Anemia and HIV in the Antiretroviral Era: Potential Significance of Testosterone

200

AIDS RESEARCH AND HUMAN RETROVIRUSESVolume 21, Number 3, 2005, pp. 200–206© Mary Ann Liebert, Inc.

Anemia and HIV in the Antiretroviral Era: Potential Significance of Testosterone

CAROLINE BEHLER,1 STARLEY SHADE,2 KELLAN GREGORY,2 DONALD ABRAMS,2

and PAUL VOLBERDING1,2

ABSTRACT

Anemia, the most common hematological disorder in human immunodeficiency virus (HIV) infection and ac-quired immunodeficiency syndrome (AIDS), is associated with decreased quality of life and survival. Hypo-gonadism is prevalent in advanced HIV disease, however, low testosterone levels have not been customarilyimplicated in HIV-associated anemia. This study was undertaken to determine whether there is a relation-ship between testosterone levels and androgen use with anemia in HIV, and to characterize other clinical cor-relates of HIV-associated anemia. This was a cross-sectional, observational study of 200 HIV-positive patientsat a public hospital HIV clinic from July 2000 to August 2001. A written questionnaire detailed previous andcurrent medication use, opportunistic infections, and malignancies. Hematological and virological parame-ters, testosterone, and erythropoietin levels were measured; CD4� T lymphocyte count and viral load nadirand peak levels were obtained from the computerized medical record. Anemia was defined as hemoglobin�13.5 g/dl in men and �11.6 g/dl in women. Twenty-four percent of women and 28% of men were anemic.Anemia was associated with lymphopenia (adjusted OR 4.0, 95% CI 1.36–11.80), high erythropoietin levels(adjusted OR 7.73, 95% CI 2.92–20.48), and low testosterone levels (adjusted OR 3.27, 95% CI 1.01–10.60).Anemia was negatively associated with female sex (adjusted OR 0.30, 95% CI 0.11–0.85), current antiretro-viral therapy (adjusted OR 0.43, 95% CI 0.20–0.95), current androgen use (adjusted OR 0.20, 95% CI0.05–0.84), and macrocytosis (adjusted OR 0.23, 95% CI 0.09–0.61). Low testosterone levels may have a pos-itive association and supplemental androgens a negative association with anemia in HIV disease.

INTRODUCTION

ANEMIA IS THE MOST COMMON HEMATOLOGICAL ABNORMAL-ITY IN HUMAN IMMUNODEFICIENCY VIRUS (HIV), with an es-

timated prevalence of 63–95%, increasing in prevalence withgreater severity of HIV disease.1–4 Although antiretroviral ther-apy has resulted in increased survival with HIV, and is associ-ated with lower rates of anemia,3 anemia remains a commonproblem even for patients treated with antiretroviral agents. Notonly is anemia associated with fatigue and decreased quality oflife,5,6 but it is also independently associated with decreasedsurvival with HIV.2,7–11 Some studies have demonstrated im-provements in fatigue and quality of life in HIV-positive pa-tients treated for anemia.12–14

Anemia of chronic disease, also called anemia of inflamma-tion, is the most common cause of anemia in HIV infection,

and is thought to be due to decreased erythropoietin produc-tion, decreased erythropoietic response to erythropoietin, andtrapping of iron in reticuloendothelial cells.15,16 Other potentialcauses of anemia include the use of antiretroviral therapy (es-pecially zidovudine and D4T) and other medications (trimetho-prim-sulfamethoxazole, ganciclovir, and amphotericin), bonemarrow infiltration with neoplastic or infectious diseases (tu-berculosis and fungal infections), parvovirus infection, nutri-tional deficiencies (vitamin B12, folate, and iron) as well asother associated diseases such as liver disease from hepatitis Bor C virus infection and chronic kidney disease.1,11,17–19 HIVinfection of bone marrow stromal cells and, rarely, autoimmunehemolytic anemia19 have also been implicated.

