|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
Histoplasmosis encompasses a spectrum of clinical forms with DH being the most severe.1,4,8 Despite the fact that primary lung infection is frequently followed by spleen and liver invasion, DH refers only to a process of intense fungus multiplication in lungs and in extra-pulmonary organs and body sites.14,911 Disseminated histoplasmosis may occur either after recent exposure or upon endogenous reactivation of latent foci.1,11,12 In fact, HIV-infected individuals who develop DH in areas not endemic for diseases were known to have lived previously in recognized disease-endemic regions.1,810,12 Before the AIDS epidemic, risk factors for DH were immunosuppressive therapies, impaired cellular immunity, hematologic and other type of malignancies, organ transplant and dialysis, as well as extreme ages (children and old persons).1,12,13 In certain individuals, heavy exposure to infected aerosols may also give rise to DH.1,3,4,8,9 The above circumstances changed with the advent of the HIV epidemic, which transformed histoplasmosis into a common and severe fungal disease.811 Presently, patients with HIV have their own risk factors, such as low CD4 lymphocyte counts (< 200/µL), an epidemiologic history of exposure to the fungus, and presence of antibodies to H. capsulatum at the time of diagnosis.1316
The present study analyzes two cohorts of DH patients, those with AIDS (cohort 1) and those not infected with HIV (cohort 2). The main objectives were to determine clinical differences, effectiveness of diagnostic methods, results of antifungal therapy in connection with viral co-infection, and influence of highly active antiretroviral therapy (HAART).
| PATIENTS AND METHODS |
|
|
|---|
The immune status of those patients in cohort 1 receiving HAART was determined by CD4 lymphocyte counts, both during and in some cases, at the end of antifungal treatment. All patients with DH received antifungal therapy appropriate to the severity of their clinical manifestations and more recently, observing the recommendations of Wheat and others.11,18 Treatment consisted of amphotericin B, itraconazole, and in some cases of fluconazole, ketoconazole, or posaconazole given in the context of clinical trials. The outcome of therapy was evaluated according to a scoring system previously reported,17 in which each of the clinical abnormalities present before therapy and any mycologic findings received an arbitrary score of 2. The sum of all these scores constituted the denominator of a fraction. During treatment (35 months) and at the end of therapy, the above abnormalities and mycologic observations were re-evaluated. If they had resolved, each was given the same score of 2 recorded at initiation of treatment; if abnormalities had improved but not resolved, a score of 1 was assigned to each one; if they had not resolved, each received a score of 0. If a patient showed clinical and/or mycologic deterioration, a negative score equivalent to deducting 2 points for each parameter evaluated before therapy was used. These scores were aggregated to form the numerator of the fraction. The resultant equation was calculated and the results expressed as follows: 1) a negative score indicated deterioration of the clinical condition; 2) zero indicated no change in the patients condition; 3) a positive score indicated minor or major improvement; and 4) a score of 1 indicated complete resolution of all abnormalities observed before therapy.
The data were processed using Excel® (Microsoft, Redmond, WA). Statistical analyses were done using Fishers exact test. The Students t-test was used for group comparisons, and the Pearson correlation test was used to evaluate response to antifungal treatment.
Appropriate informed consent was obtained from the patients and the study was reviewed and approved by the Ethics Committee of the Corporación para Investigaciones Biológicas.
| RESULTS |
|
|
|---|
The most frequent clinical manifestations are shown in Table 1
. In both cohorts, constitutional symptoms (asthenia, weight loss, anorexia, fever) predominated, being observed in more than 85% of the patients. Additional symptoms such as respiratory problems and gastrointestinal alterations were also frequently recorded (59% and 46% in cohorts 1 and 2, respectively), with no significant differences. Physical examination showed enlarged lymph nodes, hepatosplenomegaly, skin and mucosal lesions, as well as lung auscultation abnormalities (Table 1
). These signs were not significantly different among the two cohorts, with the exception of skin lesions, which were observed in 16 (53.3%) in cohort 1 and 2 (9%) in cohort 2 (P = 0.001). In 13 of cohort 1 patients (81.2%) skin lesions were widespread (face, thorax, abdomen, and upper limbs) and showed various characteristics (maculopapular, ulcerated, and/or crusted). In contrast, in cohort 2, skin lesions were nodular and less widespread.
|
Active co-morbidities observed previous to or simultaneous with the diagnosis of histoplasmosis are shown in Table 2
. They were present simultaneously with DH in 21 (70%) patients in cohort 1 with a predominance of mycotic diseases (13 of 30 patients), mostly oral candidiasis. In cohort 2, simultaneous co-morbidities were observed in 7 (31.8%) patients, with 6 showing associated conditions indicating immune response alterations (diabetes, malnutrition, and cirrhosis). Active co-morbidities observed before a diagnosis of histoplasmosis were present in 23.3% of the patients in cohort 1 and 18.2% of the patients in cohort 2.
