High frequencies of both hypouricaemia (22%) and hyperuricaemia (42%) have been described in HIV-infected individuals [1–3]. Whereas hypouricaemia has been attributed to increased renotubular loss , urate elevation has been suggested to result from increased cell turnover [3,4]. Aside from HIV infection itself, urate serum levels could potentially be altered by antiretroviral drugs. Didanosine, for example, causes hyperuricaemia  and urate measurements were even proposed as an adjuvant compliance index . This study examined whether other antiretroviral drugs could also influence urate levels.
After ethics committee approval, serum urate was prospectively measured in HIV-positive subjects at each visit. Consecutive subjects were included between January 2000 and February 2005 at our institution. Patients had to have no malignancies and were required not to be treated with tuberculostatic or urate-lowering agents. Using SPSS 13.01 (SPSS, Chicago, Illinois, USA), statistical effects were analysed by χ2-square and one-way ANOVA testing, and Pearson or Spearman correlation coefficients. Univariate variables were confirmed by multivariate regression. Serum urate is expressed is expressed in relation to the upper limit of normal (ULN, 70 mg/l).
Out of 280 subjects screened, 10 failed to meet entry criteria but 270 subjects (190 male) entered the study and a total of 2287 visits were analysed. Median age of patients at all visits was 40 years (range, 19–76), median body mass index (BMI) was 23.9 (range, 14.0–40.6). Patients were without antiretroviral treatment at 20.4% of visits and had an undetectable serum HIV load (< 50 copies/ml) at 58% of visits. The median CD4 cell count was 408 cells/μl (SD, 238). Urate did not correlate with the number of visits. Urate was ≥ ULN in 75 patients (27.8 %) at 323 visits (14.2 %). Minimal and maximal urate levels were 0.3×ULN and 1.9×ULN.
In univariate statistical analysis, women had lower urate [0.64×ULN (SD, 0.19)] than men [0.84×ULN (SD, 0.21)] (P < 0.001). Urate was ≥ ULN at 37 of 895 visits (4.1%) for women and at 286 of 1392 visits (20.5%) in men (P < 0.001). Urate was positively correlated with the BMI and the serum creatinine, but not correlated with the CD4 cell count and HIV plasma load.
Among all visits, urate levels were identical in patients with and without antiretroviral treatment [mean urate 0.76×ULN (SD, 0.22) in both populations). The use of stavudine, however, was associated with an increase in urate (Fig. 1a). The use of didanosine was also associated with an elevation of urate at all time points [0.88×ULN (SD, 0.20) among all patients treated with didanosine]. Patients treated with the combination of didanosine and stavudine had significantly higher levels than those treated with didanosine or stavudine plus another nucleoside reverse transcriptase inhibitor (NRTI) (Fig. 1b). Exposure to the purine analogue tenofovir disoproxil fumarate without didanosine was associated with diminished urate levels [0.73×ULN (SD, 0.20); P = 0.03]. When tenofovir was added to didanosine, urate levels were lower [0.74×ULN (SD, 0.20); P < 0.001] than with didanosine without tenofovir [0.91×ULN (SD, 0.18] and similar to controls without antiretroviral drugs. Abacavir did not change urate. Zalcitabine and protease inhibitors were also associated with increased urate [0.89×ULN (SD, 0.13) and 0.83×ULN (SD, 0.23), respectively], both compared with patients without antiretroviral therapy and compared with individuals treated with other antiretroviral drugs (P < 0.001 for all comparisons).
Multivariate regression analysis confirmed the independent contribution to urate elevation of male gender (P = 0.009), months of stavudine treatment (P < 0.001), increased serum creatinine (P < 0.001), elevated BMI (P = 0.002), protease inhibitor use (P < 0.001), age (P = 0.001), and didanosine exposure (P = 0.048). Gender and stavudine treatment were the strongest predictors of the variability in urate, (r2 = 0.247), whereas the contributory effect of the other variables was only 8%. HIV viral load was not an explanatory variable of urate values.
Multiple gene products and epigenetic factors influence urate levels in humans [7,8]. We confirmed the known factors for urate elevation in HIV-uninfected persons (gender, age, renal function and BMI) in our cohort but found a quantitatively important additional influence of some antiretroviral agents on urate. In accord with earlier trials [9,10], we identified an association between the use of didanosine and elevated urate. Didanosine (usual dosage 400 mg/day) is a purine analogue that can be degraded into urate and so may elevate urate pools, normally averaging approximately 1200 mg [10,11]. However the lack of an effect of the purine analogue abacavir (dosage 600 mg/day) suggests that the degradation of purine NRTI into urate cannot fully explain our findings. Tenofovir may downregulate urate by inhibiting purine nucleoside phosphorylase, an ubiquitous enzyme that degrades purines .
Urate may also be increased through the mitochondrial toxicity of some NRTI. Mitochondrial dysfunction may increase the formation of lactate, which competes with urate for tubular secretion in the kidney . Respiratory chain failure causes ATP depletion, which increases urate production in the purine nucleotide cycle [8,13]. This mechanism is the basis for the hyperuricaemia in some metabolic myopathies and may provide an explanation for the association between dideoxynucleoside analogues (stavudine, didanosine or zalcitabine) and elevated urate.