Hypogonadism is also common in HIV infection,20 and isassociated with weight loss or AIDS wasting syndrome.21–23 Itmay also contribute to HIV-associated anemia, though this has

1Department of Medicine, San Francisco Veterans Affairs Medical Center, University of California–San Francisco, San Francisco, California94143.

2Positive Health Program, San Francisco General Hospital, University of California–San Francisco, San Francisco, California 94143.

5489_02_p200-206 3/15/05 2:30 PM Page 200

not yet been investigated in detail. Patients with hypogonadismare often anemic, and the anemia usually improves with an-drogen therapy; indeed, polycythemia is a known complicationof supplemental androgens. Randomized trials of androgen ther-apy in HIV-positive men for AIDS wasting syndrome haveshown increased hemoglobin and hematocrit levels in subjectstreated with androgens compared to those in the placebogroup.24,25 Androgens are thought to influence erythropoiesisthrough increased erythropoietin production in the kidney,26 in-creased sensitivity to erythropoietin,27 increased hemoglobinsynthesis,28 and by directly stimulating erythroid precur-sors.27,29–32

This study was undertaken to investigate whether de-creased testosterone levels are associated with anemia andsupplemental androgen use is inversely correlated with ane-mia, as well as to characterize the clinical correlates of ane-mia in a population of HIV� patients of a public medical cen-ter clinic.

MATERIALS AND METHODS

Study design, setting, and participants

In this cross-sectional, observational study, HIV� patientswho had been sent for venipuncture at the San Francisco Gen-eral Hospital HIV outpatient specialty clinic in July 2000–Au-gust 2001 were consecutively asked to participate in the studyuntil a total of 200 participants were accrued. Written informedconsent was obtained from all of the participants. This studywas approved by the San Francisco General Hospital, Univer-sity of California–San Francisco human subjects committee.

Measurements

Questionnaire. Patients were asked about previous and cur-rent antiretroviral use (zidovudine, D4T, other nucleoside re-verse transcriptase inhibitors, nonnucleoside reverse tran-scriptase inhibitors, and protease inhibitors), AIDS-defining

TESTOSTERONE IN HIV-ASSOCIATED ANEMIA 201

TABLE 1. SELECTED PATIENT CHARACTERISTICSa

Anemic (%) Not Anemic (%) All (%) p value

N 55 (28) 145 (73) 200 (100)Mean age (�SD) 44 (�8) 47 (�7) 46 (�7)Gender

Male 46 (84) 115 (79) 161 (81)Female 9 (16) 29 (20) 38 (19) 0.69Transgender (MTF) 0 ( 1 (1) 1 (1)

EthnicityWhite 30 (55) 88 (61) 118 (59)African-American 15 (27) 32 (22) 47 (24)Hispanic 7 (13) 16 (11) 23 (12) 0.42Asian/Pacific Islander 2 (4) 9 (6) 11 (6)Other 1 (2) 0 ( 1 (1)

History of injection drug use 7 (13) 10 (7) 17 (9) 0.17AIDS-defining illness history: 15 (27) 32 (22) 47 (24) 0.44

Pneumocystis carinii 11 (20) 14 (10) 25 (13) 0.048Kaposi’s sarcoma 4 (7) 11 (8) 15 (8) 0.94M. avium complex 2 (4) 7 (5) 9 (5) 0.71M. tuberculosis 3 (5) 6 (4) 9 (5) 0.70Cytomegalovirus 1 (2) 6 (4) 7 (4) 0.43Non-Hodgkin’s lymphoma 1 (2) 5 (3) 6 (3) 0.55Toxoplasmosis 0 (0) 4 (3) 4 (2) 0.21

Hepatitis B infection 2 (4) 9 (6) 11 (6) 0.098Hepatitis C infection 17 (31) 34 (23) 51 (26) 0.18Medication history

