• 2008/2009 Conference Coverage
• Previous Conference Coverage
• Upcoming Conferences
• Conference Reports From Other Sources

Coverage provided by Howard Grossman, M.D.

July 10, 2001

• Switching Protease Inhibitor (PI) to Nevirapine (NVP) Leads to a Better Lipid Profile Than Switch to Efavirenz (EFV) (Poster 484)
  Authored by Imperiale, S.; Carlier, H.; Boehringer Ingelheim, Ridgefield, CT, USA

One of the treatment strategies being explored for the management of lipodystrophy has been to replace a protease inhibitor with a non-nucleoside reverse transcriptase inhibitor (NNRTI). This poster was a meta-analysis of eighteen trials evaluating the impact of switching from a protease inhibitor to either nevirapine or efavirenz on the progression of body shape changes and hyperlipidemia.

All studies published in peer-reviewed journals or presented at international conferences between 1999 and February 2001 were evaluated. Switching of NRTIs was permitted. The switch from the protease inhibitor must have been due to clinical or metabolic manifestations of lipodystrophy. There was no restriction on ARV therapy history including time on protease inhibitors.

Blood cholesterol and triglycerides data, before and after the switch to the NNRTI, were pooled in a composite analysis. Some values were derived from graphs. Changes in body shape following the switch were collected, but could have been determined by patient self-report, physician assessment, anthropometric measurement or DEXA.

In all eighteen studies, virologic suppression was maintained in the vast majority of patients (although the numbers are not given in this poster). There was a decrease in CHO and TG in all studies examining the switch from protease inhibitor to nevirapine. One study reported an increase in HDL, while another trial reported a reduction in LDL following the switch to nevirapine, but not following the switch to efavirenz. Insulin resistance normalized in two studies following the switch to nevirapine.

Elevations in CHO and TGI were observed in a number of studies following the switch to efavirenz. Five studies demonstrated elevations in total CHO. Only one study showed improvement. Four studies showed increased TG levels, while four others showed improvements. Four studies showed an increase in HDL, whereas one study showed elevations in HDL and also in LDL and glucose.

"Significant" changes in body shape were reported in three trials looking at a switch to nevirapine (although again, what "significant" means and how it was determined is not noted on the poster).

"Limited, nonsignificant" improvement in body shape was reported in two studies of efavirenz switch. Subjective improvement (patient self-report) in body shape or quality of life was reported in four studies examining the switch to nevirapine, whereas "partial" improvement was reported in two studies of the switch to efavirenz.

Figure 1. Changes in Total Cholesterol Following Switch from PI to NVP or EFV, or with Ongoing PI

Figure 2. Changes in Triglyceride Following Switch from PI to NVP or EFV, or with Ongoing PI

The authors of this poster conclude that the replacement of the protease inhibitor with nevirapine was associated with a reduction in mean blood cholesterol and triglycerides levels, leading to normalization in many patients. The switch to efavirenz was not associated with such improvement. They felt that the investigation of physical changes yielded less conclusive findings.

The authors' conclusions about lipid changes fit with the generally accepted interpretation of the studies in the past few years. The changes in triglycerides with a switch to nevirapine have been quite marked, as have the changes in cholesterol.

However, this poster suffers from several flaws. First, one must take note that this is a poster done by two people working for Boehringer Ingelheim. While there is no disputing the findings, the spin on this poster favoring nevirapine against efavirenz is unmistakable.

Second, many of the switch studies have been seriously flawed in their collection of lipid data. A number of them have collected non-fasting samples and it is not clear how the authors reconcile this kind of data with properly collected fasting levels, since they make no distinction.

Third, it is impossible to draw any real conclusions about body-shape changes given the variety of ways that data was collected, from DEXA scans to patient self-report. Certainly, no quantitative conclusions can be reached when so much subjective data is used. Most authors have concluded that there may be some resolution of fat accumulation, manifested as abdominal distention or buffalo hump, but no resolution of fat wasting, one of the most deforming aspects of lipodystrophy.

Fourth, it is well known that efavirenz can increase total cholesterol, but most studies have found that the increase is in HDL, with unknown consequences.

Despite these flaws, the authors' conclusions, listed below, were all very reasonable:

1. a switch from protease inhibitor to NNRTI improves at least some of the lipid abnormalities;

2. viral load usually remains undetectable; and

3. this is a useful strategy for doctors to follow.

Previous | Next

Click here to take our survey and let us know what you think of our coverage!