We also found a small but independent effect of protease inhibitors on urate elevation. Boosted protease inhibitors were previously associated with an increased incidence of gout . Interestingly, gout is associated with hyperlipidaemia, insulin resistance and central adiposity in HIV-uninfected persons , symptomatology that is also associated with protease inhibitor therapy .
The observational nature of our data and the small size of the cohort limit the conclusions that can be drawn. The study also selected against medications known to contribute to or lower hyperuricaemia. The number of subjects excluded from the analysis for these reasons, however, was low. Of note, we did not assess the incidence or prevalence of gout. Gout has been described in HIV-infected patients [16,17] and the annual incidence estimated to be 0.5%, an order of magnitude higher than the incidence in the normal population [8,14]. This indicates that the effect of antiretroviral drugs on urate may be of clinical relevance.
Hyperuricaemia is of multifactorial origin in HIV-infected patients. Urate levels and gout require clinical attention, research into the pathophysiological mechanisms and systematic validation in randomized trials and large cohorts.
1. Maesaka JK, Cusano AJ, Thies HL, Siegal FP, Dreisbach AW. Hypouricemia in acquired immunodeficiency syndrome. Am J Kidney Dis 1990; 15:252–257.
2. Medina-Rodriguez F, Guzman C, Jara LJ, Hermida C, Alboukrek D, Cervera H, et al
. Rheumatic manifestations in human immunodeficiency virus positive and negative individuals: a study of 2 populations with similar risk factors. J Rheumatol 1993; 20:1880–1884.
3. Manfredi R, Mastroianni A, Coronado OV, Chiodo F. Hyperuricemia and progression of HIV disease. J Acquir Immune Defic Syndr Hum Retrovirol 1996; 12:318–319.
4. Manfredi R, Chiodo F. Longitudinal assessment of serum urate levels as a marker of HIV disease progression. Int J STD AIDS 1998; 9:433–434.
5. Lambert JS, Seidlin M, Reichman RC, Plank CS, Laverty M, Morse GD, et al
. 2′,3′-dideoxyinosine (ddI) in patients with the acquired immunodeficiency syndrome or AIDS-related complex. A phase I trial. N Engl J Med 1990; 322:1333–1340.
6. Richardson D, Liou SH, Kahn JO. Uric acid and didanosine compliance in AIDS clinical trials: an analysis of AIDS Clinical Trials Group protocols 116A and 116B/117. J Acquir Immune Defic Syndr 1993; 6:1212–1223.
7. Takahashi S, Moriwaki Y, Tsutsumi Z, Yamakita J, Yamamoto T, Hada T. Increased visceral fat accumulation further aggravates the risks of insulin resistance in gout. Metabolism 2001; 50:393–398.
8. Becker MA, Roessler BJ. Hyperuricemia and gout. In The Metabolic and Molecular Basis of Inherited Disease
edn. Edited by Scriver CR, Beaudet AL, Sly WS, Valle D. New York McGraw Hill; 1995:Ch. 49.
9. Lambert JS, Seidlin M, Reichman RC, Plank CS, Laverty M, Morse GD, et al
. 2′,3′-Dideoxyinosine (ddI) in patients with the acquired immunodeficiency syndrome or AIDS-related complex. A phase I trial. N Engl J Med 1990; 322:1333–1340.
10. Richardson D, Liou SH, Kahn JO. Uric acid and didanosine compliance in AIDS clinical trials: an analysis of AIDS Clinical Trials Group protocols 116A and 116B/117. J Acquir Immune Defic Syndr 1993; 6:1212–1223.
11. Ray AS, Olson L, Fridland A. Role of purine nucleoside phosphorylase in interactions between 2′,3′-dideoxyinosine and allopurinol, ganciclovir, or tenofovir. Antimicrob Agents Chemother 2004; 48:1089–1095.
12. Ray AS, Olson L, Fridland A. Role of purine nucleoside phosphorylase in interactions between 2′,3′-dideoxyinosine and allopurinol, ganciclovir, or tenofovir. Antimicrob Agents Chemother 2004; 48:1089–1095.
13. Mineo I, Tarui S. Myogenic hyperuricemia: what can we learn from metabolic myopathies? Muscle Nerve 1995; 3:S75–S81.
14. Creighton S, Miller R, Edwards S, Copas A, French P. Is ritonavir boosting associated with gout? Int J STD AIDS 2005; 16:362–364.
15. Carr A, Samaras K, Burton S, Law M, Freund J, Chisholm DJ, Cooper DA. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 1998; 12:F51–F58.
16. Mehlhaff DL, Stein DS. Gout secondary to ritonavir and didanosine. AIDS 1996; 10:1744.
17. Disla E, Stein S, Acevedo M, Cuppari G. Gouty arthritis in the acquired immunodeficiency syndrome. An unusual but aggressive case. Arthritis Rheum 1995; 38:570–572.