- Email: [email protected]
Testicular atrophy in AIDS: A study of 57 autopsy cases
Original Contributions Testicular Atrophy in AIDS: A Study of 57 Autopsy Cases MONIQUE E. DE PAEPE,” MD, AND MARIAN WAXMAN, MD The pertinent clinical data and histologic features of the testes in 57 autopsied acquired immunodeficiency syndrome (AIDS) patients were analyzed and compared with those of 55 age-matched control patients without AIDS. The testes of the AIDS patients showed a significantly lower degree of spermatogenesis (determined by a testicular score count), as well as more prominent thickening of the basement membrane and interstitial fibrosis when compared with the controls. While the precise cause of testicular atrophy in AIDS patients remains to be determined, the chronicity of the disease, prolonged fever, malnutrition, testicular infection, and chemotherapy are all contributing factors. Since the vast majority of the studied AIDS patients were homosexual and most control patients were heterosexual, the observed testicular changes can be ascribed to AIDS and/or homosexuality. Because of a high prevalence of sexually transmitted diseases, antisperm antibodies, and possible zinc deficiency and endocrine disorders, homosexual men appear predisposed to tubular atrophy. Conversely, AIDS-related factors, such as a direct toxic effect of the human immunodeficiency virus on germinal epithelium or as yet undetermined endocrine imbalances might exert a detrimental effect on the testis independent of homosexuality. HUM PATHOL 20:210-214. 0 1989 by W.B. Saunders Company.
pita1 charts were reviewed for data regarding age, AIDS risk factors (homosexuality, intravenous [IV] drug abuse), history of sexually transmitted disease [STD], smoking, alcohol abuse, cirrhosis of the liver, chemotherapy, radiation to the pelvis, prolonged fever (more than 6 months), chronic illness (more than 6 months), and significant low weight (~70% of the reference weight according to the 1983 Metropolitan Life Insurance reference weights for height’). All available biopsy material concerning the AIDS patients was examined. The autopsy protocols and slides were reviewed for general information, main pathologic features of the AIDS patients, and main diseases of the controls. The gross descriptions and the slides of the testes were examined. In the AIDS group, one section of testis was available from 17 patients and two sections from 40 patients; in the control group, one section was studied in 20 patients and two sections in 35. All the sections were formalin-fixed, paraffin-embedded, and stained with hematoxylin-eosin; all the sections of AIDS testes were stained with Ziehl-Neelsen stain for acid-fast bacilli. The degree of spermatogenesis was determined using a modification of the Johnsen testicular score count method.8 For each case, 100 representative sections of the seminiferous tubules were evaluated and a score from one to ten was designated, according to the most mature germ cell stage identified. From these tubular scores, a mean sperm score was calculated for every subject, which proved to be an objective and reproducible value. The Leydig cells were judged as normal, increased, or decreased in number. (The Leydig cells constitute approximately 3% of the visual field in the healthy adult man.s) Other parameters noted were tubular scarring, thickening of the tubular basement membrane, interstitial fibrosis, nonspecific interstitial inflammation, abnormal thickening of the blood vessel walls, and the presence of microorganisms. Statistical analysis was performed using the Student’s t test.
Several recent autopsy series dealing with men who have died of the acquired immune deficiency syndrome (AIDS) have consistently described decreased spermatogenesis or testicular atrophy.‘-6 With one notable exception,6 none of these reports has focused on this subject or attempted to offer a satisfactory explanation for the almost universal testicular damage seen in AIDS patients. Our study was undertaken in an attempt to gain further insight into the pathogenesis of testicular atrophy and, in particular, to evaluate whether this injury is specific for AIDS or whether it is a nonspecific sequel of a wasting disease.
RESULTS MATERIALS AND METHODS
Clinical
We reviewed the material concerning 57 consecutive unselected male AIDS patients and 55 male age-matched control patients autopsied at St. Vincent’s Hospital and Medical Center (New York) from 1984 to 1987. The hos-
The age of the AIDS patients ranged from 26 to 69 years (mean, 41.1 years; SD, 10.4), and age of the control patients ranged from 24 to 64 years (mean, 44.2 years; SD, 9.7). Fifty-five AIDS patients (96%) were homosexual, of which one also had a history of IV drug abuse; two were heterosexual drug addicts. One control patient was homosexual and one had a history of IV drug abuse. Deceased IV drug users with or without AIDS are routinely referred to the medical examiner’s office and are therefore usually not autopsied at the hospital. The control group did not contain more homosexual patients because the sexual preference of non-AIDS patients is not routinely asked and almost all admitted homosexual pa-
From the Department of Pathology, St. Vincent’s Hospital and Medical Center, New York. Accepted for publication March 8, 1988. * Present address: Department of Pathology, Memorial SloanKettering Cancer Center, New York. Key words: testis, atrophy, AIDS, sperm score. Address correspondence and reprints requests to Marian Waxman, MD, Our Lady of Mercy Medical Center, Department of Pathology, 600 E 233rd St, Bronx, NY, 10466. 0 1989 by W.B. Saunders Company. 0046~8177/89/2003-0003$5.00/O
210
TESTlCULAR ATROPHY IN AIDS (De Paepe & Waxman)
tients who were autopsied died with AIDS. Venereal disease was reported in 34 of 35 AIDS patients (97%) who were specifically questioned (20 gonorrhea and syphilis, 11 gonorrhea only, three syphilis only). STD was not mentioned in the charts of the other 22 AIDS patients or in those of any of the control patients. Prolonged fever was found in 13 AIDS patients (23%) and nine controls (16%). Chronic illness was present in 40 AIDS patients (70%) and 24 controls (44%). Eighteen AIDS patients (32%) and 24 control patients (44%) were smokers. Ten AIDS patients (18%) and 16 controls (29%) had a history of alcoholism and/or cirrhosis. In both groups, 13 patients (23% to 24%) received chemotherapy (various combinations of antimetabolites, vinca alkaloids, alkylating agents, and antibiotics). In addition, 11 AIDS patients (and none of the controls) received drugs known to adversely affect the spermatogenesis: the antifungal drug ketoconazoleg (five cases), corticosteroids’O (four cases), and the antiviral agent BWB 759U (9[2-Hydroxy-1-(hydroxymethyl)ethoxymethyl] guanine, DHPG)” (two cases). No patient had received radiation treatment to the genital area. Significant underweight was found in 18 patients of the AIDS group (32%) and in eight controls (15%). In 5 1 AIDS cases (89%), death was attributed to overwhelming opportunistic infection (combined with Kaposi’s sarcoma [KS] in 16 cases and malignant lymphoma in four cases). In the remaining six AIDS patients, the causes of death were disseminated KS (three cases), malignant lymphoma (two cases), and KS with lymphoma (one case), without evidence of infection. The deaths of the control patients were attributed to malignancy (23 cases), cardiovascular disease (21 cases), infection (six cases), and complications of alcoholism (five cases).
