Skip Navigation

Osteonecrosis Complicating Highly Active Antiretroviral Therapy in Patients Infected with Human Immunodeficiency Virus

  1. Paul Monier1,
  2. Kevin McKown2, and
  3. Michael S. Bronze1
  1. 1 Department of Medicine, Division of Infectious Diseases, University of Tennessee, Memphis
  2. 2 Division of Rheumatology, University of Tennessee, Memphis
  1. Reprints or correspondence: Dr. Paul Monier, Dept. of Medicine, Room H308, Coleman Building, 956 Court Ave., Memphis, TN 38163 (pmonier{at}utmem.edu).
 
Next Section

Abstract

Few patients with osteonecrosis that complicates human immunodeficiency virus (HIV) infection have been reported. We describe 5 patients whose symptoms of osteonecrosis developed with viral suppression and improvement in CD4 lymphocyte counts as a result of antiretroviral therapy. In addition, we review previously reported cases. The mean patient age was 35 years, and HIV was the sole risk factor in only 33%. Eighteen patients (55%) were receiving antiretroviral treatment when osteonecrosis was diagnosed. For 16 patients whose CD4 lymphocyte counts were reported, the mean CD4 count was 350 cells/mm3. In 10 of these patients, the occurrence or worsening of symptoms of osteonecrosis appeared to be related to successful antiretroviral therapy. We conclude that osteonecrosis is an emerging manifestation of HIV infection and that it may be either a consequence of immunologic and virologic improvement resulting from antiretroviral therapy or a complication caused by the drugs themselves.

Osteonecrosis, which is characterized by death of the cellular elements of bone, is usually the result of an impaired arterial blood supply [1] and has been described in association with HIV infection [2–18]. In most patients, the cause of the osteonecrosis is usually associated with advanced HIV disease, the presence of antiphospholipid antibodies, or other risk factors for osteonecrosis, such as the use of alcohol, tobacco, or corticosteroids [19]. Recent reports have suggested that osteonecrosis may arise as a complication in patients whose virologic status has greatly improved as a result of highly active antiretroviral therapy (HAART), or as a result of either the metabolic complications of HAART (i.e., hyperlipidemia) or the drugs used to treat HIV [16–18, 20].

We describe 5 patients with osteonecrosis associated with HIV infection and review the pertinent medical literature. In each of these patients, the osteonecrotic activity increased with treatment-induced reductions in viral load and increases in CD4+ lymphocyte counts. We believe that with the expanding use of HAART, the frequency of this underreported complication of HIV infection may increase.

Previous SectionNext Section

Methods and Materials

We describe 5 patients seen in our HIV clinic, which houses ∼1000 patients, of whom 72% are African American and 27% are white. The male-to-female ratio is approximately 60^:^40, and 98% of patients are ⩾ 20 years of age.

Using the key words “human immunodeficiency virus” and “osteonecrosis,” we searched the English-language medical literature for case reports of osteonecrosis that complicated HIV infection. In aggregate, 28 cases were described in detail sufficient enough to allow for comparison, and these cases, along with our 5 cases, are the basis of the present report.

Previous SectionNext Section

Case Reports

Case 1

A 38-year-old white man who had HIV infection diagnosed in early 1998 was evaluated for progressive bilateral hip pain in 1996, and osteonecrosis of both femoral heads was subsequently diagnosed. At that time, he was treated conservatively with nonsteroidal anti-inflammatory drugs. After treatment, his symptoms resolved; however, it is unknown whether he had progressive changes on radiography. He admitted to smoking one-half of a pack of cigarettes per day, but denied having any history of injection drug abuse or substantial alcohol use. His HIV infection was complicated only by intermittent oral thrush, and he was treated with several antiretroviral medications, including zidovudine, lamivudine, stavudine, and indinavir.

