Toxoplasmosis after Hematopoietic Stem Cell Transplantation

Clinical Infectious Diseases, Nov 2000

Forty-one cases of toxoplasmosis were diagnosed in 15 European transplantation centers in patients who had undergone allogeneic hematopoietic stem cell transplantation (HSCT) from 1994 through 1998. Most patients (39 [94%]) were seropositive for Toxoplasma gondii before they underwent transplantation, and 30 (73%) had developed moderate to severe acute graft-versus-host disease before they developed toxoplasmosis. Thirty-five (85%) patients had Toxoplasma disease with evidence of organ involvement, whereas 6 (15%) patients had Toxoplasma infection, as defined by fever and a positive polymerase chain reaction (PCR) finding for T. gondii in blood. Nine patients were diagnosed at autopsy. Thirty patients (73%) had not received antimicrobial prophylaxis with anti-Toxoplasma activity after undergoing transplantation. The median day of onset of disease after HSCT was 64. Twenty-two (63%) patients died from toxoplasmosis, and 23 (66%) received adequate anti-Toxoplasma therapy for ⩾3 days. Among these 23 patients, 11 (48%) showed a complete response and 3 (13%) showed improvement. In univariate and multivariate analyses, having received adequate therapy and experiencing late infection (>63 days after HSCT) were associated with a lower risk of dying from toxoplasmosis. Toxoplasmosis after HSCT is a severe infection with a high mortality rate even when diagnosed soon after HSCT, and PCR may help establish the diagnosis earlier.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://cid.oxfordjournals.org/content/31/5/1188.full.pdf

Toxoplasmosis after Hematopoietic Stem Cell Transplantation

R. Martino () 1 3 4 10 J. Maertens 0 1 3 10 S. Bretagne 1 3 9 10 M. Rovira 1 2 3 10 E. Deconinck 1 3 7 10 A. J. Ullmann 1 3 6 10 T. Held 1 3 5 10 C. Cordonnier 1 3 9 10 for the European Group for Blood 1 3 10 Marrow Transplantation Infectious Diseases Working Party 1 3 8 10 0 Department of Hematology, University Hospital Gasthuisberg , Leuven, Belgium 1 Patients. Fifty-one HSCT centers from 18 countries partici- pated in this study. The total number of allogeneic and autologous HSCTs performed from 1994 to 1998 at these 51 institutions was 4231 and 6787 , respectively. Among autologous recipients, 2 cases of the classic lymphadenopathic-infectious mononucleosis-like syndrome were seen late after unmanipulated autologous bone mar- row transplantation in previously seronegative patients. Both pa- tients recovered without treatment , and the diagnosis was made retrospectively on the basis of an increase in IgM levels followed 2 Department of Hematology, Hospital Clnic i Provincial , Barcelona, Spain 3 Received 23 December 1999; revised 18 April 2000; electronically pub- lished 7 November 2000 4 Division of Clinical Hematology, Hospital de la Sant Creu i Sant Pau , Barcelona, Spain 5 Charite-Virchow Klinikum, Berlin, Germany 6 Division of Internal Medicine, Johannes Gutenberg University Hospital , Mainz 7 Department of Hematology , Hopital Jean Minjoz, Besancon, France 8 Members of the working party are listed after the text. Clnica, Hospital de la Santa Creu i Sant Pau , Av. Sant Antoni M Claret, 167 08025 Barcelona, Spain 9 Department of Hematolgy , Hopital Henri Mondor, Creteil 10 Spain: Dr. R. Martino (Hospital de la Santa Creu i Sant Pau , Barcelona) , Dr. M. Rovira (Hospital Clnic i Provincial , Bar- celona) , Dr. R. De la C amara (Hospital de la Princesa , Madrid) , and Dr. R. Parody (Hospital Virgen del Roco , Sevilla). Ger- many: Dr. T. Held (Charite-Virchow Klinikum, Berlin) , Dr. A. J. Ullmann (Johannes Gutenberg University Hospital , Mainz), and Dr. U. Platzbecker (Universitatsklinikum Carl Gustav Ca- rus, Dresden ). Belgium: Dr. J. Maertens (University Hospital Gasthuisberg , Leuven) , Dr. A. Ferrant (Universite Catholique de Louvain , Brussels) , and Dr. R. Schots (Academic Hospital, Free University Brussels). France: Dr. C. Cordonnier , Dr. S. Bretagne (H opital Henri Mondor, Creteil), and Dr. E. Decon- inck (H opital Jean Minjoz, Besancon). Italy: Dr. M. Picardi (Universita delli Studi di Napoli , Federico II, Napoli). Greece: Dr. D. Petropoulos (Santa Sophia Children's Hospital, Athens ). Slovenia: Dr. S. Zver (University Medical Centre , Ljubljana) Forty-one cases of toxoplasmosis were diagnosed in 15 European transplantation centers in patients who had undergone allogeneic hematopoietic stem cell transplantation (HSCT) from 1994 through 1998. Most patients (39 [94%]) were seropositive for Toxoplasma gondii before they underwent transplantation, and 30 (73%) had developed moderate to severe acute graft-versus-host disease before they developed toxoplasmosis. Thirty-five (85%) patients had Toxoplasma disease