Antiretroviral therapy—ever 42 (76) 133 (92) 175 (88) 0.003Antiretroviral therapy—current 28 (51) 111 (77) 139 (70) 0.0004Current other medication use

TMP/SMXb 20 (37) 39 (27) 59 (30) 0.16Androgens (men only)c 3 (7) 20 (18) 23 (15) 0.088EPOd 1 (2) 0 (0) 1 (1) 0.095G-CSFe 1 (2) 0 (0) 1 (1) 0.10

aPatient characteristics are compared among patients with and without anemia using chi-square tests of homogeneity and t teststo compare means. Percentages reported reflect the proportion of the population in the column with the particular patient char-acteristic.

bTrimethoprim sulfamethoxasole, n � 199, data missing for 1.cSupplemental androgens, n � 155, data missing for 6.dErythropoietin, n � 196, data missing for 4.eGranulocyte-colony-stimulating factor, n � 196, data missing for 4.

5489_02_p200-206 3/15/05 2:30 PM Page 201

illnesses, hepatitis B and C status, intravenous drug use, pastand current use of trimethoprim–sulfamethoxazole, androgens,erythropoietin, granulocyte colony-stimulating factor, andmethadone.

Laboratory parameters. Hemoglobin, red blood cell count,white blood cell count, mean corpuscular volume, absolute lym-phocyte count, platelet, current CD4� lymphocyte count, viralload (VL), testosterone, and erythropoietin levels were mea-sured at the time of the study. The computerized medical recordwas abstracted for CD4 count and VL nadir and peak levels,and for hepatitis B surface antigen and hepatitis C antibody sta-tus. Anemia was defined as hemoglobin �13.5 g/dl in men and�11.6 g/dl in women, thrombocytopenia as platelet count�150 � 109 platelets/liter, lymphopenia as absolute lympho-cyte count �1.0 � 109 cells/liter, normal testosterone level formen as �270 ng/dl, and normal erythropoietin level as 4.1–19.5mU/ml, based on the normal laboratory parameters at San Fran-cisco General Hospital.

Statistical analysis

All analysis was conducted using PC SAS, version 8.2. Pa-tient characteristics were compared among those with and with-out anemia using chi-square tests of homogeneity or Fisher’sexact tests for small expected cell sizes to compare categoricalvariables, and Student’s t tests to compare means. Next, we as-sessed correlations among covariates that differed among thosewith and without anemia to assess possible interaction. We usedlogistic regression to obtain unadjusted odds ratios and 95%confidence intervals (CI) for each predictor. We explored pos-sible interaction between covariates by including two variablesand their interaction term in these simple models. Next, we in-cluded variables that achieved a p value of �0.1 (�0.20 for in-

teraction terms) in a multivariable logistic regression model.From this model we obtained adjusted odds ratios and 95% CIfor variables in the model. Nadir CD4� T cell counts and peakviral load were not included in multivariable models becausethey were highly associated with patients’ most recent CD4�

T cell counts and viral load, respectively.

RESULTS

Description of population

This population was predominantly male (81%) and white(59%). Almost 90% of the subjects had used antiretroviral ther-apy at some time, and 70% reported current antiretroviral use.Of the subjects 24% reported a history of opportunistic infec-tions, half of which were Pneumocystis carinii pneumonia(PCP) (Table 1).

Anemia was the most common hematological abnormality,present in 28% of the population (24% of women and 28% ofmen), followed by thrombocytopenia in 14% and lymphopeniain 10% (Table 2). The distribution of hemoglobin levels amongmen and women is shown in Fig. 1. The mean and nadir CD4counts were 384 and 186 cells/mm3, respectively. Individualswith anemia had lower mean CD4� T lymphocyte counts, weremore likely to be lymphopenic, less likely to be on antiretrovi-ral therapy, and had higher erythropoietin levels than nonane-mic subjects (Tables 1 and 2). Anemia was present in 20% ofindividuals on antiretroviral therapy and 44% of those who werenot (p � 0.0004); excluding those who were antiretroviral-naive, anemia was found in 20% and 39%, respectively (p �0.019).

