Introduction
Antiretroviral treatment with the HIV-protease inhibitor indinavir is associated with urological complaints. Urological complications occur in approximately 8% of patients on an indinavir dosage of 800mg three times a day (tid), and vary from nephrolithiasis (3-4% of all patients) to symptomatic crystalluria with dysuria or back pain (7% of all patients) [1,2]. These adverse reactions can usually be resolved by hydration and a short treatment interruption of 1 to 3 days, but may occasionally lead to acute renal failure, which requires urological intervention to relieve the obstruction [3-5]. In a majority of indinavir users with recurrent renal colic episodes, treatment with indinavir is abandoned [6].
Independently of the indinavir concentration, up to 20% of absorbed indinavir is excreted into the urine, more than half of which is excreted in the intact form [1,7]. Because of this route of elimination, high indinavir plasma concentrations result in high indinavir urine concentrations, depending on the urine volume, which may induce indinavir precipitates if urinary solubility is exceeded. Accordingly, the calculi voided by patients are of a radiolucent, gelatinous nature and consist of indinavir base monohydrate (intact indinavir) [2,5,6,8]. Apart from the dilution of urine through a higher fluid intake, to prevent and relieve urological symptoms, the avoidance of high indinavir plasma concentrations may be an important measure for reducing the rate of urological complications. In this study, we investigated the relationship between indinavir plasma concentrations and urological complaints, which occurred during indinavir treatment in HIV-infected patients.
Methods
Case series
The patient population of the Erasmus University Medical Centre Rotterdam (EMCR) comprises approximately 500 adult HIV-infected patients, 104 (21 women, 20%) of whom currently receive indinavir treatment. Indinavir-treated patients, who presented with overt urological complaints between January and December 1997, were enrolled in the study. These patients received indinavir at a dosage of 800mg tid as part of the highly active antiretroviral therapy (HAART) for HIV infection. Urological complaints were defined as episodes of renal co!ic, flank pain, dysuria or haematuria. Information on viral load, assessed by HIV-RNA polymerase chain reaction (Amplicor HIV-1 monitor; Roche, Basel, Switzerland), urinalysis, routinely performed by semi-quantitative dipstick test (Rapignost7 total-screen L dipstick, Behringwerke AG, Marburg, Germany), and concomitant medication was retrieved retrospectively from patient records. All assessments were done during the regular outpatient clinic visit of each patient, which did not necessarily coincide with the manifestation of urological symptoms.
Control patients
Indinavir concentrations in patients with urological complaints were compared with a standard concentration curve as derived from the full pharmacokinetic curves of 14 HIV-infected patients (two women) from the University Hospital Nijmegen, taking 800mg of indinavir tid (Fig. 1). This control group had an average age of 41.4±7.3 years (range 31-52) and an average weight of 72.1±9kg (range 61-89). After an average duration of indinavir therapy of 6.9±3.3 months (range 1-12), 10 patients had less than 200 copies per ml HIV-RNA. All control patients were free of overt signs of intoxication (including urological complications). The pharmacokinetic parameters of this group (area under the concentration time curve, AUC; peak concentration, Cmax; and minimum concentration, Cmin) were consistent with the data as presented in the Crixivan® product monograph [1].
Sample collection
A blood sample was collected from each patient with overt urological complaints at variable time points after the last ingestion of indinavir. The time of the last ingestion was based on patient interview. The time lapse between the occurrence of urological symptoms and indinavir concentration assessment was not determined. In order to attain comparable indinavir concentrations between cases and controls, samples were evaluable only if drawn at least 1.5h after the last intake, to avoid variance caused by absorption, and less than 8h after the last intake, as there were no measurement points for control patients after 8h. Seven millilitres of blood was drawn from an antecubital vein into a lithium heparin vacutainer, from which plasma was separated by centrifugation and was stored in the refrigerator until transfer to the Department of Clinical Pharmacy at the University Hospital Nijmegen.
