Comparison of Rates of Lower Extremity Amputation in Patients With and Without Gout in the US Department of Veterans Affairs Health System | Orthopedics | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Risk of Undergoing Lower Extremity Amputation (LEA) in Patients With Gout vs Controls Without Gout
Risk of Undergoing Lower Extremity Amputation (LEA) in Patients With Gout vs Controls Without Gout

Adjusted hazard ratios (aHRs) and 95% CIs were generated using multivariable Cox proportional hazards regression models. Estimates provided from separate models examining any LEA (overall) and by LEA type. All models adjusted for age, sex, race, ethnicity, body mass index, smoking history, and comorbidities (hypertension, cardiovascular disease, peripheral artery disease, cancer, cerebrovascular disease, chronic lung disease, dementia, diabetes, and kidney disease).

Figure 2.  Cumulative Incidence of Undergoing Any Lower Extremity Amputation (LEA)
Cumulative Incidence of Undergoing Any Lower Extremity Amputation (LEA)

Cumulative incidence estimates using Cox proportional hazards regression models and based on the presence of gout and the dual presence of gout and/or diabetes. The overall Cox proportional hazards regression model was adjusted for age, sex, race and ethnicity, body mass index, smoking history, and comorbidities (hypertension, cardiovascular disease, peripheral artery disease, cancer, cerebrovascular disease, chronic lung disease, dementia, diabetes, and kidney disease). The model of dual associations of gout and/or diabetes was adjusted for the same covariates with exception of diabetes. Shaded areas indicate 95% CIs.