Despite treatment with these medications, the patient's virologic status failed to improve (HIV viral load, 98,982 copies/mL; CD4 lymphocyte count, 177 cells/mm3). With this virologic status, his symptoms of osteonecrosis were quiescent. Treatment with delavirdine, nelfinavir, and didanosine was begun in June 1998. In October 1998, his viral load was undetectable (<146 copies/mL). In December 1998, he presented with bilateral hip pain that was typical of symptoms previously related to osteonecrosis. Radiological studies confirmed progression of osteonecrosis of both femoral heads, and virologic studies confirmed sustained suppression of HIV.

Case 2

A 33-year-old white man with HIV infection diagnosed in 1991 developed Pneumocystis carinii pneumonia in June 1996. He subsequently began receiving zidovudine, lamivudine, and ritonavir. Within 3 months of initiation of antiretroviral therapy, his viral load was 288 copies/mL, and his CD4 lymphocyte count increased to 396 cells/mm3. In November 1996, he developed pain in both hips, the shoulders, and the right knee; the pain progressed during the next year. Radiological evaluation revealed widespread osteonecrosis. This patient had no other known risk factors for osteonecrosis.

Case 3

A 54-year-old black woman who had HIV infection diagnosed in 1996 was referred to our clinic in August 1998. Her viral load was >500,000 copies/mL. Antiretroviral therapy with zidovudine, zalcitabine, and saquinavir produced a dramatic reduction in the viral load (to undetectable levels) and an increase in the CD4 lymphocyte count (to 352 cells/mm3). Within 2 months, she experienced severe bilateral hip and thigh pain. Radiological examinations were consistent with osteonecrosis of the left femoral head.

Case 4

A 33-year-old white man with HIV infection diagnosed in 1996 had an initial HIV viral load of 28,000 copies/mL and a CD4 lymphocyte count of 374 cells/mm3. He had not had opportunistic infections, but his medical history was complicated by hypertension and hyperlipidemia. He had smoked cigarettes and frequently used cocaine. Antiretroviral treatment with stavudine, saquinavir, and lamivudine produced a marked reduction in the viral load and an increase in the CD4 lymphocyte count. However, drug intolerance resulted in an increase in the viral load to 499,000 copies/mL, and his CD4 lymphocyte count dropped to 54 cells/mm3. His medication regimen was subsequently changed to administration of didanosine, delavirdine, and nelfinavir. By September 1998, a moderate reduction in the viral load (to 36,717 copies/mL) had occurred, and the CD4 lymphocyte count had increased to 174 cells/mm3.

In October 1998, he presented at the clinic with a new onset of right hip pain, and radiographs revealed changes consistent with osteonecrosis of the right femoral head. In January 1999, he was admitted to the hospital for evaluation of chest pain. His CD4 lymphocyte count at that time was 18 cells/mm3, and the viral load was 513,623 copies/mL. There were no complaints of hip pain. After receiving counseling, he started treatment with a new antiretroviral regimen that included stavudine, lamivudine, and efavirenz, which led to sustained virologic improvement. His hip pain recurred and required intermittent treatment with narcotics for control of symptoms. Serial radiographs were not obtained once the diagnosis of osteonecrosis was made.

Case 5

A 55-year-old black man was found to be HIV positive in August 1998. He had a history of significant alcohol abuse and occasional hip pain. Because the viral load was 96,639 copies/mL, and because the CD4 lymphocyte count was 25 cells/mm3, antiretroviral therapy with zidovudine and lamivudine was initiated. One month later, his viral load had dropped to 1143 copies/mL, and his CD4 lymphocyte count had increased to 79 cells/mm3. Shortly thereafter, he presented at the clinic with a chief complaint of severe pain in the left hip; radiographs revealed changes consistent with osteonecrosis of both femoral heads.

