INTRODUCTION
Histoplasmosis is a common opportunistic infection in patients infected with human immunodeficiency virus (HIV) who reside in disease-endemic areas, such as Brazil. Disseminated histoplasmosis has been an acquired immunodeficiency syndrome (AIDS)–defining illness since 1987.1 It is reported to occur in approximately 5% of patients with AIDS who live in disease-endemic areas.2 A study performed in southern Brazil using the histoplasmin skin test showed that 89% of reactors were young men between 17 and 19 years of age.3 Comparison of AIDS-disseminated histoplasmosis patients from the United States and Brazil showed more frequent skin involvement (1–7% versus 66%) and death (5–13% versus 19%) in the Brazilian patients.4
The infection is characterized by a wide spectrum of manifestations ranging from asymptomatic illness to severe disseminated histoplasmosis.5 Most patients have respiratory symptoms. Acute histoplasmosis is usually a self-limited disease, with fever, chills, nonproductive cough, headaches, and generalized malaise.5 In AIDS patients, the disease is associated with a nonspecific clinical presentation, usually unexplained fever and weight loss. Approximately 10–20% of these patients have septic shock at presentation, with fever, hypotension, renal and hepatic failure, respiratory distress syndrome, and coagulopathy.6 Laboratory findings such as a serum creatinine level > 2.1 mg/dL and an albumin level < 3.5 g/dL have been associated with an increased risk of severe manifestations such as septic shock, respiratory failure, and death.2 A prospective study among HIV-positive patients with disseminated histoplasmosis found that dyspnea, a platelet count < 100,000/mm3, and a lactate dehydrogenase (LDH) level > 2 times the normal upper limit were independent risk factors for death.7 Better knowledge of the factors associated with high mortality in patients with disseminated histoplasmosis would result in better treatment. The objective of this study was to investigate the risk factors potentially associated with death in patients with AIDS and disseminated histoplasmosis.
MATERIALS AND METHODS
We carried out a retrospective study of 164 HIV-positive patients with disseminated histoplasmosis to determine the risk factors associated with death. All reported cases met the 1987 AIDS case definition of the Centers for Disease Control and Prevention (Atlanta, GA). Physicians caring for these patients were contacted, and consent was obtained to review medical records. The patients were admitted between January 1995 and January 2004 to a public tertiary hospital in Fortaleza in northeastern Brazil. The study protocol was reviewed and approved by the Ethical Committee of the Universidade Federal do Ceará.
A diagnosis of histoplasmosis was confirmed by at least one of the following methods: identification of Histoplasma capsulatum in peripheral blood, bone marrow aspirate, blood and bone marrow culture; lysis-centrifugation system; and histopathologic examination of different tissues. An HIV infection was confirmed by indirect immunofluorescence and/or enzyme-linked immunosorbent assay. Patients with isolated pulmonary forms of histoplasmosis were not included in the study.
Case-patients were those who died before hospital discharge (non-survivors). Data on histoplasmosis included date of diagnosis, clinical manifestations, and laboratory test results at the time of admission and before antifungal treatment (for this analysis, data were obtained from seven days before admission through seven days after admission). Demographic factors were age, sex, sexual behavior, intravenous drug abuse, and previous blood transfusion. The following complications were also observed: respiratory insufficiency, hepatic dysfunction, sepsis and acute renal failure, and neurologic manifestations (including clinical signs and symptoms caused by nervous system injury or dysfunction). Respiratory insufficiency was defined as the need for mechanical ventilation. Hepatic dysfunction was defined as a serum total bilirubin level > 2 mg/dL or another clinical condition that indicated hepatic failure. Sepsis was defined according to the American College of Chest Physicians/Society of Critical Care Medicine.8 Acute renal failure was defined as a serum creatinine level > 1.5 mg/dL or an increase in the serum creatinine concentration by more than 50% or > 0.5 mg/dL above baseline.9 Patients with histoplasmosis were treated with amphotericin B as induction therapy at a dose of 0.7–1 mg/kg/day (up to a cumulative dose of 1 gram), followed by a maintenance therapy with itraconazole, 600–800 mg/day.
Statistical analysis.