Analytical methods
Indinavir concentrations in the plasma were determined using a validated high performance liquid chromatography (HPLC) assay, which has been described previously [9]. In brief, plasma is mixed with acetonitrile before centrifugation. The subsequent supernatant is evaporated to dryness, the residue of which is dissolved in eluent and injected into the HPLC system. The system contains an octadecyl column with a mobile phase of acetonitrile-water and uses ultraviolet detection at 210nm.
The results of the indinavir plasma concentration measurements were expressed as ‚concentration ratios‚. The ratio was calculated by dividing the indinavir concentration in each patient with urological complaints by the average concentration of the control group at the same or nearest time interval after the ingestion of indinavir. For this purpose, the concentration means of the control group were extrapolated to 15min time intervals so that the maximum time difference between controls and cases was 7.5min.
Statistical analysis
Continuous variables such as age, weight and duration of therapy were tested with a Mann-Whitney U test. The average concentration ratios of patients with urological complaints was analysed for statistical significance against the expected concentration ratio of 1.0 by calculating the 95% confidence intervals around the mean ratio of patients with urological complaints, based on a student distribution, after normalizing the distribution by log transformation. A ratio of 1.0 indicated an indinavir concentration equalling the mean concentration of the control population at the same time interval after the ingestion of indinavir. A sensitivity analysis was also performed, in which the upper 95% confidence limit instead of the mean of the control group was chosen as the denominator in the ratio.
Results
Seventeen patients with overt urological complaints, 12 men and five women, representing 16% of the population treated with indinavir at the EMCR, were included in the study. Their mean age was 37.4±8.6 (range 27-57), their mean weight was 65.6±9.2kg (range 46-85) and they had been on indinavir treatment for an average duration of 7.1±4.8 (range 0.4-14.8) months (Table 1). In 15 (88%) patients the plasma HIV-RNA had been reduced to less than 500 copies per ml.
Table 2 provides an overview of the results. Twelve (71%) cases occurred during the spring and summer. Fifteen patients had an evaluable indinavir concentration measurement. Concentration ratios ranged from 0.55 to 11.49 (Fig. 1). Except for one patient, all (93%) had a concentration above the mean of the controls and 12 (80%) patients had a concentration above the upper 95% confidence limit of the control population (Fig. 1). The log-corrected mean ratio was 2.64 (95% confidence interval 1.68-4.14), which was significantly different from the expected ratio of 1.0. In the sensitivity analysis the log-corrected mean ratio was 1.96 (95% confidence interval 1.23-3.03).
The average urine pH was 5.9 (n=11, range 5-9). Urinalysis (semi-quantitative dipstick test) further revealed the presence of leucocytes (i.e. >15 leucocytes/μl) in 10 out of 11(91%) cases in the absence of infection (i.e. dipstick test negative for bacteria), and erythrocytes (i.e. >10 erythrocytes/μl) in eight out of 11 (73%) cases. Nine out of 10 (90%) patients showed slight proteinuria of 0.15-0.30g/l. All symptoms cleared without urological intervention. Concomitant medication included co-trimoxazole, diclofenac, haloperidol, cetirizine, dapsone, megestrol, metoprolol, fluconazol, clindamycin, psylliumseed and butylscopolamin, none of which is known to affect indinavir concentrations significantly. In one patient the occasional use of heroin could not be excluded.
In six patients with urological symptoms and an indinavir concentration higher than the upper 95% confidence limit (ratio range 2.25-11.49), the dose of indinavir was reduced to 600mg tid. Repeat indinavir concentration ratios, measured at 58±44 days after dose reduction, approached 1.0 (range 0.63-1.37) in all six patients (Table 2). Five out of six remained free of urological symptoms and one patient remained free of urological symptoms when having extra fluid intake. All six continued to have viral loads of less than 500 copies per ml after a reduction in the dose of indinavir for follow-up periods of 5-16 months. As for the other patients, six discontinued indinavir treatment, five because of gastrointestinal complaints, three of whom had first tried dose reductions other than 600mg tid, and one because of lipodystrophy. Two had modified the dose to a twice a day regimen and suffered residual urological complaints. One patient continued indinavir at 800mg tid without major problems, but HIV-RNA had increased to 2560 copies per ml after 8.5 months. There might have been a problem with compliance in this patient. As for the two patients with non-evaluable indinavir concentration assessments, one reduced the dose to 400mg tid, but later appeared to have a low indinavir plasma concentration and high viral load for which reason indinavir was stopped. The other patient had no dose reduction, but indinavir treatment was withdrawn because of gastrointestinal intolerance.
