Dr. Perez-Ruiz has received consulting fees (more than $10,000 per year) from IPSEN Pharma.
Using serum urate levels to determine the period free of gouty symptoms after withdrawal of long-term urate-lowering therapy: A prospective study
Article first published online: 29 SEP 2006
Copyright © 2006 by the American College of Rheumatology
Arthritis Care & Research
Volume 55, Issue 5, pages 786–790, 15 October 2006
How to Cite
Perez-Ruiz, F., Atxotegi, J., Hernando, I., Calabozo, M. and Nolla, J. M. (2006), Using serum urate levels to determine the period free of gouty symptoms after withdrawal of long-term urate-lowering therapy: A prospective study. Arthritis & Rheumatism, 55: 786–790. doi: 10.1002/art.22232
- Issue published online: 29 SEP 2006
- Article first published online: 29 SEP 2006
- Manuscript Accepted: 12 JAN 2006
- Manuscript Received: 16 AUG 2005
- Asociación de Reumatólogos del Hospital de Cruces
- Gout suppressants;
Withdrawal of urate-lowering therapy (ULT) is associated with recurrence of acute gouty arthritis and tophi, but no data are available about factors associated with recurrence of gouty symptoms. Therefore, life-long therapy prescription is usually advised, but the prospect of life-long therapy may contribute to very low compliance rates. Our objective was to ascertain the outcome of ULT withdrawal after long-term, documented control of serum urate levels.
We conducted a prospective, long-term, followup study of patients treated with ULT during a 5-year period. Both diagnosis and recurrence of gout were determined based on monosodium urate crystal identification in synovial fluid or material aspirated from tophi.
Low average serum urate levels while receiving ULT and during the followup period after ULT withdrawal were statistically associated with the longest period in which patients were free of gouty symptoms, suggesting that depletion and formation of the body's urate pool is dependent on both time and serum urate levels. Patients whose average serum urate levels were <5.05 mg/dl while receiving ULT and <8.75 mg/dl after ULT withdrawal had the longest (>4 years) time to recurrence.
Proper and long-term reduction of serum urate level is associated with long-term periods in which patients are free of gouty symptoms, probably due to the reduction of the urate pool. These results suggest that 5-year intermittent, instead of life-long, ULT could be offered to patients with good serum urate control during ULT.
Gout is a urate crystal deposition disease. Gouty symptoms are related to the acute and chronic inflammatory responses to monosodium urate monohydrate (MSU) crystals deposited in articular and periarticular tissues (1). The goal of long-term urate-lowering therapy (ULT) is to reduce extracellular urate levels to a subsaturating range as reflected by serum urate levels less than ∼7.0 mg/dl. Achievement of this goal results in dissolution of MSU crystals (2), reduction in the body's urate pool (3, 4), and reduction or disappearance of acute gouty flares and tophi (5–7) associated with the decrease of MSU crystals in synovial fluid (8, 9).
Expert opinion suggests that ULT should be continued indefinitely (10), based in part on the observation that gouty symptoms recur after ULT is discontinued (11–13). The studies reporting recurrence, however, have some shortcomings: they were not prospective or did not document either a definite diagnosis of gout or long-term control of urate levels during ULT.
By contrast, poor compliance with ULT has been documented (14), with as many as 82% of patients who were prescribed allopurinol withdrawing from therapy, never initiating therapy, or showing inadequate compliance within 24 months after the initial prescription. The prospect of life-long ULT might be one of the causes. Also, by comparison with other continuous treatment regimens prescribed for disorders that may be asymptomatic for long periods, patients appear to withdraw frequently from ULT once gouty symptoms have resolved (15), even if the duration of ULT has been <1 year (11). The current study was designed to ascertain the outcome (with regard to gouty symptoms) of ULT withdrawal after long-term, documented control of serum urate levels.
PATIENTS AND METHODS
This prospective, observational study of patients with chronic gout was carried out with approval of the Ethical Clinical Investigation Board. Inclusion criteria were as follows: definite diagnosis of gout based on MSU crystal identification by polarized, compensated, microscopic examination of samples of synovial fluid or tophus aspirate; ULT for 5 years (60 months) in patients with no palpable tophi on physical examination at baseline or for 5 years after disappearance of palpable tophi; average serum urate levels <7.0 mg/dl during the ULT followup period; consent to withdraw ULT and willingness to undergo followup examination according to protocol after withdrawal; and acute gouty attacks or appearance of tophi after ULT withdrawal confirmed by monosodium urate crystal identification. Exclusion criteria for analysis were as follows: average serum urate levels ≤7.0 mg/dl after ULT withdrawal, lost to followup, ULT prescription by any physician other than those involved with the study for any cause, and suspected gouty attack not confirmed by MSU crystal identification. The average serum urate levels during ULT followup and withdrawal followup were calculated using the trapezoidal method (5). Serum urate levels were monitored every 3–6 months during ULT according to prior serum urate levels and the presence of comorbid conditions such as renal function impairment. Serum urate levels were monitored after ULT withdrawal twice during the first year and at least once a year thereafter. Lifestyle recommendations, such as avoidance of alcoholic beverages, high protein intake, and gradual reduction of excess weight, were encouraged during ULT and especially after ULT withdrawal. Statistical analysis was performed with the SPSS 13.0 statistical package (SPSS, Chicago, IL). Survival (Kaplan-Meier) analyses were conducted using the log rank (Mantel-Cox) test and the Cox proportional hazard method.
