Nosocomial infections in medical-surgical intensive care units in Argentina: attributable mortality and length of stay

BACKGROUND: Nosocomial infections are an important public health problem in many developing countries, particularly in the intensive care unit (ICU). Limited data exists on the incidence and burden of nosocomial infection in the ICU in Argentina. METHODS: We performed baseline prospective nosocomial infection surveillance of all patients for 6 months in 3 medical-surgical ICUs (MS-ICUs) in Argentina (2 months in each ICU). Nosocomial infections were identified using the Centers for Disease Control and Prevention National Nosocomial Infections Surveillance definitions. Overall and site-specific nosocomial infection rates, attributable mortality, and excess length of hospital stay were calculated. RESULTS: The overall nosocomial infection rate was 27% and 90 per 1000 patient-days. The most common site of infection was catheter-related bloodstream infection (32%), followed by ventilator-associated pneumonia (25%), and catheter-associated urinary tract infection (23%). The rate of central catheter-associated bloodstream infection in the MS-ICU was 44.61 per 1000 device-days, with an attributable mortality of 25%, and 12 attributable extra days of hospital stay. The urinary catheter-associated urinary tract infection rate in the MS-ICU was 22.55 per 1000 urinary catheter-days, with an attributable mortality of 5%, and 5 excess extra days of hospital stay. The ventilator-associated pneumonia rate in the MS-ICU was 50.87 per 1000 ventilator-days with an attributable mortality of 35%, and 10 attributable extra days of hospitalization. CONCLUSION: Our study finds high rates of nosocomial infections in ICUs in Argentina, associated with a considerable attributable mortality and excess length of stay. Ongoing targeted surveillance and implementation of infection control strategies is necessary to control this growing problem.     

Infección nosocomial en las unidades de cuidados intensivos

Nosocomial infections (NI) still have a high incidence in intensive care units (ICUs), and are becoming one of the most important problems in these units. It is well known that these infections are a major cause of morbidity and mortality in critically ill patients, and are associated with increases in the length of stay and excessive hospital costs. Based on the data from the ENVIN-UCI study, the rates and aetiology of the main nosocomial infections have been described, and include ventilator-associated pneumonia, urinary tract infection, and both primary and catheter related bloodstream infections, as well as the incidence of multidrug-resistant bacteria. A literature review on the impact of different nosocomial infections in critically ill patients is also presented. Infection control programs such as zero bacteraemia and pneumonia have been also analysed, and show a significant decrease in NI rates in ICUs.     

Nosocomial infections in the intensive care units at a university hospital in a developing country: comparison with National Nosocomial Infections Surveillance intensive care unit rates

OBJECTIVE: As a measure of the quality of care provided to patients in the intensive care unit, comparison of nosocomial infection rates with those of the National Nosocomial Infection surveillance was completed during a 3-year observation period. DESIGN: The study design was a prospective study during 3 years between 1993 and 1995. During that period, patients at the medical/surgical and neurosurgical intensive care units and the high-risk nursery were surveyed for nosocomial infections. Device use, bloodstream infection, urinary tract infection, and ventilator-associated pneumonia nosocomial infection rates were calculated and compared with the National Nosocomial Infection Surveillance published rates for the same period. SETTING: The study setting was the medical/surgical intensive care unit, the neurosurgical intensive care unit, and the high-risk nursery at the Jordan University Hospital. RESULTS: Overall infection rates were 17.2 per 100 patients in the medical/surgical intensive care unit, 14.2 to 18.5 per 100 patients in the neurosurgical intensive care unit, and 13.4 to 73.5 per 100 patients in the high-risk nursery. When compared with the weight of the infants, these rates were 61.9 to 94 per 100 in infants weighing <1500 g, 26 to 30.8 per 100 patients in infants weighing >1500 g to 2500 g, and 11.7 to 14.4 per 100 in infants weighing >2500 g. Whereas device use was moderate, bloodstream infection and ventilator-associated pneumonia rates were >90th percentile for National Nosocomial Infection Surveillance in the high-risk nursery, and urinary tract infection was >90th percentile in the medical/surgical and neurosurgical intensive care units. Nosocomial infections at the intensive care units in developing countries need further investigation and control.     