with evidence of organ involvement, whereas 6 (15%) patients had Toxoplasma infection, as defined by fever and a positive polymerase chain reaction (PCR) finding for T. gondii in blood. Nine patients were diagnosed at autopsy. Thirty patients (73%) had not received antimicrobial prophylaxis with anti-Toxoplasma activity after undergoing transplantation. The median day of onset of disease after HSCT was 64. Twenty-two (63%) patients died from toxoplasmosis, and 23 (66%) received adequate anti-Toxoplasma therapy for >3 days. Among these 23 patients, 11 (48%) showed a complete response and 3 (13%) showed improvement. In univariate and multivariate analyses, having received adequate therapy and experiencing late infection (163 days after HSCT) were associated with a lower risk of dying from toxoplasmosis. Toxoplasmosis after HSCT is a severe infection with a high mortality rate even when diagnosed soon after HSCT, and PCR may help establish the diagnosis earlier. - Toxoplasma gondii is an intracellular protozoan parasite. Transmission to human hosts may occur transplacentally; by ingestion of raw or undercooked meat containing tissue cysts; or by exposure to oocysts in soil contaminated with cat feces, whose intestine is the parasites final host and from where oocysts are excreted [1]. Primary infection is seldom symptomatic in healthy individuals and is usually recognized serologically. The probability of having been infected (i.e., seroprevalence of IgG antibodies against T. gondii) increases with age and varies between geographic areas from 5%90% of healthy adults [1]. However, clinically symptomatic infection may present in any of 4 ways: (1) the more common mild lymphatic form that may resemble infectious mononucleosis; (2) chronic toxoplasmosis with retinochoroiditis; (3) an acute, fulminating, disseminated infection; and (4) single-organ disease, usually encephalitis. The latter 2 forms are usually seen in highly immunocompromised individuals, in whom most such infections appear to result from reactivation of latent tissue parasites in seropositive individuals. by a finding of IgG antibodies against T. gondii. However, there were no cases of Toxoplasma disease with organ involvement after autologous HSCT, thus confirming the rarity of this complication in this setting [24, 6, 7]. Forty-one cases were diagnosed in allogeneic HSCT recipients, which corresponds to a frequency of 0.97%. Fifteen centers that reported having seen >1 case of toxoplasmosis were sent detailed questionnaires, and this study is based on the data they reported. Definitions. Table 1 details the proposed definitions for toxoplasmosis diagnosed after HSCT. Classically, toxoplasmosis in this setting has been diagnosed by the histologic demonstration of tachyzoites in involved organs, either by biopsy or at autopsy [24]. In recent years, development of highly sensitive PCR techniques for the detection of small amounts of parasites in various diagnostic samples have been developed [8]. Although these techniques are not currently standardized, they are being used in many microbiology laboratories for the noninvasive diagnosis of toxoplasmosis. Thus, 8 of 15 centers in the current study reported having >1 positive PCR test during the workups for the patients. This led to the finding of a previously undescribed clinical situation with respect to toxoplasmosis in HSCT recipients. Specifically, 6 patients who had fever of unknown origin as the only sign and symptom were found to have a positive PCR test from at least 1 blood sample; the fever abated and the PCR became negative after antiToxoplasma therapy. Such patients without any evidence of organ involvement have been classified as having Toxoplasma infection by analogy with cytomegalovirus infection and disease in this setting [9]. Toxoplasma disease was defined as the presence of Toxoplasma infection plus clinical, radiological, or pathological evidence of organ involvement by T. gondii. Definite Toxoplasma disease was defined as the histologic or cytologic demonstration of tachyzoites in tissue samples obtained either by biopsy or bronchoalveolar lavage (BAL), or at autopsy. Patients who had clinical and radiologic evidence suggestive of Toxoplasma disease plus >1 positive PCR test from blood, CSF, or BAL, but who had no histologic confirmation, were classified as having probable Toxoplasma disease. Patients who had only a cranial CT scan or MRI highly suggestive of CNS toxoplasmosis (as considered by each hospitals neuroradiologists) and responded to anti-Toxoplasma therapy but had no other laboratory evidence of Toxoplasma disease were considered as having possible (presumptive) Toxoplasma disease. Disseminated toxoplasmosis was defined as clinical, radiological, or