Previous SectionNext Section

Results

We describe 5 cases of osteonecrosis associated with HIV infection. In all 5 cases, osteonecrosis occurred in association with advanced HIV disease. In each case, symptoms coincided with virologic response after initiation of aggressive antiretroviral therapy—that is, bone pain developed simultaneously with the occurrence of viral suppression and the restoration of CD4 lymphocyte counts. In 2 patients, symptoms abated as the viral loads increased and the CD4 lymphocyte counts decreased. Of the 5 patients in our series, 2 had other risk factors for osteonecrosis. Patient 4 had a history of hyperlipidemia, and patient 5 had a history of alcohol abuse. Three of our patients had multiple-site involvement, and 1 of these 3 had extensive disease that involved 5 different joints.

Although osteoporosis and osteopenia have been reported to occur in HIV-infected patients, our patients were not evaluated by means of either bone densitometry or biochemical assays of bone turnover. Therefore, it is possible that each of these patients had evidence of both osteonecrosis and osteoporosis. Furthermore, because of managed-care issues, radiological monitoring of patients whose symptoms abated was not possible. However, only 1 patient required joint-replacement therapy.

All of these patients were receiving antiretroviral therapy at the time of the diagnosis of osteonecrosis, and all responded to the regimen that they were taking at that time. Four of the 5 were receiving a protease inhibitor at the time of diagnosis, whereas all 5 were taking a nucleoside analog, and 2 were taking a nonnucleoside analog.

A review of the English-language medical literature identified 67 established cases of osteonecrosis that occurred in patients with HIV infection [2–18]. Twenty-eight of these cases were reported in detail sufficient enough to allow for comparison with our 5 cases (table 1). Including patients in our series, the mean patient age was 35 years, and 64% of the patients were male. This mean age is lower than that of our cohort of patients, but whether that changed the epidemiology of osteonecrosis in our group of patients is unknown.

View this table:
Table 1

Summary of data from cases of osteonecrosis in patients infected with HIV.

It is interesting that HIV infection was the sole risk factor for osteonecrosis in 60% of our cohort of patients, whereas only 28% of the remaining 28 patients had this as the sole risk factor. The finding of a higher percentage of men with HIV infection and osteonecrosis among our patients and among those patients whose cases were reviewed is in concert with findings of recent studies indicating that nearly all of the affected patients were male [17, 18]. When dates were reported, the diagnosis of HIV infection predated the diagnosis of osteonecrosis 73% of the time, and for 15% of cases, the 2 diagnoses were made at the same time. In 55% of the cases, osteonecrosis occurred in association with advanced HIV disease, which was indicated by a CD4 lymphocyte count of <200 cells/mm3 or by documentation of HIV-associated opportunistic infections. Bone disease involving multiple sites occurred in 70% of patients.

HIV was identified as the sole risk factor for osteonecrosis in only 33% of cases. Seven (21%) of 33 patients had received steroids before diagnosis, 21% had a history of injection drug abuse, and 12% had a history of alcohol abuse. The presence of antiphospholipid antibodies has been associated with osteonecrosis [17, 21, 22], and the titer was reported to be positive in 19%. Hyperlipidemia has also been reported to be associated with osteonecrosis [19, 23], and it was noted in 12% of these patients.

Eighteen (55%) of 33 patients were reported to have been receiving antiretroviral therapy at the time of diagnosis of osteonecrosis. Of these 18 patients, all received a nucleoside analog; 15 of them received zidovudine, 7 received a protease inhibitor, and 2 received a nonnucleoside analog.

For 16 patients, the CD4 lymphocyte count and/or HIV viral load was noted at the time of diagnosis of osteonecrosis; the mean CD4 lymphocyte count at that time was 350 cells/mm3 (range, 79–1064 cells/mm3). In 10 of the 16 patients, either the occurrence of osteonecrosis or the worsening of bone pain can be related to the success of antiretroviral therapy. Similar data were reported in abstract form by Hodak et al. [17], who described 10 patients with osteonecrosis that correlated with extensive protease-inhibitor exposure.