Low testosterone levels were found in 15% of men not us-ing supplemental androgens, and were more prevalent in men

BEHLER ET AL.202

TABLE 2. LABORATORY VALUES

Anemic Not anemic All patients p-value

Mean CD4� T lymphocyte count [cells/mm3 (�SD)]Nadir 144 (�135) 202 (�150) 186 (�148) �0.0001Most recent 267 (�204) 428 (�241) 384 (�242) �0.0001

Mean log10 viral load (�SD)Peak 5.0 (�0.8) 4.4 (�1.0) 4.6 (�1.0) �0.0001Most recent 4.0 (�1.5) 2.5 (�1.3) 2.9 (�1.5) �0.0001

Lymphopenia 11 (20) 8 (6) 19 (10) 0.008(lymphocytes �1.0 � 109/liter)

Thrombocytopenia 10 (19) 17 (12) 27 (14) 0.070(platelets �150 � 109/liter)

Microcytosis (MCVa �80 fl) 7 (13) 4 (3) 11 (6) 0.014Macrocytosis (MCV �100 fl) 10 (18) 61 (42) 71 (36) 0.002Erythropoietin levelb

High (�19.5 mU/ml) 22 (40) 35 (25) 57 (29)Normal (4.1–19.5 mU/ml) 33 (60) 105 (75) 138 (70) 0.03Low (�4.1 mU/ml) 0 1 (1) 1 (1)

Testosterone levelc

Normal/high (�270 ng/dl) 30 (75) 82 (89) 112 (85) 0.037Low (�270 ng/dl) 10 (25) 10 (11) 20 (15)

aMean corpuscular volume.bData missing for 4, n � 196.cMen only, excluding 23 with current supplemental androgen use, data missing for 6, n � 132.

5489_02_p200-206 3/15/05 2:30 PM Page 202

who were anemic (25%) than not anemic (11%) (p � 0.037)(Table 2). Subgroup analysis did not show a significant differ-ence in testosterone levels with and without supplemental an-drogen use; however, erythropoietin levels seemed to vary: 50%of men on supplemental androgens had greater than normal ery-thropoietin levels versus 21% not on supplemental androgens(p � 0.004).

Predictors of anemia

We compared patients with anemia to those without anemiausing bivariate analyses (Tables 2 and 3). Anemic patients ap-peared to have more advanced HIV diseased than patients with-out anemia. They had lower recent and nadir CD4� T lym-phocyte cell counts, as well as higher recent and peak viralloads. They were less likely to have ever taken antiretroviraltherapy and were less likely to be currently taking antiretrovi-ral therapy than those without anemia. There was no significantassociation between current zidovudine use and anemia. Withthe exception of Pneumocystis carinii pneumonia, opportunis-tic infections and hepatitis B and C were not found to be sig-nificantly associated with anemia. Erythropoietin use (only twopatients in this population) was also not significantly associatedwith anemia.

We then compared patients with anemia to other patients us-ing multivariable logistic regression (Table 3). Anemia was lesscommon in women than in men when adjusted for absolute lym-phocyte count, mean corpuscular volume, current antiretro-viral therapy, and androgen use (adjusted OR 0.30, 95% CI0.11–0.85) (Table 3). Other variables that had a negative asso-ciation with anemia include current antiretroviral therapy (ad-justed OR 0.43, 95% CI 0.20–0.95), current androgen use (ad-justed OR 0.20, 95% CI 0.05–0.84), and macrocytosis (adjustedOR 0.23, 95% CI 0.09–0.61) (Table 3). Anemia was associatedwith lymphopenia (adjusted OR for anemia 4.0, 95% CI1.36–11.80) (Table 3) and lower CD4 counts (267 cells/mm3

in anemic, 428 cells/mm3 in not anemic, p � 0.0001) (Table 2).The interaction between erythropoietin and testosterone lev-

els and their associations with anemia was examined. Usinganemia, low/normal erythropoietin (EPO) level and high/nor-

mal testosterone level as the reference variables, high EPO lev-els (adjusted OR 7.73, 95% CI 2.92–20.48) and low testosteronelevels (adjusted OR 3.27, 95% CI 1.01–10.60) were both asso-ciated with anemia (Table 3).