histologic evidence of disease in 11 organ. Toxoplasma disease was considered the main cause of death when no other major life-threatening infection or other potentially fatal complication occurred immediately before death, although it was considered as a contributing cause of death when such a complication occurred; patients were considered as having died from another cause if they had responded to therapy before death from an unrelated complication. Patients were considered evaluable for response to anti-Toxoplasma therapy if they completed at least 3 days of treatment. Complete response to therapy was defined as disappearance of all clinical signs and symptoms attributable to Toxoplasma disease and normalization or only residual findings in previously abnormal imaging studies. Patients who improved in any of these findings but did not reach optimal recovery before death or last follow-up were considered to be improved, whereas all other patients were considered as not having responded to therapy. Statistical analysis. The x2 statistic or Fishers exact test were used to establish differences in the distribution of categorical variables, and the Mann-Whitney U test was used to compare continuous variables. Survival was calculated from the initial symptoms of Toxoplasma disease until death from any cause or from time of last follow-up for survivors. Tests of significance were 2sided, with a significance level of P < .05. Results Patient characteristics are detailed in table 2. The median age was 35 years, with 3 patients aged !16 years. Most patients had received a bone marrow transplant, mostly from a human leukocyte antigen ( HLA)identical sibling donor (61%). Status of the underlying disease at transplantation was in early (standard risk) phase in 41%. Most patients (76%) were conditioned before transplantation with a total body irradiationcontaining regimen, usually in conjunction with cyclophosphamide. For most patients (73%), prophylaxis for graft-versus-host (GVH) disease was cyclosporine plus methotrexate, and 13 (32%) patients received a T celldepleted allograft. Before the allograft was performed, the IgG Toxoplasma serology of the recipient Proposed definitions for toxoplasmosis after hematopoietic stem cell transplantation. Toxoplasmosis classification Toxoplasma disease Definite Toxoplasma disease Probable Toxoplasma disease (documented by PCR) Possible Toxoplasma disease (documented by imaging) Toxoplasma infection Histologic or cytologic demonstration of tachyzoites in tissue samples obtained either by biopsy or bronchoalveolar lavage, or at autopsy. Isolation of the parasite by culture in these samples would be evidence of disease. Clinical and radiologic evidence suggestive of organ involvement plus at least one positive PCR test from blood, CSF, or BAL, but no histologic confirmation and absence of another pathogen that may explain the findings. CT or MRI highly suggestive of CNS toxoplasmosis (as considered by each hospitals neuroradiologists) and response to anti-Toxoplasma therapy, but no laboratory evidence of toxoplasmosis and absence of another pathogen that may explain the findings. Positive PCR in blood in a patient without any evidence of organ involvement or seroconversion for Toxoplasma gondii after transplantation in a previously seronegative patient (with or without fever). NOTE. Data are no. of patients or no. (%), unless otherwise indicated. AGVH, acute GVH; CGVH, chronic GVH; CyA, cyclosporine A; GVH, graftversus-host; HSCT, hematopoietic stem cell transplantation; TOXO, Toxoplasma disease. a Two cases of aplastic anemia; 1 of Hodgkins disease; 1 of thalassemia; and 1 of sickle-cell disease. b Three previous autologous HSCTs. c Nine patients developed toxoplasmosis after day 100. d Immunosuppressive therapy in the 6 weeks that preceded toxoplasmosis. and donor were unknown for 8 patients; of the other 33 patients, 31 (94%) were seropositive, and for the 2 seronegative patients, the donors serostatus was unknown. Clinical presentation. Table 2 details the development of GVH disease and the immunosuppressive therapy given before the onset of toxoplasmosis. Thirty patients developed grade IIIV acute GVH (AGVH) disease (grade II, 15 patients; grades IIIIV, 15 patients). Of the 9 patients who developed toxoplasmosis after day 100 after HSCT, 3 had prior chronic GVH disease. Not surprisingly, all 41 patients had received immunosuppressive therapy during the 6 weeks that preceded the onset of toxoplasmosis, mainly cyclosporine A (95%) and corticosteroids (78%). When the first signs and symptoms were recorded, most patients were still receiving these drugs, and a significant proportion (44%) had active GVH disease (mainly AGVH disease). After transplantation, 18 patients had