An approach to the design and implementation of clinical trials of empirical antibiotic therapy in febrile and neutropenic cancer patients

The results of many clinical trials on empirical therapy in febrile, neutropenic cancer patients cannot be readily transferred to the clinical practice, because the methodology is often flawed and definitions, study endpoints and eligibility criteria differ from trial to trial. This article critically reviews some issues related to the design and implementation of randomised clinical trials of empirical antibiotic therapy in cancer patients. Within the definition of phase III clinical trials, two approaches co-exist, based on the trial's specific aims: the "explanatory" approach and the "pragmatic" approach. The usual "explicit" aim of clinical trials of empirical therapy in febrile, neutropenic patients has been that of comparing the "efficacy" of two regimens. However, this term has been more often used with reference to the antibacterial activity of the regimen under study (explanatory aim) than to indicate the practical benefits it draws to the overall patient population treated for fever and neutropenia (pragmatic aim). These two meanings are often taken as perfectly interchangeable, while, conversely, they are completely distinct (though not independent) treatment effects. Most trials conducted in this patient population in recent years are explanatory trials, though not explicitly so, but their results have been widely applied to clinical practice, as they were pragmatic trials. In an explanatory trial the appropriate endpoint is success or failure (defined by clinical and laboratory data) among those patients affected with the specific infection for which the study drug is being given, while in pragmatic trials survival is probably the more appropriate outcome variable, since they are designed to assess the practical benefits that the overall population of febrile and neutropenic patients can obtain from the new empirical treatment. Unfortunately, survival is not a practical study endpoint for the difficulty in assessing the cause of death in this patient population and, especially, for the need for very large sample sizes, which might render the implementation problematic even for large, multicentre groups. Both types of trials need an intention to treat analysis, but this is especially crucial for pragmatic trials, which should not differentiate those cases in which success was obtained through multiple treatment modifications from those who did not require any treatment change. Obviously, this implies that no conclusion should be drawn about the antibacterial activity of the study drugs and that the number of treatment modifications should be taken into account in the interpretation of the results, especially for quality of life and cost evaluations. Information related to fever and signs of infection, age, underlying disease, neutropenia and concomitant administration of other antibiotics are crucial entry criteria that need to be clearly discussed and defined. Finally, the evaluation of toxicity is problematic in this patient population, due to the existence of a number of toxigenic factors, including the underlying disease, the type of infectious complication, the administration of chemotherapy and radiotherapy and the use of parental nutrition. All these effects tend to overlap, thus impairing the investigator's ability to detect specific drug-related side-effects.

Prophylaxis of streptococcal bacteraemia with oral penicillin V in children undergoing bone marrow transplantation

The work described aimed to evaluate the incidence of streptococcal bacteraemia in children undergoing bone marrow transplantation and receiving prophylaxis with penicillin V. From January 1991 to December 1993 oral penicillin V was administered as prophylaxis for streptococcal bacteraemia to patients undergoing bone marrow transplantation at G. Gaslini Children's Hospital, Genoa, Italy. The data were compared with those from a similar population receiving bone marrow transplantation from September 1984 to July 1990 and not receiving this kind of prophylaxis. Streptococcal bacteraemia was diagnosed in 7/17 (41%) episodes of bacteraemia observed in the period January 1991 to December 1993, while it accounted for 71% of all bacteraemias in the period from September 1984 to July 1990 and was especially frequent from January 1988 to July 1990, comprising 13/15 (87%) of observed bacteremias. The decrease of this complication observed after the beginning of the prophylaxis programme was statistically significant. Oral penicillin V is effective as prophylaxis of streptococcal bacteraemias in children receiving bone marrow transplantation in a centre with a high incidence of this complication.

[Fungal infections in pediatric oncology]

The clinical charts of cancer patients with documented fungal infections hospitalized at G. Gaslini Children's Hospital, Italy, from 1980 to 1990 were reviewed. Thirty-seven episodes developing in 37 patients were identified, based on microbiological and/or histological documentation. Patients' age ranged from 3 months to 18 years (median 7 years). Twenty patients were treated for hematological malignancy and 17 had solid tumor. Seven patients (3 with leukemia and 4 with solid tumours), developed mycosis after bone marrow transplantation procedure. A history of neutropenia in the month preceding the documentation of fungal infection was present in 76% of cases (28 of 37). However, only 16 of 28 (55%) of these patients were still neutropenic at time of diagnosis. In 40% of the cases the fungal infection developed as primary infection not preceded by any febrile and/or infectious episode. Fungemias without evident organ localization accounted for the 40% of episodes with a mortality rate of 20%. The other 22 cases (60%) were classified as invasive mycoses; 9 of these patients died (41%). Mortality was higher among patients with mold infection (5 of 7, 72%), than in those with yeast infection (7 of 29.24%). Molds infections and invasive mycoses were virtually absent in the first part of our period of observation (1980-84), but emerged in the second period (1985-90) when also the incidence rate of fungal disease increased (from 2.67/10,000 person/day to 5.93), probably in relation with extensive construction works and with the implementation of a bone marrow transplantation program.