|
As shown in Table 3
, DH was often diagnosed in both cohorts by direct microscopic observation (Wright stain) of H. capsulatum yeast cells in clinical specimens, and the fungus was also isolated in culture from these specimens. Significant differences (P < 0.05) were observed only for the latter procedure. In eight patients (four in each cohort), a diagnosis was established by the observation of yeast cells in hematoxylin and eosin- and/or Gomori-stained tissue biopsy specimens. Serologic test results with histoplasmin were significantly more reactive (P < 0.05) in cohort 2 than in cohort 1.
|
|
|
The scores of the point system evaluation were condensed into two groups (Figure 2
): patients responding adequately (complete resolution or major improvement), and patients not responding (minor improvement, worsening, or death). This system was applied during and at the end of therapy. During treatment (three months in cohort 1 and five months in cohort 2), 53.4% of cohort 1 patients showed complete resolution or major improvement. In contrast, 86.4% of cohort 2 patients showed an adequate response (P = 0.012) (Figure 2a
). At the end of therapy, the scores indicated complete resolution or major improvement in 66.7% of the patients in cohort 1 versus 95.5% of the patients in cohort 2 (P < 0.05) (Figure 2b
).
|
|
| DISCUSSION |
|
|
|---|
We conducted a retrospective study of 52 DH patients and divided them in two cohorts, 30 with AIDS and 22 not co-infected with HIV. Patients in cohort 1 were diagnosed from 1988 to 2004, whereas patients in cohort 2 were diagnosed for a longer period (19792001). The aim of this comparative study was to detect significant differences between the two cohorts to facilitate early recognition and treatment of DH in HIV-infected patients.
As previously reported in other series,6,2123 males predominated (80.7%) in cohort 1 with a male to female ratio of 29:1; in cohort 2, the sex distribution was not so markedly different (P < 0.05). Age distribution was similar in the two cohorts (young adults) as reported by others,15,21,22 but the age range of cohort 2 patients extended from childhood to old age. This indicated that in individuals not infected with HIV, DH may occur at any age,2224 whereas in those with AIDS, a rather restricted age range is to be expected.15,2123
We found few significant differences in the clinical manifestations of histoplasmosis among the two cohorts. Fever appeared to be a common symptom (more than 80% of the cases) in both DH patients with AIDS and in those not co-infected with HIV.9,10,13,21,23 Skin lesions, mostly widespread, predominated (53.3%) in cohort 1 patients and were uncommon in cohort 2 cases (P = 0.001). Such lesions were also more frequent (66%) in Brazilian DH patients co-infected with HIV compared with those in North America (7%).21 In another series of Brazilian patients, skin lesions had an intermediate (47.6%) frequency.24 No skin lesions were observed in a large multi-center study in the United States of HIV-infected histoplasmosis patients.13 We observed that cohort 2 patients differed significantly from cohort 1 only in their lower proportion of skin lesions (9%). It is interesting to note that in HIV-infected Latin American patients, the skin constitutes a more important target organ for H. capsulatum than in North American patients; however, no explanations for this can be given unless a late diagnosis or inadequate diagnostic facilities were considered.
Gastrointestinal involvement was observed in similar proportions in both cohorts, but was more frequent in Brazilian HIV-infected patients with histoplasmosis than in cases in the United States,21 a finding that has been confirmed in different series.13,22 The frequency of abdominal problems was similar in this study (46.6%) and in those of Hajjeh and others (24%)13 and Karimi and others (33%).21
Other differential findings were observed for lung lesions, with significantly more interstitial infiltrates observed in cohort 1 (63.3%). Interstitial infiltrates were also observed in half of the patients reported by Hajjeh and others,13 in 63% of the Brazilian patients studied by Karimi and others,21 and in 53.3% of French Guyana patients.16 This involvement suggests the existence of a primary lung infection that was not noticed.1,13
Anemia and leukopenia were the most common hematologic abnormalities and differed significantly between cohorts 1 and 2 (P < 0.001 for either parameter). Sedimentation rates were elevated in all cohort 1 patients but only in half of cohort 2 patients (P = 0.001). These results should considered when HIV infection in a patient with histoplasmosis is suspected.
Although DH can be diagnosed by observation of H. capsulatum yeast cells in clinical specimens, isolation of the fungus in culture varied in the two cohorts, with nearly all (96.3%) patients in cohort 1 showing positive results, but only 73.6% in cohort 2 (P < 0.05). Conversely, detection of antibodies in sera by either immunodiffusion or complement fixation was significantly lower in cohort 1 patients than in cohort 2 patients (P < 0.05). These findings have been previously demonstrated by Wheat9,12 and by Karimi and others in the Brazilian cohort.21
The number of CD4 lymphocytes in cohort 1 patients was low (3053 cells/µL) in those receiving HAART and in those not receiving HAART, suggesting that histoplasmosis occurs when the immune response is markedly impaired, as demonstrated by McKinsey and others, who found that the annual incidence of histoplasmosis in AIDS patients increases when CD4 lymphocyte counts are less than 50 cells/µL.14 This finding has also been reported in patients in the United States.13,23
In our patients HAART significantly increased CD4 lymphocyte counts (to more than 150/µL), but most importantly, improved the response to antifungal therapy, as shown by the fact that all patients thus treated achieved complete resolution or major improvement of their pre-therapy abnormalities in a manner similar to those patients not co-infected with HIV (cohort 2). In contrast, patients not receiving HAART did not respond as well (P = 0.003) to antifungal treatment. Despite the fact that antiretrovirals are known to improve defense mechanisms, thus allowing AIDS patients to overcome opportunistic infections,25 their positive influence in histoplasmosis has been recognized only recently, as shown by one case report from the Phillipines,26 and by a series of cases from Argentina,27 which indicate that HAART immune restoration effectively cooperates with antifungal therapy in controlling the mycosis.
