Nosocomial infection in patients admitted to an intensive care unit at Al-Shifa Hospital in the Gaza Strip, occupied Palestinian territory: a retrospective assessment

Abstract

<h3>Background</h3> Admissions to intensive care units (ICUs) are generally associated with an increased risk of nosocomial infections. No data have been reported for the incidence and determinants of these infections in patients admitted to ICUs in the Gaza Strip, occupied Palestinian territory. Therefore, the frequency of nosocomial infections in patients admitted to the largest ICU in the Gaza Strip at Al-Shifa Hospital was retrospectively assessed. <h3>Methods</h3> Systematic sampling was used to select the medical file of every tenth patient admitted to the ICU at Al-Shifa Hospital from Jan 1, 2006, to June 30, 2009, from the hospital's archive. All available files were included irrespective of the patient's age. Incomplete files and those of patients admitted from another ICU or who developed febrile illness before or on admission to the ICU were excluded. Information about patients was collated and entered into SPSS (version 19.0). The frequency distribution of the variables was analysed and the χ² test was used to assess differences between patients including in the duration of stay, illnesses, and the outcome of care. The study was approved by the administration of Al-Shifa Hospital. <h3>Findings</h3> Of 206 patients (mean age 36 years [SD 25]) admitted to the ICU, whose medical files were selected, 128 (62%) were men. The mean period of stay at the ICU was 6·2 days (8·7). 55 (27%) patients stayed in the ICU for up to 1 day; 43 (21%) for 2–3 days; 72 (35%) for 4–7 days; 19 (9%) for 8–14 days; and the remaining 17 (8%) for more than 2 weeks. 50 (24%) patients had nosocomial infections—28 (56%) pneumonia, seven (14%) wound infections, seven (14%) multiple causes, three (6%) bacteraemia, three (6%) undetermined, and two (4%) infections related to the urinary tract. The causative organisms of the nosocomial infections were pseudomonas in 15 (30%) patients, klebsiella in eight (16%), acinetobacter in six (12%), <i>Escherichia coli</i> in five (10%), staphylococci in two (4%), <i>Enterococcus faecalis</i> in one (2%), and more than one microorganism in six (12%); the causative microorganisms were not identified in seven (14%) patients. 29 (81%) of 36 patients who stayed in the ICU for more than 1 week had nosocomial infections compared with 15 (21%) of 72 who stayed from 4 days to 1 week, and six (6%) of 98 who stayed for up to 3 days (p<0·0001 for overall difference between groups). The risk of nosocomial infections was increased in some groups: 13 (19%) of 68 patients with an internal medical diagnosis, 12 (19%) of 64 admitted for post-surgical care, ten (20%) of 50 admitted for post-neurosurgical care, five (38%) of 13 admitted after being resuscitated for cardiopulmonary arrest, and ten (91%) of 11 with multiple trauma developed nosocomial infections (p<0·0001). The risk of death was higher in the 50 patients who acquired nosocomial infections than in the 156 who did not (42% <i>vs</i> 21%; attributable risk percentage 50%, 95% CI 17–83; p=0·013). <h3>Interpretation</h3> The high frequency and attributable risk percentage of mortality of nosocomial infections in the ICU at Al-Shifa Hospital indicate the need for additional preventive and control measures and a surveillance system for these infections in the ICUs of all hospitals in the Gaza Strip. <h3>Funding</h3> None.