Sex differences in the management of patients hospitalized with ischemic cerebrovascular disease

BACKGROUND AND PURPOSE

Previous studies suggest that the management of coronary artery disease differs for women compared with men. We examined this issue for ischemic cerebrovascular disease.

METHODS

We reviewed the use of angiography and carotid endarterectomy among patients discharged from Connecticut hospitals during 6 years over the past decade. Crude and age-adjusted rates of angiography and endarterectomy were determined for each sex.

RESULTS

Among 22,582 female and 19,729 male patients discharged, the rate of cerebral angiography was 11.8% for men and 7.2% for women; the age-adjusted odds ratio was 0.77 (95% confidence interval [CI], 0.72 to 0.82). The rate of endarterectomy was 10.6% for men and 5.7% for women; the age-adjusted odds ratio was 0.67 (95% CI, 0.62 to 0.72). The distribution of cerebrovascular disease type differed by sex, however, with carotid artery disease representing a larger proportion of men (12.2% [2415/19,729]) than women (6.9% [1554/22,582]) (chi 2 = 355.8, P < .0001). When restricted to this diagnosis, no sex differences exist (odds ratio for angiography, 1.00 [95% CI, 0.87 to 1.14] and for endarterectomy, 0.93 [95% CI, 0.81 to 1.07]).

CONCLUSIONS

Overall, women hospitalized for ischemic cerebrovascular disease undergo fewer angiograms and are less likely to have carotid endarterectomy than men. These differences are not found when analysis is restricted to subjects with carotid disease and suggest that part of the difference in management may be due to biological differences between men and women.

Primary prevention of cryptococcal meningitis by fluconazole in HIV-infected patients

To evaluate whether oral fluconazole reduces the risk of a first episode of cryptococcal meningitis in HIV-infected patients, we conducted a case-control study of patients cared for in a university teaching hospital and two urban HIV-outpatient clinics. Cases consisted of HIV-infected patients with CD4 cell counts less than 250/microL who developed a first episode of cryptococcal meningitis between July 1, 1990, and June 30, 1993. For each case (n = 18), 4 control subjects were chosen from HIV-infected patients (CD4 count < 250/microL) whose cerebrospinal fluid was negative for cryptococcal antigen and culture, and who were matched by age, sex, and time of lumbar puncture. There were no significant differences between cases and controls in age, sex, insurance status, mean CD4 count, history of oral candidosis, presence of a previous AIDS-defining illness, the number of visits to the HIV-outpatient clinic, or use of antiretroviral therapy. In the 6 months before lumbar puncture, 2 of 18 cases (11%) and 26 of 72 controls (36%) were exposed to fluconazole, a finding that gives a matched odds ratio (adjusted for race, route of HIV infection, and CD4 count) of 0.08 (95% CI 0.01-0.84; p = 0.035) and indicates a 92% protective efficacy. We conclude that fluconazole reduces the risk of a first episode of cryptococcal meningitis in HIV-infected patients with a CD4 count less than 250/microL. Although the optimum dose and duration of fluconazole could not be determined, our results suggest that less than daily use was effective in the prevention of cryptococcal meningitis.

Changing pattern of pathogens causing broviac catheter-related bacteraemias in children with cancer

The incidence of pathogens causing catheter-related bacteraemias in children undergoing antineoplastic chemotherapy with or without bone marrow transplantation at G. Gaslini Children's Hospital, Genoa, Italy, was analysed by comparing data from a retrospective study (1985-1988) with that obtained from a prospective one (1989-1992). In both periods catheter-related bacteraemias one (1989-1992). In both periods catheter-related bacteraemias were more frequent in non-neutropenic than in neutropenic patients. Among catheter-unrelated bacteraemias the pattern of infecting pathogens remained unchanged between the study periods, with Gram-positive bacteria remaining the predominant pathogens. Conversely, among catheter-related bacteraemias, the incidence of Gram-negative bacilli increased significantly from 3 to 38%, and that of Gram-positive bacteria fell from 63 to 32% (P = 0.001, chi 2 test for heterogeneity.