Discussion
The relationship between indinavir exposure and antiretroviral effect has been reported, but an exposure-toxicity relationship has not yet been described [1,10,11]. The present results suggest an association between indinavir plasma concentrations and the occurrence of urological complaints during indinavir-containing antiretroviral therapy. Patients with urological complaints in this study consistently presented with high indinavir plasma concentrations in comparison to a control group without urological complaints.
Although this study did not analyse urinary crystals, it was shown that complaints often coincided with sterile leucocyturia, haematuria and proteinuria, consistent with a pattern of mucosal irritation caused by (micro)lithiasis. As indinavir urinary concentrations and clearance were not measured in this study, it is not known how these relate to high indinavir plasma concentrations and the occurrence of indinavir-associated urological symptoms The process of indinavir crystal formation has been reported to be enhanced by high urinary pH and low urinary volume [2,12]. All patients were advised to drink at least 1.5l per day. Although most patients claimed that their intake was maximal, precise intakes and urinary volumes are not known. In fact, many urological complaints occurred during periods of hot weather when optimal hydration may have been difficult to achieve [13]. Therefore, we cannot exclude urinary volume as the cause of urological complaints in our patients.
High indinavir plasma concentrations may have a multifactorial origin. The oral absorption of indinavir is pH dependent and subsequent elimination depends on the liver cytochrome P450 system (isoenzyme CYP3A4) and urinary clearance [12,14]. None of the patients used any drugs known to have significant effects on indinavir pharmacokinetics. Two patients had a chronic active hepatitis B infection, with possibly impaired liver clearance of indinavir. None of the patients was known to have impaired renal function, but renal clearance may have been reduced by insufficient hydration. It is not known how this would affect the plasma concentration of indinavir, as the kidneys play a minor part in the clearance of indinavir. The high indinavir plasma concentrations could not be attributed to low weight, as the patients with urological complaints and patients from the control group had similar weights. A recent study [15] observed an influence of sex on the pharmacokinetics of indinavir, which would suggest a higher propensity for women to develop urological complications.
Importantly, patients remained free of urological complaints after a dose reduction to 600mg tid. Subsequent indinavir plasma concentrations fell within the limits of the control group and viral suppression was maintained during follow-up. Indinavir dose reduction may thus be a useful and safe control measure for urological complaints associated with high plasma indinavir concentrations. Although indinavir concentration may not be predictive of urological complaints, it might be worthwhile to perform a cost-benefit analysis of monitoring indinavir concentrations in all patients in the prevention of urological complaints.
In this pilot study indinavir concentrations were compared with an historical control group. More research is needed to confirm the findings and to ascertain the long-term effects of dose reduction in patients with high indinavir plasma concentrations. Further research should also be aimed at finding an explanation for the high incidence of urological symptoms seen in our patient population. The cumulative incidence was 16%, which is much higher than reported so far [1,2]. This might reflect population characteristics or may be a consequence of close patient monitoring [13,15]. In addition, the following points should be recognized. First, urological complaints occurring during indinavir treatment may not always be attributable to indinavir. Second, the possible causes of increased indinavir plasma concentrations that can be removed should be taken into account before applying any dose adjustment. Evidently, dose reductions should only be applied on the basis of indinavir plasma concentration measurements. Finally, it should be noted that this study compared single-point measurements of indinavir plasma concentration in individual patients with the average concentration of an independent control group. The results of this control group were believed to be a proper reference, but it is not known how single-point measurements of indinavir plasma concentrations relate to other pharmacokinetic parameters.