Pathologic Findings The testes were not measured or weighed routinely. Decreased stringing was described in eight AIDS patients (14%) and four control patients (7%). Two AIDS cases showed grossly small granulomas, and two others showed broad fibrous bands, Thickening of the tubular basement membrane, nonspecific interstitial inflammation (Fig l), interstitial fibrosis (Fig 2), and infection were found more often in the testes of AIDS patients than in the controls (Table 1). The evidence of a specific infection was found in 18 AIDS patients; the microorganisms were Mycobacterium avium-intracellular (MAI) in six cases, Toxoplasma in five, and cytomegalovirus (CMV) in eight (one with Toxophma). In all cases, the same organism was identified in at least one other organ at autopsy. These three organisms were seen within the seminiferous tubules and in the interstitium, and CMV inclusions were also frequently noted in the endothelial cells. Two cases of GMV showed tubular microabscesses and thrombi in small blood vessels. MA1 infection presented in three testes as aggregates of “striated blue” histiocytes and in three other cases as granulomas, associated with central necrosis in one instance. Thus, this last case was similar to tuberculous orchitis
FIGURE 1. Marked interstitial inflammatory infiltrate composed mostly of lymphocytes. (Hematoxylin-eosin stain; magnification x 100.)
except for the multitude of acid-fast bacilli characteristic of MAI. The two groups showed no significant differences in the frequency of vascular wall thickening, abnormal Leydig cell numbers, and marked tubular scarring. There was no evidence of KS or lymphoma in any testis. The mean sperm score of the testes of AIDS patients was found to be significantly lower than that of the controls (3.94 v 6.22; P < .OOl) (Table 1). Sixtyeight of the AIDS testes had mean sperm score values of 2 to 4 only (Fig 3), whereas the testes of the control patients displayed a more even distribution of the scores, with 42% of the cases above 7. The mean sperm score was significantly lower in infected than in non-infected testes of AIDS patients (3.42 v 3.94; P < .05) (Table 2).
Correlation of the Mean Sperm Score With the Clinical Features Within each study group, a history of chronic illness, prolonged fever, low weight, and chemother-
FIGURE 2. Severe interstitial fibrosis which merges with the thickened fibrotic basal laminae of the tubules. (Hematoxylin-eosin stain; magnification x100.)
211
HUMAN PATHOLOGY TABLE
Volume 20, No. 3 (March 1989)
1. Histologic Features of the Testes AIDS (n = 57)
Sections examined Tubular scarring Thickening basement membrane Interstitial fibrosis Interstitial inflammation Thickened vessel walls Specific infection Leydig cells Increased Decreased Sperm score Mean Range
Controls
TABLE
Effect of Specific
Infection
on the Sperm Score Sperm
(n = 55)
97 4
90 3
29
13
29
2
16
5
8
6
18
0
8 3
6 1
3.94 1.22-8.99
2.
AIDS All (n = 57) Specific infection (n = 18) CMV (n = 8) Toxoplasma (n = 5) MA1 (n = 6)
Score
3.94 r 1.70 3.42 ? 0.97 3.51 2.86 3.68
reflects the poor correlation between the gross and microscopic appearance of the testis, as noted by others.’ In two cases, a significant finding at gross inspection consisted of areas of granulomas, confirmed as testicular mycobacteriosis by light microscopy. A specific infection of the testis was diagnosed in 32% of our AIDS patients, slightly more than the 22% found in the series of Chabon et al.‘j Testicular infection has been reported by other investiin view of the overgators 1~2,4-6,12and is expected whelming infections often seen in AIDS patients. Our finding of CMV, Toxoplasma, and MA1 organisms within the lumina of the seminiferous tubules is significant in that it indicates the possibility of their transmission via the semen, as reported previously for CMV.13 Infection with Toxoplasmu was associated with only a few mononuclear inflammatory cells and the organisms were rather inconspicuous; the necrosis and accumulation of eosinophiles’* were missing. The much higher incidence of the nonspecific interstitial inflammation and of interstitial fibrosis in the AIDS group was noteworthy, compared with the control group (28% u 9% and 5 1% v 4%, respectively). Chabon et al reported finding interstitial inflammation in 53% of the testes of their AIDS patients.6 Although the significance of these findings is not clear at present, they may possibly represent a morphologic expression of autoimmune orchitis. Generally, we feel that one or two sections per case were representative of the significant pathology in our review; there were no significant findings limited to one testis when sections of the two were available for review. The same pertains to the patient’s
6.22 1.55-8.96
apy was associated with decreased spermatogenesis, whereas age over 50 years, alcohol abuse, cirrhosis, and smoking did not seem to affect the testes consistently (Table 3). Comparison of the mean sperm score of the AIDS and control patients with regard to chronic illness, prolonged fever, and low weight revealed significantly lower scores in the AIDS patients in all three subgroups (P < .05 to .OOl). Chemotherapy was found to have an equally adverse effect on spermatogenesis in the testes in both groups, lowering the mean sperm score to 3.71 (AIDS group) and 4.21 (control group) (P > .05). Ketoconazole and steroid intake and a history of IV drug abuse did not significantly correlate with the degree of testicular atrophy, while the two patients who had received BWB 759U displayed a mean sperm score far above the average of both patient groups. DISCUSSION The gross features of the testes were described as abnormal in only 2 1% of the AIDS patients, which
TABLE
3.
Correlation of Sperm Score With Clinical Features AIDS
FIGURE 3. Seminiferous tubules lined by Sertoli cells and spermatogonia; tubular score 3. Artifactual space separates the epithelium from the markedly thickened fibrotic basal lamlna. (Hematoxylin-eosin stain; magnification x250.)