Piperacillin-tazobactam plus amikacin versus ceftazidime plus amikacin as empiric therapy for fever in granulocytopenic patients with cancer. The International Antimicrobial Therapy Cooperative Group of the European Organization for Research and Treatment of Cancer

Gram-positive bacteria have become the predominant infecting organisms in granulocytopenic cancer patients. Empiric antibiotic regimens used in febrile neutropenic patients often include an extended-spectrum cephalosporin, but the response to therapy in gram-positive coccal bacteremia has been unsatisfactory. Thus, new antibiotics with better activity against gram-positive bacteria should be tested. The objective of this prospective randomized controlled study was to evaluate and compare the efficacy and tolerance of piperacillintazobactam plus amikacin with that of ceftazidime plus amikacin, the standard regimen of the International Antimicrobial Therapy Cooperative Group of the European Organization for Research and Treatment of Cancer, in the empiric treatment of febrile granulocytopenic cancer patients. A total of 858 episodes were eligible for this study, and 706 episodes were assessable for efficacy. The antibiotic treatment was successful in 210 (61%) of 342 episodes in the piperacillin-tazobactam-amikacin group compared with 196 (54%) of 364 episodes treated with ceftazidime plus amikacin (P = 0.05). The time to defervescence was significantly shorter (P = 0.01) and the time to failure was significantly longer (P = 0.02) in the piperacillin-tazobactam-amikacin group. A significant difference in response to bacteremic infections between the two patient groups was found: piperacillin-tazobactam plus amikacin was successful in 40 of 80 episodes (50%), and ceftazidime plus amikacin was successful in 35 of 101 episodes (35%) (P = 0.05). A multivariate analysis showed that the probability of failure was significantly greater with ceftazidime plus amikacin than with piperacillin-tazobactam plus amikacin (P = 0.02). This trial suggests that piperacillin-tazobactam plus amikacin is more effective than ceftazidime plus amikacin for the empiric treatment of fever and bacteremia in granulocytopenic cancer patients. Although cutaneous reaction was more frequently associated with piperacillin-tazobactam plus amikacin than with ceftazidime-amikacin, this unwanted effect was relatively mild and its incidence was comparable to that of other penicillin compounds.

Factors associated with bacteraemia in febrile, granulocytopenic cancer patients. The International Antimicrobial Therapy Cooperative Group (IATCG) of the European Organization for Research and Treatment of Cancer (EORTC)

The objective of this investigation was to determine factors predictive of bacteraemia at presentation in febrile, granulocytopenic cancer patients in order to estimate the probability of bacteraemia in each patient, and to compare factors associated with a diagnosis of gram-positive or gram-negative bacteraemia. Retrospective analysis of two sets of data (derivation and validation sets) randomly obtained from a large prospective study was conducted in a multicentre study of febrile, granulocytopenic cancer patients admitted for empiric antibacterial therapy. Within the derivation set, prognostic factors (clinical and laboratory data) likely to be associated with a generic diagnosis of bacteraemia and with a specific diagnosis of gram-positive or gram-negative bacteraemia were analysed by means of three backward, stepwise, logistic regression analyses. The predictive probability of bacteraemia was calculated using the logistic equation. The discriminating ability of the model in predicting bacteraemia was evaluated in the derivation and validation sets using receiver-operating characteristic curves. The predictive probability of gram-positive or gram-negative bacteraemia was not calculated. In the derivation set, 157 of 558 episodes (28%) were microbiologically documented bacteraemias. Predicting factors were antifungal prophylaxis, duration of granulocytopenia before fever, platelet count, highest fever, shock and presence and location of initial signs of infection. The variables institution, antibacterial prophylaxis and underlying disease showed borderline associations with bacteraemia. Shock was associated with gram-negative bacteraemia, while signs of infection at catheter site were predictive of gram-positive bacteraemia. Quinolone prophylaxis was negatively associated with gram-negative bacteraemia. When tested in the validation set, the model was poorly predictive, although a small subgroup of episodes (representing only 16% of the total sample size) with low risk of bacteraemia was identified. Factors predictive of bacteraemia can be identified, with discrimination between gram-positive and gram-negative aetiology. Further studies are warranted in order to improve the discriminant ability of the model.