The diagnosis of osteonecrosis was made radiographically. Typical features on plain radiographs include cystic and sclerotic changes due to bone necrosis and a “crescent sign” created by subchondral collapse. Imaging with technetium bone scans revealed increased uptake in affected bone, but this method tends to be less sensitive than MRI. Findings compatible with osteonecrosis were reported for 96% of patients evaluated by plain radiography and for all patients evaluated by either bone scanning or MRI (data not shown).

Previous SectionNext Section

Discussion

Osteonecrosis, which is commonly referred to as “avascular necrosis of bone,” is a disorder in which skeletal pain develops from the death of cellular elements of bone [1, 19]. It most commonly affects the femoral head, and it is bilateral ∼60% of the time [19]. The exact mechanism of this disorder is poorly understood, but there are established risk factors for its development, including prolonged treatment with steroids, alcohol abuse, and injection drug abuse [19]. The presence of antiphospholipid antibodies and hyperlipidemia are among many factors thought to be associated with its development, although a direct causal relationship has not been established. The mechanism of development of osteonecrosis is poorly understood but is generally thought to be related to interruption of the blood supply to the affected bone [1, 19].

During the past decade, several reports of an association between HIV infection and osteonecrosis have been published, all of which have a common message: HIV testing should be part of the evaluation for osteonecrosis, particularly if no other risk factors are present. Several theories regarding mechanisms for the occurrence of osteonecrosis in patients infected with HIV have been proposed. Antiphospholipid antibody levels are commonly elevated in HIV-infected individuals and may predispose them to intraosseous platelet aggregation and subsequent bone necrosis [11, 24, 25]. HIV-infected patients may acquire antibodies that are directed against, and therefore may cause a deficiency of, protein S, thereby predisposing them to thrombosis [11, 26–29]. Infection-related vasculitis has been suggested as a possible mechanism as well [3]. More recently, hyperlipidemia associated with treatment with protease inhibitors has been suggested as a possible factor in the development of osteonecrosis in patients infected with HIV [16].

Therefore, the mechanisms by which HAART could cause osteonecrosis are unclear, but they involve (1) increased production of anti-protein S or antiphospholipid antibodies as a result of enhanced humoral immunity; (2) hyperlipidemia due to protease inhibitors; and (3) abnormal fat distribution, including cushingoid features, which are also associated with protease inhibitors. Alteration in lipid metabolism is thought to be one mechanism involved in glucocorticoid-associated osteonecrosis [30].

Osteonecrosis is reported to occur most often in association with advanced HIV disease. However, our observation that symptoms related to osteonecrosis occurred in each of our 5 patients after viral suppression and improvement in CD4 lymphocyte counts raises the question of whether this association is a consequence of the immunologic and virologic improvement resulting from HAART or whether it occurs as a metabolic complication of HAART.

After evaluation of the data for our 5 patients with osteonecrosis associated with HIV infection, and after reviewing the aforementioned cases in the literature, we made several conclusions and observation. First, HIV testing should be performed on any patient found to have osteonecrosis, particularly in the absence of other risk factors or associations. Second, health care providers who routinely care for HIV-infected patients should be aware of this association, especially in individuals with advanced disease who are receiving HAART. Third, the incidence of osteonecrosis in patients infected with HIV is probably underestimated, and if (as our findings suggest) osteonecrosis develops as a consequence or complication of HAART, its occurrence may become more frequent. Health care providers should have a heightened index of suspicion for HIV infection as a risk factor for osteonecrosis. Conversely, HIV-infected patients who present with skeletal or articular signs should be evaluated for osteonecrosis.