Table 1.  Characteristics of US Veteran Patients With and Without Gout at Index Datea
Characteristics of US Veteran Patients With and Without Gout at Index Datea
Table 2.  Crude Incidence Rates of Lower Extremity Amputation in Patients With and Without Gouta
Crude Incidence Rates of Lower Extremity Amputation in Patients With and Without Gouta
Table 3.  Associations of Serum Urate Control and Urate-Lowering Therapy Administration With LEA in Patients With Gouta
Associations of Serum Urate Control and Urate-Lowering Therapy Administration With LEA in Patients With Gouta
1.
Chen-Xu  M, Yokose  C, Rai  SK, Pillinger  MH, Choi  HK.  Contemporary prevalence of gout and hyperuricemia in the United States and decadal trends: the National Health and Nutrition Examination Survey, 2007-2016.   Arthritis Rheumatol. 2019;71(6):991-999. doi:10.1002/art.40807PubMedGoogle ScholarCrossref
2.
Garg  R, Sayles  HR, Yu  F,  et al.  Gout-related health care utilization in US emergency departments, 2006 through 2008.   Arthritis Care Res (Hoboken). 2013;65(4):571-577. doi:10.1002/acr.21837PubMedGoogle ScholarCrossref
3.
Singh  JA, Yu  S.  Time trends, predictors, and outcome of emergency department use for gout: a nationwide US study.   J Rheumatol. 2016;43(8):1581-1588. doi:10.3899/jrheum.151419PubMedGoogle ScholarCrossref
4.
Wertheimer  A, Morlock  R, Becker  MA.  A revised estimate of the burden of illness of gout.   Curr Ther Res Clin Exp. 2013;75:1-4. doi:10.1016/j.curtheres.2013.04.003PubMedGoogle ScholarCrossref
5.
FitzGerald  JD, Dalbeth  N, Mikuls  T,  et al.  2020 American College of Rheumatology Guideline for the Management of Gout.   Arthritis Rheumatol. 2020;72(6):879-895. doi:10.1002/art.41247PubMedGoogle ScholarCrossref
6.
Richette  P, Doherty  M, Pascual  E,  et al.  2018 Updated European League Against Rheumatism evidence-based recommendations for the diagnosis of gout.   Ann Rheum Dis. 2020;79(1):31-38. doi:10.1136/annrheumdis-2019-215315PubMedGoogle ScholarCrossref
7.
Mikuls  TR, Farrar  JT, Bilker  WB, Fernandes  S, Schumacher  HR  Jr, Saag  KG.  Gout epidemiology: results from the UK General Practice Research Database, 1990-1999.   Ann Rheum Dis. 2005;64(2):267-272. doi:10.1136/ard.2004.024091PubMedGoogle ScholarCrossref
8.
Krishnan  E, Baker  JF, Furst  DE, Schumacher  HR.  Gout and the risk of acute myocardial infarction.   Arthritis Rheum. 2006;54(8):2688-2696. doi:10.1002/art.22014PubMedGoogle ScholarCrossref
9.
Rho  YH, Lu  N, Peloquin  CE,  et al.  Independent impact of gout on the risk of diabetes mellitus among women and men: a population-based, BMI-matched cohort study.   Ann Rheum Dis. 2016;75(1):91-95. doi:10.1136/annrheumdis-2014-205827PubMedGoogle ScholarCrossref
10.
Cai  M, Xie  Y, Bowe  B,  et al.  Temporal trends in incidence rates of lower extremity amputation and associated risk factors among patients using Veterans Health Administration services from 2008 to 2018.   JAMA Netw Open. 2021;4(1):e2033953. doi:10.1001/jamanetworkopen.2020.33953PubMedGoogle Scholar
11.
Johnson  RJ.  Why focus on uric acid?   Curr Med Res Opin. 2015;31(suppl 2):3-7. doi:10.1185/03007995.2015.1087979PubMedGoogle ScholarCrossref
12.
Perez-Ruiz  F, Becker  MA.  Inflammation: a possible mechanism for a causative role of hyperuricemia/gout in cardiovascular disease.   Curr Med Res Opin. 2015;31(suppl 2):9-14. doi:10.1185/03007995.2015.1087980PubMedGoogle ScholarCrossref
13.
Yu  W, Cheng  JD.  Uric acid and cardiovascular disease: an update from molecular mechanism to clinical perspective.   Front Pharmacol. 2020;11:582680. doi:10.3389/fphar.2020.582680PubMedGoogle Scholar
14.
Skeete  K, Hess  EP, Clark  T, Moran  S, Kakar  S, Rizzo  M.  Epidemiology of suspected wrist joint infection versus inflammation.   J Hand Surg Am. 2011;36(3):469-474. doi:10.1016/j.jhsa.2010.10.033PubMedGoogle ScholarCrossref
15.
Blumberg  G, Long  B, Koyfman  A.  Clinical mimics: an emergency medicine-focused review of cellulitis mimics.   J Emerg Med. 2017;53(4):475-484. doi:10.1016/j.jemermed.2017.06.002PubMedGoogle ScholarCrossref
16.
Rousseau  I, Cardinal E  E, Raymond-Tremblay  D, Beauregard  CG, Braunstein  EM, Saint-Pierre  A.  Gout: radiographic findings mimicking infection.   Skeletal Radiol. 2001;30(10):565-569. doi:10.1007/s002560100392PubMedGoogle ScholarCrossref
17.
Patel  GK, Davies  WL, Price  PP, Harding  KG.  Ulcerated tophaceous gout.   Int Wound J. 2010;7(5):423-427. doi:10.1111/j.1742-481X.2010.00700.xPubMedGoogle ScholarCrossref
18.
Moxey  PW, Gogalniceanu  P, Hinchliffe  RJ,  et al.  Lower extremity amputations: a review of global variability in incidence.   Diabet Med. 2011;28(10):1144-1153. doi:10.1111/j.1464-5491.2011.03279.xPubMedGoogle ScholarCrossref
19.
Hachmöller  A.  Outcome of minor amputations at the diabetic foot in relation to bone histopathology: a clinical audit [in Germany].   Zentralbl Chir. 2007;132(6):491-496.PubMedGoogle ScholarCrossref
20.
Sariyilmaz  K, Eren  I, Ozkunt  O, Sungur  M, Kilicoglu  OI, Dikici  F.  Bilateral first-ray amputation of the foot due to severe tophaceous gout complicated by infection and discharged sinus: a case report.   J Am Podiatr Med Assoc. 2018;108(1):58-62. doi:10.7547/16-110PubMedGoogle ScholarCrossref
21.
Yu  KH, Luo  SF, Liou  LB,  et al.  Concomitant septic and gouty arthritis—an analysis of 30 cases.   Rheumatology (Oxford). 2003;42(9):1062-1066. doi:10.1093/rheumatology/keg297PubMedGoogle ScholarCrossref
22.
Zhou  C, Xue  C, Yang  B,  et al.  Amputation of the first metatarsophalangeal joint due to a giant gouty tophi: a case report.   Medicine (Baltimore). 2017;96(43):e8441. doi:10.1097/MD.0000000000008441PubMedGoogle Scholar
23.
LaMoreaux B, Francis-Sedlak M, Neville S, Holt R. Gout is an independent risk factor for undergoing an amputation procedure [abstract]. Arthritis Rheumatol. 2020;72(suppl 10):s673.
24.
von Elm  E, Altman  DG, Egger  M, Pocock  SJ, Gøtzsche  PC, Vandenbroucke  JP; STROBE Initiative.  The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.   Ann Intern Med. 2007;147(8):573-577. doi:10.7326/0003-4819-147-8-200710160-00010PubMedGoogle ScholarCrossref
25.
Helget  LN, England  BR, Roul  P,  et al. The incidence, prevalence, and burden of gout in the Veterans Health Administration. Arthritis Care Res (Hoboken). 2021;73(9):1363-1371. doi:10.1002/acr.24339
26.
Icks  A, Scheer  M, Morbach  S,  et al.  Time-dependent impact of diabetes on mortality in patients after major lower extremity amputation: survival in a population-based 5-year cohort in Germany.   Diabetes Care. 2011;34(6):1350-1354. doi:10.2337/dc10-2341PubMedGoogle ScholarCrossref
27.
Peters  EJ, Childs  MR, Wunderlich  RP, Harkless  LB, Armstrong  DG, Lavery  LA.  Functional status of persons with diabetes-related lower-extremity amputations.   Diabetes Care. 2001;24(10):1799-1804. doi:10.2337/diacare.24.10.1799PubMedGoogle ScholarCrossref
28.
Remes  L, Isoaho  R, Vahlberg  T, Viitanen  M, Koskenvuo  M, Rautava  P.  Quality of life three years after major lower extremity amputation due to peripheral arterial disease.   Aging Clin Exp Res. 2010;22(5-6):395-405. doi:10.1007/BF03337735PubMedGoogle ScholarCrossref
29.
Mayfield  JA, Reiber  GE, Sanders  LJ, Janisse  D, Pogach  LM; American Diabetes Association.  Preventive foot care in diabetes.   Diabetes Care. 2004;27(suppl 1):S63-S64. doi:10.2337/diacare.27.2007.S63PubMedGoogle ScholarCrossref
30.
Monteiro-Soares  M, Vale-Lima  J, Martiniano  J, Pinheiro-Torres  S, Dias  V, Boyko  EJ.  A systematic review with meta-analysis of the impact of access and quality of diabetic foot care delivery in preventing lower extremity amputation.   J Diabetes Complications. 2021;35(4):107837. doi:10.1016/j.jdiacomp.2020.107837PubMedGoogle Scholar
31.
Coburn  BW, Michaud  K, Bergman  DA, Mikuls  TR.  Allopurinol dose escalation and mortality among patients with gout: a national propensity-matched cohort study.   Arthritis Rheumatol. 2018;70(8):1298-1307. doi:10.1002/art.40486PubMedGoogle ScholarCrossref
32.
Neogi  T, Jansen  TL, Dalbeth  N,  et al.  2015 Gout Classification Criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative.   Arthritis Rheumatol. 2015;67(10):2557-2568. doi:10.1002/art.39254PubMedGoogle ScholarCrossref
33.
Chhana  A, Dalbeth  N.  The gouty tophus: a review.   Curr Rheumatol Rep. 2015;17(3):19. doi:10.1007/s11926-014-0492-xPubMedGoogle ScholarCrossref
34.
Rashid  N, Coburn  BW, Wu  YL,  et al.  Modifiable factors associated with allopurinol adherence and outcomes among patients with gout in an integrated healthcare system.   J Rheumatol. 2015;42(3):504-512. doi:10.3899/jrheum.140588PubMedGoogle ScholarCrossref
35.
Vargas-Santos  AB, Neogi  T, da Rocha Castelar-Pinheiro  G, Kapetanovic  MC, Turkiewicz  A.  Cause-specific mortality in gout: novel findings of elevated risk of non-cardiovascular-related deaths.   Arthritis Rheumatol. 2019;71(11):1935-1942. doi:10.1002/art.41008PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Views 1,905
    Citations 0
    Original Investigation
    Rheumatology
    January 6, 2022