CT of invasive pulmonary aspergillosis in children with cancer

In treating cases of malignancy, the use of chemotherapy carries a high risk of lower respiratory tract infections, especially fungal pneumonopathy. This complication is a major cause of mortality and is often difficult to diagnose because of non-specific clinical or radiological changes, but the early recognition of invasive fungal disease is imperative. CT is an important non-invasive method for the detection and evaluation of opportunistic fungal infections. In these patients an improved survival rate can be achieved when early detection by CT leads to the prompt institution of high-dose antifungal therapy. We illustrate the spectrum of CT findings of invasive pulmonary aspergillosis encountered in children with cancer. These patients had previously been treated with high-dose chemotherapy with or without bone marrow rescue, and underwent radiological examinations because of clinical evidence of pneumonopathy. Representative cases demonstrate the clinical applications of CT in the evaluation and management of invasive fungal disease.

Hepatic and splenic mycosis in children with acute leukemia

Three cases of hepatic and/or splenic mycosis in children with acute leukemia are reported. Patients presented with fever not responding to broad spectrum antibiotics during or after prolonged and profound neutropenia. Noteworthy, in 1 case no abnormality in liver function tests was detectable at diagnosis. CT scan showed focal hepatic lesions in all patients. In 1 patient ultrasounds failed to detect hepatic lesions and showed only splenic lesions, while CT scan detected hepatic and not splenic lesions. Definitive diagnosis was performed by open liver biopsy and histological demonstration of fungal invasion. Cultures were negative in all cases. All patients survived with prolonged amphotericin B treatment (from 36 to 40 mg/kg), combined with flucytosine in two cases. In conclusion, this clinical entity should be taken into account by physicians taking care of neutropenic patients, even in the absence of abnormalities in liver function. CT seems to be the method of choice for diagnosis, although ultrasounds might be useful for detecting concomitant splenic lesions.

Myeloablative therapy and unpurged autologous bone marrow transplantation for poor-prognosis neuroblastoma: report of 34 cases

From October 1984 to November 1987, 34 patients aged from 1 year 1 month to 7 years 7 months with resistant or relapsed neuroblastoma (NB) (group 1, 10 patients), unselected disseminated NB (group 2, 14 patients), or selected disseminated NB (group 3, 10 patients) received myeloablative therapy (MAT) followed by unpurged autologous bone marrow transplantation (ABMT) at the end of an intensive protocol, which included high-dose chemotherapy and surgery to the primary tumor. Median time from diagnosis to MAT and ABMT was 6 months (5 months from last relapse to MAT and ABMT in the relapsed patients). The MAT regimen included vincristine, fractionated total body irradiation (TBI), and melphalan. Seventeen patients were grafted in complete remission (CR), five in very good partial remission (VGPR), 10 in partial remission (PR), and two in progressive disease (PD). The acute toxic death rate was 2.9%. The overall progression-free survival was 29%. The median progression-free survival was 20 months for the 17 patients grafted in CR, 6 months for the five patients grafted in VGPR, and 12 months for the 10 patients grafted in PR.

Serum concentrations and safety of single daily dosing of amikacin in children undergoing bone marrow transplantation

The serum concentrations and safety of single daily dosing of amikacin were studied in 16 episodes of fever and granulocytopenia in children undergoing bone marrow transplantation. Amikacin (20 mg/kg) was administered as a 20 min iv infusion once daily for 7 to 19 days; all patients received concomitant therapy with ceftazidime. Peak amikacin concentrations measured immediately after infusion on day 1 and 4 of therapy averaged 72.29 +/- 11.6 mg/l and 74.02 +/- 19.29 mg/l respectively. A slight but statistically significant increase 30 minute post-infusion concentrations was observed in most patients during therapy. Serum amikacin were less than 3 mg/l in all trough samples and within 6 h post-infusion in most patients. A significant increase in serum creatinine was observed in one patient, who was also receiving cyclosporin A. Auditory function was evaluated in 10 patients and showed no changes. Although other drugs were added to cover Gram-positive or fungal pathogens, all patients survived. These data combined with the recent experience of single daily dose amikacin in adults supports further evaluation of novel dosage regimens of aminoglycosides in children.

