Nosocomial infections after cardiac surgery in infants and children: incidence and risk factors.

This prospective study was undertaken to determine the spectrum, sites and main risk factors for hospital-acquired infections (HAI) in our paediatric cardiothoracic intensive care unit (PCICU), and to determine the main organisms causing bloodstream infection in this setting. All patients admitted between January and December 1999 were prospectively followed for the development of HAI. To define risk factors, patients were grouped by age, complexity score, length of stay in PCICU, and whether the patient's chest was open or closed postoperatively. Three hundred and thirty-five patients underwent cardiac surgery. Fifty-five patients acquired 69 HAIs (HAI patient rate 16.4%). The most common HAI were bloodstream and surgical wound infection in 10 and 8%, respectively. The main causative organisms were Klebsiella spp.,Enterobacter spp. and Pseudomonas spp. in 22, 17 and 16% of episodes, respectively. Staphylococcus spp. accounted for 16% of episodes. The main risk factors for developing HAI were: neonatal age [P < 0.05, odds ratio (OR): 5.89, 95% confidence interval (CI): 2.96-11.58] prolonged PCICU stay (P < 0.05, OR: 6.82, 95% CI: 3.37-14.48), open chest postoperatively (P < 0.05, OR: 3.44, 95% CI: 1.31-8.52) and high complexity score (P < 0.05, OR: 4.03 95% CI: 1.87-8.43). The main causative organisms of bloodstream infections in children hospitalized in the PCICU differ from those in adult and pediatric general intensive care units (ICUs) and include mainly Gram-negative bacilli. High complexity score, neonatal age, prolonged ICU stay, and open chest postoperatively are risk factors of HAI in this patient population.

Prediction of mortality in patients with bacteremia: the importance of pre-existing renal insufficiency.

Pre-existing renal insufficiency serves as a common risk factor in the development of acute renal failure. Acute renal failure is a common finding in patients with bacteremia and is associated with poor prognosis. A total of 2722 consecutive patients 18 years old or older, fulfilling strike criteria of bacteremia or fungemia were prospectively evaluated to establish the prognostic importance of pre-existing renal insufficiency in bacteremic patients. They were classified according to serum creatinine levels upon admission into three groups. 915 patients had normal creatinine levels (< or = 1.0 mg/dL), 1528 had mild to moderate renal failure (creatinine 1.1-3 mg/dL) and 279 patients had severe renal failure upon admission (creatinine > 3.0 mg/dL). Mild to severe renal failure upon admission was associated with old age, male gender, diabetes mellitus, ischemic heat disease, hypertension and congestive heart failure. The serum albumin in patients with severe renal failure was significantly low, with a mean of 2-9 mg/dL. Urinary tract infections were more prevalent in patients with mild to severe renal failure, while intravenous line infections, bacterial endocarditis and soft and skin tissue infections were more common in patients with normal renal function. In the 279 patients with severe renal failure the mortality rate was significantly higher (50%) compared to patents with mild to moderate renal failure and patients with normal renal function (21% and 26% respectively, p = 0.0001). Multiple regression analysis revealed that pre-existing serum creatinine > 3 mg/dL was significantly associated with death attributable to bacteremia (OR = 1.7, 95% CI 1.0-2.7). In conclusion, adult bacteremic patients with pre-existing serum creatinine above 3 mg/dL upon admission are at increased risk of mortality due to bacteremia than patients with normal or mild to moderate renal failure.

Five-year prospective study of bacteraemic urinary tract infection in a single institution.