Nosocomial infection and fatality in medical and surgical intensive care unit patients

We prospectively studied 526 patients admitted to the medical intensive care unit (MICU) and 799 patients admitted to the surgical intensive care unit (SICU) at a municipal hospital over a 20-month period. Rates of nosocomial infection were higher in the SICU patients (31% vs 24%). The SICU patients had more urinary tract infections, bacteremias, and wound infections, and the MICU patients were older, had higher acute physiology scores on admission and were more often admitted with shock or coma. The SICU patients were more likely to have received prior antibiotic therapy and had significantly higher numbers of endotracheal tubes, arterial lines, central venous lines, and indwelling bladder catheters. Of the 23 variables univariately associated with nosocomial infection, only five remained significant after entry into step-wise regression models. The MICU patients had a higher fatality rate in the MICU than did the SICU patients (18% vs 10%), but the relative risk of a death following nosocomial infection was 3.5 for both groups. Thirty variables were significantly associated with hospital fatality; nine remained significant after analysis by stepwise logistic regression.     

Nosocomial infection in a neonatal intensive care unit: a prospective study in Taiwan

BACKGROUND: We performed a prospective analysis to determine the prevalence of nosocomial infection and associated risk factors in our neonatal intensive care unit (NICU). METHODS: Data were collected prospectively on underlying diagnoses, therapeutic interventions/treatments, infections, and outcomes at 9 am every day from November 2004 through October 2005. Prevalence of nosocomial infection and infection site definitions were according to the National Nosocomial Infections Surveillance system of the Centers for Disease Control and Prevention. RESULTS: Among 528 infants enrolled, 60 (11.4%) had 97 nosocomial infections. The survival rate was 92%. The prevalence of nosocomial infections was 17.5%: bloodstream infection, 4.7%, clinical sepsis, 6.3%, pneumonia, 5.1%, urinary tract infections (UTIs), 0.7%, surgical site infection, 0.7%. Intervention-associated infection rate: central intravascular catheter-associated bloodstream infection, 13.7%, TPN-associated bloodstream infection, 15.8%, ventilator-associated pneumonia, 18.6%, surgical site infection 13.7%, urinary catheter-associated UTI, 17.3%. Cut-off values of onset of central intravascular catheter-associated bloodstream infection and ventilator-associated pneumonia were 6 days and 10 days after intervention, respectively. Patients with a birth weight <1000 g (relative risk, 11.8, 95% confidence interval, 7.66-18.18; P < .001) were at the greatest risk for nosocomial infection. CONCLUSIONS: This study revealed the high prevalence of nosocomial infections in NICU patients, and the urgent need for a national surveillance and more effective prevention interventions.     

Nosocomial infections in critically ill infectious disease patients: Results of a 7-year focal surveillance

Summary An incidence study on nosocomial infections in critically ill infectious disease patients was carried out in the intensive care unit (ICU) of a university hospital for infectious diseases over a 7-year period (1 January 1990 to 31 December 1996). A total 660 patients who stayed in the ICU for over 48 h were prospectively observed. The patients were divided into two groups: one with central nervous system infections (442 patients) and the other with other severe infections (218 patients). The risk of nosocomial sepsis and pneumonia was significantly higher in patients suffering from severe central nervous system infections. The incidence of sepsis was 24.2% vs 11.4% (relative risk 1.95; 95% confidence interval 1.32–2.89); the incidence of pneumonia was 30.5% vs 14.7% (relative risk 2.09; 95% confidence interval 1.47–2.96). The incidence of urinary tract infection was 14.3% vs 13.3% (relative risk 1.07; 95% confidence interval 0.71–1.61). Density rates of nosocomial septic episodes were 21.1±37.1 vs 11.7±32.4 episodes/100 central venous-line days (P<0.006). Nosocomial pneumonia occurred only in mechanically ventilated patients (36.9±61.2 vs 28.5±65.8 episodes per 1000 ventilatory days, P=0.012). Nosocomial urinary tract infection occurred only in patients with urinary catheters (11.6±60.7 episodes/1000 urinary catheter days vs 18.7±90.1, P=0.886). Multivariate regression analysis identified age, diagnosis of CNS infection, duration of urinary tract catheterization, the use of central venous lines and mechanical ventilation as independent risk factors of nosocomial sepsis. Duration of mechanical ventilation, use of steroids and diagnosis of CNS infection were independent risk factors of nosocomial pneumonia. A subanalysis identified tetanus patients to be at particular risk of nosocomial infections.     