    Comparison of Rates of Lower Extremity Amputation in Patients With and Without Gout in the US Department of Veterans Affairs Health System

    Author Affiliations
    • 1Veterans Affairs Nebraska–Western Iowa Health Care System, Omaha, Nebraska
    • 2Division of Rheumatology, Department of Internal Medicine, University of Nebraska Medical Center, Omaha
    • 3Department of Biostatistics, College of Public Health, University of Nebraska Medical Center, Omaha
    • 4Section of Rheumatology, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts
    • 5Horizon Therapeutics, Deerfield, Illinois
    JAMA Netw Open. 2022;5(1):e2142347. doi:10.1001/jamanetworkopen.2021.42347
    Key Points

    Question  Are lower extremity amputations (LEAs) performed more often in patients with gout?

    Findings  In this cohort study of 5 924 918 patients with and without gout, patients with gout had higher rates of LEA than those without gout after adjusting for comorbidities and other risk factors; this increase was apparent across different types of LEA. Among those with gout, poor serum urate control, but not the administration of urate-lowering therapy, was associated with a higher incidence of LEA.

    Meaning  These findings suggest that because gout is associated with an increased rate of LEA, efforts to better understand the degree to which these procedures might be preventable is warranted.

    Abstract

    Importance  Cardiometabolic and other risk factors could render patients with gout more likely to undergo lower extremity amputation (LEA).

    Objective  To examine the rate of and factors associated with LEA in patients with gout.

    Design, Setting, and Participants  In this matched cohort study using national administrative data, multivariable Cox proportional hazards regression models were used to examine the associations of gout with LEA. In analyses limited to patients with gout, attributes of serum urate control and treatment with urate-lowering therapy were examined as factors associated with LEA. This study included patients who used US Department of Veterans Affairs services from January 1, 2000, to July 31, 2015. Patients with gout were identified using diagnostic codes and matched with up to 10 controls by age, sex, and year of benefit enrollment. Data analysis was performed from January 26, 2021, to September 3, 2021.

    Exposures  Gout classification served as the primary independent variable of interest. In analyses limited to patients with gout, factors associated with serum urate control and urate-lowering therapy were examined.

    Main Outcomes and Measures  Overall LEA, as well as toe, transmetatarsal, below-the-knee, and above-the-knee amputation.

    Results  This cohort study included 5 924 918 patients, 556 521 with gout (mean [SD] age, 67 [12] years; 550 963 (99.0%) male; 88 853 [16.0%] Black non-Hispanic; 16 981 [4.3%] Hispanic/Latinx; 345 818 [62.1%] White non-Hispanic; 80 929 [14.5%] with race and ethnicity data missing; and 23 940 [4.3%] classified as other) and 5 368 397 without gout (mean [SD] age, 67 [12] years; 5 314 344 [99.0%] male; 558 464 [10.4%] Black non-Hispanic; 204 291 [3.0%] Hispanic/Latinx; 3 188 504 [59.4%] White non-Hispanic; 1 257 739 [23.4%)] with race and ethnicity data missing; and 159 399 [3.0%] classified as other). Compared with patients without gout, patients with gout were more likely to undergo amputation, an increased rate that remained after adjustment (adjusted hazard ratio, 1.20; 95% CI, 1.16-1.24) and was highest for below-the-knee amputation (adjusted hazard ratio, 1.59; 95% CI, 1.39-1.81). In those with gout, poor serum urate control (mean >7 mg/dL during the preceding year) was associated with a 25% to 37% increase in the rate of amputation. In contrast, treatment with urate-lowering therapy was not associated with the LEA rate.

    Conclusions and Relevance  In this matched cohort study, patients with gout were more likely to undergo LEA. This increase was independent of other comorbidities that have been associated with amputation, including diabetes and peripheral vascular disease. Serum urate control was independently associated with the LEA rate, suggesting the possibility that lower extremity amputation may be preventable in some patients.

    Introduction

    Gout is the most common form of inflammatory arthritis worldwide, affecting up to 4% of all adults with higher prevalence rates in older men and in members of minoritized racial and ethnic minority groups.1 Characterized by hyperuricemia and articular deposition of monosodium urate, the clinical course of gout consists of painful arthritis flares separated by asymptomatic intercritical periods of variable duration. In the absence of effective urate-lowering therapy (ULT), flare frequency and severity increase over time, with some patients developing advanced gout characterized by tophi and chronic systemic inflammation. As such, gout poses a substantial and increasing burden, resulting annually in more than 200 000 emergency department visits2,3 and total health care costs exceeding $6 billion in the US alone.4 A significant proportion of costs are preventable, reflecting missed treatment opportunities. With an increasingly well-understood pathogenesis, gout represents a highly treatable condition with management strategies focused on anti-inflammatory agents for flares and ULT to reduce or even halt disease progression.5,6

    Although gout alone poses a substantial burden, it also has strong associations with comorbidities and increased mortality. Cardiovascular disease, hypertension, chronic kidney disease (CKD), and diabetes are all overrepresented in patients with gout.7 Compared with the general population, patients with gout demonstrate a greater than 25% increase in the rate of acute myocardial infarction8 and are more likely to develop diabetes.9 Notably, these comorbidities also portend an increased risk of undergoing lower extremity amputation (LEA). Diabetes, for example, is associated with a 6-fold greater likelihood of receiving LEA, whereas advanced forms of CKD portend between a 2- to 4-fold increase in the rate of amputation.10

