Cryptococcosis is a potentially life-threatening disease caused by Cryptococcus neoformans, an encapsulated yeast. It occurs in individuals with cellular immune deficiencies such as those caused by HIV infection. The availability of highly active antiretroviral treatments (HAART) has been associated in developed countries with dramatic declines in the incidence of new AIDS cases  and of various opportunistic infections, including cryptococcosis in HIV-infected patients [2–4]. Under HAART, CD4 T lymphocyte counts rise and their function is improved, even allowing the safe withdrawal of maintenance antifungal therapy after successful treatment of cryptococcosis [5,6]. Among the complications recorded during HAART, paradoxical reactions resembling exacerbations of the primary infection and ascribed to immune reconstitution have been observed during mycobacterial and cytomegalovirus infections, viral hepatitis [7,8], Pneumocystis jiroveci pneumonia , histoplasmosis  or cryptococcosis [8,11–18]. However, little is known of the incidence and risk factors predisposing to immune reconstitution inflammatory syndrome (IRIS) in the context of C. neoformans infection. The nationwide surveillance data for cryptococcosis in France were used to describe and analyse the incidence and risk factors associated with IRIS-related cryptococcosis [6,19].
Using the ongoing nationwide surveillance of cryptococcosis managed by the French National Reference Centre for Mycoses (NRCM) [4,20], a retrospective multicentre cohort was designed in France including HIV-infected adults who developed a first episode of culture-confirmed cryptococcosis between March 1996 and June 2000 with a follow-up of at least 3 months until June 2002 . Standardized questionnaires were used. All questionnaires were reviewed by two of the group to uncover unsuspected IRIS and to retrieve missing data. IRIS was defined as the occurrence of new clinical symptoms consistent with an inflammatory process in patients with immunological and/or virological response following the initiation of HAART together with negative fungal cultures from all sites sampled and no new evolutive opportunistic infections, HIV-related neoplasia or HIV drugs-related disorders. These criteria are in accord with those previously reported . The agreement of both investigators was necessary for a diagnosis of IRIS. HAART consisted initially of two nucleoside reverse transcriptase inhibitors plus at least one protease inhibitor or a non-nucleoside reverse transcriptase inhibitor. A systematic extensive work-up to assess dissemination of the fungal infection was performed at the time of antifungal therapy initiation in 179 of the 240 patients treated for cryptococcosis and enrolled in the cohort study. Of these 179 patients, 120 were treated with HAART during the study period and were, therefore, at risk of developing an IRIS.
C. neoformans isolates were serotyped at the NRCM . Sections of tissue samples obtained during the work-up in the IRIS patients were stained with haematoxylin and eosin and Alcian blue and carefully reviewed for this study. Some paraffin-embedded sections were subjected to immunohistochemical analysis using a murine monoclonal IgG1 antibody (E1) reactive with cryptococcal capsular polysaccharide . In one case, anti-CD68 antibody (PGM1 clone; Dakopatt, Trappes, France) was used to detect the presence of macrophages. Proinflammatory cytokines (tumour necrosis factor-α, interleukin-6, interleukin-8) and anti-inflammatory markers (interleukin-10, soluble tumour necrosis factor receptor II) in cerebrospinal fluid (CSF) and plasma samples collected at the time of cryptococcosis diagnosis were measured by enzyme-linked immunosorbent assays (R&D Systems, Abingdon, UK).
In accordance with French law, the database was approved by the Commission Nationale de l’Informatique et des Libertés. All data were coded. Epi-Info software [version 6.04c, 1997; Centers for Disease Control and Prevention, Atlanta, Georgia, USA, and World Health Organization, Geneva, Switzerland) was used. Study subjects were considered at risk of IRIS from the day they had experienced both cryptococcosis and HAART. The period at risk continued until development of IRIS, death, loss to follow-up or end of the study period, whichever came first. The incidence of IRIS was calculated as the ratio of the number of cases of IRIS divided by the person-time at risk. The 95% confidence interval (CI) of the incidence was estimated assuming a Poisson distribution of IRIS cases. Risk factors for the development of IRIS were identified by the comparison of sociodemographic, clinical and biological variables in patients with and without IRIS, using Fisher's exact test and the Mann–Whitney U-test for categorical and continuous variables, respectively. Odds ratios (OR) and their 95% CI, pertaining to the association between IRIS and exposure variables, were determined by means of logistic regression analysis. All variables with P value < 0.25 in univariate analysis (i.e., cryptococcosis revealing HIV infection, HAART started between days 0 and 60 of the diagnosis of cryptoccosis, fungaemia at the time of cryptococcosis, disseminated cryptococcosis, and absence of sterilization at week 2 of antifungal treatment) were entered simultaneously into the full model. CD4 cell counts at the time of cryptococcosis, because of their clinical relevance, were also added to the model. Variables with higher P values were removed following a backwards-stepwise selection procedure, leaving only variables with P < 0.05 in the final model. Analyses were performed using Stata computer package version 8 (Stata Statistical Software, Stata Corp., College Station, Texas, USA).