These results did not change significantly if results for menand women are analyzed separately. The combined effect ofhigh EPO and low testosterone levels did not appear to have astatistically significant association with anemia.

DISCUSSION

This study describes characteristics associated with anemiain an HIV� population in an urban HIV subspecialty clinic.Anemia was less prevalent in this population than in previousstudies, though still common despite a high rate of antiretrovi-ral use. Some of the findings in this study concur with thosepublished previously, including the association of normal he-moglobin levels with current antiretroviral therapy and macro-cytosis, and an association between lymphopenia and lowerCD4 counts with anemia3,4,33 The negative association ofmacrocytosis with anemia is most likely a reflection of currentantiretroviral use, as this is a known effect of zidovudine andd4T. Anemia appeared to be less prevalent in women than inmen in this population; this may in part be due to the fact thata lower hemoglobin level was used to define anemia in women.Figure 1 shows that hemoglobin levels tended to distribute at alower range in women than in men.

Low testosterone levels were associated with anemia in menin this population. Not only were men with low testosteronelevels more likely to be anemic but current androgen use hada negative correlation with anemia. Subgroup analysis did notshow a significant difference in testosterone levels with andwithout supplemental androgen use. This suggesting that sup-plemental androgens brought low testosterone levels into thenormal range; however, we do not have testosterone levels pre-and postandrogen therapy to confirm this. Men on supplemen-tal androgens had higher erythropoietin levels than those notusing androgens. The cause of this association is not apparent;


0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

<8

8.0-

8.9

9.0-

9.9

10.0

-10.

9

11.0

-11.

9

12.0

-12.

9

13.0

-13.

9

14.0

-14.

9

15.0

-15.

916

.0+

Hemoglobin (g/dL)

Per

cen

t o

f to

tal p

op

ula

tio

n

Male (n=161)Female (n=38)

FIG. 1. Hemoglobin level by gender.

5489_02_p200-206 3/15/05 2:30 PM Page 203

one hypothetical mechanism is the stimulation of erythropoi-etin production by supplemental androgens (to a greater degreethan endogenous androgens) resulting in increased hemoglobinlevels, though this is speculative. The negative association ofsupplemental androgen use with anemia could also be relatedto a confounding factor, such as a higher degree of complianceor health care utilization among these participants.

In this population, anemia was correlated with high erythro-poietin levels, suggesting that the erythropoietin response wasappropriate, but the degree of response was not sufficient. Ifthis is true, then increasing erythropoietin levels by exogenouserythropoietin or androgen administration may be of benefit.Alternately, the association of high erythropoietin levels withanemia could indicate causes other than a blunted erythropoi-etin response, such as iron deficiency, bone marrow infiltration,or acute blood loss, though these causes of anemia were notspecifically investigated. The combined effect of high erythro-poietin and low testosterone levels did not have a statisticallysignificant association with anemia, though this may be due toa low number of participants in this category.