    In addition to concomitant comorbidity and potential adverse effects of serum urate and inflammation on vascular function,11-13 arthritis manifestations might also influence LEA rates in gout. It is recognized, for example, that gout mimics infectious processes of septic arthritis,14 cellulitis,15 and osteomyelitis16 and can masquerade as a nonhealing diabetic foot ulcer,17 which is a frequent indication for LEA.18 Numerous reports18-22 have described digital and/or lower limb amputations in the context of gout, some21,22 citing infection in the setting of gout, and another19 reported a gout diagnosis made postoperatively after histopathologic examination of resected tissues. Whether patients with gout undergo LEA at a higher rate than those without gout is unknown.23 Addressing this knowledge gap and identifying LEA determinants would inform future efforts to reduce amputations and improve outcomes in gout.

    The goal of this study was to examine the rate of and factors associated with LEA in patients with gout. To begin to understand the degree to which LEA might be preventable, we also examined whether amputation in gout was associated with measures of serum urate control or ULT administration. We hypothesized that patients with gout would have higher rates of LEA than patients without gout, independent of comorbidity and other risk factors, and among those with gout, higher serum urate concentrations and suboptimal ULT would be associated with increased amputation rates.

    Methods

    This national, matched cohort study was approved by the institutional review board at the Veterans Affairs (VA) Nebraska–Western Iowa Health Care System. This study involved the use of identifiable data. The requirement for informed consent was waived per institutional policy because there was no prospective data collection or patient contact. This report adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.24

    Study Population and Data Source

    This study leveraged administrative data from the national Veterans Health Administration (VHA). We used inpatient and outpatient data in the VA Corporate Data Warehouse, accessed through the VA Informatics and Computing Infrastructure. Data elements included pharmacy-dispensing, laboratory data, Current Procedural Terminology, Fourth Edition (CPT-4) and International Classification of Diseases, Ninth Revision (ICD-9) codes corresponding to electronic health record data from January 1, 2000, through July 31, 2015. We limited cohort creation to avoid misclassification related to International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) implementation initiated in October 2015. We identified patients with gout using an algorithm that required patients to have an ICD-9 code of 274.xx from 2 or more encounters 30 or more days apart with the second date serving as the index date.25 Each case was then matched with up to 10 patients without gout by birth year, sex, and year of VHA enrollment (the latter to account for temporal health care trends10 and to ensure similar durations of preindex observation). Comparators without gout had no prior ICD-9 codes of gout or receipt of previous ULT (allopurinol, febuxostat, probenecid, or pegloticase) and were assigned the same index date as their matched case counterpart. Potential cases and controls were excluded if they received LEA in the VHA before the index date. Study participants were followed up from the index date until first LEA, death, or the end of follow-up (September 2015), whichever occurred first. Some patients without gout were censored at the time they fulfilled the gout algorithm, after which they were crossed over and contributed to gout observation after matching with up to 10 new controls. Data analysis was performed from January 26, 2021, to September 3, 2021.

    Lower Extremity Amputation

    Lower extremity amputation was defined using a combination of CPT4, ICD-10 Procedure Coding System, and International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes as described by Cai et al10 (eTable 1 in the Supplement). Only the first LEA event was examined to reduce bias associated with multistage amputation procedures or revisions. Procedures were further classified by type: toe, transmetatarsal, below the knee, or above the knee. Crude incidence rates (IRs) of LEA and 95% CIs were calculated in patients with gout and comparators without gout by dividing the number of events by the total patient-years of follow-up in each group.

    Suboptimal Serum Urate Control and ULT

    To examine associations of gout care with LEA, we examined 2 components of surveillance and management during each year of follow-up among patients with gout: the receipt of conventional oral ULT and serum urate concentration. Using an a priori definition, patients were conservatively considered to have received adequate ULT with 2 or more pharmacy dispensing-episodes together totaling at least 90 days of coverage with allopurinol, febuxostat, or probenecid within each 1-year follow-up interval. Individuals receiving none of these agents were considered to have received suboptimal ULT. Those dispensed ULT during the year but not satisfying criteria for adequate treatment were categorized as indeterminate. Using an a priori definition for serum urate measures, patients with mean urate values less than 6 mg/dL (to convert to millimoles per liter, multiply by 0.0595) during each year of follow-up were considered to have adequate serum urate control, consistent with gout management guidelines,5,6 whereas those with mean levels greater than 7 mg/dL were considered to have suboptimal control. Missing values for a given year were imputed using available values for up to 2 prior years, recognizing that patients with well-controlled gout may undergo infrequent testing and that testing frequency in a real-world setting is variable. Patients with gout with serum urate levels of 6 to 7 mg/dL or for whom no serum urate measure was available were categorized as having indeterminate serum urate control.

    Statistical Analysis

    Group characteristics were compared using a 2-tailed, unpaired t test for continuous variables or a χ2 test for categorical variables. In primary analyses, the association of gout with LEA was examined using unadjusted Cox proportional hazards regression followed by multivariable Cox proportional hazards regression. Covariates included in multivariable models were selected a priori, were ascertained using administrative and electronic health data, and included age, sex, race, ethnicity, body mass index (calculated as weight in kilograms divided by height in meters squared), smoking history, hypertension, cardiovascular disease, peripheral artery disease, cancer, cerebrovascular disease, chronic lung disease, dementia, diabetes, and kidney disease. Race and ethnicity information was taken from electronic health care data and included as a covariate given the variability in gout prevalence and comorbidities across categories. Race and ethnicity reported as other included American Indian or Alaska Native, Asian, multiple race, or Native Hawaiian or other Pacific Islander. Body mass index was calculated from the closest weight value preceding the index date and modal height within the electronic health record. Smoking information was collected from routine clinical notes (VHA health factors) and modeled as current, former, never, or missing, with missing included because missing values may be informative (eg, death prohibiting medical record documentation). Comorbidities were defined based on the presence of at least 2 corresponding ICD-9 or ICD-9-CM codes occurring at any time before the index date (eTable 2 in the Supplement). The association of gout with LEA was quantified using hazard ratios (HRs) and 95% CIs, where the cause-specific hazard is interpreted as the hazard of amputation while the patient is alive and free of LEA. Two additional sensitivity analyses were conducted to evaluate missing covariates or loss of follow-up. Specifically, associations of gout with overall LEA were examined after removing individuals with missing data for race and ethnicity or smoking (ie, complete case approach) and censoring patients with more than 365 consecutive days without active VHA follow-up (defined by the occurrence of an outpatient visit or hospitalization).