  • Received December 15, 1999.
  • Revision received May 4, 2000.
  • Accepted November 29, 2000.
Previous Section
 

References

  1. ↵
    1. Klippel J,
    2. Dieppe P
    1. Mazieres B
    . Osteonecrosis. In: Klippel J, Dieppe P, editors. Rheumatology. 2d ed. London: Mosby; 1998. p. 847.
  2. ↵
    1. Solomon G,
    2. Brancato L,
    3. Winchester R
    . An approach to the human immunodeficiency virus positive patient with a spondyloarthropathic disease. Rheum Dis Clin North Am 1991;17:43-58.
    MedlineWeb of Science
  3. ↵
    1. Gerster JC,
    2. Camus JP,
    3. Chave JP,
    4. Koeger AC,
    5. Rappoport G
    . Multiple site avascular necrosis in HIV infected patients. J Rheumatol 1991;18:300-2.
    MedlineWeb of Science
  4. ↵
    1. Belmonte MA,
    2. Garcia-Portales R,
    3. Domenech I,
    4. Fernandez-Nebro A,
    5. Camps MT,
    6. De Ramon E
    . Avascular necrosis of bone in human immunodeficiency virus infection and antiphospholipid antibodies. J Rheumatol 1993;20:1425-8.
    MedlineWeb of Science
  5. ↵
    1. Chevalier X,
    2. Larget-Piet B,
    3. Hernigou P,
    4. Gherardi R
    . Avascular necrosis of the femoral head in HIV-infected patients. J Bone Joint Surg Br 1993;75:160.
    Medline
  6. ↵
    1. Molina J,
    2. Citera G,
    3. Rosler D,
    4. et al
    . Coexistence of HIV virus infection and systemic lupus erythematosus. J Rheumatol 1995;22:347-50.
    MedlineWeb of Science
    1. Koeger AC,
    2. Banneville B,
    3. Gerster J,
    4. et al
    . Avascular osteonecrosis in HIV infected patients: 10 cases [abstract]. Arthritis Rheum 1995;38(Suppl):S199.
  8. ↵
    1. Stovall D,
    2. Young T
    . Avascular necrosis of the medial femoral condyle in HIV-infected patients. Am J Orthopedics 1995;24:71-3.
  9. ↵
    1. Cheung NT,
    2. Saklatvala J,
    3. Dawes PT
    . Multiple joint avascular necrosis: beware of tuberculosis and human immunodeficiency virus a rare but important cause. Br J Rheumatol 1995;34:387-91.
    Abstract/FREE Full Text
  10. ↵
    1. Tigges S,
    2. Meli R
    . Osteonecrosis associated with HIV infection. Can Assoc Radiol J 1995;46:280-4.
    MedlineWeb of Science
  11. ↵
    1. Rademaker J,
    2. Dobro JS,
    3. Solomon G
    . Osteonecrosis and human immunodeficiency virus infection. J Rheumatol 1997;24:601-4.
    MedlineWeb of Science
    1. Gerster J,
    2. Rossetti G
    . Aseptic avascular osteonecrosis mimicking arthritis in HIV infection. J Rheumatol 1998;25:604-5.
    MedlineWeb of Science
  13. ↵
    1. Olive A,
    2. Queralt C,
    3. Sirera G,
    4. Centelles M,
    5. Force L
    . Osteonecrosis and HIV infection: 4 more cases. J Rheumatol 1998;25(Suppl 6):1243-4.
    Medline
  14. ↵
    1. Llauger J,
    2. Palmer J,
    3. Roson N,
    4. Fernandez A,
    5. Camins A
    . Osteonecrosis of the knee in an HIV-infected patient. AJR Am J Roentgenol 1998;171:987-8.
    FREE Full Text
    1. Koeger AC
    . Osteonecrosis and HIV infection. J Rheumatol 1999;26:752.
    MedlineWeb of Science
  16. ↵
    1. Meyer D,
    2. Behrens G,
    3. Schmidt RE,
    4. Stoll M
    . Osteonecrosis of the femoral head in patients receiving HIV protease inhibitors. AIDS 1999;13(Suppl 9):1147-8.
    CrossRefMedlineWeb of Science
  17. ↵
    1. Hodak SP,
    2. Fluhme D,
