Immunocompromised children or adults patients are a population at high risk of infection, thus constituting a major cause of morbidity and mortality. They are 2 times more common in oncology and hematology patients (9-12%) than in not cancerous patients (6-7%).The incidence of hematology nosocomial infections (NI) is around 20%. Although data on NI are available, those about healthcare associated infections (HAI) are poorly studied.


Our aim was to define the incidence rate and risk factors of HAI of hematology patients over a period of 7 months from February to August 2012 in preliminary study to create a local register for HAI and establish strategies for hygiene and preventive policy of our unit.

Patients & methods

We took forward all aged over 18 patients hospitalized and monitored for hematologic malignancies in the adult and transplant unit at hematology and oncology center at the 20th August 1953 hospital in Casablanca. Informations were collected and validated by dealing doctors using a farm and data analysis was done by biostatisticians team using Epi Info software. Rates of HAI were calculated for patients- 1000 hospitalization day.


Among this population of 237 patients, we diagnosed 183 HAI in 94 patients, witch 62.8% were followed for acute myeloid leukemia (AML). Incidence rate of HAI was 42.1% with confidence interval (CI) of 95% at [37.4-46.9%] against 29.9% for NI and incidence density of HAI by 1000 hospitalization day was 31.6 versus 24.6 for NI. Risk factors have implicated significantly neutropenia 87.4% (p<0.001), chemotherapy 83% (p<0.001) and temporary central catheter 60% (p<0.001). HAI were microbiologically documented in 41% witch bacteremia in 28%. Pneumonia was noted in the third of cases. Infection was of unknown origin in 25%. Germs isolated were mainly bacteria (93%), specially Staphylococci (32%) and Enterobacteriaceae (29%). Majority of our isolates have one or many antibiotic resistance.


Incidence rate of HAI was high in our study relative to NI, so they should be preferred because they’re covering a broader definition and therefore can be used as a quality indicator for prevention programs. Our study presented a preliminary step in the creation of a HAI register and the initiation of a prevention policy adjusted on ecological profile of our unit.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.