Cryptococcosis-associated immune reconstitution inflammatory syndrome in France
Among the 120 HIV-infected patients with treated cryptococcosis and HAART, 10 developed IRIS during 239 person-years of follow-up (estimated incidence, 4.2/100 person-years; 95% CI, 2.2–7.8). Two additional patients with IRIS notified to the NRCM but not included in the database corresponding to the study period were added for the description of the disease and the identification of risk factors. All 12 patients considered to have IRIS were men; mostly Caucasian (n = 7) and cryptococcosis was the AIDS -defining illness in 82% (Table 1). When cryptococcosis was diagnosed, they all had meningitis and the yeast infection was disseminated in 11 (positive culture of blood for 10, urine for six, bronchoalveolar lavage or sputum for two, gastric and bone marrow biopsies for one each). Brain imaging (computed tomography or magnetic resonance imaging) was abnormal in two (pontine grey matter abnormalities in two, together with dilated Virchow–Robin spaces without parenchymal oedema in one). Chest radiographs were abnormal for five patients (showing diffuse interstitial pneumonia in four, excavated nodules with enlarged mediastinal lymphadenopathies in one). Initial antifungal therapy consisted of amphotericin B (≥ 0.5 mg/kg body weight daily) or its liposomal derivative for 10 in combination with flucytosine for five and fluconazole for two. After 2 weeks of treatment, 7 of the 10 CSF specimens cultured still contained viable yeasts. None of the specimens tested after 3 months of antifungal treatment grew C. neoformans. Subsequent consolidation and maintenance antifungal therapy beyond 3 months consisted of prolonged administration of fluconazole (≥ 200 mg daily).
Risk factors for cryptococcosis-associated immune reconstitution inflammatory syndrome
Baseline (at the time of cryptococcosis diagnosis) characteristics of the 12 patients who would subsequently develop IRIS were compared with those of 110 patients who did not develop IRIS (Table 1). According to the univariate analysis, those developing IRIS were less often aware of their HIV infection, more frequently had fungaemia and disseminated cryptococcosis at baseline, had more frequently started HAART within 8 weeks following cryptococcosis diagnosis and more frequently had persistently positive CSF cultures after 2 weeks of antifungal therapy (Table 1). Other baseline parameters, including CSF cellularity and protein concentration, did not differ significantly. Baseline cytokine responses in the CSF and blood did not differ between groups (data not shown). Multivariate analysis showed that independent risk factors for the subsequent occurrence of IRIS in this population were HIV infection that was revealed by cryptococcosis (OR, 4.8; 95% CI, 1.0–21.7; P = 0.04), CD4 cell count < 7 × 106 cells/l (lower quartile) (OR, 4.0; 95% CI, 0.9–17.2; P = 0.06), fungaemia at baseline (OR, 6.1; 95% CI, 1.1–35.2; P = 0.04) and initiation of treatment within 2 months of cryptococcosis (OR, 5.5; 95% CI, 1.0–29.6; P = 0.05).
Major characteristics of immune reconstitution inflammatory syndrome
HAART was started a median of 12 days (range, 39 days before to 60 days after) following the diagnosis of cryptococcosis (Table 1) and consisted initially of two nucleoside reverse transcriptase inhibitors plus at least one protease inhibitor in 10 patients or a non-nucleoside reverse transcriptase inhibitor in two patients. The median time between initiation of HAART and the onset of IRIS was 8 months (range, 2–37). At the time of IRIS, the median total dosages of amphotericin B and fluconazole that had been administered were 1.5 g (range, 0.3–5.1) and 74 g (range, 8–572), respectively; the median CD4 T cell count was 215 × 106 cells/l (range, 43–640) and the virus load was undetectable (< 1.7 log10 copies/ml; range, 1.7–3.8) in 58%.