A randomized crossover study of silver-coated urinary catheters in hospitalized patients

BACKGROUND: Urinary tract infections (UTIs) account for 30% to 40% of nosocomial infections resulting in morbidity, mortality, and increased length of hospital stay. OBJECTIVE: To assess the efficacy of a silver-alloy, hydrogel-coated latex urinary catheter for the prevention of nosocomial catheter-associated UTIs. METHODS: A 12-month randomized crossover trial compared rates of nosocomial catheter-associated UTI in patients with silver-coated and uncoated catheters. A cost analysis was conducted. RESULTS: There were 343 infections among 27,878 patients (1.23 infections per 100 patients) during 114,368 patient-days (3.00 infections per 1000 patient-days). The relative risk of infection per 1000 patient-days was 0.79 (95% confidence interval, 0.63-0.99; P =.04) for study wards randomized to silver-coated catheters compared with those randomized to uncoated catheters. Infections occurred in 291 of 11,032 catheters used on study units (2.64 infections per 100 catheters). The relative risk of infection per 100 silver-coated catheters used on study wards compared with uncoated catheters was 0.68 (95% confidence interval, 0.54-0.86; P =.001). Fourteen catheter-associated UTIs (4.1%) were complicated by secondary bloodstream infection. One death appeared related to the secondary infection. Estimated hospital cost savings with the use of the silver-coated catheters ranged from $14,456 to $573,293. CONCLUSIONS: The risk of infection declined by 21% among study wards randomized to silver-coated catheters and by 32% among patients in whom silver-coated catheters were used on the wards. Use of the more expensive silver-coated catheter appeared to offer cost savings by preventing excess hospital costs from nosocomial UTI associated with catheter use. Arch Intern Med. 2000;160:3294-3298.     

APACHE II and ISS scores as predictors of nosocomial infections in trauma patients

BACKGROUND: Nosocomial infections affect more than 2 million patients annually in the United States at a cost of $4.5 billion. The aim of this study is to identify the role of the APACHE II score and the Injury Severity Scale (ISS) as independent predictors of nosocomial infections in trauma patients admitted to the intensive care unit (ICU). METHODS: A retrospective chart review of 113 trauma patients admitted to the ICU was conducted by an infectious disease physician. Demographic data and incidence of nosocomial infections were recorded. Multivariate logistic regression analysis was used to determine variables that are predictive of the occurrence of nosocomial infections. RESULTS: Presence or absence of intubation, ICU length of stay, APACHE II score, and ISS were related to the presence of infections; however, only the ICU length of stay was an independent predictor of a nosocomial infection, with an odds ratio of 1.81. By linear regression, 17% of the variance in the ICU duration of stay was a result of the APACHE II score in patients with a score >/=5. CONCLUSION: APACHE II score and ISS score were not good predictors of the incidence of nosocomial infections in trauma patients admitted to the ICU, but the APACHE II score has a modest correlation with the duration of stay in the ICU. A stratified cohort study could identify the subset of patients for which the APACHE II score predicts a prolonged stay in the ICU, thus an increased risk of infection.     

Annual incidence and risk factors for nosocomial bacterial infections in an acute care geriatric unit

PURPOSE: Elderly inpatients are particularly exposed to the risk of nosocomial infections, thus the study of their risk factors and consequences is of interest. METHODS: Among 1,565 subjects referred to a short-term geriatric unit, patients hospitalised for a year for an acute event and unable to move themselves were followed up for the occurrence of nosocomial infections. RESULTS: Among these 402 immobilised patients (age: 86.3 +/- 7.6 years), 102 nosocomial infections occurred in 91 patients (22.6%), whereas the estimation of the incidence in the total hospitalised population (1,565 subjects, age: 85.1 +/- 6.2 years) was 9.4% (95% confidence interval [CI] 8.3-11.2). Forty-seven point seven percent of nosocomial infections were urinary tract nosocomial infections, 27.5% were lower respiratory nosocomial infections, 9.2% were cutaneous nosocomial infections, 7.3% were septicaemia and 8.2% were of unknown origin. The relative risk (RR) of NI linked to functional dependency for mobility was 5.5 (95% CI: 3.93-7.7, P < 0.001). Other risk factors were: for all nosocomial infections: cancer diagnosis (RR 1.1, 95% CI: 1.1-1.2, P = 0.01); and respectively for urinary tract NI: bladder indwelling (RR 4.8, 95% CI: 2.9-7.7, P < 0.001), pulmonary NI: swallowing disorders (RR 5.4, 95% CI: 2.8-10.5, P < 0.001); and septicaemia: venous catheter (RR 5.4, 95% CI: 1.3-23.3, P = 0.002). NI were associated with an increased length of stay (22.1 +/- 11.7 days in infected patients vs 16.3 +/- 9.5 days in immobilised non-infected subjects, P < 0.001). The mean length of stay for the 1,565 subjects was 10.3 +/- 7.6 days. Death was attributed to nosocomial infections in 13 subjects. In conclusion, functional dependency for mobility, bladder indwelling, venous catheter, swallowing disorders and diagnosis of cancer were risk factors for nosocomial infections in hospitalised elderly subjects in an acutecare setting.     

