the Huber-White sandwich method for HRs ().
The baseline characteristics of the study population are shown in Table 1. Participants with worse renal function had a higher prevalence of known risk factors for gout, such as elevated blood pressure, obesity, abnormal blood glucose levels, and hyperlipidemia, as well as higher levels of alcohol consumption, older age, and use of diuretics. This was consistent in both study arms.
|Serum creatinine 0.5–0.99 mg/dl (n = 1,706)||Serum creatinine 1.00–1.09 mg/dl (n = 3,373)||Serum creatinine 1.10–1.19 mg/dl (n = 3,567)||Serum creatinine 1.20–2.00 mg/dl (n = 4,220)|
|Estimated GFR, ml/minute/1.73 m2||103 ± 5||90 ± 4||80 ± 3||68 ± 6|
|Age, years||45.8 ± 5.91||46.0 ± 5.96||45.9 ± 5.99||46.7 ± 5.93|
|African American, %||4.92||5.60||5.52||10.9|
|Current smoker, %||77.2||70.6||63.5||52.9|
|Body mass index, kg/m2||27.3 ± 3.74||27.5 ± 3.52||27.7 ± 3.39||28.0 ± 3.35|
|Daily fructose consumption, gm||19.7 ± 17.4||19.8 ± 17.1||20.1 ± 17.1||20.4 ± 17.4|
|Daily alcohol consumption, gm||27.1 ± 38.5||25.5 ± 35.4||25.6 ± 36.7||23.7 ± 35.6|
|Daily calories from alcohol, %||7.86||7.31||7.26||6.88|
|Systolic blood pressure, mm Hg||138 ± 15.5||138 ± 15.3||138 ± 14.8||138 ± 15.4|
|Diastolic blood pressure, mm Hg||92.1 ± 9.68||92.2 ± 9.50||93.0 ± 9.47||93.5 ± 9.48|
|Hypertensive at baseline||57.6 ± 49.4||59.2 ± 49.2||61.6 ± 48.6||67.2 ± 47.0|
|Daily aspirin use, %||4.5||4.7||3.8||4.6|
|Serum cholesterol, mg/dl||251 ± 37.7||252 ± 36.9||254 ± 36.7||256 ± 35.7|
|LDL cholesterol, mg/dl||158 ± 37.2||159 ± 35.6||161 ± 36.0||161 ± 35.8|
|HDL cholesterol, mg/dl||43.3 ± 13.6||42.3 ± 11.7||42.0 ± 11.7||41.5 ± 11.1|
|Plasma triglyceride, mg/dl||187 ± 150||194 ± 158||195 ± 141||197 ± 135|
|Hematocrit, %||46.2 ± 3.8||46.4 ± 3.8||46.6 ± 3.7||46.5 ± 3.9|
|Serum creatinine, mg/dl||0.88 ± 0.05||1.00 ± 0.00||1.10 ± 0.00||1.27 ± 0.10|
|Serum urate, mg/dl||6.37 ± 1.28||6.59 ± 1.27||6.75 ± 1.24||7.15 ± 1.36|
|Metabolic syndrome, %||17||17||18||22|
|Antihypertensive medications, %||17||17||18||24|
|Diuretic use, %||14||15||16||21|
Incidence and prevalence of gout
There were 352 incident cases of gout in the usual care group and 370 cases in the special intervention group. Overall, the incidence of gout was 9.4 per 1,000 person-years (95% CI 9.0–9.9). The rates were similar between the usual care group and the special intervention group (9.2 per 1,000 person-years [95% CI 8.3–10.2] and 9.7 per 1,000 person-years [95% CI 8.7–10.7], respectively).
As glomerular function decreased, the cumulative incidence of gout increased (Figure 1). Patients with severe renal dysfunction had a ∼15-fold higher incidence rate of gout compared to patients in the normal category. The SIRs were similarly higher with worsening renal function (Table 2). Kidney damage, as evinced by proteinuria and/or hematuria, was associated with higher incidence rates of gout (9.2 per 1,000 person-years [95% CI 8.76–9.66] among those without kidney damage and 13.38 [95% CI 13.11–13.64] among those with kidney damage).