There are some notable weaknesses to this study. The cross-sectional design does not allow for the determination of causeand effect or the time course of any associations. The popula-tion studied was a convenience sample selected from patientsseen in clinic and sent for laboratory work who agreed to par-ticipate. This may be a more compliant or healthier populationthan those who did not meet these criteria, limiting the gener-

alizability of the results. Data on comorbidities were incom-plete or largely based on self-report; antiretroviral use was alsoprimarily ascertained by self-report. Only total testosterone lev-els were measured, and not free testosterone, a more sensitiveindicator of androgen deficiency. Fluctuation in sex hormone-binding globulin (SHBG) levels from acute illness or hepaticdysfunction could affect the total testosterone level without af-fecting free testosterone. Additionally, erythropoietin may in-crease SHBG levels, as has been shown in a study of recom-binant erythropoietin.34

This study is unique in that it describes testosterone levelsand exogenous androgen use in patients with HIV-associatedanemia. An interaction between androgens and anemia in HIVhas biological plausibility; possible mechanisms include in-creased erythropoietin production, increased hemoglobin syn-thesis, and direct stimulation of erythroid precursors.

If HIV-positive patients with anemia are screened and treatedfor low testosterone levels, not only could the underlying causeof anemia be effectively treated, these patients could also ben-efit from the appropriate treatment for hypogonadism (andavoid other sequelae such as decreased lean body mass osteo-porosis and erectile dysfunction). Testosterone therapy has beenshown to increase lean body mass and quality of life for HIV-infected individuals with hypogonadism.24,25 In some smallstudies of patients with anemia and end-stage renal disease, ad-ministration of androgens has been shown to decrease the doseof erythropoietin needed to treat anemia,35 suggesting a possi-

BEHLER ET AL.204

TABLE 3. PREDICTORS OF ANEMIAa,b

Unadjusted odds Adjustedratio odds ratio

n Anemic (%) (95% Cl) (95% Cl)

GenderFemale 38 23.7 0.78 (0.34, 1.78) 0.30 (0.11, 0.85)Male/MTF transgender 162 28.4 — —

Absolute lymphocyte countLymphopenia (�1.0 � 109/liter) 19 57.9 3.05 (1.24, 7.51) 4.00 (1.36, 11.80)Normal (�1.0 � 109/liter) 178 24.7 — —

Mean corpuscular volumeMacrocytosis (�100 fl) 71 14.1 0.34 (0.16, 0.74) 0.23 (0.09, 0.61)Normal (80–100 fl) 117 32.5 — —Microcytosis (�80 fl) 11 63.6 2.91 (0.87, 9.77) 2.02 (0.50, 8.13)

Current antiretroviral therapyYes 139 20.1 0.32 (0.17, 0.61) 0.43 (0.20, 0.95)No 61 44.3 — —

Current androgen useYes 23 13.0 0.36 (0.10, 1.27) 0.20 (0.05, 0.84)No 171 28.7 — —

Erythropoietin (EPO) � testosterone interactionc

High EPO—high normal testosterone 46 41.3 2.50 (1.23, 5.11) 7.73 (2.92, 20.48)Low/Normal EPO—low testosterone 24 37.5 2.39 (0.89, 6.39) 3.27 (1.01, 10.60)High EPO—low testosterone 11 27.3 0.33 (0.04, 2.50) 0.17 (0.012, 2.36)Low/normal EPO—high/normal testosterone 115 20.9 — —

aUnadjusted odds ratios and 95% confidence intervals obtained by logistic regression.bVariables that achieved a p value of �0.1 in tests of interaction were included in a multivariable logistic regression model,

which was used to obtain adjusted odds ratios and 95% confidence intervals. Nadir CD4� T cell counts and peak viral load werenot included in the multivariable models, since they were highly associated with most recent CD4� T cell counts and viral load,respectively.

cLow/normal EPO and high/normal testosterone were defined as the baseline variables.

5489_02_p200-206 3/15/05 2:30 PM Page 204

ble role for the use of supplemental androgens for treatment ofanemia in patients even with normal testosterone levels. Thisbenefit and cost savings over erythropoietin therapy alonewould have to be weighed against potential harms of supple-mental androgens such as hypertension, hyperlipidemia, and po-tentiation of benign prostate hyperplasia and prostate cancer,among others.

