The Body Covers: The 6th Conference on Retroviruses and Opportunistic Infections
Abstract No. 680: Avascular Necrosis in HIV-Infected Patients: A Potential Link to Protease Inhibitors
February 3, 1999
This study was of particular interest to me because I have had three patients in my own practice who developed this particular complication in the last 18 months. Prior to that I had never seen this and, talking to other HIV physicians, this was a rarity. At this conference, as I read this poster, doctor after doctor stopped and stated that they had 2-3 cases as well. This is an important observation and bears close monitoring. It may be that there should be a pooling of the information to see if we can tease out the relative risks.
In brief summary, all the patients were on HAART therapy and in a population of 600, there were 8 identified cases for an estimated prevalence of 1.33%. On review of their own cases from Georgetown University Medical Center in Washington, there were no cases identified in their medical records from 1990 thru 1996. Three cases were diagnosed in 1997, and five cases in 1998. The patient populations consisted of seven males and one female with a mean age of 39 years at the time of diagnosis. The mean CD4 cell count at the time of diagnosis was 289, and five of the eight patients carried a diagnosis of AIDS. All the patients were receiving protease inhibitor containing regimens at the time of diagnosis; five patients were exposed to indinavir prior to the diagnosis of AVN. Two patients were exposed to nelfinavir prior the diagnosis of AVN. Two patients were exposed to a combination PI therapy, such as ritonavir plus saquinavir; and, one patient received ritonavir alone and one patient received saquinavir alone prior to diagnosis. The mean duration from initiation of PI therapy to the onset of symptoms was 59.7 weeks. The mean duration from initiation of corticosteroid exposure and comorbid factors for AVN were as follows. Up to five patients would receive corticosteroid therapy. Only two patients had received a prolonged course (greater than 12 weeks), the same two patients who were receiving corticosteroids at the time of diagnosis of AVN. Two patients were never exposed to corticosteroids, and had no significant comorbid factors for AVN. Five patients had random triglycerides levels available at some point prior to the diagnosis of AVN; all were elevated in the range of 205 to 465. The patient with the triglycerides level of 465 also had clinical findings consistent with lipodystrophy. Of those patients with steroid history, the type of steroids were quite varied, ranging from short course of Prednisone, testosterone replacement, Florinef and Megace. Of particular note, the duration of symptoms prior to diagnosis was 5.9 weeks.
In their discussions, the authors note that the pathogenic mechanisms remain unclear. However, they suggest the following hypothesis. Numerous morbid conditions exist that can also be associated with the development of non-traumatic AVN. These diverse etiologies are postulated to operate by a common mechanism of compromised circulation to the affected bone. These include: 1) hyperviscosity syndromes, 2) intravascular coagulation, 3) venostasis, and 4) syndromes resulting in increased intraosseous pressure. Cytotoxic factors have also been implicated in the development of AVN. Corticosteroid-induced AVN appears to occur as a result of increased adipogenesis with increase of the fat cell mass within the marrow of the femoral head. This increase in fat cell mass results in increase intraosseous pressure with subsequent hypoperfusion. Fat embolization has also been postulated as a potential mechanism for MEM. Hypertriglyceridemia has been recognized in the orthopedic literature as a comorbid factor for AVN. PI therapy has been associated with dyslipidemia and fat redistribution, and the authors postulate that this could result in fat embolization and/or increased adipogenesis within the femoral heads. In their study, they were able to identify five patients with elevated serum triglycerides. As these were random (not fasting) triglycerides levels, they cannot comment on the direct association with protease inhibitor induced hypertriglyceridemia and AVN. There is not a causal relationship fully established in protease inhibitor therapy and AVN and, other than my experience and in discussing this with other physicians looking at this presentation, no one saw it in the era prior to HAART. Whether this is directly related to protease inhibitors or other co-factors that are stimulated, remains to be elaborated. As a special caution to all people who take care of HIV patients on HAART, the duration of the development of symptoms to diagnosis is of concern. This is undoubtedly because it is such a rare occurrence that it was not thought to get diagnostic studies earlier. I think now, with heightened awareness, x-rays should be obtained as soon as possible when patients on HAART complain of persistent hip pain.
Authored by: J. Timpone, et al.
This article was provided by The Body PRO. Copyright © Body Health Resources Corporation. All rights reserved.