212
P
Controls
All patients
3.94 (n = 57)
6.22 (n = 55)
<.OOl
Age over 50 yr Chronic illness Prolonged fever Underweight Alcoholism/cirrhosis Tobacco Chemotherapy
3.93 3.76 3.64 3.57 3.68 3.92 3.71
6.62 5.09 5.42 5.38 6.84 6.25 4.21
<.OOl <.OOI <.05 <.05 <.OOI <.OOl NS
Other medications Ketoconazole, corticosteroids BWB 759U
3.56 (n = 9) 8.79 (n = 2)
IV drug
3.94 (n = 3)
abuse
(n (n (n (n (n (n (n
= = = = = = =
8) 40) 13) 18) IO) 18) 13)
(n (n (n (n (n (n (n
= = = = = = =
21) 24) 9) 8) 16) 24) 13)
5.06 (n = 1)
NS
TESTICULARATROPHYIN AIDS (De Paepe & Waxman)
main sperm score which was generally the same when both testes were examined and compared. Despite widespread KS diagnosed in 35% of the AIDS patients, we did not see the tumor in any testis. Only a few AIDS patients with KS involving the epididymis,’ tunica albuginea,5 and testis (not further specified)4 have been described. Testes in our AIDS patients showed significantly impaired spermatogenesis and increased interstitial fibrosis and thickening of basement membrane, as compared with those of the age-matched autopsy population, confirming the association of AIDS and testicular atrophy noted by other observers.‘-6 The exact pathogenesis of the testicular atrophy is unclear; a wide variety of known causes of decreased spermatogenesis applies to AIDS patients. When possible, we correlated the degree of spermatogenesis with the histologic and clinical features of the patients in order to determine the relative importance of the various factors examined. Our two groups were age-matched and contained comparable numbers of patients with a history of alcoholism and smoking; none of these known causes of testicular atrophy (old age,14 alcoholism/cirrhosis,15r16 smoking”) showed a distinct correlation with the sperm score (Table 3). The number of Leydig cells was essentially normal in the vast majority of our cases and showed no correlation with the degree of spermatogenesis, confirming the observations of other investigators.5*6 Despite this, endocrine imbalances cannot be excluded as an etiology for the testicular atrophy in AIDS patients since pituitary and hypothalamic disorders may manifest primarily as affecting the seminiferous tubules.r8 Since AIDS is almost inherently associated with chronic illness, stress, prolonged fever, and malnutrition, it is tempting to ascribe the observed testicular changes to these classic causes of testicular atrophy. lgTz3The degree of spermatogenesis in AIDS patients with chronic illness and a low body weight at the time of death was lower than the average of the entire AIDS group (Table 3), which confirms the described detrimental effect of these factors on the testis. Since the same factors were found to be associated with less severe atrophy in the control group (Table 3), it is suggested that the chronicity of the disease contributes to the decreased spermatogenesis in these patients, although it is not the primary cause. The presence of infiltrative disease of hypothalamus and pituitary gland (eg, infection, lymphoma) could not be evaluated since sections of these organs were not routinely examined. Since most of the AIDS patients in our study are homosexual, the endocrine implications of homosexuality should be considered. The literature on this subject is conflicting.24-26 We are unable to draw any conclusions in this regard from our study because our control group contained only one homosexual. Several mechanisms could be proposed to explain the low sperm score among the AIDS patients with testicular infections (Table 2). A direct effect on the germinal epithelium has been ascribed to mumps
and other viruses,*’ and could also exist for the organisms identified in the testes in this series and for the human immunodeficiency virus (HIV) itself. Retrovirus-like particles, consistent with HIV, have been demonstrated ultrastructurally in the testis of AIDS patients.28 CMV infection of the male genital tract has been associated with alterations in the pH of the seminal fluid and a reduction in the sperm count.2g In our series, two CMV-infected testes showed multiple thrombi which could lead to ischemic damage to the testis, as postulated by others.’ Several investigators have proposed immune phenomena as possible mechanisms for the testicular atrophy in homosexual AIDS patients,2r4s5*6 based on the finding of elevated antisperm antibody titers in the serum of homosexual men compared with heterosexual men.3o Our finding of a high incidence of interstitial inflammation and fibrosis might support this theory (see above). The hypothesis of homosexuality-associated autoimmune orchitis as a major cause of testicular damage in AIDS can be challenged, however, since several autopsy series describe the same degree of testicular atrophy in AIDS patients whether they are homosexual or not.3,4,6 Thirteen AIDS patients (23%) had received chemotherapy. In all cases, at least one drug with known adverse effects on the testis of humans or laboratory animals was included in the therapeutic regimen.3’ Since there was no significant difference in the degree of testicular atrophy in AIDS patients and controls with chemotherapy, it can be inferred that testicular changes in this subgroup are secondary to the chemotherapy and not due to AIDS. Some of the AIDS patients received drugs with known adverse effects on spermatogenesis (corticosteroids,r” ketoconazole,g and the experimental drug BWB 759U1’), but in view of the small sample sizes, no conclusions can be drawn concerning the effect of these drugs on the testis in AIDS patients. No AIDS patient in our study had a history of radiation treatment, which is known to adversely affect spermatogenesis.32 The blood vessels of the testes of AIDS patients were not found to differ from those of the testes of the control patients; therefore, a vascular compromise with secondary ischemia is not a likely cause of the testicular dysfunction in AIDS, with the possible exception of testes infected with CMV. The testes of our AIDS patients with a history of IV drug abuse had a similar appearance to those of homosexual AIDS patients, an experience shared by Chabon et al6 The small number of cases precluded any conclusions regarding detrimental effects of opiatesS3 and other recreational drugs on the testis of AIDS patients. behavior According to some, 34,35 homosexual could lead to zinc deficiency, another known cause of decreased spermatogenesis. Many of the AIDS patients studied had a history of venereal disease, which confirms the reported high incidence of STDs among homosexuals.36 Genital tract infection has been associated with decreased fertility,37*38 but the impact of venereal disease on the testes of AIDS patients could 213