Despite the aforementioned limitations to this study, thesedata imply that androgens could play a role in HIV-associatedanemia. Further studies are warranted to investigate this asso-ciation prospectively, as well to examine as the utility of testos-terone levels as part of the work-up for HIV-associated anemiaand the use of androgens therapeutically.

ACKNOWLEDGMENTS

This study was funded by the Center for AIDS Research(Grant P30 AI 27763-11) and by OrthoBiotech. The Center forAIDS Research and Ortho Biotech did not participate in the de-sign and conduct of the study; in the collection, analysis, andinterpretation of the data; or in the preparation, review, or ap-proval of the manuscript.

REFERENCES

1. Coyle T: Hematologic complications of human immunodeficiencyvirus infection and the acquired immunodeficiency syndrome. MedClin North Am 1997;81:449–470.

2. Sullivan PS, Hanson DL, Chu SY, et al.: Epidemiology of anemiain human immunodeficiency virus (HIV)-infected persons: Resultsfrom the multistate adult and adolescent spectrum of HIV DiseaseSurveillance Project. Blood 1998;91:301–308.

3. Moore RD and Forney D: Anemia in HIV-infected patients re-ceiving highly active antiretroviral therapy. J Acquir Immune DeficSyndr 2002;19:54–57.

4. Semba RD, Shah N, and Vlahov D: Improvement of anemia amongHIV-infected injection drug users receiving highly active anti-retroviral therapy. J Acquir Immune Defic Syndr 2001;26:315–319.

5. Breitbart W, McDonald MV, Rosenfeld B, et al.: Fatigue in am-bulatory AIDS patients. J Pain Symptom Manage 1998;15:159–167.

6. Volberding P: The impact of anemia on quality of life in humanimmunodeficiency virus-infected patients. J Infect Dis 2002;185(Suppl. 2):S110–114.

7. Saah AJ, Hoover DR, He Y, et al.: Factors influencing survival af-ter AIDS: Report from the Multicenter AIDS Cohort Study(MACS). J Acquir Immune Defic Syndr 1994;7:287–295.

8. Moore RD, Keruly JC, and Chaisson RE: Anemia and survival inHIV infection. J Acquir Immune Defic Syndr Hum Retrovirol1998;19:29–33.

9. Mocroft A, Kifk O, Barton SE, et al.: Anaemia is an independentpredictive marker for clinical prognosis in HIV-infected patientsfrom across Europe. AIDS 1999;13:943–950.

10. Semba RD, Shah N, Klein RS, et al.: Prevalence and cumulativeincidence of and risk factors for anemia in a Multicenter CohortStudy of human immunodeficiency virus-infected and -uninfectedwomen. Clin Infect Dis 2002;34:260–266.

11. Volberding P, for The Anemia in HIV Working Group: Consen-sus statement: Anemia in HIV infection—current trends, treat-ment options, and practice strategies. Clin Ther 2000;22:1004–1020.

12. Abrams DI, Steinhart C, Frascino R: Epoetin Alfa therapy foranaemia in HIV-infected patients: Impact on quality of life. Int JSTD AIDS 2000;11:659–665.

13. Revicki DA, Brown RE, Henry DH, et al.: Recombinant humanerythropoietin and health-related quality of life of AIDS patientswith anemia. J Acquir Immune Defic Syndr 1994;7:474–484.

14. Henry DH, Gildon NB, Benson CA, et al.: Recombinant humanerythropoietin in the treatment of anemia associated with humanimmunodeficiency virus (HIV) infection and zidovudine therapy.Ann Intern Med 1992;117:739–748.

15. Means RT: Advances in the anemia of chronic disease. Int J Hema-tol 1999;70:7–12.

16. Means RT and Krantz SB: Progress in understanding the patho-genesis of the anemia of chronic disease. Blood 1992;80:1639–1647.

17. Hambleton J: Hematologic complications of HIV infection. On-cology 1996;10:671–680.

18. Claster S: Biology of anemia, differential diagnosis and treatmentoptions in human immunodeficiency virus infection. J Infect Dis2002;185(Suppl. 2):S105–109.