Ceftazidime plus amikacin versus ceftazidime plus vancomycin as empiric therapy in febrile neutropenic children with cancer

Two antibiotic regimens, ceftazidime plus amikacin and ceftazidime plus vancomycin, were compared in a prospective, randomized clinical trial as empiric therapy in febrile granulocytopenic children with cancer. The rate of response was similar in the two groups (66% vs. 77%). The prevalence of secondary gram-negative bacteremia was higher--but not significantly higher--in the group receiving vancomycin. Adverse reactions also occurred more often in the latter group (35% vs. 4%). Mortality did not differ significantly in the two groups. Adjustment for independent predictors of response to treatment by means of multivariate analysis confirmed the lack of any remarkable difference between the responses to the two regimens. We conclude that the use of vancomycin instead of amikacin in combination with ceftazidime does not significantly improve the outcome of treatment of fever and infection in granulocytopenic children with cancer and that the use of vancomycin is associated with an increased frequency of both secondary infections due to gram-negative bacteria and adverse reactions.

Prospective randomized evaluation of ciprofloxacin versus piperacillin plus amikacin for empiric antibiotic therapy of febrile granulocytopenic cancer patients with lymphomas and solid tumors. The European Organization for Research on Treatment of Cancer International Antimicrobial Therapy Cooperative Group

Empiric therapy for febrile granulocytopenic patients is mandatory, but whether monotherapy is a safe alternative and whether fluoroquinolones are useful agents for this indication are still controversial issues. The use of monotherapy with intravenous ciprofloxacin (200 to 300 mg every 12 h) was evaluated against combined therapy with piperacillin plus amikacin in febrile granulocytopenic patients with solid tumor or lymphoma. The study was discontinued prematurely because patients treated with ciprofloxacin had a significantly lower overall success rate than patients treated with piperacillin plus amikacin (31 of 48 patients [65%] versus 48 of 53 patients [91%], P = 0.002). Patients with gram-positive coccal bacteremia had a particularly poor outcome: therapy failed for six of eight patients (75%) treated with ciprofloxacin, while therapy failed for none of four patients treated with piperacillin plus amikacin. Death from primary infection during initially randomized protocol therapy occurred in 7 of 48 patients (14.5%) treated with ciprofloxacin and in 3 of 53 (6%) treated with piperacillin plus amikacin. This study does not support the use of this dose of intravenous ciprofloxacin as empiric monotherapy for fever in granulocytopenic patients.

Infections in the compromised child

Children receiving chemotherapy for malignant diseases show different patterns of infection depending on their underlying disease and its therapy. Granulocytopenia carries the risk of bacterial infection, and also, if prolonged, of fungal infection. Impairment of cell-mediated immunity predisposes to infections with Pneumocystis carinii and is thought to be responsible for severe primary infections with varicella and measles, as well as the severe cytomegalovirus infections seen after allogeneic bone marrow transplantation. Absence or impairment of splenic function predisposes to overwhelming septicaemia with encapsulated organisms, while defects in the normal mechanical barriers to infection provide routes for bacterial and fungal invasion. Despite the lack of physical signs of a normal inflammatory response, clinical evaluation may be critical to the localization of infection in the immunocompromised child. Blood culture and biopsy remain pivotal investigations in the achievement of a microbiological diagnosis. Empirical treatment with a combination of antibiotics has been shown in comparative studies to be effective in initial management of the febrile neutropenic patient: continuing studies are evaluating the role of monotherapy and of different antibiotic combinations, particularly in the light of changing patterns of bacterial infections. Empirical antifungal therapy has been shown to be necessary for persistent or recurrent fever, particularly as persistent fungal infection may compromise the outcome of continuing cytotoxic therapy. Continuing uncertainties over many aspects of management of the infected immunocompromised child provide scope for clinical trials in parallel with trials evaluating new anticancer regimens. The use of new diagnostic methods, the role of prophylaxis, the most appropriate empirical regimen, the evaluation of new antimicrobial agents, all require careful evaluation for efficacy and safety. Perhaps the greatest dilemma of all is how far results from trials in adults can be extrapolated to paediatric practice.

Fluconazole in the treatment of candidiasis in immunocompromised children

Fluconazole, a new triazole derivative, was evaluated in a pilot study of 34 episodes of candidiasis in 24 children. All the patients had predisposing conditions, such as human immunodeficiency virus infection, cancer, organ or bone marrow transplantation, neonatal age and malnutrition, and obstructive uropathy. The drug was administered at 6 mg/kg (body weight) once daily either orally or intravenously. Two patients with fungemia due to Candida parapsilosis required an increased dosage of 12 mg/kg. Clinical and microbiological success was achieved in 30 of 34 cases (88%). Drug-related transaminase increases were observed in two cases (6%). Fluconazole may represent an effective alternative to amphotericin B in the treatment of candidiasis in children. Comparative trials are necessary to assess optimal dosages and efficacy.