HUMAN PATHOLOGY
Volume 20, No. 3 (March 1989)
cytomegalovirus retinitis with 9-[2-hydroxy-1-(hydroxy methyl) ethoxymethyl] guanine. Ann Intern Med 103:377-380, 1985 12. Nistal M, Santana A, Paniaqua R, et al: Testicular toxoplasmosis in two men with the acqmred immunodeficiency syndrome (AIDS). Arch Path01 Lab Med 110:744-746. 1986 13.‘ Biggar RJ, Andersen HK, Ebbesen P, et Seminal fluid excretion of cytomegalovirus related to immunosuppression in homosexual men. Br Med J 286:2010-2013, 1983 14. Warner BA, Dufau ML, Santen RJ: Effects of aging and illness on the pituitary testicular axis in men: Qualitative as well as quantitative changes in luteinizing hormone. J Clin Endocrinol Metab 60:263-268, 1985 15. Galvao-Teles A, Goncalves L, Carvalho H, et al: Alterations of testicular morphology in alcoholic disease. Alcoholism (NY) 7: 144-148, 1983 16. Anderson RA Jr, Willis BR, Oswald C, et al: Male reproductive tract sensitivity to ethanol: A critical overview. Pharmacol Biochem Behav 18:305-310, 1983 (suppl 1) 17. Handelsman DJ, Conway AJ,-Boylan LM, et al: Testicular function in potential sperm donors: Normal ranges and the effects of smoking and varicocele. Int ] Androl 7:369-382, 1984 18. Ree GH, Martin F, M$es K, et al: Hormonal changes in human leprosy. Lepr Rev 52:121-126, 1981 19. Morley JE, Melmed S: Gonadal dysfunction in systemic disorders. Metabolism 28: 1051-1073, 1979 20. Chlebowski RT, Heber D: Hypogonadism in male patients with metastatic cancer prior to chemotherapv. Cancer Res 42:24952498, 1982 21. Collu R, Gibb W, Ducharme JR: Effects of stress on the cronadal function. I Endocrinol Invest 7:529-537. 1984 22. Macleod J, Hotchkiss RS: The effect of hyperpyrexia y;;; spermatozoa counts in men. Endocrinology 28:780-784,
not be determined because of the absence of history of sexually transmitted diseases in the control group. SUMMARY
ai:
This study has revealed statistically significant morphologic differences between the testes of homosexual AIDS patients and age-matched heterosexual control patients without AIDS. The chronicity of the disease, prolonged fever, malnutrition, testicular infection, and chemotherapy all contribute to the testicular atrophy of AIDS patients. With the exception of chemotherapy, however, none of these factors can entirely explain the severity of the testicular damage. Due to the absence of equal numbers of homosexual patients in the control group, the effects of homosexuality and AIDS on the testis cannot be clearly separated. Homosexuals, the largest AIDS risk group, are in theory particularly predisposed to testicular atrophy: venereal disease, antisperm antibodies, and possible zinc deficiency and endocrine changes have all been associated with homosexuality and are known causes of decreased spermatogenesis. Testicular atrophy in AIDS patients is not limited to homosexuals and has been reported to occur with the same frequency and severity in AIDS patients of other risk groups. Factors more specifically related to AIDS may therefore underlie the testicular atrophic changes; a direct adverse effect of HIV on the germinal epithelium and as yet undetermined endocrine phenomena can be postulated.
L
.
,
”
23. Steinberger E, Dixon WJ: Some observations on the effect of heat on the testicular germinal epithelium. Fertil Steril 10:578595, 1959 24. Tourney G, Petrilli AJ, Hatfield LM: Hormonal relationshins in homosexual men. Am I Psvchiatrv 132:288-290. 1975 ’ 25. Doerr P, Pirke KM, K&k&t G, et al: Further studies on sex hormones in male homosexuals. Arch Gen Psychiatry 33:61 l614, 1976 26. Kolodny RC, Masters WH, Hendry XJ, et al: Plasma testosterone and semen analysis in male homosexuals. N Engl J Med 285:1170-1174, 1971 27. Mikuz G, Damjanov I: Inflammation of the testis, epididymis, peritesticular membranes and scrotum. Pathol Annu 17:101128, 1982 28. Lecatsas G, Houff S, Macher A, et al: Retrovirus-like particles in salivary glands, prostate and testes of AIDS patients. Proc Sot Exp Biol Med 178:653-655, 1985 29: Lang DJ, Kummer JF, Hartley DP: Cytomegalovirus in semen. Persistence and demonstration in extracellular fluids. N Engl J Med 291:121-123, 1974 30. Witkin SS, Sonnabend J: Immune responses to spermatozoa in homosexual men. Fertil Steril 39:337-342, 1983 3 1. Chapman RM: Gonadal injury resulting from chemotherapy. Am J Ind Med 4:149-161, 1983 32. Ash P: The influence of radiation on fertility in man. Br J Radio1 53:271-278, 1980 33. Wang C, Chan V, Yeung RTT: The effect of heroin addiction on oituitarv-testicular function. Chn Endocrinol9:455-461. 1978 ‘ ’ 34. Weiner RG: AIDS and zinc deficiency. JAMA 252: 14091410, 1984 35. Apgar J: Zinc and reproduction. Annu Rev Nutr 5:43-68, 1985 36. Judson FN, Penley KA, Robinson ME, et al: Comparative prevalence rates of sexually transmitted diseases in heterosexual men. Am J Epidemiol 112:836-843, 1980 37. Quesada EM, Dukes CD, Deen GH, et al: Genital infection and sperm agglutinating antibodies in infertile men. J Urol99: 106108, 1968 38. Witkin SS, Toth A: Relationship between genital tract infections, sperm antibodies in seminal fluid and infertility. Fertil Steril 40:805-808, 1983
Acknowledgment. The authors thank Dr Francois Luks for assistance in preparing the manuscript, Denise Mitchell for secretarial work, and Dr Julius Weber for photography.