Clinical features associated with IRIS were fever and acute headache in almost all patients (n = 10). Signs of recent onset, and unrecorded at baseline, were seizures (n = 4), hemiplegia, facial paralysis, paraplegia, dysarthria and acute dyspnoea, occurring each in one patient, and enlarged lymphadenopathy in four patients (cervical, retropharyngeal, subclavicular or mediastinal lymph nodes, each in one patient). Chest computed tomography showed pulmonary nodules and infiltrates in two patients (Fig. 1a,b). Cerebral magnetic resonance imaging was abnormal in four of five patients examined (including two in whom no abnormality had been found at the time of cryptococcosis diagnosis), showing diffuse microabscesses, marked oedema and abnormal contrast media uptake in the Virchow–Robin spaces or meninges (Fig. 1c,d). The patient with paraplegia had a dorsal intramedullary abscess. Direct examination of tissue or fluid samples obtained at the time of IRIS was negative for seven patients but encapsulated yeasts were present for five. All mycological cultures were negative as were all other microbiological investigations. Serum and/or CSF cryptococcal antigen was detected in 10 patients; it was not found in the CSF of one with no concomitant serum sample available and it was not tested in the last patient. For the seven CSF samples tested, cellularity and protein concentrations were variable [median 48 × 106 cells/l (range, 1–406) and median 0.68 g/l (range, 0.3–1.1), respectively]. Tissue biopsies (lung, bronchii, leptomeninges and intramedullary abscess from one patient each and lymph nodes from two patients) showed lesions containing lymphocytes, plasma cells, macrophages and/or histiocytes forming granulomas with or without necrosis (Fig. 2a). Phagocytosed yeasts labelled by monoclonal antibody E1 were seen inside macrophages in two samples (one pulmonary and one cerebral) (data not shown). A predominance of CD68 macrophages was seen in a pulmonary nodule (Fig. 2b).
Treatment and outcome of immune reconstitution inflammatory syndrome
The diagnosis of IRIS was rapidly made for eight patients and assessed retrospectively for the remaining four. However, for all 12 patients, a relapse of cryptococcosis was initially suspected, which led to changes of antifungal therapy for all patients but one: administration of amphotericin B or its liposomal derivative with or without flucytosine to seven patients (for up to 2 years for one patient with persistent necrotic adenopathies despite negative fungal cultures); fluconazole alone for two and fluconazole combined with flucytosine for two. HAART was continued in 11/12 and discontinued in one patient who had a CD4 cell count of 350 × 106 cells/l. Three patients were hospitalized in intensive care units. One died shortly after admission without administration of anti-inflammatory drugs for refractory seizures related to acute lymphocytic meningoencephalitis. The other two, who had severe vigilance disturbances associated with hemiparesia for one and dysarthria for the other, survived. Both received steroids (1 or 2 mg/kg daily) and improved markedly in 3–15 days. Among the other nine patients, one did not receive any anti-inflammatory treatment and died with enlarged necrotic lymph nodes containing numerous yeasts internalized in granuloma cells. The other eight patients survived, including five who did not receive any anti-inflammatory treatment and completely recovered spontaneously within 1 (four patients) to 2 months (one patient). Two patients received specific therapy for IRIS: thalidomide for 4 months, with a favourable progressive return to normal of lymph-node size, or steroids for severe lymphadenopathy (1 mg/kg daily), with dramatic improvement within 10 days, and then lower doses for a total of 8 months. The patient with the intramedullary abscess was successfully treated surgically. One patient was lost to follow-up after IRIS but none of the survivors developed an IRIS or cryptococcal relapse during subsequent follow-up (median post-IRIS follow-up 22 months; range, 6–41). Eight of the nine survivors had persistent control of their HIV loads and prolonged immune reconstitution (CD4 cell count stabilization).
IRIS has been reported for a limited number of patients with various opportunistic non-fungal infections [8,24] and even more rarely during P. jiroveci[8,9], Histoplasma capsulatum or C. neoformans[8,11–18] infections.
Two distinct clinical entities can be identified in the context of cryptococcosis. The first is the discovery of an unknown fungal infection shortly after HAART introduction during the first weeks following the diagnosis of HIV infection. The clinical picture can include positive yeast cultures and marked meningeal reaction [8,11,16], sometimes associated with elevated intracranial pressure . This entity is probably often misdiagnosed and none of the clinicians participating in the French Cryptococcosis Study Group diagnosed it during the study period, nor did we after careful review of the questionnaires. The second entity, corresponding to our 12 IRIS cases, occurs later after HAART initiation when the fungal infection is already controlled, as assessed by negative cultures. It has previously been described anecdotically during cryptococcosis [12–15,17,18].
