One quarter of people starting antiretroviral therapy in Zambia between 2004 and 2007 had impaired kidney function, and even mild impairment of kidney function increased the risk of death by 40% to 70%, according to findings presented this week at the 2008 HIV Implementers’ Meeting in Kampala, Uganda.
The findings are of particular significance because Zambia is one of the countries in Africa which is now moving towards first-line treatment with tenofovir, a drug which carries a higher risk of kidney toxicity in people who already have impaired kidney function, also known as renal insufficiency.
Capacity for diagnosing and managing impaired kidney function is limited outside major hospitals in developing countries.
Renal insufficiency may be caused directly by HIV damage to the kidney tubules, a condition known as HIV nephropathy, or by opportunistic infections or the use of drugs toxic to the kidneys. Cohort studies in the United States have shown that the risk of death in women is increased when renal insufficiency is present, even after controlling for the degree of immune suppression.
However, HIV suppression with antiretroviral therapy has been associated with improved kidney function in people with more advanced renal impairment, particularly those with low CD4 cell counts (see report)
The present study, presented by Carolyn Bolton of the Center for Infectious Disease Research in Lusaka, Zambia, looked at renal status in 25,249 patients starting antiretroviral therapy between May 2004 and September 2007 at 18 treatment sites run by PEPFAR grantees in Zambia. Tenofovir treatment began to be introduced in July 2007, but the number of tenofovir treated patients in this study was very modest.
Patients were eligible to start treatment if they had a CD4 cell count below 200 cells/mm3 or WHO stage 3 or 4 disease with a CD4 cell count below 350 cells/mm3. Data on patients were comprehensive and collected through a computerised system, making analysis relatively easy.
Patient vital status had been ascertained by patient tracing and family reporting in the case of death, and renal function was assessed by calculating the Cockcroft-Gault score, an equation which uses a patient’s age, weight and serum creatinine to calculate creatinine clearance (click here for a calculator).
Of the 25,249 patients studied 66% had normal renal function, 25% had mild renal insufficiency, 8% had moderate renal insufficiency and 1% had severe renal insufficiency. Mild renal insufficiency was defined as creatinine clearance of 61-90ml/min.
After an average of 720 days of follow-up, 19% of 560 patients with moderate renal insuffiency had died, compared to 12% of 1546 patients with mild renal insufficiency had died and 7% of those with normal kidney function. Almost 40% of those with severe renal insufficiency had died within 240 days of starting treatment.
The relative risk of death, after adjusting for CD$ count, WHO disease stage and haemoglobin, was 3.7 after 90 days in those with severe impairment, 1.9 in those with moderate impairment and 1.4 in those with mild impairment.
The risk of death was greater during the first 90 days of treatment: 1.7-fold greater for those with mild renal insufficiency, 2.3 fold for those with moderate insufficiency and 4.3 fold for those with severe impairment. All relative risks were statistically significant.
Multivariate analysis showed that female sex (AOR 1.2), haemoglobin 2 (AOR 1.7) were significant risk factors.
However the study is limited by the lack of longitudinal follow-up on creatinine clearance and the lack of information on cause of death, said Carolyn Bolton. Nevertheless, she noted, the study had found decreased survival even in those with mild renal insufficiency.
“Zambia only has one site where dialysis can be performed, so we need to limit renal failure as far as possible,” she told the conference.
Bolton C et al. Renal insufficiency and risk of death among HIV-infected adults initiating antiretroviral therapy in Lusaka, Zambia. 2008 HIV Implementers’ Meeting, Lusaka, Zambia, abstract 532.