Epidemiología general de las infecciones nosocomiales. Sistemas y programas de vigilancia

Infections related to the health-care system are those associated with health care practices in hospitalized patients as well as in out-patients with health-care contact. Nosocomial infections affect 5% of in-patients, and carry a high morbidity, mortality and economic cost. The main types of nosocomial infections are related to invasive procedures, and include respiratory tract infection, surgical site infections, urinary tract infections, and vascular catheter bacteremia. It has been shown that the application of checklists and a bundle of measures are useful in preventing these infections. Epidemiological surveillance, defined as the gathering of information to take actions, is the basis of infection control programs. These have evolved from a global surveillance targeted at processes and indicators of nosocomial infection. The comparison of these indicators can be useful in establishing preventive measures.     

Periodic Surveillance of Nosocomial Infections in a Neurosurgery Intensive Care Unit

Abstract Background: We assessed data on the epidemiology of nosocomial infections (NIs) in a 14-bed neurosurgical intensive care unit (NSICU) and used surveillance data for the promotion of quality improvement activities. Patients and Methods: Prospective periodic surveillance was performed over five 3-month periods between July 1998 and October 2002 on all patients admitted with a length of stay > 24 hours. Results: 763 patients with a total of 4,512 patient days and a mean length of stay of 5.9 days were enrolled within the 15-month study period. A total of 93 NIs were identified in 82 patients. Urinary tract infections (24.7%), pneumonia (23.6%), and bloodstream infections (17.2%) were the most frequent NIs recorded. Device-associated incidence rates were 6.0 (3.8–9.0, CI_95%) for urinary tract infection, 4.4 (2.4–7.4, CI_95%) for bloodstream infection, and 10.3 (6.3–15.9, CI_95%) for pneumonia per 1,000 days at risk. For improvement of infection control-related processes, evidence-based infection control guidelines were established and an NSICU nurse was designated to be responsible for infection control issues on the ward. In addition, several infection control problems arose during the observation periods and were rapidly responded to by introducing specific intervention strategies. Conclusion: Periodic surveillance is a valuable tool for assessing the epidemiology of NIs in the NSICU setting as well as for promoting the initiation of quality improvement activities.     

A large outbreak of infections caused by a strain of staphylococcus aureus resistant to oxacillin and aminoglycosides

An extensive outbreak of nosocomial infections caused by oxacillinand aminoglycoside-resistant Staphylococcus aureus (OARSA) occurred over a 16 month period. A total of 349 isolates of OARSA were obtained from 174 patients. Colonization with OARSA was found in 92 patients. There were 120 infections in 82 patients; 50 were surgical wound infections, 13 were nonsurgical wound infections, six were pneumonias, 15 were urinary tract infections, 12 were intravenous site infections, and there were 19 episodes pf bacteremia (seven transient, 12 persistent). In patients with persistent bacteremia, the mortality rate was 33 percent. In patients treated for persistent bacteremia with vancomycin, the survival rate was 80 percent. Infections were highly associated with the surgical intensive care unit, and 90 percent of the isolates of OARSA tested had the same phage-type. Elderly patients with significant underlying disease, a history of previous surgery or of prior antimicrobial therapy appeared to be at increased risk for OARSA infections. OARSA were resistant to multiple antibiotics besides oxacillin, but all isolates were sensitive to vancomycin and rifampin. Three surgical intensive care unit nurses were found to be nasal carriers of OARSA, and one nurse had dermatitis of both hands colonized with OARSA. Following the removal of these nurses from the surgical intensive care unit and the institution of strict infection control measures, the number of OARSA infections and colonizations decreased to less than one per month. OARSA produces serious nosocomial disease, and epidemiologic intervention was effective in controlling this outbreak.     

