Both infection with the human immunodeficiency virus type 1 (HIV) and zidovudine (formerly called azidothymidine [AZT]) cause myopathy. To identify criteria for distinguishing zidovudine-induced myopathy from that caused by primary HIV infection, we reviewed the histochemical, immunocytochemical, and electron-microscopical features of muscle-biopsy specimens from 20 HIV-positive patients with myopathy (15 of whom had been treated with zidovudine) and compared the findings with the patients' clinical course and response to various therapies. Among the zidovudine-treated patients, the myopathy responded to prednisone in four, to the discontinuation of zidovudine in eight, and to nonsteroidal anti-inflammatory drugs in two. Numerous "ragged-red" fibers, indicative of abnormal mitochondria with paracrystalline inclusions, were found in the biopsy specimens from the zidovudine-treated patients but not in those from the other patients. The number of these fibers appeared to correlate with the severity of the myopathy. All the patients, regardless of whether they had been treated with zidovudine, had inflammatory myopathy characterized by degenerating fibers, cytoplasmic bodies, and endomysial infiltrates consisting of CD8+ cells (mean +/- SD, 60.7 +/- 6.4 percent) and macrophages (39.2 +/- 6.4 percent) associated with Class I major histocompatibility complex (MHC-I) antigens (HLA-A, -B, and -C antigens) in the muscle fibers. The numbers and percentages of CD8+ cells and macrophages were similar in both the zidovudine-treated and the untreated HIV-positive patients. Specimens obtained on repeat muscle biopsy from two patients in whom the myopathy responded to the discontinuation of zidovudine showed remarkable histologic improvement. We conclude that long-term therapy with zidovudine can cause a toxic mitochondrial myopathy, which coexists with a T-cell-mediated inflammatory myopathy that is restricted to MHC-I antigen, and is indistinguishable from the myopathy associated with primary HIV infection or polymyositis in HIV-seronegative patients.

Five cases of a new disease presented with muscular weakness or cramping after exercise; three of the cases also had an elevated serum creatine phosphokinase. Muscle biopsies were histologically normal but lacked adenylate deaminase by stain and solution assay, while the erythrocyte isozyme was normal. A clinical diagnostic test has been developed, and the human enzyme was separated by acrylamide-gel electrophoresis.

Within tested strains of the genera Naegleria and Acanthamoeba the ability to grow at high temperatures seems directly related to virulence, with nonvirulent strains unable to grow at normal or elevated body temperatures. Outside these genera, nonvirulent Hartmannella and Tetramitus do grow at elevated temperatures, which suggests a barrier to pathogenicity other than temperature sensitivity. The high optimal temperature of pathogenic Naegleria apparently explains previous difficulty in obtaining isolates from the aquatic environment.

We used very sensitive and specific monoclonal-antibody sandwich assays for human chorionic gonadotropin (hCG) and human luteinizing hormone (hLH) to measure both hormones in serum samples from normal men and women. When single serum samples from 92 men were studied, 73 percent had detectable hCG. In normal men, the amount of detectable hCG averaged 8.9 pg per milliliter, with a range of less than 3.0 to 160 pg per milliliter (biologic potency = 13,450 IU per milligram). In postmenopausal women the hCG level averaged 111 pg per milliliter and ranged from 32 to 510. In women of reproductive age the hCG level varied with the menstrual cycle. Gonadotropin-releasing hormone administered to 10 normal men increased both hCG and hLH. When daily serum samples were studied throughout a normal menstrual cycle, hCG concentrations paralleled those of hLH; follicular-phase concentrations were higher than those of the luteal phase, and there was a midcycle ovulatory peak of hCG coincident with the hLH peak. When hCG was measured every 10 minutes for six hours in eight postmenopausal women, distinct pulses were detected in parallel with those of hLH: hLH pulsed at a mean (+/- SEM) frequency of 0.56 +/- 0.08 pulses per hour; hCG pulsed at 0.54 +/- 0.07 pulses per hour. The mean pulse durations were 89 +/- 22 and 56 +/- 20 minutes for hLH and hCG, respectively. We conclude that hCG is produced in a pulsatile fashion, probably by the pituitary, in all normal adults.

No abstract