Results were expressed mean ± SD or median (range) for quantitative variables. Univariate and multivariate analysis of clinical and laboratory data was conducted with SPSS version 10.0 (SPSS Inc., Chicago, IL) and Epi-Info version 6.04b (Centers for Disease Control and Prevention) software. Comparison of parameters was done with Student’s t-test and Fischer’s exact test. Analysis of associations between death and categorized risk factors was done with Fischer’s exact test and Pearson’s chi-square test. A logistic regression model was used for quantitative variables. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. A multivariate logistic regression was performed to analyze the possible risk factors for death. The factors included in the multivariate model were those that showed a significance level < 20% in the univariate analysis (by Mann-Whitney test and chi-square test). P values < 0.05 were statistically significant.
RESULTS
A total of 164 patients had a confirmed diagnosis of infection with HIV and disseminated histoplasmosis. The sex ratio (men:women) was 4:1, and the mean ± SD age of the patients was 34.2 ± 8.19 years. Although differences in sex were not observed, sex was a decreased risk factor for death in heterosexual patients (P = 0.025). The demographic characteristics of the patients are shown in Table 1. The diagnosis of histoplasmosis was made through identification of H. capsulatum in bone marrow aspirate (81.1% of the cases), in peripheral blood smears by a lysis-centrifugation system (79.5%), culture of bone marrow aspirates (47.8%) and histopathologic examination of different tissues (57.9%).
All patients had disseminated manifestations; some also had pulmonary disease manifestations. The mean ± SD duration of illness before hospitalization was 1.94 ± 2.58 months in survivors and 2.21 ± 2.91 months in non-survivors, but this difference was not statistically significant. Diarrhea was observed at admission in 100 patients, of whom 43 died. Unusual manifestations included cerebral histoplasmosis, confirmed by autopsy, and pericarditis in one patient each. Of the 164 patients, 52 had respiratory insufficiency, of whom 34 (65.4%) died; 16 patients had sepsis, of whom 14 (87.5%) died; and 97 patients had acute renal failure, of whom 39 (42.2%) died during hospitalization.
The laboratory abnormalities in survivors and non-survivors before antifungal treatment are shown in Table 2. Analysis of CD4 cell counts before admission was not possible because the diagnosis of histoplasmosis occurred within one month of the diagnosis of HIV infection in many patients and because data were missing for patients, most of whom died.
Results of univariate logistic regression are shown in Table 3. Diarrhea, neurologic disorders, a hemoglobin level < 8.0 g/dL, a serum urea concentration > 2 times the normal upper limit, a serum creatinine concentration > 1.5 mg/dL, and an aspartate aminotransferase (AST) level > 2.5 times the normal upper limit were associated with increased risk of death. During hospitalization, respiratory insufficiency, sepsis, and acute renal failure were associated with an increased risk of death. Heterosexuality was a protective factor against death.
All variables found to be associated with increased risk of death by univariate analysis included in the initial model were analyzed by multivariate logistic regression. The final multivariate logistic regression model showed that a hemoglobin level < 8.0 g/dL (OR = 3.8, 95% CI = 1.4–10.5, P = 0.008), an AST > 2.5 times the normal upper limit (OR = 1.0, 95% CI = 1.001–1.005, P = 0.007), acute renal failure (OR = 2.96, 95% CI = 1.24–7.06, P = 0.015), and respiratory insufficiency (OR = 12.2, 95% CI = 4.4–33.3, P = 0.01) were significantly associated with an increased risk of death (Table 4).
DISCUSSION
Histoplasmosis is the first opportunistic infection observed in 22–85% of HIV-infected patients in disease-endemic areas.10–12 Disseminated histoplasmosis usually occurs when CD4 cell counts are < 100/mm3.13
Some factors potentially associated with death that were identified by univariate and multivariate analysis in our study differed from those identified by Wheat and others2 and Couppié and others,7 which are the only other published studies on prognostic factors for AIDS-associated disseminated histoplasmosis. The prospective study by Couppié and others,7 which included 82 patients to determine the prognostic factors associated with death within one month after starting antifungal treatment, found associations by multivariate analysis between AIDS-associated disseminated histoplasmosis and dyspnea, thrombocytopenia, and an increased LDH levels. Wheat and others2 in a retrospective study of 155 patients treated with amphotericin B found that high levels of serum creatinine, low levels of serum albumin, and previous use of zidovudine were associated with septic shock, respiratory insufficiency, and death. In our study, the presence of acute renal failure was associated with an increased risk of death.