Cryptococcosis associated-IRIS can be responsible for various clinical presentations . The most common signs were enlarged necrotic lymph nodes and acute central nervous system (CNS)-related symptoms. However, two major observations can be made. First, a previously uninfected (or apparently not infected) organ can be the site of new symptoms or signs, such as the intramedullary abscess seen in one patient or lymphadenopathies. Second, the clinical presentation can be so severe (recurrent seizures, vigilance disorders) as to require admission to an intensive care unit. Death can even be directly attributed to evolutive IRIS (two of our patients), an outcome rarely reported in the context of other opportunistic infections . We estimated the incidence of cryptococcosis-associated IRIS in France to be 4.2/100 person-years of all HAART-treated patients with a past history of cryptococcosis. However, this figure may be an underestimation. Indeed, we may have missed IRIS cases, especially those that would have occurred during the active phase of the fungal infection soon after introduction of HAART, because clinicians were not aware of this particular entity at that time. Unfortunately, no other published data on cryptococcocis are available for comparison. Comparison with other opportunistic infections associated with IRIS is difficult because of the different defining criteria applied. However, a prevalence of up to 36% has been advanced for Mycobacterium tuberculosis-associated IRIS .
Patients who would subsequently develop IRIS had profound immune deficiencies at the time of cryptococcosis diagnosis because antiretroviral treatment had not yet been prescribed or had been prescribed too recently to have an effect. At the time of IRIS, however, the virus load had already fallen sharply and the CD4 T lymphocyte counts risen in almost all patients, but a low CD4 cell count or a detectable virus load did not rule out the diagnosis of IRIS, as previously reported [7,8,27]. Symptoms of IRIS reflect a strong inflammatory response. Indeed, brain and thoracic imaging showed abnormal contrast medium uptake into the lesions. In a few cases, CSF cell counts and protein concentrations were higher than those usually encountered in the setting of HIV infection, and histological examination found granulomas, in agreement with some published case reports [8,12,16]. These granulomas were mostly composed of macrophages containing internalized cryptococci that did not grow in culture. Such an inflammatory reaction is extremely rare, especially in the CNS of patients with AIDS during the active phase of C. neoformans infection, and it is more commonly observed in HIV-seronegative patients and in experimental models of cryptococcosis after remission of the fungal infection [28,29].
Given the incidence and the occasionally fatal outcome of cryptococcosis-associated IRIS, identifying the risk factors contributing to its development seems important. Previous studies suggested that the more-immunocompromised patients are more prone to develop the syndrome because of an enhanced effect of HAART in more active HIV infection . According to our univariate analysis, previously undiagnosed HIV infection, fungaemia, disseminated cryptococcosis, lack of CSF sterilization after 2 weeks of antifungal treatment and introduction of HAART within 2 months after the diagnosis of cryptococcosis differed significantly between patients who would subsequently develop IRIS and those who would not. The multivariate analysis retained previously undiagnosed HIV infection, low CD4 cell counts, fungaemia and HAART introduction within 2 months after cryptococcosis diagnosis as independent variables associated with IRIS occurrence. These characteristics seem to indicate that a subgroup of patients might be at greater risk of developing IRIS during the course of cryptococcosis than others and that they can be identified provided that an extensive work-up is done at the time the first episode of cryptococcosis is diagnosed.
Given the occasional fatality, as documented in one of our patients who died immediately from acute inflammatory encephalitis, adequate management of IRIS patients is crucial. Although most patients recovered spontaneously, severe IRIS required specific anti-inflammatory treatment and sometimes admission to intensive care units. Various anti-inflammatory treatments have been reported in the literature and in our series, including non-steroidal agents for necrotic lymphadenopathies [14,30], high-dose steroids for severe CNS manifestations , and even thalidomide, an anti-tumour necrosis factor-α agent. Surgery can also be considered, either to obtain tissue samples for a definitive diagnosis of IRIS as opposed to mycological relapse, or to remove enlarged necrotized lymphadenopathies or abscesses . Finally, the development of IRIS in the context of C. neoformans infection raises the question of the most appropriate time to initiate HAART after cryptococcosis diagnosis in an untreated HIV-infected patient with a profound immune deficiency. A safety zone of 4 weeks was recently suggested for tuberculosis . Based on our data, we would also recommend avoiding HAART introduction at least until cryptococcosis is microbiologically controlled. Most importantly, patients with the identified risk factors associated with IRIS should be monitored particularly carefully for months after the introduction of HAART.