|Overall||Estimated GFR by category, ml/minute/1.73 m2|
|Observation time, person-years||76,602||31,600||42,905||2.097|
|Incident gout, no.||722||212||453||57|
|Incidence rate, per 1,000 person-years||9.4 (9.0–9.9)||6.7 (6.2–7.3)||10.6 (10.5–10.6)||27.2 (23.1–32.4)|
|SIR (95% CI)||745 (714–778)||581 (540–623)||792 (788–796)||1,759 (1,479–2,122)|
|Unadjusted HR (95% CI)||–||1.00 (referent)||1.5 (1.5–1.6)||3.9 (3.4–4.5)|
|Age-adjusted HR (95% CI)||–||1.00 (referent)||1.5 (1.5–1.6)||3.8 (3.3–4.5)|
|HR (95% CI), first multivariable Cox model||–||1.00 (referent)||1.4 (1.3–1.6)||3.3 (2.9–3.9)|
|HR (95% CI), second multivariable Cox model||–||1.00 (referent)||1.4 (1.2–1.5)||2.8 (1.3–6.0)|
Cox regression models
Estimated GFR and gout
When the estimated GFR was assessed as a continuous variable in Cox regressions, each standard deviation (14 ml/minute/1.73 m2) decrease in estimated GFR was associated with an unadjusted HR of 1.43 (95% CI 1.35–1.51) for gout incidence. The age-adjusted HR was nearly identical (1.44 [95% CI 1.34–1.54]). In the first multivariable Cox model, each standard deviation decrease in estimated GFR was associated with an HR of 1.38 (95% CI 1.33–1.44).
Table 2 shows the results of Cox models with HRs calculated based on severity strata of renal function. There was a steep increase in the HR as renal dysfunction worsened; when data from the patients who were categorized as having severe renal disease (estimated GFR <30 ml/minute/1.73 m2, with 1 incident case of gout over 15 person-years of observation) were analyzed separately, that disease category was associated with an HR of 13.2 (95% CI 3.5–49.1) in the first multivariable Cox model and 9.8 (95% CI 8.1–11.8) in the second Cox model.
Urinary evidence of kidney damage and the risk of gout
To study the impact of different parameters of kidney damage on the risk of gout, separate regression analyses were repeated, using hematuria and proteinuria as the covariate of interest, after adjustment for nonrenal covariates (as in the first Cox model) (Table 2). Table 3 shows the results of these regressions. In the multivariable model, kidney damage was associated with an HR of 1.24 (95% CI 1.18–1.31).
|Unadjusted for glomerular function||Adjusted for glomerular function|
|HR||95% CI||HR||95% CI|
|None||1.00 (referent)||–||1.00 (referent)||–|
|1+ (30 mg/dl)||1.11||0.86–1.45||1.10||0.86–1.41|
|2+ (100 mg/dl)||0.93||0.53–1.63||0.88||0.50–1.56|
|≥3+ (≥300 mg/dl)||3.46||2.01–5.98||2.89||1.73–4.83|
|None||1.00 (referent)||–||1.00 (referent)||–|
|Moderate or large||2.41||2.29–2.54||2.18||2.08–2.29|
|Absent||1.00 (referent)||–||1.00 (referent)||–|
Chronic kidney disease and the risk of gout
Table 4 shows the incidence of gout by chronic kidney disease status. The unadjusted HR was 1.87 (95% CI 1.86–1.88). Adjustment for covariates did not significantly change the HR. Sensitivity analyses confirmed the robustness of these findings (Table 5).
|Incident gout among patients with chronic kidney disease, no.||97|
|Incidence rate (95% CI)||16.7 (16.4–17.1)|
|SIR (95% CI)||1,217 (1,191–1,244)|
|Unadjusted HR (95% CI)||1.87 (1.86–1.88)|
|Age-adjusted HR (95% CI), chronic kidney disease vs. no chronic kidney disease||1.83 (1.83–1.83)|
|HR (95% CI), first multivariable Cox model of chronic kidney disease vs. no chronic kidney disease||1.61 (1.60–1.61)|
|HR (95% CI), second multivariable Cox model of chronic kidney disease vs. no chronic kidney disease||1.51 (1.23–1.85)|
|Case definition of gout||Incident cases of gout, no.||Overall incidence rate per 1,000 person-years (95% CI)||SIR (95% CI)||Multivariable-adjusted HR (95% CI), chronic kidney disease vs. no chronic kidney disease|
|Self-reported||1,476||20.2 (19.8–20.6)||2,016 (1,773–2,350)||1.4 (1.3–1.5)|
|Self-reported and physician diagnosis||600||7.7 (7.3–8.1)||1,073 (1,050–1,091)||1.8 (1.6–1.9)|
|Self-reported, physician diagnosis, and either use of gout medication or serum urate level >7.0 mg/dl||529||6.8 (6.1–7.5)||9,776 (8,436–11,350)||1.74 (1.69–1.80)|
The present study is the first to show that objective measures of chronic kidney disease are risk factors for incident gout independent of hyperuricemia. Overall, those with an estimated GFR of <60 ml/minute/1.73 m2 had a 2-fold higher incidence of gout than those with a higher estimated GFR. Presence of kidney damage, as evinced by proteinuria, increased the risk of gout independent of serum urate levels and estimated GFR.