Clinical and laboratory factors associated with an increase risk of death included diarrhea, neurologic disorders, a hemoglobin level < 8.0 g/dL, a serum urea concentration > 2 times the normal upper limit, a serum creatinine concentration > 1.5 mg/dL, an AST level > 2.5 times the normal upper limit, respiratory insufficiency, sepsis, and acute renal failure. Heterosexual behavior (man or women) was associated with a decreased risk of death by univariate analysis, but we could not provide any explanation for this finding.
Multivariate analysis showed associations between death and a hemoglobin level < 8.0 g/dL, an increase in AST levels, acute renal failure, and respiratory insufficiency. The association between low hemoglobin levels and death may be due to hemorrhagic disorders or bone marrow suppression. Several studies have shown the occurrence of hepatomegaly and/or high levels of AST and alanine aminotransferase in patients with histoplasmosis.12,14–17 Respiratory insufficiency is known to be a classic risk factor for death in patients with disseminated histoplasmosis, and this was confirmed in our study. Univariate analysis showed that serum creatinine levels > 1.5 mg/dL on hospital admission were associated with death. Some patients had signs of dehydration at admission, which could have contributed to the genesis of acute renal failure. High fever, extreme weight loss, thromboctyopenia, and coagulopathy were not associated with death, as previously reported.2 High levels of LDH have been associated with increased risk of death.7 In our study, LDH levels were higher in non-survivors, but this was not significantly associated with death in the logistic regression model. The presence of sepsis could be an important risk factor of death. It was associated with death in the univariate analysis, but this factor could not be included in the multivariate analysis because only a small number of patients had this complication.
The usefulness of this study in identifying additional predictors of death was limited by the small sample size, the retrospective study design, and missing data. However, the main contribution of this study was identification of laboratory abnormalities associated with death, which resulted in early recognition of the cases that must be treated more aggressively.
In conclusion, the presence of low hemoglobin levels, increased serum AST levels, acute renal failure, and respiratory insufficiency were independent risk factors for death in AIDS patients with disseminated histoplasmosis. This was the first study to identify risk factors for death in Brazilian AIDS patients with disseminated histoplasmosis. Further prospective studies are necessary to establish additional risk factors.
Demographic characteristics of 164 survivors and non-survivors with acquired immunodeficiency syndrome and disseminated histoplasmosis*
| Patients | |||
|---|---|---|---|
| Survivors | Nonsurvivors | P | |
| * NS = not significant. | |||
| † Median (range). | |||
| Number | 111 | 53 | |
| Age, years | 34 (20–53) | 34 (19–52) | NS |
| Male, % | 77 | 85 | NS |
| Heterosexual, % | 56 | 36 | 0.025 |
| Intravenous drug abuse, % | 7.2 | 7.5 | NS |
| Previous blood transfusion, % | 4.5 | 1.9 | NS |
| Duration of illness, months† | 1.0 (0.0–12.0) | 1.0 (0.0–15.0) | NS |
Laboratory values of survivors and non-survivors with acquired immunodeficiency syndrome and disseminated histoplasmosis at admission*
| Survivors (n = 111) | Non-survivors (n = 53) | P | |
|---|---|---|---|
