Atazanavir associated with significant risk of kidney stones and renal impairment

This article is more than 9 years old. Click here for more recent articles on this topic

Studies presented at last week’s British HIV Association (BHIVA) conference found that kidney stones were four times more common in people taking the protease inhibitor atazanavir than in people taking other drugs. However they were still relatively uncommon – about one case in 50 patients who were being investigated for possible kidney stones turned out to have them.

A parallel study found a significant association between atazanavir and signs of renal failure, though it also found associations between impairment and other protease inhibitors too.

Atazanavir and kidney stones

In the first study Neesha Rockwood and colleagues from the Chelsea and Westminster Hospital in  west London did a retrospective case note review of patients attending the clinic between May 2006 and February 2010, singling out those who had had an abdominal X-ray or CT scan, a renal ultrasound scan, or an intravenous urogram (in which a contrast medium is injected and then x-rays are taken of it being excreted via the urinary system). These four procedures indicate that kidney stones are being investigated as a possible diagnosis.

She only included in this study patients who underwent one of the above investigations and were taking either the non-nucleoside drug (NNRTI) efavirenz (Sustiva, and in Atripla) or one of the protease inhibitors (PIs) atazanavir, lopinavir or darunavir (Reyataz, Kaletra, Prezista).


kidney stone

Stone-like lumps that develop in the kidneys. Made up of crystals which form as the kidneys clear waste products from the blood. 


Relating to the kidneys.

multivariate analysis

An extension of multivariable analysis that is used to model two or more outcomes at the same time.

protease inhibitor (PI)

Family of antiretrovirals which target the protease enzyme. Includes amprenavir, indinavir, lopinavir, ritonavir, saquinavir, nelfinavir, and atazanavir.

exclusion criteria

Defines who cannot take part in a research study. Eligibility criteria may include disease type and stage, other medical conditions, previous treatment history, age, and gender. For example, many trials exclude women who are pregnant, to avoid any possible danger to a baby, or people who are taking a drug that might interact with the treatment being studied.

There have been case reports of kidney stones in patients taking atazanavir and analysis of these stones showed that they consisted of drug that has dissolved out of urine: atazanavir is less soluble in alkaline fluids, which is why it must also not be taken with antacids.

A sub-analysis excluded people with previous exposure to the PI indinavir (Crixivan), which is now hardly used because of its tendency to cause kidney stones. In all over 6,000 patients’ notes were reviewed. Of these 1206 people had taken atazanavir, 2803 efavirenz, 828 lopinavir and 818 darunavir, and 206 had previously taken indinavir.

Two per cent of patients taking atazanavir developed kidney stones, at an annual rate of 0.73% of patients per year (an annual event rate of one case per 137 patients). In contrast 0.54% of patients on the other three drugs developed kidney stones, an event rate of 0.19% or one event per 526 patients.

When patients with previous indinavir exposure were excluded, this brought the percentage of patients with kidney stones on atazanavir to 1.5%, with an annual event rate of 0.46% (one in 217 patients a year). The annual event rate in patients on the other three drugs who had not taken indinavir was 0.12% (one in 833 patients a year).

This meant that patients on atazanavir were 3.85 times as likely to get kidney stones as patients on the other drugs, and although excluding previous indinavir users brought the rates down, the relative difference remained the same.

Patients on darunavir were somewhat more likely to get kidney stones than patients on lopinavir or efavirenz (annual event rates 0.45%, 0.19% and 0.15% respectively).

Patients who had kidney stones were ten times more likely than others to have histories of chronic renal impairment and their levels of the liver waste product bilirubin (which causes jaundice, another side-effect associated with atazanavir) were on average twice as high.

Dr Rockwood said her study probably underestimated the prevalence of kidney stones in patients as some stones were transparent to X-rays and cases in patients who had transferred from other hospitals could have been missed. Atazanavir could have consequences after it was stopped: in one case, a kidney stone containing atazanavir was found in a patient who had stopped the drug 21 months previously.

Protease inhibitors and renal impairment

In a second study, this time of atazanavir’s effect on the kidneys’ ability to filter waste products, the same team found a 52% increased risk of this kind of chronic kidney disease in patients on atazanavir compared to the whole patient group. However this was lower than the risk associated with lopinavir (69% increased risk) and there was some association with darunavir use too, though this became statistically insignificant in multivariate analysis.

Chronic kidney disease was defined as an estimated glomerular filtration rate (eGFR) of less than 60 millilitres per minute per 1.72m2. EGFR is based on levels of the waste product creatinine in the blood and defines the amount of blood the kidneys are able to filter.

This study assessed the rate of kidney impairment in patients prescribed the same four drugs as in the previous study. Out of 2115 patients (78% male, 60% with an undetectable viral load) prescribed these therapies, 386 (18%) developed renal impairment as defined above.

Women were over 50% more likely to develop renal failure, and older patients considerably more likely, as seen in other studies (kidney failure becomes more frequent with age). Patients taking atazanavir, darunavir or lopinavir were, respectively, 27%, 53% and 71% more likely to develop kidney failure. Other risk factors included a history of tenofovir use (68% more likely) and chronic hepatitis B infection (21% more likely).

The risk associated with current tenofovir use appeared to accumulate with time (9% raised risk for each year on tenofovir).

Viral load or CD4 count at baseline, ethnic origin and hepatitis C co-infection were not associated with kidney failure.

Patients currently taking efavirenz had a 40% reduced risk of renal impairment compared with the average patient, though this advantage disappeared if they stopped efavirenz.

In multivariate analysis, taking lopinavir was associated with a 69% higher risk of kidney impairment than taking efavirenz. Patients taknig atazanavir had a 52% higher risk of renal impairment than patients taking efavirenz.. The risk with darunavir (versus efavirenz) was raised 31% but this became statistically non-significant (i.e. the effect of other variables became more important).

There was some recovery after the protease inhibitors were withdrawn. Half the patients who reached an eGFR below 60ml/min/1.73m2 had an eGFR above this figure a year after stopping. Patients who were on tenofovir had a 30% greater rate of renal recovery if they stopped this drug in addition to their protease inhibitors; however stopping tenofovir by itself was not associated with significant improvement.

Atazanavir side effects are likely to come under more scrutiny in the future as the London HIV Consortium has recently made an agreement with HIV clinicians that it will be the first PI of choice in patients who need to take this class of drugs, and some patients will be encouraged to switch to it.


Rockwood N et al. Atazanavir exposure is associated with increased rate of renal stones compared with efavirenz, lopinavir and darunavir. Seventeenth annual BHIVA Conference, Bournemouth. Abstract O4. 2011.

Rockwood N et al. A comparative analysis of risk factors associated with efavirenz, darunavir/ritonavir, lopinavir/ritonavir, atazanavir/ritonavir and renal impairment. Seventeenth annual BHIVA Conference, Bournemouth. Abstract O1. 2011.

Related topics