Conclusion
This study showed that urological complaints, occurring during indinavir treatment, are associated with elevated indinavir plasma concentrations. If urological complaints persist despite maximal fluid intake and the indinavir plasma concentration is above the upper 95% confidence limit, dose adjustment may be considered, so that patients can safely continue taking indinavir. In our patients, a dose reduction to 600mg tid was found to be effective. More research is needed to elucidate further the relationship between indinavir plasma concentrations and urological complaints and to determine the long-term effects of dose reductions. The influence of climate on indinavir pharmacokinetics and indinavir-related urological complaints should be clarified.
Acknowledgement
We are grateful to Drs M.C.J.M. Sturkenboom, Pharm D, and B.H.Ch. Stricker, MB PhD, for their critical comments.
References
1. Crixivan® (indinavir sulfate) product monograph. Whitehouse Station, NJ, USA: Merck and Co., 1996.
2. Kopp JB, Miller KD, Mican JA, et al.: Crystalluria and urinary tract abnormalities associated with indinavir. Ann Intern Med1997, 127:119-125.
3. Berns JS, Cohen RM, Silverman M, Turner J: Acute renal failure due to indinavir crystalluria and nephrolithiasis: report of two cases. Am J Kidney Dis1997, 30:558-560.
4. Sutherland SE, Reigle MD, Seftel AD, Resnick MI: Protease inhibitors and urolithiasis. J Urol 1997, 158:31-33.
5. Bruce RG, Munch LC, Hoven AD, et al.: Urolithiasis associated with the protease inhibitor indinavir. Urology 1997, 50:513-518.
6. Gentle DL, Stoller ML, Jarrett TM, Ward JF, Geib KS, Wood AF: Protease inhibitor-induced urolithiasis. Urology 1997, 50:508-511.
7. Balani SK, Woolf EJ, Hoagland VL, et al.: Disposition of indinavir, a potent IIIV-1 protease inhibitor, after an oral dose in humans. Drug Metab Dispos 1996, 24:1389-1394.
8. Daudon M, Estepa L, Viard JP, Joly D, Jungers P: Urinary stones in HIV-1-positive patients treated with indinavir. Lancet 1997, 349:1294-1295.
9. Burger DM, De Graaff M, Wuis EW, Koopmans PP, Hekster Y: Determination of indinavir, an HIV-protease inhibitor, in human plasma by reversed-phase high-performance liquid chromatography. J Chromatogr B 1997, 703:235-241.
10. Stein DS, Fish DG, Bilello JA, Preston SL, Martineau GL, Drusano GL: A 24-week open-label phase I/II evaluation of the HIV protease inhibitor MK-639 (indinavir). AIDS 1996, 10:485-492.
11. Burger DM, Koopmans PP, Brinkman K, et al.: Therapeutic drug monitoring of the HIV protease inhibitor indinavir. 37th International Conference on Antimicrobial Agents and Chemotherapy. Toronto, September 1997 [abstract A-19].
12. Lin JH, Chen IW, Vastag KJ, Ostovic D: pH-dependent oral absorption of L-735,524, a potent HIV protease inhibitor, in rats and dogs. Drug Metab Dispos 1995, 23:730-735.
13. Bach MC, Godofsky EW: Indinavir nephrolithiasis in warm climates. J Acquir Immune Defic Syndr Hum Retrovirol 1997, 14:296-298.
14. Chiba M, Hensleigh M, Lin JH: Hepatic and intestinal metabolism of indinavir, an HIV protease inhibitor, in rat and human microsomes. Major role of CYP3A. Biochem Pharmacol 1997, 53:1187-1195.
15. Burger DM, Koopmans PP, Brinkman K: Influence of gender on indinavir pharmacokinetics. Sixth European Conference on Clinical Aspects and Treatment of HIV-Infection. Hamburg, October 1997 [abstract 252].
© 1999 Lippincott Williams & Wilkins, Inc.