    Recognizing diabetes as the strongest factor associated with LEA occurrence among US veterans,10 we subsequently examined patients in 4 groups defined by the presence or absence of gout and diabetes as part of preplanned secondary analyses, again using Cox proportional hazards regression models and the aforementioned covariates. To explore the degree to which amputation events might be associated with factors related to gout-specific health care, we examined components of surveillance and management as determinants of LEA in analyses restricted to patients with gout. Those with adequate or indeterminate care were combined into a single group for ease of interpretation. Patients were then categorized during each year of follow-up into 1 of 4 groups based on these 2 components of gout care: (1) adequate or indeterminate serum urate control and adequate or indeterminate ULT (Reference group), (2) adequate or indeterminate serum urate control and suboptimal ULT, (3) suboptimal serum urate control and adequate or indeterminate ULT, or (4) suboptimal serum urate control and suboptimal ULT. Associations of this combined variable with LEA during the next year of follow-up were examined using multivariable Cox proportional hazards regression models as described above, with the index calendar year serving as an additional covariate in each model (eFigure in the Supplement). Note that the serum urate and ULT status variable could change from year to year within a patient, but all other covariates were fixed index values. All analyses were performed using Stata software, version 15 (StataCorp) within the VA Informatics and Computing Infrastructure. A 2-tailed P < .05 was considered statistically significant.

    Results

    This cohort study included 5 924 918 patients, 556 521 with gout (mean [SD] age, 67 [12] years; 550 963 (99.0%) male; 88 853 [16.0%] Black non-Hispanic; 16 981 [4.3%] Hispanic/Latinx; 345 818 [62.1%] White non-Hispanic; 80 929 [14.5%] with race and ethnicity data missing; and 23 940 [4.3%] classified as other) and 5 368 397 without gout (mean [SD] age, 67 [12] years; 5 314 344 [99.0%] male; 558 464 [10.4%] Black non-Hispanic; 204 291 [3.0%] Hispanic/Latinx; 3 188 504 [59.4%] White non-Hispanic; 1 257 739 [23.4%)] with race and ethnicity data missing; and 159 399 [3.0%] classified as other) (Table 1). Compared with patients without gout, individuals with gout were more likely to be Black or African American, to have obesity, and to have comorbidities, with the exception of dementia, which was slightly more frequent in those without gout. In addition to being less likely to have missing data for race and ethnicity, patients with gout were also less likely than those without gout to have missing data for smoking status (31 965 [5.7%] vs 868 838 [16.2%]).

    From January 1, 2000, to July 31, 2015, a total of 4970 LEAs were performed during 3.4 million patient-years of follow-up in patients with gout (IR, 1.46 [95% CI, 1.42-1.50] procedures per 1000 patient-years), and 24 583 LEAs were performed during 32.1 million patient-years of follow-up in comparators (IR, 0.77 [95% CI, 0.76-0.78] procedures per 1000 patient-years) (Table 2). Transmetatarsal procedures were the most common procedure, followed in descending order by above-the-knee, toe, and below-the-knee amputations. Crude IRs were universally higher for patients with gout than patients without gout overall and across LEA categories. Results from unadjusted Cox proportional hazards regression are given in Table 2, showing an increased association of LEA with gout both overall (HR, 1.85; 95% CI, 1.80-1.91) and for each procedure type.

    Results from multivariable Cox proportional hazards regression models are shown in Figure 1. After adjustment, gout remained associated with a significant increase in LEA rate (adjusted HR [aHR], 1.20; 95% CI, 1.16-1.24). Rates were highest for below-the-knee amputations (aHR, 1.59; 95% CI, 1.39-1.81) but were also elevated for toe (aHR, 1.27; 95% CI, 1.18-1.37), transmetatarsal (aHR, 1.11; 95% CI, 1.06-1.16), and above-the-knee (aHR, 1.22; 95% CI, 1.15-1.30) procedures. Results from full models are detailed in eTable 3 in the Supplement. Of other factors examined, the presence of diabetes was most strongly associated with overall LEA (aHR, 3.21; 95% CI, 3.13-3.30). Results for gout were similar in sensitivity analyses using a complete case approach (aHR, 1.18; 95% CI, 1.14-1.22). The association of gout was also similar (aHR, 1.13; 95% CI, 1.09-1.16), albeit slightly attenuated, in sensitivity analyses that censored patients without a VHA observation for 1 year.

    Results based on gout and diabetes status are shown in Figure 2. In the absence of diabetes, patients with gout demonstrated a 1.56-fold increased rate (95% CI, 1.52-1.66) of LEA (eTable 4 in the Supplement). The highest rate was observed in patients with gout and diabetes (aHR, 3.36; 95% CI, 3.02-3.75), similar to the rate observed in those with diabetes alone (aHR, 3.21; 95% CI, 3.00-3.43).

    In subsequent analyses limited to patients with gout, we examined associations of predefined measures of suboptimal ULT administration and suboptimal serum urate control with LEA. Groups characterized by suboptimal serum urate control, compared with those with adequate or indeterminate values for serum urate control and ULT administration, had universally higher LEA rates during the following year of observation (aHRs, 1.17-1.52) (Table 3). In contrast, suboptimal ULT did not appear to be associated with LEA occurrence. Frequencies in different groups at first observation were as follows: adequate or indeterminate serum urate and adequate or indeterminate ULT, 199 125 (40.2%); suboptimal serum urate and adequate or indeterminate ULT, 86 766 (17.5%); adequate or indeterminate serum urate and suboptimal ULT, 149 650 (30.2%); and suboptimal serum urate and suboptimal ULT, 59 591 (12.0%). Frequencies in different groups at last observation were as follows: adequate or indeterminate serum urate and adequate or indeterminate ULT, 194 439 (39.3%); suboptimal serum urate and adequate or indeterminate ULT, 53 917 (10.9%); adequate or indeterminate serum urate and suboptimal ULT, 211 519 (42.7%); and suboptimal serum urate and suboptimal ULT, 35 257 (7.1%).