REFERENCES 1. Reichert CM, O’Leary TJ, Levens DL, et al: Autopsy pathologv in the acquired immune deficiencv svndrome. Am I Path01 I 112:3%7-382, 1983 2. Welch K. Finkbeiner W. Aloers GE. et al: Autoosv fmdings in the acquired immune de’ficiency syndrome: JAMA 252:1152-l 159, 1984 3. Mobley K, Rotterdam HZ, Lerner CW, et al: Autopsy findings in the acquired immune deficiency syndrome. Path01 Annu 20:45-65, 1985 4. Niedt GW, Schinella RA: Acquired immunodeficiency syndrome. Arch Path01 Lab Med 109:727-734, 1985 5. Schmidts HL, Muller H, Falk S: Obduktionsbefunde beim erworbenen Immundefektsyndrom (AIDS). Pathologe 7:8-21, 1986 6. Chabon AB, Stenger RJ, Grabstald H: Histopathology of testis in acquired immune deficiency syndrome. Urology 29:658663, 1987 7. Russell RM: Nutritional assessment, in Wyngaarden JB, Smith LH Jr (eds): Cecil Textbook of Medicine. Philadelphia, Saunders, 1985, pp 1179-l 183 8. Johnsen SC: Testicular biopsy score count-A method for registration of spermatogenesis in human testes: Normal values and results in 335 hypogonadal males. Hormones 1:2-25, 1970 9. Santen RJ, Van den Bossche H, Symoens J, et al: Site of action of low dose ketoconazole on androgen biosvnthesis in men. J Clin Endocrinal Metab 57:732-736, 1985 ’ 10. Handelsman DJ, Swerdloff RS: Male gonadal dysfunction. Clin Endocrinol Metab 14:89-124, 1985 11. Felsenstein D, D’Amico DJ, Hirsch MS, et al: Treatment of
214
pita1 charts were reviewed for data regarding age, AIDS risk factors (homosexuality, intravenous [IV] drug abuse), history of sexually transmitted disease [STD], smoking, alcohol abuse, cirrhosis of the liver, chemotherapy, radiation to the pelvis, prolonged fever (more than 6 months), chronic illness (more than 6 months), and significant low weight (~70% of the reference weight according to the 1983 Metropolitan Life Insurance reference weights for height’). All available biopsy material concerning the AIDS patients was examined. The autopsy protocols and slides were reviewed for general information, main pathologic features of the AIDS patients, and main diseases of the controls. The gross descriptions and the slides of the testes were examined. In the AIDS group, one section of testis was available from 17 patients and two sections from 40 patients; in the control group, one section was studied in 20 patients and two sections in 35. All the sections were formalin-fixed, paraffin-embedded, and stained with hematoxylin-eosin; all the sections of AIDS testes were stained with Ziehl-Neelsen stain for acid-fast bacilli. The degree of spermatogenesis was determined using a modification of the Johnsen testicular score count method.8 For each case, 100 representative sections of the seminiferous tubules were evaluated and a score from one to ten was designated, according to the most mature germ cell stage identified. From these tubular scores, a mean sperm score was calculated for every subject, which proved to be an objective and reproducible value. The Leydig cells were judged as normal, increased, or decreased in number. (The Leydig cells constitute approximately 3% of the visual field in the healthy adult man.s) Other parameters noted were tubular scarring, thickening of the tubular basement membrane, interstitial fibrosis, nonspecific interstitial inflammation, abnormal thickening of the blood vessel walls, and the presence of microorganisms. Statistical analysis was performed using the Student’s t test.
Several recent autopsy series dealing with men who have died of the acquired immune deficiency syndrome (AIDS) have consistently described decreased spermatogenesis or testicular atrophy.‘-6 With one notable exception,6 none of these reports has focused on this subject or attempted to offer a satisfactory explanation for the almost universal testicular damage seen in AIDS patients. Our study was undertaken in an attempt to gain further insight into the pathogenesis of testicular atrophy and, in particular, to evaluate whether this injury is specific for AIDS or whether it is a nonspecific sequel of a wasting disease.
RESULTS MATERIALS AND METHODS
Clinical
We reviewed the material concerning 57 consecutive unselected male AIDS patients and 55 male age-matched control patients autopsied at St. Vincent’s Hospital and Medical Center (New York) from 1984 to 1987. The hos-
The age of the AIDS patients ranged from 26 to 69 years (mean, 41.1 years; SD, 10.4), and age of the control patients ranged from 24 to 64 years (mean, 44.2 years; SD, 9.7). Fifty-five AIDS patients (96%) were homosexual, of which one also had a history of IV drug abuse; two were heterosexual drug addicts. One control patient was homosexual and one had a history of IV drug abuse. Deceased IV drug users with or without AIDS are routinely referred to the medical examiner’s office and are therefore usually not autopsied at the hospital. The control group did not contain more homosexual patients because the sexual preference of non-AIDS patients is not routinely asked and almost all admitted homosexual pa-
From the Department of Pathology, St. Vincent’s Hospital and Medical Center, New York. Accepted for publication March 8, 1988. * Present address: Department of Pathology, Memorial SloanKettering Cancer Center, New York. Key words: testis, atrophy, AIDS, sperm score. Address correspondence and reprints requests to Marian Waxman, MD, Our Lady of Mercy Medical Center, Department of Pathology, 600 E 233rd St, Bronx, NY, 10466. 0 1989 by W.B. Saunders Company. 0046~8177/89/2003-0003$5.00/O
210
TESTlCULAR ATROPHY IN AIDS (De Paepe & Waxman)
tients who were autopsied died with AIDS. Venereal disease was reported in 34 of 35 AIDS patients (97%) who were specifically questioned (20 gonorrhea and syphilis, 11 gonorrhea only, three syphilis only). STD was not mentioned in the charts of the other 22 AIDS patients or in those of any of the control patients. Prolonged fever was found in 13 AIDS patients (23%) and nine controls (16%). Chronic illness was present in 40 AIDS patients (70%) and 24 controls (44%). Eighteen AIDS patients (32%) and 24 control patients (44%) were smokers. Ten AIDS patients (18%) and 16 controls (29%) had a history of alcoholism and/or cirrhosis. In both groups, 13 patients (23% to 24%) received chemotherapy (various combinations of antimetabolites, vinca alkaloids, alkylating agents, and antibiotics). In addition, 11 AIDS patients (and none of the controls) received drugs known to adversely affect the spermatogenesis: the antifungal drug ketoconazoleg (five cases), corticosteroids’O (four cases), and the antiviral agent BWB 759U (9[2-Hydroxy-1-(hydroxymethyl)ethoxymethyl] guanine, DHPG)” (two cases). No patient had received radiation treatment to the genital area. Significant underweight was found in 18 patients of the AIDS group (32%) and in eight controls (15%). In 5 1 AIDS cases (89%), death was attributed to overwhelming opportunistic infection (combined with Kaposi’s sarcoma [KS] in 16 cases and malignant lymphoma in four cases). In the remaining six AIDS patients, the causes of death were disseminated KS (three cases), malignant lymphoma (two cases), and KS with lymphoma (one case), without evidence of infection. The deaths of the control patients were attributed to malignancy (23 cases), cardiovascular disease (21 cases), infection (six cases), and complications of alcoholism (five cases).