In conclusion, IRIS-associated cryptococcosis must be differentiated from microbiological relapse and severe cases may require specific therapeutic interventions using anti-inflammatory agents. IRIS is a potentially fatal complication of AIDS-associated cryptococcosis that occurs in a well-identified subset of HAART-treated patients, who require careful monitoring.
The authors also acknowledge the active collaboration of Cécile Droz and Geneviève Chêne (INSERM U593, Bordeaux), all the French mycologists and clinicians who contributed data to CryptoStop and CryptoA/D studies and Janet Jacobson for reviewing the English language.
Note: This work was presented in part at the Fifth Conference on Cryptococcus and Cryptococcosis, Adelaide, March 2002 [abstract PO 38] and at the 44th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, DC October–November 2004 [abstract M-1054].
Conflict of interest statement: The authors declare that they have no conflict of interest with the present study.
1. Palella FJ, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA, et al
. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. N Engl J Med 1998; 338:853–860.
2. Chen S, Sorrell T, Nimmo G, Speed B, Currie B, Ellis D, et al
. Epidemiology and host- and variety-dependent characteristics of infection due to Cryptococcus neoformans
in Australia and New Zealand. Clin Infect Dis 2000; 31:499–508.
3. Mirza SA, Phelan M, Rimland D, Graviss E, Hamill R, Brandt ME, et al
. The changing epidemiology of cryptococcosis: an update from population-based active surveillance in 2 large metropolitan areas, 1992–2000. Clin Infect Dis 2003; 36:789–794.
4. Dromer F, Mathoulin-Pelissier S, Fontanet A, Ronin O, Dupont B, Lortholary O, et al
. Epidemiology of HIV-associated cryptococcosis in France (1985–2001): comparison of the pre- and post-HAART
5. Mussini C, Pezzotti P, Miro JM, Borghi V, Monforte A, Govoni A, et al
. Discontinuation of maintenance therapy for cryptococcal meningitis in patients with AIDS
treated with highly active antiretroviral therapy: an international observational study. Clin Infect Dis 2004; 38:565–571.
6. Lortholary O, Droz C, Sitbon K, Zeller V, Neuville S, Alvarez M, et al
. Long-term outcome of HIV-associated cryptococcosis (CC) at the time of HAART. Results of the Multicenter Cohort Cryptostop.43rd Interscience Conference on Antimicrobial Agents and Chemotherapy
. Chicago, September 2003 [abstract M-1752].
7. DeSimone JA, Pomerantz RJ, Babinchak TJ. Inflammatory reactions in HIV-1-infected persons after initiation of highly active antiretroviral therapy. Ann Intern Med 2000; 133:447–454.
8. Shelburne SA III, Hamill RJ, Rodriguez-Barradas MC, Greenberg SB, Atmar RL. Immune reconstitution inflammatory syndrome: emergence of a unique syndrome during highly active antiretroviral therapy. Medicine 2002; 81:213–227.
9. Wislez M, Bergot E, Antoine M, Parrot A, Carette MF, Mayaud C, et al
. Acute respiratory failure following HAART
introduction in patients treated for Pneumocystis carinii
pneumonia. Am J Respir Crit Care Med 2001; 164:847–851.
10. Breton G, Adle-Biassette H, Therby A, Ramanoelina J, Choudat L, Bissuel F, et al
. Immune reconstitution inflammatory syndrome in HIV-infected patients with disseminated histoplasmosis.AIDS
11. Woods ML, MacGinley R, Eisen DP, Allworth AM. HIV combination therapy: partial immune restitution unmasking latent cryptococcal infection. AIDS
12. Jenny-Avital ER, Abadi M. Immune reconstitution cryptococcosis after initiation of successful highly active antiretroviral therapy. Clin Infect Dis 2002; 35:128–133.
13. Breton G, Seilhean D, Cherin P, Herson S, Benveniste O. Paradoxical intracranial cryptococcoma in a human immunodeficiency virus-infected man being treated with combination antiretroviral therapy. Am J Med 2002; 113:155–157.
14. Blanche P, Gombert B, Ginsburg C, Passeron A, Stubei I, Rigolet A, et al
. HIV combination therapy: immune restitution causing cryptococcal lymphadenitis dramatically improved by anti-inflammatory therapy. Scand J Infect Dis 1998; 30:615–616.