    Discussion

    In this cohort study of US veterans, we found that gout was associated with a 20% increase in the rate of LEA. This increase was independent of comorbidities, such as diabetes, CKD, peripheral vascular disease, and others that serve as established risk factors for LEA and are more common in gout. This finding has prognostic implications for patients with gout, with 5-year mortality approaching 70% in patients with diabetes after LEA and only slightly lower rates among those without diabetes.26 In addition to portending reduced survival, LEA is associated with worse physical functioning, lower health-related quality of life, and depression, with up to half of patients institutionalized in long-term care facilities postoperatively.27,28 Extrapolating to a general population that includes 9.2 million adults in the US with gout,1 our results suggest an approximately 2% cumulative incidence during a period of 15 years (Figure 2) and suggest that as many as 184 000 patients with gout are at risk of undergoing LEA during the coming years in the US alone. With gout and LEA frequency increasing in the VHA,10,25 an implication of these observations is that the burden posed to the health care system is likely to increase in this population over time.

    In addition to being independent of comorbid conditions, our results found that the association of gout with LEA was higher in those without diabetes compared with the overall VA population. Although this may reflect the significantly lower risk of LEA among a comparator population without diabetes, it is also possible that recommended surveillance and standardized foot care applied in the setting of diabetes (but not routinely implemented in gout alone)29 might lead to reductions in LEA attributable to gout among those with both conditions. In patients with diabetes, structured foot care services have been shown to reduce the odds of LEA by more than 50%.30

    Whether a proportion of LEA procedures in gout are preventable, similar to what has been shown in diabetes,30 is unknown. The association of LEA with suboptimal urate control should not be interpreted as causal because serum urate control could serve as a surrogate for gout severity or of other unmeasured or inadequately measured cofactors, such as severity of diabetes or CKD. The association of serum urate control is juxtaposed to results showing no association between attributes of ULT administration and LEA. These null findings may reflect real-world health care practices, which rarely involve ULT dose escalation and infrequent achievement of serum urate goals31 as recommended in management guidelines.5,6 A prior histopathologic survey19 showing gout as an unrecognized culprit for LEA, masquerading in preoperative assessments as nonhealing wounds or deep tissue infections, suggest that at least some amputations could be avoided. For instance, systematic screening for gout and/or hyperuricemia in patients being considered for LEA could identify patients who might benefit from noninvasive imaging studies, such as musculoskeletal ultrasonography or dual energy computed tomography, which can help to discriminate gout from its mimics.32 Our results, however, do not support tophaceous deposition as the sole explanation for excess amputations that occur in gout. The risk of LEA attributable to gout in this study was higher for below-the-knee than for toe or transmetatarsal procedures, with the toe or transmetatarsal being more common anatomical locations affected by tophus formation.33 Likewise, poor serum urate control, which could serve as a marker of tophaceous burden, rendered similar associations with LEA across procedure types. Whether implementation of systematic screening for gout in patients being considered for amputation or patient and health care professional education would improve long-term outcomes is unknown and will require further study.

    Limitations

    This study has limitations. Inherent in its observational design and reliance on administrative data, the possibility of bias from misclassification (of exposure, outcome, and covariates), missing data, and loss to follow-up exists. To reduce misclassification, we required 2 or more gout diagnoses separated in time. In addition, prior work7,34,35 from our group demonstrated rates of ULT dispensing, comorbidity, and mortality that were similar to values reported in separate non-VHA gout populations. Likewise, sensitivity analyses limited to individuals without missing data and censoring with a loss of follow-up that exceeded 1 year suggest that our results are robust and not meaningfully affected by these potential sources of bias. Although we adjusted for relevant comorbid conditions, we were unable to assess the severity of these comorbidities, which may have resulted in residual confounding. Furthermore, comorbidities were assessed at baseline, but their status and severity may change over time. It is also possible that findings in the VHA may not be generalizable to other populations. However, as the largest integrated health system in the US with a patient population enriched for gout and related cardiometabolic conditions, the VHA provides an ideal context for examining health outcomes in gout.

    Conclusions

    This study found that US veterans with gout were 20% more likely to undergo LEA than those without gout. This increase was independent of comorbidities and other risk factors and was apparent across different LEA types. Among those with gout, suboptimal serum urate control was associated with a higher rate of LEA. Further investigation is needed to understand the indications for LEA procedures conducted in gout in addition to identifying potential means of prevention as a way of ultimately improving long-term outcomes in this population.

    Back to top
    Article Information

    Accepted for Publication: November 11, 2021.

    Published: January 6, 2022. doi:10.1001/jamanetworkopen.2021.42347

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Mikuls TR et al. JAMA Network Open.

    Corresponding Author: Ted R. Mikuls, MD, MSPH, Division of Rheumatology, Department of Internal Medicine, University of Nebraska Medical Center, 986270 Nebraska Medical Center, Omaha, NE 68198-6270 (tmikuls@unmc.edu).

    Author Contributions: Dr Mikuls and Mr Sayles had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Mikuls, Soto, Petro, Helget, LaMoreaux, England.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: Mikuls, Soto, Helget, Cope.

    Critical revision of the manuscript for important intellectual content: Mikuls, Soto, Petro, Helget, Roul, Sayles, Neogi, LaMoreaux, O’Dell, England.

    Statistical analysis: Mikuls, Sayles, England.

    Obtained funding: Mikuls.

    Administrative, technical, or material support: Petro, Helget, Roul, Cope, England.