In order to determine the epidemiology, microbiology, and outcome of bacteraemia originating in the urinary tract in hospitalised patients, a prospective study was conducted in a large general hospital in Israel. Data from all patients with bacteraemia were collected prospectively, and a subgroup of patients with bacteraemia secondary to urinary tract infection was analysed. There were 702 episodes of bacteraemia secondary to urinary tract infection during a five-year period (33.9% of all episodes of bacteraemia). The mean age of the patients was 76 years, and the male:female ratio was 0.9:1.0. The most common pathogens were Escherichia coli (52%), Klebsiella spp. (14%), and Proteus spp. (9%). Pseudomonas spp. were isolated from 8% of all patients, from 19% of those who had received antibiotics, and from 15% of males. Enterococcus spp. were isolated from 4% of males but from no females. Five percent of the episodes were polymicrobial, and 16% of the infections were hospital acquired. On logistic multivariate regression analysis, predictors of mortality were: hospitalisation in a medical department, hospital-acquired infection, inappropriate empiric antibiotic treatment, presence of decubitus ulcer(s), respiratory or renal failure, and elevated urea and decreased albumin levels.

Factors associated with increased risk in inappropriate empiric antibiotic treatment of childhood bacteraemia.

UNLABELLED: To identify bacteraemic children who are at increased risk of inappropriate empiric antibiotic therapy, we performed univariate and multivariate analyses of prospectively-studied bacteraemic episodes. Appropriateness of therapy was defined according to the in vitro susceptibility of the isolate. Inappropriate empiric therapy was found in 38% of 516 bacteraemic episodes and was associated with higher mortality. The rate of inappropriate treatment was lower in neonates and infants (28% and 33%, respectively) but higher in children 1- to 5-years old (51%, P = 0.0029). The rate was dependent on the source of bacteraemia (range, 18%-70%, P = 0.0092), underlying conditions (range, 26%-53%, P = 0.0001), the specific paediatric section in which the child was hospitalized (range, 24%-70%, P = 0.0002), and the causative micro-organism (range, 15%-75%, P < 0.0001). Four clinical variables that independently and significantly affected the rate of inappropriate antibiotic treatment were identified by multivariate stepwise logistic regression analysis (odds ratios in parentheses): hospital-acquired bacteraemia (2.3), age of 1- to 5-years (2.1), cytotoxic therapy (1.8) and presence of central i.v. line (1.6). CONCLUSION: We defined bacteraemic children who are at risk of inappropriate empiric antibiotic therapy. Special efforts are needed to improve their treatment and consequently their outcome.

A prospective study of Gram-negative bacteremia in children.

BACKGROUND: Hospital- and community-acquired Gram-negative bacteremia is a significant cause of mortality and morbidity in pediatric medical centers. Gram-negative organisms are isolated in > 50% of pediatric patients with bacteremia. OBJECTIVES: To analyze clinical and epidemiologic variables associated with Gram-negative bacteremia in a tertiary children's medical center. METHODS: A 6-year prospective study of children with Gram-negative bacteremia in a tertiary care children's medical center in Israel. RESULTS: Three hundred seventy-four episodes of Gram-negative bacteremia were studied during 6 years. The predominant isolates were Klebsiella pneumoniae, Pseudomonas aeruginosa and Escherichia coli, which accounted for 109, 81 and 79 episodes (26, 20 and 19%), respectively. Of all episodes 43% occurred in neonates and infants younger than 2 years and 47% were hospital-acquired. Underlying conditions mainly acute leukemia and lymphoma, were present in 55% of the patients. Urinary tract infection followed by lower respiratory tract infection were the most common identified sources of bacteremia. Central intravenous catheters were associated with 53% of the episodes. The crude mortality was 11.4%. Increased mortality was significantly associated with acute leukemia, neutropenia, hospital-acquired infections and previous corticosteroid therapy (P = 0.03, 0.003, 0.006 and 0.01, respectively). Increased antibiotic resistance of hospital-acquired vs community-acquired isolates was noted; 44 to 77% resistance of nosocomial Klebsiella and Enterobacter sp. to second and third generation cephalosporins and 18% were resistant to amikacin. CONCLUSIONS: Klebsiella pneumoniae is currently the most common organism causing Gram-negative bacteremia in children. Because of the relatively high resistance of Gram-negative organisms to second and third generation cephalosporins, we suggest that empiric antibiotic therapy for Gram-negative bacteremia include a combination of an aminoglycoside and an anti-Pseudomonas beta-lactam.

Incidence and severity of group A streptococcal bacteremia in a medical center in Israel, 1981-1994.

To determine recent trends in the incidence and severity, group A streptococcal (GAS) bacteremia was studied over the last 14 years (1981-1994). There were 116 events of GAS bacteremia, representing 1.7% of all bacteremic episodes, without an increase in recent years. A total of 108 patients were available for study. Underlying conditions were found in 95 patients (88%), including mainly malignant diseases, chronic obstructive pulmonary disease, congestive heart failure and diabetes mellitus. A source of the bacteremia was noted in 71 patients (66%), with skin and soft tissue infection being the major portal of entry. All isolates were susceptible to penicillin. Overall mortality was 21%. Mortality had not increased in recent years, but depended significantly on several clinical factors: increased age; admission temperature; source of bacteremia (highest for GAS bacteremia without an identified source); and underlying conditions (highest for diabetes mellitus and chronic pulmonary disease, absent for patients with no underlying disease). This study shows that neither the incidence nor the severity of GAS bacteremia has increased in recent years. Severity is significantly affected by the source of bacteremia and the presence of underlying conditions.

Bacteraemia in a community and a university hospital.

Four hundred and forty-one and 1048 episodes of bacteraemia were prospectively surveyed over a period of 18 months in two hospitals, a 450 bed community hospital and a 900 bed tertiary care urban university hospital. Incidence of bacteraemia was 2.18 per 1000 hospitalization days (10.1 per 1000 admissions) in the community hospital and 2.64 per 1000 hospitalization days (12.0 per 1000 admissions (P < 0.004)) in the university hospital. Sixty six and 62% of episodes of bacteraemia were community acquired. The majority of bacteraemic episodes originated on the internal medicine wards of both hospital--46.7% and 58.7% respectively; the incidence of bacteraemia in the medical divisions of both hospitals was 23.1 and 17.5 per 1000 admissions respectively (P < 0.01). Overall mortality rates were 22% and 26.7% respectively. 39.9% and 44% of all isolates were Gram-positive pathogens. Escherichia coli was the commonest Gram-negative pathogen in both hospitals, particularly the community hospital--47.5% vs 32.8% (P < 0.005) of all Gram-negative pathogens, while Pseudomonas spp. were significantly more common in the university hospital--18.5% vs 11.8% (P < 0.02). Non-enterococcal streptococci were more common in the community hospital while enterococci were far more common at the university hospital--15.1% vs 1% of all Gram-positive pathogens (P < 0.05). Staphylococcus epidermidis was more common among the community hospital Gram-positive bacteraemias--31.1% vs 18.6% (P < 0.005). For almost all genera and species, antibiotic resistance was higher at the university hospital. Twenty nine point four per cent of Staphylococcus aureus isolates from the university hospital were methicillin resistant compared to 2.4% at the community hospital (P < 0.005). 29.4% of all Streptococcus pneumoniae isolates at the university hospital were penicillin resistant while no resistance was found at the community hospital. A high resistance rate to ofloxacin was found at the university hospital among S. aureus and Pseudomonas sp. Sources of bacteraemia did not differ significantly between the two hospitals. In conclusion, although outcome did not differ significantly for the two hospitals, there were significant differences between blood culture isolates in these two different settings. These differences may influence clinical decision-making about antibiotic therapy for patients in these hospitals.

Prognostic significance of the neutrophil count in immunocompetent patients with bacteraemia.

To examine the prevalence of neutropaenia in immunocompetent, bacteraemic patients, and whether it carries an independent risk for mortality, we surveyed 2096 bacteraemic patients without malignant diseases, and who were not receiving cytotoxic drugs. The granulocyte count on the day of the first positive blood culture was < 1 x 10(9) cells/l in 33 patients (1.7%, group 1); 1.0-4.0 x 10(9) cells/l in 154 patients (7.9%, group 2); 4.0-8.0 x 10(9) cells/l in 564 patients (29%, group 3); 8.0-20.0 x 10(9) cells/l in 1034 patients (53%, group 4); and > 20.0 x 10(9) cells/l in 163 patients (8.4%, group 5). The mortality rates in the five groups were 39.4%, 18.8%, 18.1%, 25.7% and 25.8%, respectively (p = 0.0001). The main pathogens in group 1 were Staphylococcus aureus in 25% of patients and Pseudomonas sp. in 23%. Mortality in group 1 patients was higher than in the other patients (odds ratio 1.4, 95% CI 1.1-1.9]. Mortality was also significantly higher in group 2 patients with high blood urea nitrogen. The percentage of neutropaenia, septic patients without known risk factors for neutropaenia is small, but their mortality is high. Overall mortality in patients with relative neutropaenia (1.0-4.0 x 10(9) cells/l) is low, but a subgroup of patients with high blood urea nitrogen is at considerable risk for a fatal outcome. High leucocyte counts are also a marker of increased risk for mortality, but this association is not an independent prognostic factor.

Childhood bacteremia in Israel: causes, age relation, predisposing factors and source.

To delineate the spectrum of childhood bacteremia in a tertiary medical center in Israel and to define the historical, clinical and environmental variables that affect it, 339 bacteremic episodes were studied. Ten of the episodes (3%) were polymicrobial and 148 (44%) were hospital acquired. Staphylococcus epidermidis (17%), Staph. aureus (10%), gram-negative bacilli and Haemophilus influenzae (7%) were the most frequent etiologic pathogens. Some organisms (e.g., H. influenzae, Streptococcus pneumoniae) caused mainly community-acquired bacteremia, while Klebsiella sp., Enterococcus faecalis, and Acinetobacter sp. caused mainly nosocomial bacteremia. Underlying conditions were noted in 72% of the bacteremic children. A source of the bacteremia was identified in 60% of the episodes; the most common was i.v.-line infection. Age, underlying condition, source and location in the hospital markedly affected the profile of microorganisms causing childhood bacteremia. Each of these variables defined 3-5 organisms that were most prominent. In each episode of suspected bacteremia, these variables should be considered, thus defining the most likely causative pathogen(s), which should be covered by appropriate empiric antimicrobial treatment.

Bacteremia in patients without known underlying disorders.

Of 2030 consecutive patients with bacteremia, only 102 were free from underlying disorders. 43 were males, and the median age was 66 years. The sources of infection were the urinary tract (in 48%), lower respiratory tract (13%), endocarditis (7%), biliary tract (6%) and the meninges (5%). The most common pathogens were Escherichia coli (in 45% of patients), Streptococcus pneumoniae (21%), Staphylococcus aureus (9%) and hemolytic streptococci (9%). Overall mortality rate was 13%, 4% in patients with urinary tract infection and 19% in patients with other sources. Half of the deaths occurred within 2 days of hospitalization, and 75% of them within 4 days. All patients with septic shock and all patients with meningitis died. Other factors related to mortality were residence in a nursing home, low functional capacity, advanced age, high blood urea nitrogen and creatinine and low albumin, and infections caused by Staphylococcus aureus, Neisseria meningitidis and polymicrobial infections. A protective effect of appropriate antimicrobial antibiotic therapy could not be demonstrated. In conclusion, bacteremic patients with no known underlying disorder and source of their infection other than the urinary tract should be given maximum supportive treatment and should be closely watched.

Bloodstream infections in patients older than eighty years.

During a period of 3 years in a University Hospital in Israel, 339 episodes of bacteraemia were observed in patients 80 years of age or older, and 658 episodes in patients 60-79 years of age. Patients older than 80 were more often residents of nursing homes, frequently had a history of a cerebrovascular accident, but were less often neutropenic. Twenty-four per cent of bacteraemia episodes in the very old were hospital acquired compared with 40% in the old patients. The most common source of bacteraemia was the urinary tract, 50% of episodes in the very old, and 34% of episodes in the old. The percentage of episodes in which anaerobic bacteria were isolated was 5% in the very old and 1% in the old, and the difference was significant when corrected for the sources of bacteraemia. All cases of community-acquired bacterial endocarditis in patients of 80 or over were caused by pathogens originating from the gut. Thirty-five per cent of patients of 80 and over and 30% of patients aged 60-79 years died during hospitalization. Fatality was not associated with advanced age in the very old. Factors significantly and independently associated with fatality in both groups were a hospital-acquired infection, shock, low serum albumin, renal dysfunction and inappropriate antibiotic treatment.

Patterns of multiple resistance to antibiotics in gram-negative bacteria demonstrated by factor analysis.

Principal component analysis was used to demonstrate the main associations between patterns of resistance to antibiotic drugs in 670 gram-negative bacteria consecutively isolated from blood cultures over a period of two years. Six factors were derived, which accounted for 84% of the total variance of the original matrix. Each factor represented an association between resistance to certain antibiotics as follows: factor 1: aztreonam, third generation cephalosporins and aminoglycosides; factor 2: first and second generation cephalosporins; factor 3: tetracycline and chloramphenicol; factor 4: ampicillin and ureidopenicillins; factor 5: trimethoprim/sulfamethoxazole; factor 6: fluoroquinolones. On two-way analysis of variance the difference in the factor scores was significant between bacteria for all factors except factor 5. The difference in factor scores between community and hospital acquired strains was significant only for factors 1, 2 and 6. Only the score of factor 6 showed a clear trend to increase with time during the two-year study period. Patients who were treated with antibiotics prior to bacteremia had higher scores for all factors, the difference being most marked in patients treated with fluoroquinolones. Factor analysis can be used to describe phenotypic associations between resistance to antibiotics, and the factor score used to compare groups of isolates and to demonstrate temporal and other trends.

Risk factors for mortality due to bacteremia and fungemia in childhood.

To define risk factors for mortality due to bacteremia and fungemia of childhood, 242 episodes (for which the mortality rate was 19%) were studied prospectively by univariate and multivariate analyses. The mortality rate was higher in neonates (23%) and in individuals 10-18 years old (26%) than in infants and young children (10%-16%). The mortality rate was 29% for children who had neutropenia, 29% for those who had received therapy with steroids, 26% for those who had received antibiotics, and 75% for those who were in septic shock. The fatality rates for polymicrobial bacteremia (40%), recurrent bacteremia (67%), and hospital-acquired bacteremia (28%) were higher than those for other types of bacteremia; the fatality rate was related to inappropriate empiric antibiotic treatment or to the specific organism isolated (mortality rates associated with the latter ranged from 0 to 60%). Seven variables that independently and significantly affected mortality were defined with use of multivariate logistic regression analysis: septic shock (odds ratio [OR], 26.4); polymicrobial (OR, 5.4), recurrent (OR, 4.5), or hospital-acquired (OR, 4.3) bacteremia; candidemia (OR, 3.6); inappropriate antibiotic treatment (OR, 2.4); and neutropenia (OR, 2.3). These variables should be considered for adequate management of bacteremic patients who are at high risk for death.

Patients at risk for inappropriate antibiotic treatment of bacteraemia.

In order to define patients at high risk for inappropriate antibiotic treatment of bacteraemia, we compared 682 bacteraemic patients, treated with an antibiotic drug to which the infecting micro-organism was susceptible, with 419 patients who were inappropriately treated. On a multivariate logistic regression analysis including only clinical variables, four factors were found to be both significantly and independently associated with inappropriate antibiotic treatment: hospital-acquired bacteraemia (odds-ratio (OR) of 1.9), antibiotic treatment in the month prior to the bacteraemia (OR 1.9), residence in a nursing home (OR 1.8), and the presence of a central line (OR 1.7). A second model, including bacteriological data, showed four micro-organisms to be independently associated with inappropriate antibiotic treatment: Candida sp. (OR 14.2), Acinetobacter sp. (OR 5.0), Enterococcus sp. (OR 3.6) and Pseudomonas sp. (OR 2.2). In this model, only two clinical features were included: hospital-acquired infection and previous antibiotic treatment. Special efforts should be made to improve empirical antibiotic treatment in the groups defined above, and to facilitate early laboratory diagnosis of the micro-organisms associated with inappropriate treatment.

Bacteremia and fungemia of unknown origin in adults.

Two hundred fifteen (23%) of 955 episodes of bacteremia (defined as including fungemia) detected in adult patients during 2 years were of unknown origin. Sixty-six percent of episodes of unknown origin were hospital acquired. The median age of patients with bacteremia of unknown origin was 65 years, and their most common underlying disorders were solid malignancy (28% of patients) and diabetes mellitus (18%). Only three factors were associated with bacteremia of unknown origin (as opposed to episodes with a known source): peripheral venous catheterization, hemodialysis, and plasmapheresis. Gram-negative bacteria were isolated from the blood in 62% of episodes of unknown origin; 10% of episodes were polymicrobial. Staphylococci were isolated from 67% of patients undergoing hemodialysis and from 37% of those with diabetes; Pseudomonas species from 15% of patients with hospital-acquired episodes; and Candida species from 21% of patients with a central venous catheter. Fifteen percent of episodes in cancer patients were polymicrobial. Empirical antibiotic treatment was inappropriate in 49% of episodes of unknown origin and in 35% of episodes with a known source (P less than .001). Death rates were 44% and 25% in episodes of unknown and known origin, respectively. An unknown source of bacteremia was independently associated with a fatal outcome.

Bacteremia in adult diabetic patients.

OBJECTIVE: To compare the microbiology, sources, complications, and outcome of bacteremia in diabetic and nondiabetic patients. RESEARCH DESIGN AND METHODS: A prospective study was conducted of all episodes of bacteremia in hospitalized diabetic and nondiabetic patients. The study consisted of patients greater than or equal to 18 yr of age with bacteremia detected within a 19-mo interval. RESULTS: We compared 124 episodes of bacteremia in 119 diabetic patients to 508 episodes in 480 nondiabetic patients. Diabetic patients were older than nondiabetic patients (median age 74 vs. 68 yr, P = 0.0001). In patients with an indwelling urinary catheter and bacteremic urinary tract infection, the percentage of Klebsiella in diabetic patients was 60% (6 of 10) and in nondiabetic patients was 17% (4 of 23, P = 0.04). In patients without an obvious source of bacteremia, the percentage of staphylococcal isolates in diabetic patients was 29% (10 of 35) and in nondiabetic patients was 14% (24 of 176, P = 0.04). Staphylococci were a common cause of bacteremic infections of the extremities in diabetic patients (12 of 19, 63%) and nondiabetic patients (20 of 50, 40%). Septic shock was the only complication that was more common in diabetic patients. The mortality in diabetic and nondiabetic patients was 28 and 29%, respectively. CONCLUSIONS: Our results represent elderly patients with non-insulin-dependent diabetes mellitus. In this group, empirical treatment for suspected bacteremic urinary tract infection in patients with a urinary catheter should include coverage for Klebsiella. Empiric treatment for suspected bacteremia of unknown origin or caused by infection of the extremities should include an antistaphylococcal drug. The prognosis of bacteremia in diabetic and nondiabetic patients was similar.

Predictive index for optimizing empiric treatment of gram-negative bacteremia.

In a survey of 296 episodes of gram-negative bacteremia in 286 patients (aged 13-99 years), four clinical variables were found to predict both significantly and independently the subsequent isolation of a multiresistant strain; hospital acquisition of the infection, antibiotic treatment before the bacteremic episode, endotracheal intubation, and thermal trauma as the cause of hospitalization. These variables were combined in an index that served to classify the patients into four groups with an increasing prevalence of multiresistant strains, Pseudomonas isolates, and isolates resistant to each of the antibiotic drugs in common use. For example, the percentage of isolates susceptible to cefuroxime in the four groups were 79%, 56%, 34% and 25%, and to gentamicin, 89%, 79%, 46%, and 33% (P less than .001 for both comparisons). The performance of the index was validated in a second group of 144 episodes of gram-negative bacteremia. The index kept its discriminative power. Compared with the prescriptions of the attending physicians, the index could probably have improved empiric antibiotic treatment in 24% of patients.