19. Bain BJ: Pathogenesis and pathophysiology of anemia in HIV in-fection. Curr Opin Hematol 1999;6:89–93.

20. Dobs AS, Dempsey MA, Ladenson PW, et al.: Endocrine disor-ders in men infected with human immunodeficiency virus. Am JMed 1988;84:611–616.

21. Coodley GO, Loveless MO, Nelson HD, et al.: Endocrine functionin the HIV wasting syndrome. J Acquir Immune Defic Syndr1994;7:46–51.

22. Rietschel P, Corcoran C, Stanley T, et al.: Prevalence of hypogo-nadism among men with weight loss related to human immunode-ficiency virus infection who were receiving highly active anti-retroviral therapy. Clin Infect Dis 2000;31:1240–1244.

23. Dobs A and Brown T: Metabolic abnormalities in HIV disease andinjection drug use. J Acquir Immune Defic Syndr 2002;31(Suppl.2):S70–77.

24. Grinspoon S, Corcoran C, Anderson E, et al.: Sustained anaboliceffects of long-term androgen administration in men with AIDSwasting. Clin Infect Dis 1999;28:634–636.

25. Bhasin S, Storer TW, Asbel-SethiN, et al.: Effects of testosteronereplacement with a nongenital, transdermal system, Androderm, inhuman immunodeficiency virus-infected men with low testosteronelevels. J Clin Endocrinol Metab 1998;83:3155–3162.

26. Navarro JF and Mora C: Androgen therapy for anemia in elderlyuremic patients. Int Urol Nephrol 2001;32:549–557.

27. Udupa KB, Crabtree HM, and Lipschitz DA: In vitro culture of proerythroblasts: Characterization of proliferative response to erythropoietin and steroids. Br J Haematol 1986;62(4):705–714.

28. Modder B, Foley JE, and Fisher JW: The in vitro and in vivo ef-fects of testosterone and steroid metabolites on erythroid colonyforming cells (CFU-E). J Pharmacol Exp Ther 1978;207:1004–1012.

29. Ohno Y and Fisher JW: Effects of androgens on burst forming units(BFU-E) in normal rabbit bone marrows. Life Sci 1978;22:2031–2036.

30. Urabe A, Sassa S, and Kappas A: The influence of steroid hor-mone metabolites on the in vitro development of erythroid coloniesderived from human bone marrow. J Exp Med 1979;149:1314–1325.

31. Claustres M and Sultan C: Androgen and erythropoiesis: Evidencefor an androgen receptor in erythroblasts from human bone mar-row cultures. Horm Res 1988;29:17–22.

32. Gardner FH and Besa EC: Physiologic mechanisms and the he-matopoietic effects of the androstanes and their derivatives. CurrTop Hematol 1983;4:123–195.


5489_02_p200-206 3/15/05 2:30 PM Page 205

33. Levine AM, Berhane K, Masri-Lavine L, et al.: Prevalence andcorrelates of anemia in a large cohort of HIV-infected women:Women’s Interagency HIV Study. J Acquir Immune Defic Syndr2000;26:28–35.

34. Lawrence IG, Price DE, Howlett TA, et al.: Erythropoietin and sex-ual dysfunction. Nephrol Dial Transplant 1997;12(4):741–747.

35. Ballal SH, Domoto DT, Polack DC, et al.: Androgens potentiatethe effects of erythropoietin in the treatment of anemia of end-stagerenal disease. Am J Kidney Dis 1991;17:29–33.

Address reprint requests to:Caroline Behler

Department of MedicineUniversity of California-San Francisco

Box 0654San Francisco, California 94143

E-mail: [email protected]

BEHLER ET AL.206

5489_02_p200-206 3/15/05 2:30 PM Page 206

Documents

Anemia and HIV in the Antiretroviral Era: Potential Significance of Testosterone