National nosocomial infection surveillance system-based study in Iran: additional hospital stay attributable to nosocomial infections

BACKGROUND: Nosocomial infection is a serious health and financial problem. The purpose of this study was to determine the extra hospital stay attributable to nosocomial infections for patients undergoing surgery. METHOD: All patients undergoing surgery admitted from March 1, 1999, to February 28, 2000, to the 38-bed general surgery ward within a university hospital in Shiraz, Iran, were included in this study. The study was planned as a pairwise-matched case-control study nested in cohort design. A case was defined as any patient with 1 of 4 of the following nosocomial infections: urinary tract infection; surgical site infection; bloodstream infection; or pneumonia, whereby definitions for the nosocomial infections were on the basis of National Nosocomial Infection Surveillance system definitions. For each patient, an appropriate match was selected, which resulted in 69 pairs of study patients. RESULTS: The total incidence of nosocomial infection during the study period was 17.59%. The mean extra length of hospitalization as a result of all major kinds of nosocomial infections was 6.62 days total, which was obtained using 4.4, 5.33, 8.73, and 9.2 extra days for urinary tract infection, pneumonia, surgical site infection, and bloodstream infection, respectively. CONCLUSION: Nosocomial infections add considerable costs to the health care system in Iran. Therefore, the development of strategies and concepts to reduce the incidence of nosocomial infections is cost-effective and warranted, and an appropriate surveillance system on the basis of international criteria is the cornerstone for this task.     

The attributable mortality and costs of primary nosocomial bloodstream infections in the intensive care unit.

Primary nosocomial bloodstream infection (BSI) is a common occurrence in the intensive care unit (ICU) and is associated with a crude mortality of 31.5 to 82.4%. However, an accurate estimate of the attributable mortality has been limited because of confounding by severity of illness. We undertook this study to assess the attributable mortality and costs associated with an episode of BSI. Infected patients were defined as those who had an episode of BSI during the study period. Uninfected control subjects were matched to the infected patients based upon a number of factors, including predicted mortality on the day prior to infection. The main outcome measures were crude ICU mortality, length of stay, and costs. We found no difference in the crude mortality for the infected and the uninfected patients (35.3 and 30.9%, respectively, p = 0.51). However, among survivors, the patients with nosocomial bloodstream infections did have excess length of stay (mean, 10 d; median, 5 d; p = 0.007) and increased direct costs (mean difference, $34,508; p = 0.008). After matching for severity of illness, we could not detect an association between primary nosocomial bloodstream infections and increased ICU mortality. We did find that primary nosocomial bloodstream infections increased ICU length of stay and costs.     

Nosocomial infection of the urinary tract

Urinary tract infections appear to be responsible for 35 percent of all hospital-acquired infections, occurring in approximately two patients per 100 admissions. The great majority of infections are associated with urinary tract instrumentation. Female sex, advanced age and debilitating underlying illness appear to be associated with an increased risk of infection, but other risk factors have been poorly defined and case-control studies assessing excess morbidity and mortality associated with nosocomial bacteriuria have not been made. In most instances, the hospitalized patients are the reservoirs for the etiologic organisms, but cross infection from other infected patients also occurs. Current preventive efforts have been primarily directed at aseptic catheter care techniques and reducing catheter use. Further developments in our prevention and control of these infections require an understanding of the mechanisms responsible for colonization of the urethra and bladder with gram-negative organisms and ways to discriminate patients at especial risk.     

Risk factors for nosocomial imipenem-resistant Acinetobacter baumannii infections.

OBJECTIVES: To identify the risk factors for nosocomial imipenem-resistant Acinetobacter baumannii (IRAB) infections. METHODS: A prospective case-control study, set in an 1100-bed referral and tertiary-care hospital, of all patients who had nosocomial A. baumannii infections between January 1 and December 31, 2004. Only the first isolation of A. baumannii was considered. RESULTS: IRAB was isolated from 66 (53.7%) patients and imipenem-sensitive Acinetobacter baumannii (ISAB) was isolated from 57 (46.3%) patients during the study period. The mean duration of hospital stay until A. baumannii isolation was 20.8+/-13.6 days in IRAB infections, whereas it was 15.4+/-9.4 days in ISAB infections. Of the patients, 65.2% with IRAB infections and 40.4% with ISAB infections were followed at the intensive care unit (ICU). Previous carbapenem use was present in 43.9% of the patients with IRAB and 12.3% of the patients with ISAB infection. In univariate analysis female sex, longer duration of hospital stay until infection, ICU stay, emergent surgical operation, total parenteral nutrition, having a central venous catheter, endotracheal tube, urinary catheter or nasogastric tube, previous antibiotic use, and previous administration of carbapenems were significant risk factors for IRAB infections (p<0.05). In multivariate analysis, longer duration of hospital stay until A. baumannii isolation (odds ratio (OR) 1.043; 95% confidence interval (CI) 1.003-1.084; p=0.032), previous antibiotic use (OR 5.051; 95% CI 1.004-25.396; p=0.049), and ICU stay (OR 3.100; 95% CI 1.398-6.873; p=0.005) were independently associated with imipenem resistance. CONCLUSIONS: Our results suggest that the nosocomial occurrence of IRAB is strongly related to an ICU stay and duration of hospital stay, and that IRAB occurrence may be favored by the selection pressure of previously used antibiotics.     