received no prophylaxis for Pneumocystis carinii pneumonia, 12 received pentamidine administered by inhalation, and 11 received trimethoprim/sulfamethoxazole (TMP/SMZ). Thus 30 patients (73%) had not received P. carinii pneumonia prophylaxis with anti-Toxoplasma activity after transplantation, and among the 11 who had received TMP/SMZ, only 3 were receiving it at onset of toxoplasmosis. Table 3 shows the main clinical and radiological findings in the 35 patients with Toxoplasma disease. The onset of disease was early after transplantation in most patients, with a median of 64 days (range, 4516 days) for the initial signs and symptoms and 75 days (range, 12676) for its diagnosis. Five cases occurred before day 30, and 27 before day 100; only 8 had a late onset, after day 100. The CD41 lymphocyte count was known in only 10 patients and was !200 cells/mL in 9 patients. Most patients (80%) had fever attributable to Toxoplasma disease, and the main other clinical symptoms were neurological (80%), pulmonary (49%), or both, usually with bilateral alveolar or interstitial infiltrates. Seventeen patients had isolated CNS involvement, and only 3 had isolated pneumonia; the other 15 patients had 11 involved organ. For 28 patients with neurologic signs and symptoms, either CT (25 patients) or MRI (16) scans were performed; both were performed for 13 patients. In 24 patients, abnormal CNS findings were found on >1 of these: a single nodular lesion in 8 patients and bilateral lesions in 16. MRI appeared to be more sensitive than CT, because all 16 MRI studies were abnormal, including those for 6 of 10 patients with a previously normal CT scan. Of the 25 CT scans, 15 were abnormal (confirmed in 7 by MRI) and 10 were normal (MRI not done in 4). Of note, all CT scans were performed before MRI scans when both were done, although the exact number of days between them was not recorded. Twenty-nine patients (83%) had definite (22 patients) or probable (7) Toxoplasma disease, and 6 (17%) had possible (presumptive) disease, as detailed in table 4. Of the 22 definite cases, 11 were diagnosed at autopsy, including 9 cases in patients who were not suspected to have Toxoplasma disease before death. However, a PCR test was not performed before death for any of these 9 patients. The other 2 definite postmortem diagnoses were found to be probable cases during life on the basis of a positive PCR test of CSF and blood. Diagnostic methods. Table 4 summarizes the clinical and laboratory findings that led to the diagnosis of Toxoplasma Clinical characteristic Bilateral lesionse Retinochoroiditisf NOTE. Data are no. of patients or no. (%), unless otherwise indicated. HSCT, hematopoietic stem cell transplantation. a If not suspected or diagnosed before death. b ! 0.5 3 109 neutrophils/L. c Available for 10 patients. d Three patients had isolated lung involvement. e Nine patients had alveolar infiltrates and 5 were described as having interstitial pneumonia. f Funduscopic examination was done in 14 patients. disease in the 29 patients with definite or probable disease. Histologic examination of involved organs was the main diagnostic method. In 2 patients with pneumonia, cytologic examination of BAL samples with direct immunofluorescence with antiT. gondii antibodies was the diagnostic test. It is of note that a positive PCR test was found in 16 patients, mostly in blood samples, and 5 patients had a positive PCR result from 2 different sites (blood and CSF for 3 patients, blood and BAL for 2). Therapy and outcome. Twenty-two patients (63%) died from Toxoplasma disease, either as the major cause (17 patients) or a contributing cause to death (5 patients), and their median survival was 11 days (range, 363 days). Thirteen patients (37%) survived Toxoplasma disease, but 7 died from another cause at a median of 74 days (range, 33530 days) from initial symptoms, and 6 were alive at a median of 630 days (range, 2441936 days). Fourteen patients (40%) had another concomitant severe infection during the course of toxoplasmosis, of which 8 contributed to the patients death (3 cytomegalovirus pneumonia, 4 invasive aspergillosis, and 1 adenovirus pneumonia). Twenty-three patients (66%) received at least 6 days of antiClinical presentation of 35 patients with Toxoplasma disease Toxoplasma therapy and were considered evaluable for response. The other 12 patients received no therapy (10 patients) or !6 days of treatment (2 patients) and died from Toxoplasma disease. Treatment details are shown in table 4. All patients who received anti-Toxoplasma therapy did so at the recommended doses for pyrimethamine (50100 mg/day orally), sulfadiazine (1 g every 6 h orally or iv) plus folinic acid, clindamycin, or both (600 mg every 6 h orally or iv), except