| * NS = not significant; LDH = lactate dehydrogenase; AST = aspartate aminotransferase. Missing data (survivors, non-survivors): hemoglobin (4, 0); hematocrit (4, 0); leukocyte count (4, 0); platelet count (8, 3); serum creatinine (5, 0); serum urea (6, 0); LDH (22, 10); AST (24, 13); albumin (65, 30). | |||
| Hemoglobin, g/dL | 8.92 ± 2.15 8.6 (3.3–13.5) | 7.49 ± 2.18 7.3 (3.0–11.9) | < 0.001 |
| Hematocrit, % | 26.57 ± 6.47 25.5 (10.1–40.4) | 22.29 ± 6.58 21.4 (8.9–37.4) | < 0.001 |
| Leukocyte count, cells/mm3 | 4,698 ± 4,796 2,800 (3.0–27,400) | 5,827 ± 4,105 4,500 (400–17,800) | NS |
| Platelet count ×103, cells/mm3 | 117 ± 93 960 (40–482) | 108 ± 101 82 (40–572) | NS |
| Serum creatinine, mg/dL | 1.42 ± 1.38 1.0 (0.2–6.9) | 1.92 ± 1.58 1.2 (0.5–7.0) | 0.048 |
| Serum urea, mg/dL | 51.24 ± 49.75 35 (11–278) | 79.98 ± 65.01 54 (10–272) | 0.005 |
| LDH, IU/L | 3,665 ± 3,495 2,205 (170–12,212) | 6,110 ± 6,610 3,120 (228–35,300) | 0.012 |
| AST, IU/L | 167 ± 190 140 (15–1,520) | 363 ± 368 264 (26–1,860) | 0.002 |
| Albumin, g/dL | 2.94 ± 0.87 2.9 (1.2–5.5) | 2.47 ± 0.55 2.5 (1.0–3.4) | 0.029 |
Risk factors for death in patients with acquired immunodeficiency syndrome and disseminated histoplasmosis by univariate logistic regression*
| Survivors (n = 111) | Nonsurvivors (n = 53) | OR (95% CI) | P | |
|---|---|---|---|---|
| * OR = odds ratio; CI = confidence interval; AST = aspartate aminotransferase. | ||||
| At Admission | ||||
| Diarrhea,% | 51 | 81 | 3.9 (1.8–8.7) | < 0.001 |
| Neurologic manifestation, % | 7 | 32 | 5.8 (2.3–14.7) | < 0.001 |
| Hemoglobin < 8.0 g/dL,% | 32 | 58 | 2.7 (1.4–5.4) | 0.004 |
| Leukocyte count < 4,000 cells/mm3, % | 61 | 39 | 0.4 (0.2–0.7) | 0.01 |
| Serum urea > 2× upper normal limit, % | 10 | 39 | 5.0 (2.2–11.4) | < 0.001 |
| Serum creatinine > 1.5 mg/dL, % | 19 | 43 | 2.9 (1.4–6.0) | 0.005 |
| AST > 2.5× upper normal limit, % | 44 | 60 | 3.1 (1.2–7.5) | 0.01 |
| During hospitalization | ||||
| Respiratory insufficiency, % | 16 | 64 | 9.7 (4.5–21.1) | < 0.001 |
| Sepsis, % | 18 | 26 | 20 (4.3–93.3) | < 0.001 |
| Acute renal failure, % | 52 | 73 | 2.5 (1.2–5.2) | 0.011 |
Independent variables associated with death in patients with AIDS and disseminated histoplasmosis by multivariate logistic regression
| OR | 95% CI | P | |
|---|---|---|---|
| * AIDS = acquired immunodeficiency syndrome; OR = odds ratio; CI = confidence interval; AST = aspartate aminotransferase. | |||
| Hemoglobin < 8.0 g/dL | 3.8 | 1.4–10.5 | 0.008 |
| AST > 2.5× upper normal limit | 1.0 | 1.001–1.005 | 0.007 |
| Acute renal failure | 2.96 | 1.24–7.06 | 0.015 |
| Respiratory insufficiency | 12.2 | 4.4–33.3 | 0.01 |
Address correspondence to Geraldo Bezerra da Silva Ju′ nior, Rua 25 de Marco, 997, Fortaleza, CE, Brazil CEP 60060-120. E-mail: geraldobezerrajr@yahoo.com.br or efdaher@for.terra.com.br
Authors’ addresses: Elizabeth de Francesco Daher, Fernando Antonio de Sousa Barros, Geraldo Bezerra da Silva Júnior, Marúsia Thomaz Ferreira, Julieta Cunha Martins, Soraya Alves Jacinto Oliveira, and Oswaldo Augusto Gutiérrez-Adrianzén, Department of Internal Medicine, Faculdade de Medicina, Hospital Universitário Walter Cantídio, Universidade Federal do Ceará, Fortaleza, Ceará, Brazil, E-mails: efdaher@for.terra.com.br and geraldobezerrajr@yahoo.com.br. Christianne Fernandes Valente Takeda, Hospital São José de Doenças Infecciosas. Fortaleza, Ceará, Brazil. Rosa Maria Salani Mota, Department of Statistics, Universidade Federal do Ceará, Fortaleza, Ceará, Brazil.
Acknowledgments: We are very grateful to the physicians, residents, medical students, and nurses from the Hospital São José de Doenças Infecciosas for assistance provided to the patients and technical support.
