Pre-treatment HIV severity, not HAART associated with peripheral neuropathy

The incidence of peripheral neuropathy amongst HIV-positive individuals has declined since the introduction of HAART, according to data from the US HOPS study published in the on-line version of the January 1^st edition of the Journal of Infectious Diseases.

The HOPS investigators also established that host factors, and the severity of HIV disease before the commencement of antiretroviral therapy were independently associated with the risk of developing peripheral neuropathy in the first year of HAART. Drugs from all the three main classes of antiretrovirals were associated with peripheral neuropathy, but the investigators found that after 12 months, HAART was protective against any further risk of developing peripheral neuropathy.

Study background

Before HAART became available peripheral neuropathy was a common and increasingly prevalent condition amongst HIV-positive individuals. Several studies showed an association between peripheral neuropathy and the severity of HIV disease. Since HAART became available, peripheral neuropathy has been associated with the use of nucleoside analogues, and the mitochondrial toxicities which this class of drugs can cause has been one of the causes suggested for the occurrence of this side-effect.

Investigators from the HOPS study (an ongoing cohort study which has recruited over 7,000 HIV-positive patients in seven US cities since 1992), had previously shown that a nadir CD4 cell count below 200 cells/mm³ was associated with fat loss whilst taking HAART. They wished to see which factors were associated with peripheral neuropathy in both the pre-HAART and HAART eras.

Study design – which factors caused peripheral neuropathy in the pre-HAART and HAART eras?

A little under 4,400 patients were included in the investigators’ analysis. Individuals with peripheral neuropathy were identified by a retrospective chart review. Data were also gathered from medical records on age, race, nadir CD4 cell count, viral load prior to commencing HAART, and antiretroviral treatment history.

The investigators then performed univariate and multivariate logistic regression analyses.

Results – HAART has led to a reduction in peripheral neuropathy

The incidence of peripheral neuropathy peaked at 25 cases per 100 patient years in 1995 before falling sharply with the increased use of HAART to five cases per 100 patient years in 2002. Nadir CD4 cell count was associated with an increased incidence of peripheral neuropathy throughout the period of the study, even after the widespread use of HAART.

In multivariate logistic regression analysis the investigators found that the following factors were associated with peripheral neuropathy:

• Age above 40 years at the time of nadir CD4 cell count (p = 0.02).
• White race (p = 0.02).
• Diabetes mellitus (p = 0.01).
• Nadir CD4 cell count below 50 cells/mm³ (p = 0.002).
• Nadir CD4 cell count 50 – 199 copies/mm³ (p = 0.01).
• First viral load above 10,000 copies (p = 0.005).

In the first year of antiretroviral therapy, the following anti-HIV drugs were associated with peripheral neuropathy:

• ddI (p
• d4T above 40mg twice daily (p
• Indinavir (p
• Ritonavir (p
• Nelfinavir (p = 0.002).
• Saquinavir (p
• Nevirapine (p = 0.02).

Furthermore, the investigators found that individuals with a nadir CD4 cell count below 50 cells/mm³ and a first viral load measurement above 10,000 copies/ml were at the greatest risk of developing peripheral neuropathy during the first year of treatment for all regimens.

After the first year of HAART the investigators failed to find any association between the continued use of any antiretroviral drug (except with a regimen of higher-dose d4T, 3TC and efavirenz) and an increased risk of peripheral neuropathy. Indeed, they found that continuing HAART use reduced the risk of peripheral neuropathy developing.

A case for the earlier initiation of HAART?

“We found that, since 1997, the incidence of [peripheral neuropathy] has been progressively decreasing. If [peripheral neuroparhy] did not develop within the first year of use of any antiretroviral agent, it was less likely to occur in subsequent years”, write the investigators.

They add, that with the exception of higher-dose d4T, HIV disease severity measured by nadir CD4 cell count and viral load at the time of first measurement “were a nearly equivalent or stronger predictor of developing [peripheral neuropathy] than any type of antiretroviral drug used.”

Indeed, they suggest that patients who develop peripheral neuropathy whilst taking HAART have been “predisposed to it because of injury to nerve tissue” caused by severe HIV disease.

The investigators note with some surprise that several protease inhibitors were revealed by their study to be associated with peripheral neuropathy. They suggest that this could be because they were acting “synergistically” with nucleoside analogues to cause the side-effect.

HAART rather than leading to an increase in the incidence of peripheral neuropathy, has led to a reduction in its incidence. As patients who develop the condition had more severe HIV disease prior to experiencing peripheral nerve damage, the investigators believe that their findings support the earlier initiation of HAART. However, these findings need to be viewed in the context of current treatment guidelines: whilst a substantial proportion of participants in this cohort initiated HAART at CD4 cell counts below 200 cells/mm³ for historical reasons, the findings do not support changes to current US and UK recommendations that treatment should commence before the CD4 cell count falls below 200 cells/mm³.