15. Trevenzoli M, Cattelan AM, Rea F, Sasset L, Semisa M, Lanzafame M, et al
. Mediastinitis due to cryptococcal infection: a new clinical entity in the HAART
era. J Infect 2002; 45:173–179.
16. King MD, Perlino CA, Cinnamon J, Jernigan JA. Paradoxical recurrent meningitis following therapy of cryptococcal meningitis: an immune reconstitution syndrome after initiation of highly active antiretroviral therapy. Int J STD AIDS
17. Cattelan AM, Trevenzoli M, Sasset L, Lanzafame M, Marchioro U, Meneghetti F. Multiple cerebral cryptococcomas associated with immune reconstitution in HIV-1 infection. AIDS
18. Boelaert JR, Goddeeris KH, Vanopdenbosch LJ, Casselman JW. Relapsing meningitis caused by persistent cryptococcal antigens and immune reconstitution after the initiation of highly active antiretroviral therapy. AIDS
19. Kwon-Chung KJ, Sorrell TC, Dromer F, Fung E, Levitz SM. Cryptococcosis: clinical and biological aspects. Med Mycol 2000; 38(suppl. 1):205–213.
20. Dromer F, Mathoulin S, Dupont B, Laporte A, and the French Cryptococcosis Study Group. Epidemiology of cryptococcosis in France: 9-year survey (1985–1993). Clin Infect Dis 1996; 23:82–90.
21. Shelburne SA III, Hamill RJ. The immune reconstitution inflammatory syndrome. AIDS
Rev 2003; 5:67–79.
22. Dromer F, Guého E, Ronin O, Dupont B. Serotyping of Cryptococcus neoformans
by using a monoclonal antibody specific for capsular polysaccharide. J Clin Microbiol 1993; 31:359–363.
23. Dromer F, Salamero J, Contrepois A, Carbon C, Yeni P. Production, characterization, and antibody specificity of a mouse monoclonal antibody reactive with Cryptococcus neoformans
capsular polysaccharide. Infect Immun 1987; 55:742–748.
24. Hirsch HH, Kaufmann G, Sendi P, Battegay M. Immune reconstitution in HIV-infected patients. Clin Infect Dis 2004; 38:1159–1166.
25. Cinti SK, Armstrong WS, Kauffman CA. Case report. Recurrence of increased intracranial pressure with antiretroviral therapy in an AIDS
patient with cryptococcal meningitis. Mycoses 2001; 44:497–501.
26. Narita M, Ashkin D, Hollender ES, Pitchenik AE. Paradoxical worsening of tuberculosis following antiretroviral therapy in patients with AIDS
. Am J Respir Crit Care Med 1998; 158:157–161.
27. Cooney EL. Clinical indicators of immune restoration following highly active antiretroviral therapy. Clin Infect Dis 2002; 34:224–233.
28. Lee SC, Dickson DW, Casadevall A. Pathology of cryptococcal meningoencephalitis: analysis of 27 patients with pathogenetic implications. Hum Pathol 1996; 27:839–847.
29. Chrétien F, Lortholary O, Kansau I, Neuville S, Gray F, Dromer F. Pathogenesis of cerebral Cryptococcus neoformans
infection after fungemia. J Infect Dis 2002; 186:522–530.
30. Manfredi R, Calza L, Chiodo F. AIDS
infection before and after the highly active antiretroviral therapy era: emerging management problems. Int J Antimicrob Agents 2003; 22:449–452.
Members of the French Cryptococcosis Study Group who participated in this study (listed in alphabetical order by city). Claude Bazin (Hôpital de la Côte-de-Nacre, Caen); Emmanuel Mortier (Hôpital Louis Mourier, Colombes); Christine Jacomet (Hôtel-Dieu, Clermont-Ferrand); Odile Salmon (Hôpital Louise-Michel, Evry); Philippe Bossi and Guillaume Breton (Hôpital de la Pitié-Salpêtrière, Paris), Christophe Piketty (Hôpital Européen Georges-Pompidou, Paris), Dominique Salmon (Hôpital Cochin, Paris); André Boibieux (Hôpital de La Croix-Rousse, Lyon); Cédric Arvieux (Hôpital Pontchaillou, Rennes).
31. Blumberg HM, Burman WJ, Chaisson RE, Daley CL, Etkind SC, Friedman LN, et al
. American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: treatment of tuberculosis. Am J Respir Crit Care Med 2003; 167:603–662.