A total of 104 patients entered into the study, and 89 patients completed the trial. Fifteen participants were excluded from analyses for the following reasons: inability to obtain samples for synovial fluid analysis during a possible gout attack (6 patients); average serum urate levels ≤7.0 mg/dl after ULT withdrawal (3 patients); ULT prescription by other physicians after ULT withdrawal for asymptomatic hyperuricemia, but not for gouty symptoms (4 patients); and lost to followup (2 patients). Characteristics of the study patients are shown in Table 1. On baseline examination (prior to ULT), 23 patients (26%) had at least 1 palpable tophus, 79 (89%) had arthritic involvement of >1 joint, and 20 (23%) had ≥4 arthritic joints. In 43 patients (48%), allopurinol was the predominant or only urate-lowering agent used; in 45 patients (51%), benzbromarone was the predominant treatment; and in 1 patient, allopurinol and benzbromarone were administered in combination. The median period of ULT in patients with tophi was 89 months (range 73–95 months).
|Characteristic||Mean ± SD||Median||Range|
|Age, years||54.05 ± 11.35||54.00||24.00–84.00|
|Time from onset, years||7.30 ± 6.43||5.00||1.00–28.00|
|Baseline body mass index, kg/m2||28.66 ± 3.28||28.11||21.53–43.71|
|Serum urate (mg/dl), baseline||8.76 ± 1.19||8.49||7.10–12.90|
|Serum urate (mg/dl), average while receiving ULT||4.88 ± 0.98||5.05||2.80–6.99|
|Serum urate (mg/dl), average after withdrawal||9.07 ± 1.26||8.75||7.15–12.40|
|Followup after ULT, months||26.39 ± 16.14||24.00||1.00–78.00|
Thirty-seven patients (41.6%) had crystal-proven recurrence of gout in the period following withdrawal of ULT: 1 patient developed a tophus but did not experience an acute arthritis attack and 36 patients had an attack of acute gouty arthritis. Recurrence of gout appeared during 6–60 months of followup; the distribution of patients with and without recurrence during each year of followup is shown in Figure 1. All but 1 gouty arthritis recurrence were monoarticular, with first metatarsophalangeal joint involvement in 19 instances (52%). An additional 5 patients were evaluated for suspected acute gouty attacks, which could not be verified and ultimately could be explained by circumstances other than gout: 4 patients had internal meniscal tears confirmed by magnetic resonance imaging, absence of urate crystals in synovial fluid samples, or macroscopic urate deposits in arthroscopic repair surgery, and 1 patient had a stress fracture of the fifth metatarsal.
By Cox regression hazard analysis, time from the onset of gout, average serum urate levels while receiving ULT, and average serum urate levels after ULT withdrawal demonstrated a P value less than 0.10 in relation to freedom from gout recurrence, but only serum urate levels during and after ULT reached statistical significance (P < 0.05). Neither the presence of tophi nor sex, age, or affected joint distribution at baseline were related to duration of freedom from gout recurrence.
Kaplan-Meier survival analysis was performed by stratifying groups according to the median of average serum urate levels while receiving ULT (5.05 mg/dl) and after ULT withdrawal (8.75 mg/dl). The mean survival time was 34 months for patients with higher serum urate levels while receiving ULT and 49 months for patients with lower levels (Table 2, Figure 2). Conversely, patients in the lower range of hyperuricemia after ULT withdrawal had a mean survival time of nearly 47 months compared with 34 months for patients in the higher range (Table 2, Figure 2).
|Serum urate (mg/dl)||Mean†||Median|
|Estimate||Standard error||95% CI||Estimate||Standard error||95% CI|
Analysis of the symptom-free duration of the 4 groups, obtained by stratifying patients by median of average serum urate levels while receiving ULT and after withdrawal of ULT (Table 3), demonstrated that patients in the highest ranges of serum urate levels had the poorest outcome (just over 2 years without symptomatic gout), in contrast to patients in the lowest range of serum urate levels who had symptom-free durations averaging >4 years.
|Serum urate level, mg/dl||Mean†||Median|
|Estimate||95% CI||Estimate||95% CI|
|>8.75 after withdrawal|
|<5.05 while receiving ULT||42.45||34.06–50.85||38.00||33.24–42.75|
|≥5.05 while receiving ULT||28.45||21.04–35.85||24.00||20.49–27.50|
|<8.75 after withdrawal|
|<5.05 while receiving ULT||50.04||40.17–59.91||48.00||43.41–52.58|
|≥5.05 while receiving ULT||38.77||30.18–47.37||46.00||25.25–66.74|
The recommendation to maintain ULT indefinitely is based on the clinical perception that patients with gout will have recurring gouty symptoms after ULT withdrawal. Although the risk of developing gout appears to be a function of the magnitude and duration of hyperuricemia, other still unknown, predisposing factors are likely because not every patient with sustained hyperuricemia develops gout (16).