The role of serum urate in the association between chronic kidney disease and gout is intriguing. Urate is freely filtered in the glomerulus, and tubular secretion is a rate-limiting step for urate excretion (). When the GFR decreases, corresponding declines in the urate content of the filtrate are matched by an adaptive increase in fractional excretion of urate and urate excretion per nephron ([5, 6]). When the inulin clearance drops to <15 ml/minute, renal and nonrenal secretory function is overloaded, and serum urate levels increase rapidly (). Consistent with these prior observations, in the present study there were large increases in the incidence of gout among patients with severe chronic kidney disease with an estimated GFR <30 ml/minute/1.73 m2 relative to those patients who had a higher estimated GFR. This is consistent with the 2009–2010 National Health and Nutrition Examination Survey data that showed a 6-fold increase in the prevalence of gout and a 20-fold increase in the prevalence of hyperuricemia among those with an estimated GFR of <30 ml/minute/1.73 m2 (). Although residual confounding by other risk factors may partially explain the link between chronic kidney disease and gout, other renal factors, as well as shared genetic factors, merit further study ().
Despite the pathophysiologic links, relatively few studies have examined the prevalence of gout among patients with renal impairment. Two large studies from France in the 1960s and 1980s showed that the prevalence of gout among those with renal impairment was no more than 1% ([12, 13]). In 1975, a large US study demonstrated that of 1,700 patients with gout, only 84 (4.9%) had primary renal disease that preceded the onset of gout. The investigators concluded that primary kidney disease was seldom a cause of gout (). More recently, data from a cohort of 18,358 diabetes patients in New Zealand suggested that declines in renal function were associated with an increased risk of gout (). Cross-sectional analyses of data from general practice registers in the UK suggested that receiving a diagnosis of chronic renal failure was associated with a 2.5-fold higher likelihood of concomitant diagnosis of gout ([15, 17, 46]). Due to underdiagnosis of chronic kidney disease in general clinical practice, some of these studies may have underestimated the true magnitude of the risk of gout ([20, 47]).
These findings must be interpreted in light of the limitations inherent in the study design and the data collected. No women were included in this study; the magnitude of association between chronic kidney disease and gout among women may be quite different from that among men. The serum creatinine assays that were used in the MRFIT were not calibrated to current standards; the resulting systematic under/overestimate of the estimated GFR may affect generalizability (although, not internal validity). The case definition of chronic kidney disease was primarily based on measures of renal glomerular function and not on measures of tubular function. Assessment of gout was not blinded and hence study physicians might have been influenced by the perceived association between gout and chronic kidney disease. Such a systematic bias was not evident in our sensitivity analyses of the case definition, although random error of misclassification might have increased the statistical variance. Based on a small number of cases of gout, the risk of gout was 9–13 times higher among those with an estimated GFR of <30 ml/minute/1.73 m2 compared to those with a normal estimated GFR. This was based on a relatively brief observation time (15 person-years) in the patients with an estimated GFR of <30 ml/minute/1.73 m2, and confidence intervals were wide. These findings, along with the exclusion of patients with a creatinine level of >2.0 mg/dl from the study, indicate that the magnitude of our risk estimate should be used as preliminary data for more definitive future studies.
After obesity and hypertension, chronic kidney disease, which affects ∼17% of the US population, is the most common risk factor for gout (). In patients with chronic kidney disease, gout is harder to treat because of absolute and relative contraindications to the medications, as well as a lack of drug efficacy. There is an unmet need for safe and effective treatment of the disease in patients with severe chronic kidney disease ().
Dr. Krishnan drafted the article, revised it critically for important intellectual content, approved the final version to be published, and takes responsibility for the integrity of the data and the accuracy of the data analysis.
- 2The kidney in gout and hyperuricemia.Mount Kisco (NY):Futura Publishing Co;1982., .
- 9Centers for Disease Control and Prevention.Prevalence of chronic kidney disease and associated risk factors—United States, 1999-2004.MMWR Morb Mortal Wkly Rep2007;56:161–5.
- 13Goutte secondaire a l'insuffisiance renale Travaus de Congres International de la Goutte et de la Lithiase Urique.Paris:Evian;1964. p.244–55..
- 31K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification.Am J Kidney Dis2002;39 Suppl 1:S1–266.
- 39An introduction to survival analysis using stata.College Station (TX):Stata Press;2002., , .
- 40Statistical methods in cancer research.Lyon:IARC Scientific Publications;1980., .
- 43An introduction to stata for health researchers.College Station (TX):Stata Press;2006..