Listeria monocytogenes brain abscesses in a girl with acute lymphoblastic leukaemia after late central nervous system relapse

A case of Listeria monocytogenes bacteraemia and meningitis with intracerebral abscesses in a girl with acute lymphoblastic leukaemia in relapse is reported. The clinical features included subacute onset with fever and marked irritability followed by seizures, meningism and confusion. The pathogen was isolated from blood and cerebrospinal fluid. Computerised tomography of the brain showed two intracerebral parenchymal localisations, in the left frontal lobe and in the right occipital lobe, respectively. The patient survived this severe infection without neurological sequelae. 2 months later she underwent allogeneic bone marrow transplantation without major complications. This case report should alert pediatric oncologists about the possible occurrence of severe intracerebral listerial infections in the immunocompromised child and suggests that this infection can be treated successfully and should not necessarily preclude continuation of antineoplastic treatments.

Infection with Fusarium species in two children with neuroblastoma

Two cases of Fusarium infection in children with neuroblastoma are reported. One of the patients had an overwhelming infection and the diagnosis was based on isolation of Fusarium moniliforme from blood and skin biopsy, and histological findings. The second patient developed chronic polyarthritis and Fusarium solani was cultured from synovial fluid samples taken from two different joints four months apart. No histological documentation of infection was obtained. The response to antifungal therapy was unfavourable. Both patients died, but in the second case the relationship between fungal infection and death was not established.

[Liver cirrhosis in childhood. Considerations on 22 cases with different etiology]

Although rather uncommon and multifactorial in etiology, liver cirrhosis is a severe and often rapidly fatal disease in pediatrics. In our institution, during the last 15 years, 22 children with liver cirrhosis have been followed. The underlying predisposing condition was HBV infection (8 cases), CMV perinatal infection (2 cases), Wilson's disease (4 cases), chronic cholestasis (2 cases) and alcohol abuse (2 cases); in 4 cases no predisposing condition was evident. In all cases the histological examination of the liver was the diagnostic cornerstone. The mean age at diagnosis was 6 years and 8 months, with an early onset especially in the posthepatitis cirrhosis. In 10 out of 22 patients, cirrhosis was not preceded by an history of chronic liver disease. Poor subjective symptomatology was present in 13 of the cases, hepatomegaly in all, splenomegaly in 18 cases, signs of hepatic failure in 13 cases. In all patients various impairments of hepatocellular synthesis were detectable, especially during the period preceding the development of hepatic insufficiency. The mean time to cirrhosis was 5 years. The average duration of the follow up was 3 years and 4 months: during the follow up 6 patients improved, 5 patients showed no clinical or functional modifications of their hepatic disease, 3 patients worsened and 8 died. In order to perform suitable treatment of liver cirrhosis the need of early diagnosis and etiological definition should be emphasized.

Role of Broviac catheters in infections in children with cancer

In a 3-year period 157 single lumen Broviac catheters were inserted in 145 children with various neoplastic diseases. The overall duration of the catheter courses was 30,533 days (median, 171; range, 2 to 647). Sixty-five percent of the catheter courses (102 of 157) were complicated by at least 1 febrile episode, for a total of 157 episodes. According to European Organization for Research on Treatment of Cancer criteria, 79 febrile episodes (50%) were classified as microbiologically documented infections, 57 (36%) with and 22 (14%) without septicemia; 31 (20%) as clinically documented infections; and 47 (30%) as possible infections. Of the 79 microbiologically documented infections 21 were catheter-related infections (CRI), 32 were catheter-unrelated infections and 26 were infections of unknown source. Only 48% of CRI occurred during neutropenia (less than 1000 neutrophils/mm3), compared with 88% of catheter-unrelated infections and 96% of infections of unknown origin (P = 0.00007). Gram-positive microorganisms (56% staphylococci) accounted for 78% of all isolates in CRI, 47% in catheter-unrelated infections and 43% in infections of unknown origins (P = 0.03). Fungi represented 12% of all isolates. Clinical and microbiologic resolution without removal of the catheter was achieved in 12 of 21 CRI (57%) and no patient died from a CRI. This study indicates that over 3 of 4 of CRI are caused by Gram-positive bacteria, occur in neutropenic and non-neutropenic patients (approximately 50%) and can be successfully treated without removing the catheter.