for one patient who received therapy with high-dose trimethoprim/sulfamethoxazole because the initial suspicion was P. carinii pneumonia. The intended duration of therapy was 46 weeks. Of the 23 treated patients, 14 (61%) responded and 9 (39%) did not. Table 5 shows the relationship between death from Toxoplasma disease and the treatment given, as well as other clinical parameters that have potential prognostic impact. By univariate analyses, having received treatment and having late-onset Toxoplasma disease (defined here as onset after the median day of onset for the population, which is day 64 after HSCT) were associated with a higher probability of not dying from toxoplasmosis. These 4 variables were analyzed in multivariate analysis with stepwise logistic regression, and both having received treatment (P ! .0001) and having Toxoplasma disease that was discovered late (P p .02) were still associated with a lower probability of dying from Toxoplasma disease. On the other hand, the interval from initial symptoms and the start of therapy in the 23 patients who received therapy had no apparent influence on outcome (median 3 vs. 5.5 days for survivors and nonsurvivors from Toxoplasma disease, respectively). Of the 14 respondents, 7 patients received secondary prophylaxis with TMP/SMZ, and there were no instances of reactivation. Among the 12 untreated patients, the median survival from onset of symptoms was only 8 days (range, 315). Organ distribution and autopsy findings. Table 6 details the organ distribution of toxoplasmosis in all 35 patients and in the 17 who died and had a postmortem examination. The proportion of disseminated disease was somewhat higher at autopsy than in the entire patient sample (65% of patients vs. 43%), especially for multiple-organ disease (41% vs. 20%). The CNS and the lungs were by far the organs most frequently involved. Myocarditis was only diagnosed at autopsy and was relatively frequent (35% of patients); liver, kidney, and other rare sites were also only found at autopsy, whereas chorioretinitis (2 patients) was diagnosed during life, although funduscopic examination was done in only 14 patients during life and probably in few, if any, autopsies. Toxoplasma infection without disease. Six patients were classified as suffering from Toxoplasma infection without evidence of organ involvement. There were 5 men, with a median age of 35 years (range, 1351 years). Four had received a transplant from an HLA-matched sibling, and 2 from a mismatched family member. Three patients had developed prior AGVH disease (1 grade I, 1 grade II, and 1 grade IV), which was active in 2 patients at the time of diagnosis of toxoplasmosis. One Diagnosis, outcome, or treatment Diagnosis Definite Toxoplasma disease Premortem definitive diagnosis Premortem suspicion, confirmed at autopsy No premortem suspicion, autopsy finding Probable Toxoplasma disease Possible Toxoplasma diseasea Diagnostic methods used in definitive or probable cases (n p 29) Histology PCR analysis Blood CSF BAL Cytology and direct immunofluorescence from BAL Outcome Died from TOXO Time after onset of symptoms, median d (range) TOXO as a major cause of death TOXO as a contributing cause of death Autopsy performed Survived TOXO Died from other causes Time after onset of symptoms, median d (range) Other concomitant severe infection Cytomegalovirus pneumonia Cytomegalovirus infection (without disease) Invasive aspergillosis Adenovirus pneumonia Treatment Received >6 d of anti-TOXO therapy Pyrimethamine/sulfadiazine Pyrimethamine/sulfadiazine plus clindamycin Pyrimethamine plus clindamycin Trimethoprim/sulfamethoxazole Response to therapy No response Complete response Improvement 23 (66) 16 4 2 1 NOTE. Data are no. of patients or no. (%), unless otherwise indicated. BAL, bronchoalveolar lavage; TOXO, Toxoplasma disease. a Premortem suspicion, but patient survived or autopsy was not performed. b Five patients had >1 positive finding by PCR performed on blood and CSF (n p 3) or BAL (n p 2). patient was neutropenic, and the CD41 lymphocyte count was known in 2 and was low in both (40 and 100 cells/mL). All 6 patients had fever of unknown origin, with a median onset of 35 days after transplantation (range, 12167 days) plus >1 blood specimen positive by PCR, with no other signs or symptoms attributable to toxoplasmosis. Despite this lack of symptoms, a cranial CT scan was done for staging in 5 patients, an MRI in 2, and a fundus oculi exam in 5, all of which were normal. All 6 patients received drugs active on toxoplasmosis; 4 received TMP/SMZ, 1 received pyrimethamine/clindamycin, and 1 received pyrimethamine/sulfadiazine. All 6 responded with defervescence, and PCR performed on blood samples was negative. Three patients received