Pathologic Findings The testes were not measured or weighed routinely. Decreased stringing was described in eight AIDS patients (14%) and four control patients (7%). Two AIDS cases showed grossly small granulomas, and two others showed broad fibrous bands, Thickening of the tubular basement membrane, nonspecific interstitial inflammation (Fig l), interstitial fibrosis (Fig 2), and infection were found more often in the testes of AIDS patients than in the controls (Table 1). The evidence of a specific infection was found in 18 AIDS patients; the microorganisms were Mycobacterium avium-intracellular (MAI) in six cases, Toxoplasma in five, and cytomegalovirus (CMV) in eight (one with Toxophma). In all cases, the same organism was identified in at least one other organ at autopsy. These three organisms were seen within the seminiferous tubules and in the interstitium, and CMV inclusions were also frequently noted in the endothelial cells. Two cases of GMV showed tubular microabscesses and thrombi in small blood vessels. MA1 infection presented in three testes as aggregates of “striated blue” histiocytes and in three other cases as granulomas, associated with central necrosis in one instance. Thus, this last case was similar to tuberculous orchitis
FIGURE 1. Marked interstitial inflammatory infiltrate composed mostly of lymphocytes. (Hematoxylin-eosin stain; magnification x 100.)
except for the multitude of acid-fast bacilli characteristic of MAI. The two groups showed no significant differences in the frequency of vascular wall thickening, abnormal Leydig cell numbers, and marked tubular scarring. There was no evidence of KS or lymphoma in any testis. The mean sperm score of the testes of AIDS patients was found to be significantly lower than that of the controls (3.94 v 6.22; P < .OOl) (Table 1). Sixtyeight of the AIDS testes had mean sperm score values of 2 to 4 only (Fig 3), whereas the testes of the control patients displayed a more even distribution of the scores, with 42% of the cases above 7. The mean sperm score was significantly lower in infected than in non-infected testes of AIDS patients (3.42 v 3.94; P < .05) (Table 2).
Correlation of the Mean Sperm Score With the Clinical Features Within each study group, a history of chronic illness, prolonged fever, low weight, and chemother-
FIGURE 2. Severe interstitial fibrosis which merges with the thickened fibrotic basal laminae of the tubules. (Hematoxylin-eosin stain; magnification x100.)
211
HUMAN PATHOLOGY TABLE
Volume 20, No. 3 (March 1989)
1. Histologic Features of the Testes AIDS (n = 57)
Sections examined Tubular scarring Thickening basement membrane Interstitial fibrosis Interstitial inflammation Thickened vessel walls Specific infection Leydig cells Increased Decreased Sperm score Mean Range
Controls
TABLE
Effect of Specific
Infection
on the Sperm Score Sperm
(n = 55)
97 4
90 3
29
13
29
2
16
5
8
6
18
0
8 3
6 1
3.94 1.22-8.99
2.
AIDS All (n = 57) Specific infection (n = 18) CMV (n = 8) Toxoplasma (n = 5) MA1 (n = 6)
Score
3.94 r 1.70 3.42 ? 0.97 3.51 2.86 3.68
reflects the poor correlation between the gross and microscopic appearance of the testis, as noted by others.’ In two cases, a significant finding at gross inspection consisted of areas of granulomas, confirmed as testicular mycobacteriosis by light microscopy. A specific infection of the testis was diagnosed in 32% of our AIDS patients, slightly more than the 22% found in the series of Chabon et al.‘j Testicular infection has been reported by other investiin view of the overgators 1~2,4-6,12and is expected whelming infections often seen in AIDS patients. Our finding of CMV, Toxoplasma, and MA1 organisms within the lumina of the seminiferous tubules is significant in that it indicates the possibility of their transmission via the semen, as reported previously for CMV.13 Infection with Toxoplasmu was associated with only a few mononuclear inflammatory cells and the organisms were rather inconspicuous; the necrosis and accumulation of eosinophiles’* were missing. The much higher incidence of the nonspecific interstitial inflammation and of interstitial fibrosis in the AIDS group was noteworthy, compared with the control group (28% u 9% and 5 1% v 4%, respectively). Chabon et al reported finding interstitial inflammation in 53% of the testes of their AIDS patients.6 Although the significance of these findings is not clear at present, they may possibly represent a morphologic expression of autoimmune orchitis. Generally, we feel that one or two sections per case were representative of the significant pathology in our review; there were no significant findings limited to one testis when sections of the two were available for review. The same pertains to the patient’s
6.22 1.55-8.96
apy was associated with decreased spermatogenesis, whereas age over 50 years, alcohol abuse, cirrhosis, and smoking did not seem to affect the testes consistently (Table 3). Comparison of the mean sperm score of the AIDS and control patients with regard to chronic illness, prolonged fever, and low weight revealed significantly lower scores in the AIDS patients in all three subgroups (P < .05 to .OOl). Chemotherapy was found to have an equally adverse effect on spermatogenesis in the testes in both groups, lowering the mean sperm score to 3.71 (AIDS group) and 4.21 (control group) (P > .05). Ketoconazole and steroid intake and a history of IV drug abuse did not significantly correlate with the degree of testicular atrophy, while the two patients who had received BWB 759U displayed a mean sperm score far above the average of both patient groups. DISCUSSION The gross features of the testes were described as abnormal in only 2 1% of the AIDS patients, which
TABLE
3.
Correlation of Sperm Score With Clinical Features AIDS
FIGURE 3. Seminiferous tubules lined by Sertoli cells and spermatogonia; tubular score 3. Artifactual space separates the epithelium from the markedly thickened fibrotic basal lamlna. (Hematoxylin-eosin stain; magnification x250.)