Prolonged ULT for patients with gout has been supported by several reports demonstrating that gout symptoms frequently recur after ULT withdrawal (11–13). However, the conclusions of these reports do not entirely agree: 2 reports conclude that ULT should be taken indefinitely (11, 13), but another suggests that long-term control of urate levels may induce complete depletion of urate deposits and put patients into an asymptomatic hyperuricemic state for a prolonged period (12).
Consideration of the methodologies and results of these reports may permit insight into the robustness of the conclusions. First, only a limited number of patients were studied: 64 patients overall in the 3 reports. Second, only 2 of the studies were prospective (11, 12), so that one-third of the entire study group was potentially subject to retrospective review bias. Third, the diagnosis of gout was confirmed by crystal identification in only 1 of the reports (12). The major criterion for inclusion was that patients were asymptomatic and willing to withdraw from ULT. In one study (11), some patients stopped allopurinol due to adverse events: at least 50% of the previously reported patients had tophaceous gout, and duration of ULT ranged from 1 to 19 years in patients with tophi (12, 13). In one series, serum urate levels while receiving ULT ranged from 2.6 mg/dl to 9.2 mg/dl (13), so that some patients did not achieve average urate levels under the solubility limit for monosodium urate while receiving ULT.
Recurrence rates ranged from 30% to 81%, and followup periods ranged from 3 to 107 months in the series with the highest recurrence rate (11–13). Serum urate levels during ULT in all series were numerically lower in patients who did not have a recurrence of gouty symptoms compared with patients who did have a recurrence of gout (11–13); also, serum urate levels after ULT withdrawal were higher (12, 13) in patients who experienced recurrence than in patients with no further gouty symptoms. Unfortunately, statistical analysis of these data was not provided in any report. Thus, previous reports in the literature have not adequately assessed the impact that withdrawing ULT has on patients with gout. A first consideration was that patients should have a definite diagnosis of gout based on urate crystal identification both at baseline and at the time of suspected recurrences. Although quite restrictive, crystal-based diagnosis is considered to be the gold standard technique (17). Indeed, for 5 patients with a suspected recurrence of gout in our series, alternative explanations were identified for painful episodes that otherwise might have been misdiagnosed as gouty recurrences. Despite the restrictive criteria for inclusion in our series, we were able to study more patients than was previously reported overall in the literature. Also, our study was conducted under the belief that ULT should be administered long term to assure depletion of the body's urate pool in patients without identifiable tophi (3, 4), and further prolonged in patients with tophaceous gout. This consideration was based on studies demonstrating that urate pool size is greater in patients with tophaceous gout than in patients with nontophaceous gout (18, 19), and consequently would take longer to deplete. We therefore chose to study patients treated for at least 5 years in the absence of clinically demonstrable tophi, and to exclude patients who did not have average serum urate levels <7 mg/dl while receiving therapy because they would not be likely to reach the desired outcome (depletion of urate deposits) or patients who had serum urate levels <7 mg/dl after withdrawal because they would not be at risk for recurrence of gouty arthritis or tophi.
The results of this study demonstrate that serum urate levels during ULT and after withdrawal of ULT are determinants of freedom from gouty symptoms after ULT withdrawal, although the limited number of postwithdrawal serum urate samples may be a limitation. These findings may have implications for clinical practice. First, serum urate levels should be lowered enough for a prolonged period of time to deplete the body's urate pool. Second, patients with average serum urate levels <7.0 mg/dl for at least a 5-year period can be regarded as likely to achieve a prolonged period without gout symptoms after ULT withdrawal. We believe that, in the first instance, the goal of ULT should be reduction of average serum urate levels to subsaturating levels for a prolonged period, and that subsequent ULT withdrawal could be considered if a substantially low serum urate level is achieved. In patients with tophi, primary treatment with ULT is necessary, but withdrawal thereafter is also tenable.
To achieve the results that we believe are possible, several conditions are necessary: a reduction of serum urate levels to the subsaturating range; compliance with ULT to achieve complete depletion of the body's urate pool; close followup of patients with gouty symptoms during ULT; and changes in lifestyle, avoiding factors known to promote hyperuricemia and gout (20), and taking actions that promote urate lowering, especially if withdrawal of ULT is considered. In this overall regard, we believe that patient compliance with ULT would be enhanced if, at the initiation of treatment, the likelihood of a determinate duration of ULT could be presented to patients in place of the prospect of life-long treatment.
The authors thank Dr. Michael A. Becker, University of Chicago (Chicago, IL) for his review of the manuscript and expert suggestions. We also thank Mrs. Rosario Lopez and Ms Inmaculada Iriondo for their help during the followup and monitoring of patients.
- 5A retrospective study of the relationship between serum urate level and recurrent attacks of gouty arthritis: evidence for reduction of recurrent gouty arthritis with antihyperuricemic therapy. Arthritis Rheum 2004; 51: 321–5., , .
- 10Gout and hyperuricemia. In: HarrisEDJr, editor. Kelley's textbook of rheumatology. 7th ed. Philadelphia: Elsevier Saunders; 2005. p. 1402–29., .