secondary prophylaxis with TMP/SMZ, and 1 of the patients who received prophylaxis had a reactivation of the infection. This patient received a 28-day course of iv TMP/SMZ with complete response and later received prophylactic TMP/SMZ for 5.5 months, but 72 days later he developed Toxoplasma encephalitis, which, at the time of his last follow-up examination, had not responded to therapy. Three patients died from an unrelated cause 133203 days after initial symptoms, and 3 were alive at 284485 days. Three of these patients are reported elsewhere [10]. Discussion This study represents the largest series of patients suffering from toxoplasmosis after HSCT and the first to attempt to establish a set of definitions for this infection. Previous reports have included single patients or few cases from single institutions, with an apparent variability in the incidence between institutions. The seroprevalence for T. gondii varies greatly between and even within countries, ranging from !15% in some North American studies [11] and in pediatric wards to 50%80% of adult HSCT recipients in countries with high endemicity, such as France [2, 12, 13]. This varying seroprevalence is probably the main reason for the great variability in the frequency of diagnosed cases of toxoplasmosis after HSCT, which has been estimated to average 0.8% [14], with !0.4% in areas of low endemicity to 2%3% in those with high antibody prevalence. The disease, however, may be underdiagnosed, because around one-half of the patients reported in the literature were diagnosed at autopsy [24]. Additionally, in some recent reports of high-risk allogeneic HSCT from Europe and North America, toxoplasmosis occurred in 9 of 509 patients who underwent transplantation [1519]. In another study, 27 of 655 HSCT recipients developed a CNS infection after transplantation, of which 20 (74%) were considered as Toxoplasma encephalitis [20], although most (16 patients) were only possible cases by our criteria. Analysis of these data indicates that after allogeneic HSCT, toxoplasmosis may not be as rare as previously suggested. Results from this study largely agree with the data obtained from literature reviews published recently [24]. The disease occurs mainly in allogeneic transplant recipients, although cases in autologous transplant recipients have been described [3, 6, 7]. Around 95% of patients are seropositive before HSCT, indicating that reactivation of latent tissue cysts in previously infected individuals is the usual mechanism implicated, as has been demonstrated in patients with AIDS [1]. It is thus important to determine a patients serostatus before transplantation. The disease usually begins early after transplantation, with 95% of the cases occurring within the first 6 months after the procedure, although late-onset cases have been reported [9, 21, 22]. AGVH disease has been suggested as a possible predisposing factor for Toxoplasma disease [24], and in our study, 77% of patients developed moderate to severe AGVH disease or chronic GVH disease, a high incidence that suggests that prior GVH disease is a risk factor for toxoplasmosis after HSCT. Table 5. Survival of patients with Toxoplasma disease according to various clinical variables. The CNS is the main site of disease, but pneumonitis and myocarditis are also frequent findings, particularly when the diagnosis is made at autopsy. In fact, myocarditis, nephritis, and involvement of other deep organs are rarely found clinically but are frequent findings at autopsy [11]. Unlike toxoplasmosis in AIDS, which may have a subacute or insidious onset, this infection is rapidly fatal after HSCT. The prognosis of this infection has been considered to be very poor on the basis of the limited published data, with nearly 90% of patients dying from toxoplasmosis [24, 23]. This contrasts with the 70%80% response rates observed in patients with AIDS [24]. However, our results suggest that if appropriately treated, up to 60% of patients may show clinical-radiological improvement or even a complete response. In a review of toxoplasmosis in the nonAIDS-immunocompromised host, receiving appropriate antimicrobial therapy was the main factor for survival: 67% of 48 treated patients improved, whereas (99%) of 89 untreated patients died from toxoplasmosis [23]. Although these data may appear obvious, they highlight the importance for a high index of suspicion for toxoplasmosis in immunocompromised patients in order to perform the appropriate diagnostic tests and in order to start therapy as soon as possible. Analysis of our data also indicates that infections found late after transplantation may have a better prognosis than infections found early. The low rate of later reactivation is probably because immune recovery