Due to its retrospective character, this study had several limitations, among them lack of information on CD4 lymphocyte counts in the HIV-infected population both before and after therapy. Additionally, variations in treatment modalities during the course of the study interfered with a more precise evaluation of the clinical responses, especially in the DH patients co-infected with HIV, whose treatment is currently defined by expert guidelines.18 However, the extended period of the study (19792001) did not alter the way in which diagnosis was established in our laboratory since newer modalities (e.g., antigenemia) have not been used regularly in our institution and have been previously used only for standardization purposes, not for diagnosis.28
This study shows that the prognosis of DH is greatly influenced by co-infection with HIV since response rates to antifungal treatment were lower in co-infected patients. Nonetheless, HAART appears to improve this unfavorable condition by restoring the immune response so that even in the presence of AIDS, antifungal therapy can be successful. Therefore, it is mandatory to suspect mycosis as early as possible and initiate antifungal and antiretroviral therapies promptly to improve the prognosis of DH patients co-infected with HIV.
Received February 24, 2005. Accepted for publication May 17, 2005.
Acknowledgments: We thank the patients for participating in the study and the physicians for referring their cases to our institution. We also thank Dr. Elizabeth Castañeda (Instituto Nacional de Salud, Bogotá, Colombia) for cooperation.
Financial support: This study was supported by the Corporación para Investigaciones Biológicas (Medellín, Colombia).
Disclosure: None of the authors have any conflicts of interest.
* Address correspondence to Angela Restrepo, Corporación para Investigaciones Biológicas, Carrera 72A # 78B-141, Medellín, Colombia. E-mail: angelares{at}geo.net.co ![]()
Authors addresses: Angela M. Tobón, Catalina de Bedout, Alejandra Zuluaga, and Angela Restrepo, Corporación para Investigaciones Biológicas, Carrera 72A # 78B-141, Medellín, Colombia and Hospital La Maria, Medellín, Colombia, Fax: 57-4-441-0855, E-mails: atobon{at}cib.org.co, cbedout{at}cib.org.co, azuluaga{at}cib.org.co, and angelares{at}geo.net.co. Carlos A. Agudelo and Juan E. Ochoa, Universidad Pontificia Bolivariana, Calle 78B # 72A-109, Medellín, Colombia, Fax: 57-4-257-2428, E-mails: carlosaguedelo{at}yahoo.com and clio{at}geo.net.co. David S. Rosero, Policlínica Villarrobledo, c/o Senda Molinera 02600, Villarobledo, Albacete, Spain, Fax: 34-96-714-5959, E-mail: roserocuesta{at}hotmail.com. Myrtha Arango, Facultad de Medicina, Universidad de Antioquia, Medellín, Colombia and Corporación para Investigaciones Biológicas, Carrera 72A # 78B-141, Medellín, Colombia, Fax: 57-4-441-0855, E-mail: myrtaa{at}geo.net.co. Luz E. Cano, Escuela de Bacteriología, Universidad de Antioquia, Medellín, Colombia and Corporación para Investigaciones Biológicas, Carrera 72A # 78B-141, Medellín, Colombia, Fax: 57-4-441-0855, E-mail: lcano{at}cib.org.co. Jaime Sampedro, Hospital La María, Calle 92 EE # 67-61, Medellín, Colombia, Fax 57-4-237-1963.
Reprint requests: Angela Restrepo, Corporación para Investigaciones Biológicas, Carrera 72A # 78B-141, Medellín, Colombia.
| REFERENCES |
|
|
|---|
This article has been cited by other articles:
![]() |
E. Lazar-Molnar, A. Gacser, G. J. Freeman, S. C. Almo, S. G. Nathenson, and J. D. Nosanchuk The PD-1/PD-L costimulatory pathway critically affects host resistance to the pathogenic fungus Histoplasma capsulatum PNAS, February 19, 2008; 105(7): 2658 - 2663. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. D. Lindsley, H. L. Holland, S. L. Bragg, S. F. Hurst, K. A. Wannemuehler, and C. J. Morrison Production and Evaluation of Reagents for Detection of Histoplasma capsulatum Antigenuria by Enzyme Immunoassay Clin. Vaccine Immunol., June 1, 2007; 14(6): 700 - 709. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |