Surveillance of nosocomal sepsis in intenvive care unit

II. ABSTRACTSURVEILLANCE OF NOSOCOMIAL SEPSIS IN INTENSIVE CARE UNITSeveral problems are present nosocomial sepsis which still remainsan important health problem in Turkish intensive care units (ICUs). Theaim of the clinical practise is to decrease the mortality rate of nosocomialsepsis in ICUs. This study sought to find mortality-related risk factors inICUs.The prospective study was performed from 1 Mart 2004 to 1 Mart2005 in the Chest Heart Vessel Surgery-ICU and Anesthesia andReanimation- ICU of Firat University Firat Medicine Center. The patientswho were followed in the ICUs for at least 48 hours were enrolled in thestudy. Those patients who died within 48 hours of ICU discharge wereincluded in the mortality analysis. Nosocomial infections were definedaccording to the Center for Disease Control and Prevention (CDC) criteriaand were categorised by specific infection site.Among the 470 patients, 81 had ICU- acquired sepsis episodes. In-ICUs nosocomial was sepsis rate 17.2% (81/470) and mortality rate 63%(51/81). The age (p= 0.02), chronic health evaluation (APACHE II score)(p<0.05) were significantly longer for patients with a fatal outcome than forsurviors. Needs for mechanical ventilation, entubation, nasogastriccatheter, urinary catheters episodes of sepsis were significantlyassociated with fatalities (OR=0.26, %95 CI 0.1-0.63, p=0.000;OR=0.31,%95 CI 0.14-0.67, p= 0.00; OR=0.12, %95 CI 0.02-0.83, p=0.000;OR=2.96, %95 CI 2.16-4.05, p=0.007, respectively). No differences4between fatalities and surviors were found in sex, the mean lengths of stayin ICU, sepsis determining days, need for a total parenteral nutrition,enteral nutrition, chest tube, abdominal drain, nonoperativ invaziv devicesand days-number of invaziv devices factor.Respiratory failure, renal failure was associated with a fataloutcome (OR=0.50 %95 CI 0.26-0.97, p=0.008; OR=1.63 %95 CI 0.97-2.75, p=0.027 respectively). Hepatic failure, solid tumor, heart disease,diabetes mellitus, blood sugar disregülation, choronic obstructive lungsdisease, peripheri vessel disease, vasculitis, bone marrow transference,body trauma was not associated with a fatal outcome.H2 receptor antagonist was associated with a fatal outcome(OR=0.60 %95 CI 0.31-1.12, p=0.046). No differences between fatalitiesand surviors were found in antiacid, proton pump inhibitors, steroid andother immunosuppression drugs. Days of H2 receptor antagonist wasassociated with a fatal outcome (p=0.04, p=0.00, respectively). Mortalityincreased not with days of steroid, proton pump inhibitor and otherimmunosuppression drugs.Emergency surgery, elective surgery, prophylaxis, wound type andblood tranfüsion were not significantly associated with a fatal outcome.The most frequent primary sources of nosocomial sepsis werelungs 38.3%, urinary 29.7% and intra-abdominal 12.4%. Cultures werepositive of the patients with common microorganisms were Pseudomonasspp. 22.3% (including 14.5% P.aeroginosa), E.coli 17.1%, S.aureus 17%(including 9% methicillin-resistant), Acinetobacter spp. 13.1%.5The most frequent antibiotic used for the patients with nosocomialsepsis were respectively meropenem 30.8%, teicoplanin 25.9%,ceftriaxone 22.2%, imipenem 16%, cefotaxime 14.8%, cefazolin 13.6%,sulbactam-cefoperazone 11.1%, vancomycin 11.1%, piperacillin-tasobactam 7.4%. Third-generation cephalosporins were the mostempirical antibiotics used in nosocomial sepsis. The empirical antibioticswere seemly 74% and not seemly 26%.The most important risk factors of mortality were observed olderage, high APACHE II score, mechanical ventilation, intubation, nasogastriccatheterization, urinary catheterization, respiratory failure, renal failure,used H2 receptor antagonist and days of H2 receptor antagonist.Key Words: Intensive care unit, nosocomial sepsis, surveillance,mortality, risk factors6