improves with time after HSCT, as opposed to patients with advanced AIDS whose immune functions continue to deteriorate after suffering toxoplasmosis until the advent of highly active antiretroviral therapy. Additional studies are warranted to determine the role of secondary prophylaxis and late primary prophylaxis in highrisk patients. Patients with a history of chronic toxoplasmosis or with prior infection, disease, or both after transplantation who develop recurrent GVH disease should probably receive secondary prophylaxis, as suggested by some case reports [21, 25] and by the example of 1 of our patients, especially when considering the relatively low toxicity of prophylaxis and the high morbidity and mortality of established disease. Because toxoplasmosis is so difficult to diagnose histologically in these patients, noninvasive diagnostic tests would be of utmost importance. Isolation of the parasite from blood or body fluids with use of rodent models or cell culture techniques is time consuming and expensive, and such services are available in few routine microbiology laboratories [1]. PCR techniques were developed for diagnosis of neonatal infections and for the noninvasive diagnosis of cerebral toxoplasmosis in patients with AIDS [8, 26]. These techniques are applicable in blood, CSF, and BAL, which are the usual samples that are available in HSCT recipients with this infection. However, PCR techniques are not standardized, and thus the results of published studies are difficult to interpret. In AIDS patients with brain lesions, PCR performed on blood and CSF has a sensitivity of 50%65% and a specificity of 95%100% for toxoplasmosis [26], but there are no data available in HSCT recipients other than case reports. In our study, 46% of the patients with Toxoplasma disease and all 6 with infection had >1 positive PCR result, thus confirming the widespread use of this diagnostic technology in clinical practice [5]. Particularly interesting were the patients with positive PCR tests from blood samples without evidence of disease [10]. These patients probably represent a transitional state between the local reactivation of tissue cysts into tachyzoites and the establishment of localized or disseminated tissue destruction by replicating tachyzoites favored by the intense cellular immunosuppression after transplantation or during GVH disease. This observation would be somehow similar to the early detection of cytomegalovirus infection by PCR or the pp65 antigenemia test [7]. Unlike the latter, however, the clinical significance of detecting Table 6. Distribution of Toxoplasma disease among 35 patients, as diagnosed with premortem and postmortem data, and distribution among 17 patients who died and had postmortem examinations performed. All patientsa Autopsy findings NOTE. Data are no. of patients or no. (%). a Includes organ involvement diagnosed or suspected in all 35 patients when findings of clinical, radiological, microbiological, and histological methods were considered together. b Seventeen patients had isolated CNS involvement. c Three patients had isolated pulmonary involvement. T. gondii DNA in blood is currently unknown, and it may well represent a common phenomenon in seropositive patients. The earlier onset of Toxoplasma infection (median day 35; range, 1351 days) than disease (median day 64; range, 4516 days) in our study suggests that infection may indeed precede disease in many patients. Additionally, because our study has shown a high morbidity and mortality of established disease, early (or preemptive) therapy could reduce the rate of overt clinical disease and improve the prognosis of toxoplasmosis. Thus research efforts to establish the role of PCR in this setting are clearly warranted. Determining the recipients pretransplantation serostatus for T. gondii is essential because the risk of toxoplasmosis in seronegative recipients appears to be very low. In summary, we propose definitions for toxoplasmosis that occurs after HSCT that may be applicable to other highly immunocompromised hosts. Toxoplasmosis is an early complication after transplantation that is life threatening, and efforts to improve its early diagnosis and treatment, as well as the use of prophylaxis in seropositive recipients after a high-risk allogeneic HSCT, are justified. Working Party Members


This is a preview of a remote PDF: https://cid.oxfordjournals.org/content/31/5/1188.full.pdf

R. Martino, J. Maertens, S. Bretagne, M. Rovira, E. Deconinck, A. J. Ullmann, T. Held, C. Cordonnier, for the European Group for Blood and Marrow Transplantation Infectious Diseases Working Party. Toxoplasmosis after Hematopoietic Stem Cell Transplantation, Clinical Infectious Diseases, 2000, 1188-1194, DOI: 10.1086/317471