Financial support: This research was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico – CNPq, Brazil. Geraldo Bezerra da Silva Júnior and Marúsia Thomaz Ferreira were supported by a grant from CNPq.
REFERENCES
- 1↑
Centers for Disease Control, 1987. Revision of the CDC surveillance case definition for the acquired immunodeficiency syndrome. MMWR Morb Mortal Wkly Rep 36 :1S–15S.
- 2↑
Wheat LJ, Chetchotisakd P, Williams B, Connolly P, Shutt K, Hajjeh R, 2000. Factors associated with severe manifestations of histoplasmosis in AIDS. Clin Infect Dis 30 :877–881.
- 3↑
Zembrzuski MM, Bassanesi MC, Wagner LC, Severo LC, 1996. Inquérito intradérmico com histoplasmina e paracoccidiodina em duas regiões do Rio Grande do Sul. Rev Soc Bras Med Trop 29 :1–3.
- 4↑
Karimi K, Wheat LJ, Connolly P, Cloud G, Hajjeh R, Wheat E, Alves K, Lacaz S, Keath E, 2002. Differences in histoplasmosis in patients with acquired immunodeficiency syndrome in United States and Brazil. J Infect Dis 186 :1655–1660.
- 5↑
Cano MVC, Hejjeh RA, 2001. The epidemiology of histoplasmosis: a review. Semin Respir Infect 16 :109–118.
- 6↑
Hajjeh RA, 1995. Disseminated histoplasmosis in persons infected with human immunodeficiency virus. Clin Infect Dis 21 (Suppl 1):S108–S110.
- 7↑
Couppie P, Sobesky M, Aznar C, Bichat S, Clyti E, Bissuel F, El Guedj M, Alvarez F, Demar M, Luovel D, Pradinaud R, Carme B, 2004. Histoplasmosis and acquired immunodeficiency syndrome: a study of prognostic factors. Clin Infect Dis 38 :134–138.
- 8↑
American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference, 1992. Definitions for sepsis and organ failure and guidelines for the need of innovative therapies in sepsis. Crit Care Med 20 :864–874.
- 9↑
Brady HR, Clarkson MR, Lieberthal W, 2004. Acute renal failure. Brenner BM, ed. Brenner and Rector’s The Kidney. Seventh edition. London: Elsevier, 1215–1292.
- 10↑
Borges AS, Ferreira MS, Silvestre MTA, Nishioka SA, Rocha A, 1997. Histoplasmose em pacientes imunodeprimidos: estudo de 18 casos observados em Uberlândia, MG. Rev Soc Bras Med Trop 30 :119–124.
- 11
McKinsey DS, Spiegel RA, Hutwagner L, Stanford J, Driks MR, Brewer J, Gupta MR, Smith DL, O’Connor MC, Dall L, 1997. Prospective study of histoplasmosis in patients infected with human immunodeficiency virus: incidence, risk factors, and pathophysiology. Clin Infect Dis 24 :1195–1203.
- 12↑
Pietrobon D, 2004. Negro-Marquínez, Kilstein J, Galindez J, Greca A, Battagliotti. Histoplasmosis diseminada y sida en un hospital argentino: manifestaciones clínicas, diagnóstico y tratamiento. Enferm Infecc Microbiol Clin 22 :156–159.
- 13↑
Sarosi GA, Johnson PC, 1992. Disseminated histoplasmosis in patients infected with human immunodeficiency virus. Clin Infect Dis 14 (Suppl 1):S60–S67.
- 14↑
Anonymous, 1997. Fever and renal failure in a 31-year-old male with AIDS. Am J Med 102 :310–315.
- 15
Burke DC, Emancipator SN, Smith MC, Salata RA, 1997. Histoplasmosis and kidney disease in patients with AIDS. Clin Infect Dis 25 :281–284.
- 16
Goswami RP, Pramanik N, Banerjee D, Raza MM, Guka SK, Maiti PK, 1999. Histoplasmosis in eastern India: the tip of the iceberg? Trans R Soc Trop Med Hyg 93 :540–542.
- 17↑
Nand N, Aggarwal HK, Singh M, Arora BR, Sen J, 2001. Renal failure in a case of histoplasmosis. J Assoc Physicians India 49 :833–834.