212
P
Controls
All patients
3.94 (n = 57)
6.22 (n = 55)
<.OOl
Age over 50 yr Chronic illness Prolonged fever Underweight Alcoholism/cirrhosis Tobacco Chemotherapy
3.93 3.76 3.64 3.57 3.68 3.92 3.71
6.62 5.09 5.42 5.38 6.84 6.25 4.21
<.OOl <.OOI <.05 <.05 <.OOI <.OOl NS
Other medications Ketoconazole, corticosteroids BWB 759U
3.56 (n = 9) 8.79 (n = 2)
IV drug
3.94 (n = 3)
abuse
(n (n (n (n (n (n (n
= = = = = = =
8) 40) 13) 18) IO) 18) 13)
(n (n (n (n (n (n (n
= = = = = = =
21) 24) 9) 8) 16) 24) 13)
5.06 (n = 1)
NS
TESTICULARATROPHYIN AIDS (De Paepe & Waxman)
main sperm score which was generally the same when both testes were examined and compared. Despite widespread KS diagnosed in 35% of the AIDS patients, we did not see the tumor in any testis. Only a few AIDS patients with KS involving the epididymis,’ tunica albuginea,5 and testis (not further specified)4 have been described. Testes in our AIDS patients showed significantly impaired spermatogenesis and increased interstitial fibrosis and thickening of basement membrane, as compared with those of the age-matched autopsy population, confirming the association of AIDS and testicular atrophy noted by other observers.‘-6 The exact pathogenesis of the testicular atrophy is unclear; a wide variety of known causes of decreased spermatogenesis applies to AIDS patients. When possible, we correlated the degree of spermatogenesis with the histologic and clinical features of the patients in order to determine the relative importance of the various factors examined. Our two groups were age-matched and contained comparable numbers of patients with a history of alcoholism and smoking; none of these known causes of testicular atrophy (old age,14 alcoholism/cirrhosis,15r16 smoking”) showed a distinct correlation with the sperm score (Table 3). The number of Leydig cells was essentially normal in the vast majority of our cases and showed no correlation with the degree of spermatogenesis, confirming the observations of other investigators.5*6 Despite this, endocrine imbalances cannot be excluded as an etiology for the testicular atrophy in AIDS patients since pituitary and hypothalamic disorders may manifest primarily as affecting the seminiferous tubules.r8 Since AIDS is almost inherently associated with chronic illness, stress, prolonged fever, and malnutrition, it is tempting to ascribe the observed testicular changes to these classic causes of testicular atrophy. lgTz3The degree of spermatogenesis in AIDS patients with chronic illness and a low body weight at the time of death was lower than the average of the entire AIDS group (Table 3), which confirms the described detrimental effect of these factors on the testis. Since the same factors were found to be associated with less severe atrophy in the control group (Table 3), it is suggested that the chronicity of the disease contributes to the decreased spermatogenesis in these patients, although it is not the primary cause. The presence of infiltrative disease of hypothalamus and pituitary gland (eg, infection, lymphoma) could not be evaluated since sections of these organs were not routinely examined. Since most of the AIDS patients in our study are homosexual, the endocrine implications of homosexuality should be considered. The literature on this subject is conflicting.24-26 We are unable to draw any conclusions in this regard from our study because our control group contained only one homosexual. Several mechanisms could be proposed to explain the low sperm score among the AIDS patients with testicular infections (Table 2). A direct effect on the germinal epithelium has been ascribed to mumps
and other viruses,*’ and could also exist for the organisms identified in the testes in this series and for the human immunodeficiency virus (HIV) itself. Retrovirus-like particles, consistent with HIV, have been demonstrated ultrastructurally in the testis of AIDS patients.28 CMV infection of the male genital tract has been associated with alterations in the pH of the seminal fluid and a reduction in the sperm count.2g In our series, two CMV-infected testes showed multiple thrombi which could lead to ischemic damage to the testis, as postulated by others.’ Several investigators have proposed immune phenomena as possible mechanisms for the testicular atrophy in homosexual AIDS patients,2r4s5*6 based on the finding of elevated antisperm antibody titers in the serum of homosexual men compared with heterosexual men.3o Our finding of a high incidence of interstitial inflammation and fibrosis might support this theory (see above). The hypothesis of homosexuality-associated autoimmune orchitis as a major cause of testicular damage in AIDS can be challenged, however, since several autopsy series describe the same degree of testicular atrophy in AIDS patients whether they are homosexual or not.3,4,6 Thirteen AIDS patients (23%) had received chemotherapy. In all cases, at least one drug with known adverse effects on the testis of humans or laboratory animals was included in the therapeutic regimen.3’ Since there was no significant difference in the degree of testicular atrophy in AIDS patients and controls with chemotherapy, it can be inferred that testicular changes in this subgroup are secondary to the chemotherapy and not due to AIDS. Some of the AIDS patients received drugs with known adverse effects on spermatogenesis (corticosteroids,r” ketoconazole,g and the experimental drug BWB 759U1’), but in view of the small sample sizes, no conclusions can be drawn concerning the effect of these drugs on the testis in AIDS patients. No AIDS patient in our study had a history of radiation treatment, which is known to adversely affect spermatogenesis.32 The blood vessels of the testes of AIDS patients were not found to differ from those of the testes of the control patients; therefore, a vascular compromise with secondary ischemia is not a likely cause of the testicular dysfunction in AIDS, with the possible exception of testes infected with CMV. The testes of our AIDS patients with a history of IV drug abuse had a similar appearance to those of homosexual AIDS patients, an experience shared by Chabon et al6 The small number of cases precluded any conclusions regarding detrimental effects of opiatesS3 and other recreational drugs on the testis of AIDS patients. behavior According to some, 34,35 homosexual could lead to zinc deficiency, another known cause of decreased spermatogenesis. Many of the AIDS patients studied had a history of venereal disease, which confirms the reported high incidence of STDs among homosexuals.36 Genital tract infection has been associated with decreased fertility,37*38 but the impact of venereal disease on the testes of AIDS patients could 213