Nosocomial infections in the elderly. Increased risk per hospital day

Elderly patients have been shown to have an increased risk of acquiring nosocomial infection per hospital admission. To determine if the length of stay accounts for this risk, daily infection rates were computed per decade of life and rates for patients over and under 60 were compared using risk ratios. Four thousand thirty-one nosocomial infections in 2,567 patients were identified for a 1980 through 1984 admission cohort in an acute-care hospital. The daily infection rates were 0.59 percent in patients over age 60 and 0.40 percent in younger patients (relative risk = 1.49). The daily incidences of urinary tract infections, respiratory infections, and septicemias were all significantly increased in elderly patients with risk ratios of 2.78, 2.07, and 1.36, respectively. Further analysis revealed that elderly patients experienced significantly more nosocomial infections for each day of hospitalization after Day 7. These data show that elderly patients experience an increased daily rate of nosocomial infection, and suggest that efforts be directed at identifying clinical conditions that predispose this population to hospital-acquired infections.     

Epidemiology of nosocomial infections in medicine intensive care unit at a tertiary care hospital in northern India

This prospective observational study describes the rates of nosocomial infections (NI), the sites of infection, the pathogens involved, their antibiogram and the risk factors at a tertiary care hospital in northern India. In 62 of the 182 enrolled patients 95 episodes of NI were recorded (incidence rate 28.6/1000 person days): pneumonia (77%); urinary tract infection (24%) and blood stream infection (24%). All isolates of Acinetobacter, Pseudomonas and Klebsiella and 83.3% of Escherichia coli were resistant to the third generation cephalosporins. An increased duration of the time spent in intensive care units and days of intervention were associated with incident NI.     

重症监护室耐药非发酵革兰阴性菌院内感染的危险因素及治疗进展研究

多耐药非发酵革兰阴性菌已成为重症监护室院内感染的重要病原菌,近年来其检出率在全球有逐年上升的趋势,对其危险因素和治疗进展的探讨具有重要意义.入院前有抗生素使用史、反复插管、最初不恰当抗生素治疗即病原菌对所用抗菌药物不敏感、重症监护室住院时间延长等是耐药菌院内感染常见的危险因素.关于耐药菌院内感染的治疗,采用早期适当联合用药原则,控制感染.     

The risk factors for infections acquired by cerebral hemorrhage and cerebral infarct patients in a neurology intensive care unit in Turkey

Few studies have investigated the risk factors for nosocomial infections developed in neurology intensive care units (ICUs). In this study, the risk factors for ICU-acquired infections in patients with cerebral hemorrhage and cerebral infarct who were treated for more than 24 h at the Ankara Training and Research Hospital were prospectively evaluated over a study period of 14 months. Of 171 patients included in the study, 71 (41.5%) were found to have acquired 163 infections in the ICU unit throughout 1,867 patient days. The rate of infection per 100 patients admitted was 95.3, and per 1,000 patient days, 87.3. The most common nosocomial infections were urinary tract infection (42.9%), pneumonia (27%) and primary bacteremia (19%). Multivariate logistic regression analysis revealed age > or = 70 (P < 0.05), the presence of a central venous catheter (P=0.004), and parenteral nutrition (P=0.02) as ICU-acquired infection risk factors. The presence of infection on admission was identified as a factor decreasing the risk of ICU-acquired infection (P < 0.001). The high infection rates found in this study may be due to lack of full compliance to infection control measures. In conclusion, each type of ICU has its own epidemiological findings for nosocomial infections and thus needs to determine the risk factors using periodical surveillance studies to guide control measures.