    Supervision: Mikuls, LaMoreaux, O’Dell, England.

    Conflict of Interest Disclosures: Dr Mikuls reported receiving grants from and serving as a consultant for Horizon Therapeutics outside the submitted work and serving in the Core Leadership group for the 2020 American College of Rheumatology Gout Guidelines. Dr Neogi reported receiving consulting fees from Alnylam. Dr LaMoreaux reported being an employee and stockholder of Horizon Therapeutics. Dr England reported receiving personal fees from Boehringer Ingelheim outside the submitted work. No other disclosures were reported.

    Funding/Support: Funding for this project was provided by the Nebraska Arthritis Outcomes Research Center and the University of Nebraska Medical Center Division of Rheumatology and Immunology. Dr Mikuls is supported by grant BX004600 from the Veterans Affairs Biomedical Laboratory Research and Development and grant U54 GM115458 from the National Institutes of Health. Dr Neogi is supported by grant K24 AR070892 from the National Institutes of Health. Dr England is supported by grant IK2 CX002203 from the Veterans Affairs Clinical Sciences Research and Development and the Rheumatology Research Foundation.

    Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Disclaimer: The views expressed in this manuscript are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs of the US government. All information and materials in the manuscript are original.

    References
    1.
    Chen-Xu  M, Yokose  C, Rai  SK, Pillinger  MH, Choi  HK.  Contemporary prevalence of gout and hyperuricemia in the United States and decadal trends: the National Health and Nutrition Examination Survey, 2007-2016.   Arthritis Rheumatol. 2019;71(6):991-999. doi:10.1002/art.40807PubMedGoogle ScholarCrossref
    2.
    Garg  R, Sayles  HR, Yu  F,  et al.  Gout-related health care utilization in US emergency departments, 2006 through 2008.   Arthritis Care Res (Hoboken). 2013;65(4):571-577. doi:10.1002/acr.21837PubMedGoogle ScholarCrossref
    3.
    Singh  JA, Yu  S.  Time trends, predictors, and outcome of emergency department use for gout: a nationwide US study.   J Rheumatol. 2016;43(8):1581-1588. doi:10.3899/jrheum.151419PubMedGoogle ScholarCrossref
    4.
    Wertheimer  A, Morlock  R, Becker  MA.  A revised estimate of the burden of illness of gout.   Curr Ther Res Clin Exp. 2013;75:1-4. doi:10.1016/j.curtheres.2013.04.003PubMedGoogle ScholarCrossref
    5.
    FitzGerald  JD, Dalbeth  N, Mikuls  T,  et al.  2020 American College of Rheumatology Guideline for the Management of Gout.   Arthritis Rheumatol. 2020;72(6):879-895. doi:10.1002/art.41247PubMedGoogle ScholarCrossref
    6.
    Richette  P, Doherty  M, Pascual  E,  et al.  2018 Updated European League Against Rheumatism evidence-based recommendations for the diagnosis of gout.   Ann Rheum Dis. 2020;79(1):31-38. doi:10.1136/annrheumdis-2019-215315PubMedGoogle ScholarCrossref
    7.
    Mikuls  TR, Farrar  JT, Bilker  WB, Fernandes  S, Schumacher  HR  Jr, Saag  KG.  Gout epidemiology: results from the UK General Practice Research Database, 1990-1999.   Ann Rheum Dis. 2005;64(2):267-272. doi:10.1136/ard.2004.024091PubMedGoogle ScholarCrossref
    8.
    Krishnan  E, Baker  JF, Furst  DE, Schumacher  HR.  Gout and the risk of acute myocardial infarction.   Arthritis Rheum. 2006;54(8):2688-2696. doi:10.1002/art.22014PubMedGoogle ScholarCrossref
    9.
    Rho  YH, Lu  N, Peloquin  CE,  et al.  Independent impact of gout on the risk of diabetes mellitus among women and men: a population-based, BMI-matched cohort study.   Ann Rheum Dis. 2016;75(1):91-95. doi:10.1136/annrheumdis-2014-205827PubMedGoogle ScholarCrossref
    10.
    Cai  M, Xie  Y, Bowe  B,  et al.  Temporal trends in incidence rates of lower extremity amputation and associated risk factors among patients using Veterans Health Administration services from 2008 to 2018.   JAMA Netw Open. 2021;4(1):e2033953. doi:10.1001/jamanetworkopen.2020.33953PubMedGoogle Scholar
    11.
    Johnson  RJ.  Why focus on uric acid?   Curr Med Res Opin. 2015;31(suppl 2):3-7. doi:10.1185/03007995.2015.1087979PubMedGoogle ScholarCrossref
    12.
    Perez-Ruiz  F, Becker  MA.  Inflammation: a possible mechanism for a causative role of hyperuricemia/gout in cardiovascular disease.   Curr Med Res Opin. 2015;31(suppl 2):9-14. doi:10.1185/03007995.2015.1087980PubMedGoogle ScholarCrossref
    13.
    Yu  W, Cheng  JD.  Uric acid and cardiovascular disease: an update from molecular mechanism to clinical perspective.   Front Pharmacol. 2020;11:582680. doi:10.3389/fphar.2020.582680PubMedGoogle Scholar
    14.
    Skeete  K, Hess  EP, Clark  T, Moran  S, Kakar  S, Rizzo  M.  Epidemiology of suspected wrist joint infection versus inflammation.   J Hand Surg Am. 2011;36(3):469-474. doi:10.1016/j.jhsa.2010.10.033PubMedGoogle ScholarCrossref
    15.
    Blumberg  G, Long  B, Koyfman  A.  Clinical mimics: an emergency medicine-focused review of cellulitis mimics.   J Emerg Med. 2017;53(4):475-484. doi:10.1016/j.jemermed.2017.06.002PubMedGoogle ScholarCrossref