HUMAN PATHOLOGY
Volume 20, No. 3 (March 1989)
cytomegalovirus retinitis with 9-[2-hydroxy-1-(hydroxy methyl) ethoxymethyl] guanine. Ann Intern Med 103:377-380, 1985 12. Nistal M, Santana A, Paniaqua R, et al: Testicular toxoplasmosis in two men with the acqmred immunodeficiency syndrome (AIDS). Arch Path01 Lab Med 110:744-746. 1986 13.‘ Biggar RJ, Andersen HK, Ebbesen P, et Seminal fluid excretion of cytomegalovirus related to immunosuppression in homosexual men. Br Med J 286:2010-2013, 1983 14. Warner BA, Dufau ML, Santen RJ: Effects of aging and illness on the pituitary testicular axis in men: Qualitative as well as quantitative changes in luteinizing hormone. J Clin Endocrinol Metab 60:263-268, 1985 15. Galvao-Teles A, Goncalves L, Carvalho H, et al: Alterations of testicular morphology in alcoholic disease. Alcoholism (NY) 7: 144-148, 1983 16. Anderson RA Jr, Willis BR, Oswald C, et al: Male reproductive tract sensitivity to ethanol: A critical overview. Pharmacol Biochem Behav 18:305-310, 1983 (suppl 1) 17. Handelsman DJ, Conway AJ,-Boylan LM, et al: Testicular function in potential sperm donors: Normal ranges and the effects of smoking and varicocele. Int ] Androl 7:369-382, 1984 18. Ree GH, Martin F, M$es K, et al: Hormonal changes in human leprosy. Lepr Rev 52:121-126, 1981 19. Morley JE, Melmed S: Gonadal dysfunction in systemic disorders. Metabolism 28: 1051-1073, 1979 20. Chlebowski RT, Heber D: Hypogonadism in male patients with metastatic cancer prior to chemotherapv. Cancer Res 42:24952498, 1982 21. Collu R, Gibb W, Ducharme JR: Effects of stress on the cronadal function. I Endocrinol Invest 7:529-537. 1984 22. Macleod J, Hotchkiss RS: The effect of hyperpyrexia y;;; spermatozoa counts in men. Endocrinology 28:780-784,
not be determined because of the absence of history of sexually transmitted diseases in the control group. SUMMARY
ai:
This study has revealed statistically significant morphologic differences between the testes of homosexual AIDS patients and age-matched heterosexual control patients without AIDS. The chronicity of the disease, prolonged fever, malnutrition, testicular infection, and chemotherapy all contribute to the testicular atrophy of AIDS patients. With the exception of chemotherapy, however, none of these factors can entirely explain the severity of the testicular damage. Due to the absence of equal numbers of homosexual patients in the control group, the effects of homosexuality and AIDS on the testis cannot be clearly separated. Homosexuals, the largest AIDS risk group, are in theory particularly predisposed to testicular atrophy: venereal disease, antisperm antibodies, and possible zinc deficiency and endocrine changes have all been associated with homosexuality and are known causes of decreased spermatogenesis. Testicular atrophy in AIDS patients is not limited to homosexuals and has been reported to occur with the same frequency and severity in AIDS patients of other risk groups. Factors more specifically related to AIDS may therefore underlie the testicular atrophic changes; a direct adverse effect of HIV on the germinal epithelium and as yet undetermined endocrine phenomena can be postulated.
L
.
,
”
23. Steinberger E, Dixon WJ: Some observations on the effect of heat on the testicular germinal epithelium. Fertil Steril 10:578595, 1959 24. Tourney G, Petrilli AJ, Hatfield LM: Hormonal relationshins in homosexual men. Am I Psvchiatrv 132:288-290. 1975 ’ 25. Doerr P, Pirke KM, K&k&t G, et al: Further studies on sex hormones in male homosexuals. Arch Gen Psychiatry 33:61 l614, 1976 26. Kolodny RC, Masters WH, Hendry XJ, et al: Plasma testosterone and semen analysis in male homosexuals. N Engl J Med 285:1170-1174, 1971 27. Mikuz G, Damjanov I: Inflammation of the testis, epididymis, peritesticular membranes and scrotum. Pathol Annu 17:101128, 1982 28. Lecatsas G, Houff S, Macher A, et al: Retrovirus-like particles in salivary glands, prostate and testes of AIDS patients. Proc Sot Exp Biol Med 178:653-655, 1985 29: Lang DJ, Kummer JF, Hartley DP: Cytomegalovirus in semen. Persistence and demonstration in extracellular fluids. N Engl J Med 291:121-123, 1974 30. Witkin SS, Sonnabend J: Immune responses to spermatozoa in homosexual men. Fertil Steril 39:337-342, 1983 3 1. Chapman RM: Gonadal injury resulting from chemotherapy. Am J Ind Med 4:149-161, 1983 32. Ash P: The influence of radiation on fertility in man. Br J Radio1 53:271-278, 1980 33. Wang C, Chan V, Yeung RTT: The effect of heroin addiction on oituitarv-testicular function. Chn Endocrinol9:455-461. 1978 ‘ ’ 34. Weiner RG: AIDS and zinc deficiency. JAMA 252: 14091410, 1984 35. Apgar J: Zinc and reproduction. Annu Rev Nutr 5:43-68, 1985 36. Judson FN, Penley KA, Robinson ME, et al: Comparative prevalence rates of sexually transmitted diseases in heterosexual men. Am J Epidemiol 112:836-843, 1980 37. Quesada EM, Dukes CD, Deen GH, et al: Genital infection and sperm agglutinating antibodies in infertile men. J Urol99: 106108, 1968 38. Witkin SS, Toth A: Relationship between genital tract infections, sperm antibodies in seminal fluid and infertility. Fertil Steril 40:805-808, 1983
Acknowledgment. The authors thank Dr Francois Luks for assistance in preparing the manuscript, Denise Mitchell for secretarial work, and Dr Julius Weber for photography.
REFERENCES 1. Reichert CM, O’Leary TJ, Levens DL, et al: Autopsy pathologv in the acquired immune deficiencv svndrome. Am I Path01 I 112:3%7-382, 1983 2. Welch K. Finkbeiner W. Aloers GE. et al: Autoosv fmdings in the acquired immune de’ficiency syndrome: JAMA 252:1152-l 159, 1984 3. Mobley K, Rotterdam HZ, Lerner CW, et al: Autopsy findings in the acquired immune deficiency syndrome. Path01 Annu 20:45-65, 1985 4. Niedt GW, Schinella RA: Acquired immunodeficiency syndrome. Arch Path01 Lab Med 109:727-734, 1985 5. Schmidts HL, Muller H, Falk S: Obduktionsbefunde beim erworbenen Immundefektsyndrom (AIDS). Pathologe 7:8-21, 1986 6. Chabon AB, Stenger RJ, Grabstald H: Histopathology of testis in acquired immune deficiency syndrome. Urology 29:658663, 1987 7. Russell RM: Nutritional assessment, in Wyngaarden JB, Smith LH Jr (eds): Cecil Textbook of Medicine. Philadelphia, Saunders, 1985, pp 1179-l 183 8. Johnsen SC: Testicular biopsy score count-A method for registration of spermatogenesis in human testes: Normal values and results in 335 hypogonadal males. Hormones 1:2-25, 1970 9. Santen RJ, Van den Bossche H, Symoens J, et al: Site of action of low dose ketoconazole on androgen biosvnthesis in men. J Clin Endocrinal Metab 57:732-736, 1985 ’ 10. Handelsman DJ, Swerdloff RS: Male gonadal dysfunction. Clin Endocrinol Metab 14:89-124, 1985 11. Felsenstein D, D’Amico DJ, Hirsch MS, et al: Treatment of
214