    3. Kumar P,
    4. Nascone J,
    5. Evans B,
    6. Timone JG
    . Program and abstracts of the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy (San Francisco). Washington, DC: American Society for Microbiology; 1999. Avascular necrosis and protease inhibitor exposure [abstract 1312]; p. 506.
  18. ↵
    1. Scribner AN,
    2. Skiest DJ,
    3. Marcantionio D,
    4. Hamid F,
    5. Troai-Cancio PV,
    6. Keiser P
    . Program and abstracts of the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy (San Francisco). Washington, DC: American Society for Microbiology; 1999. A case control study of osteonecrosis in HIV [abstract 1311]; p. 505.
  19. ↵
    1. Steinberg ME
    . Avascular necrosis: diagnosis, staging, and management. J Musculoskel Med 1997;14(11):13-25.
  20. ↵
    1. Carr A,
    2. Samaras K,
    3. Burton S,
    4. et al
    . A syndrome of peripheral lipodystrophy, hyperlipidaemia, and insulin resistance in patients receiving HIV-protease inhibitors. AIDS 1998;12:51-8.
  21. ↵
    1. Love PE,
    2. Santoro SA
    . Antiphospholipid antibodies: anticardiolipin and the lupus anticoagulant in systemic lupus erythematosus (SLE) and in non-SLE disorders. Ann Intern Med 1990;112:682-98.
    Abstract/FREE Full Text
  22. ↵
    1. Asherson RA,
    2. Liote F,
    3. Page B,
    4. et al
    . Avascular necrosis of bone and antiphospholipid antibodies in systemic lupus erythematosus. J Rheumatol 1993;20:284-8.
    MedlineWeb of Science
  23. ↵
    1. Moskal JT,
    2. Topping RE,
    3. Franklin LL
    . Hypercholesterolemia: an association with osteonecrosis of the femoral head. Am J Orthop 1997;26:609-12.
    Medline
  24. ↵
    1. Stimmler MM,
    2. Quismorio FP,
    3. Mcgehee WG,
    4. Boylen T,
    5. Sharma OP
    . Anticardiolipin antibodies in acquired immunodeficiency syndrome. Arch Intern Med 1989;149:1833-935.
    Abstract/FREE Full Text
  25. ↵
    1. Aguilar JL,
    2. Berman A,
    3. Espinoza LR,
    4. Blitz B,
    5. Lockey R
    . Autoimmune phenomena in human immunodeficiency virus infection. Am J Med 1988;85:283-4.
    Medline
  26. ↵
    1. Sorice M,
    2. Griggi T,
    3. Arcieri P,
    4. et al
    . Protein S and HIV infection. Thromb Res 1994;73:165-75.
    CrossRefMedlineWeb of Science
    1. Lafeuillade A,
    2. Alessi M-C,
    3. Poizet-Martin I,
    4. et al
    . Protein S deficiency and HIV infection. N Engl J Med 1991;324:1220.
    MedlineWeb of Science
    1. Pulik M,
    2. Lionnet F,
    3. Coudere LJ,
    4. Matheron S,
    5. Saimot AG
    . Thromboembolic disease and human immunodeficiency virus infection. Blood 1993;82:2931.
    FREE Full Text
  29. ↵
    1. Stahl HCP,
    2. Wideman CS,
    3. Spira TJ,
    4. Haff EC,
    5. Hixon GJ,
    6. Evatt BL
    . Protein S deficiency in men with long-term human immunodeficiency virus infection. Blood 1993;81:1801-7.
    Abstract/FREE Full Text
  30. ↵
    1. Wang GL,
    2. Lennox DW,
    3. Reger SI,
    4. Stamp WG,
    5. Hubbard SL
    . Cortisone-induced intrafemoral head pressure change and its response to a drilling decompression method. Clin Orthop 1981;159:274-8.
    Medline
| Table of Contents

This Article

  1. Clin Infect Dis. (2000) 31 (6): 1488-1492. doi: 10.1086/317503
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most