    16.
    Rousseau  I, Cardinal E  E, Raymond-Tremblay  D, Beauregard  CG, Braunstein  EM, Saint-Pierre  A.  Gout: radiographic findings mimicking infection.   Skeletal Radiol. 2001;30(10):565-569. doi:10.1007/s002560100392PubMedGoogle ScholarCrossref
    17.
    Patel  GK, Davies  WL, Price  PP, Harding  KG.  Ulcerated tophaceous gout.   Int Wound J. 2010;7(5):423-427. doi:10.1111/j.1742-481X.2010.00700.xPubMedGoogle ScholarCrossref
    18.
    Moxey  PW, Gogalniceanu  P, Hinchliffe  RJ,  et al.  Lower extremity amputations: a review of global variability in incidence.   Diabet Med. 2011;28(10):1144-1153. doi:10.1111/j.1464-5491.2011.03279.xPubMedGoogle ScholarCrossref
    19.
    Hachmöller  A.  Outcome of minor amputations at the diabetic foot in relation to bone histopathology: a clinical audit [in Germany].   Zentralbl Chir. 2007;132(6):491-496.PubMedGoogle ScholarCrossref
    20.
    Sariyilmaz  K, Eren  I, Ozkunt  O, Sungur  M, Kilicoglu  OI, Dikici  F.  Bilateral first-ray amputation of the foot due to severe tophaceous gout complicated by infection and discharged sinus: a case report.   J Am Podiatr Med Assoc. 2018;108(1):58-62. doi:10.7547/16-110PubMedGoogle ScholarCrossref
    21.
    Yu  KH, Luo  SF, Liou  LB,  et al.  Concomitant septic and gouty arthritis—an analysis of 30 cases.   Rheumatology (Oxford). 2003;42(9):1062-1066. doi:10.1093/rheumatology/keg297PubMedGoogle ScholarCrossref
    22.
    Zhou  C, Xue  C, Yang  B,  et al.  Amputation of the first metatarsophalangeal joint due to a giant gouty tophi: a case report.   Medicine (Baltimore). 2017;96(43):e8441. doi:10.1097/MD.0000000000008441PubMedGoogle Scholar
    23.
    LaMoreaux B, Francis-Sedlak M, Neville S, Holt R. Gout is an independent risk factor for undergoing an amputation procedure [abstract]. Arthritis Rheumatol. 2020;72(suppl 10):s673.
    24.
    von Elm  E, Altman  DG, Egger  M, Pocock  SJ, Gøtzsche  PC, Vandenbroucke  JP; STROBE Initiative.  The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.   Ann Intern Med. 2007;147(8):573-577. doi:10.7326/0003-4819-147-8-200710160-00010PubMedGoogle ScholarCrossref
    25.
    Helget  LN, England  BR, Roul  P,  et al. The incidence, prevalence, and burden of gout in the Veterans Health Administration. Arthritis Care Res (Hoboken). 2021;73(9):1363-1371. doi:10.1002/acr.24339
    26.
    Icks  A, Scheer  M, Morbach  S,  et al.  Time-dependent impact of diabetes on mortality in patients after major lower extremity amputation: survival in a population-based 5-year cohort in Germany.   Diabetes Care. 2011;34(6):1350-1354. doi:10.2337/dc10-2341PubMedGoogle ScholarCrossref
    27.
    Peters  EJ, Childs  MR, Wunderlich  RP, Harkless  LB, Armstrong  DG, Lavery  LA.  Functional status of persons with diabetes-related lower-extremity amputations.   Diabetes Care. 2001;24(10):1799-1804. doi:10.2337/diacare.24.10.1799PubMedGoogle ScholarCrossref
    28.
    Remes  L, Isoaho  R, Vahlberg  T, Viitanen  M, Koskenvuo  M, Rautava  P.  Quality of life three years after major lower extremity amputation due to peripheral arterial disease.   Aging Clin Exp Res. 2010;22(5-6):395-405. doi:10.1007/BF03337735PubMedGoogle ScholarCrossref
    29.
    Mayfield  JA, Reiber  GE, Sanders  LJ, Janisse  D, Pogach  LM; American Diabetes Association.  Preventive foot care in diabetes.   Diabetes Care. 2004;27(suppl 1):S63-S64. doi:10.2337/diacare.27.2007.S63PubMedGoogle ScholarCrossref
    30.
    Monteiro-Soares  M, Vale-Lima  J, Martiniano  J, Pinheiro-Torres  S, Dias  V, Boyko  EJ.  A systematic review with meta-analysis of the impact of access and quality of diabetic foot care delivery in preventing lower extremity amputation.   J Diabetes Complications. 2021;35(4):107837. doi:10.1016/j.jdiacomp.2020.107837PubMedGoogle Scholar
    31.
    Coburn  BW, Michaud  K, Bergman  DA, Mikuls  TR.  Allopurinol dose escalation and mortality among patients with gout: a national propensity-matched cohort study.   Arthritis Rheumatol. 2018;70(8):1298-1307. doi:10.1002/art.40486PubMedGoogle ScholarCrossref
    32.
    Neogi  T, Jansen  TL, Dalbeth  N,  et al.  2015 Gout Classification Criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative.   Arthritis Rheumatol. 2015;67(10):2557-2568. doi:10.1002/art.39254PubMedGoogle ScholarCrossref
    33.
    Chhana  A, Dalbeth  N.  The gouty tophus: a review.   Curr Rheumatol Rep. 2015;17(3):19. doi:10.1007/s11926-014-0492-xPubMedGoogle ScholarCrossref
    34.
    Rashid  N, Coburn  BW, Wu  YL,  et al.  Modifiable factors associated with allopurinol adherence and outcomes among patients with gout in an integrated healthcare system.   J Rheumatol. 2015;42(3):504-512. doi:10.3899/jrheum.140588PubMedGoogle ScholarCrossref
    35.
    Vargas-Santos  AB, Neogi  T, da Rocha Castelar-Pinheiro  G, Kapetanovic  MC, Turkiewicz  A.  Cause-specific mortality in gout: novel findings of elevated risk of non-cardiovascular-related deaths.   Arthritis Rheumatol. 2019;71(11):1935-1942. doi:10.1002/art.41008PubMedGoogle ScholarCrossref
    ×