Failure of the overgown to prevent nosocomial infection in a pediatric intensive care unit.

In a pediatric intensive care unit we conducted a 1-year prospective study of 454 patients to determine whether wearing a gown decreased the overall nosocomial infection rate, incidence of intravascular catheter colonization, breaks in handwashing technique, and traffic. The overall infection rate was 26 (13%) of 198 admissions during the gown-wearing periods v 23 (9%) of 256 admissions for the periods when gowns were not worn (P less than .25). Of 348 intravascular catheter tips cultured 16 (4.6%) were colonized during gown-wearing periods compared with 21 (6.3%) of 330 when no gowns were worn (P less than .25). Of 78 patient contacts 54 (69%) were followed by no handwashing during gown-wearing periods and 59 (70%) of 84 contacts were followed by no handwashing during periods when no gowns were worn. The mean occurrence of visits per patient per hour and total visits per hour differed between gown-wearing and no-gown-wearing periods by analysis of variance, P less than .01 and P less than .005, respectively. Although traffic was decreased during periods of gown use, overgowns are an expensive, ineffective method of decreasing nosocomial infection rates, vascular catheter colonization rates, and breaks in handwashing technique.

Short-course amphotericin B therapy for candidemia in pediatric patients.

OBJECTIVE: To determine the efficacy of short-course (7 to 14 days of therapy after the last positive blood culture) amphotericin B therapy for candidemia in children. DESIGN: Case series. SETTING: Tertiary care university medical center in Virginia. PATIENTS: Thirty patients younger than 17 years of age who had candidemia between 1983 and 1990. MEASUREMENTS AND RESULTS: The charts of 30 children with 31 episodes of candidemia were retrospectively reviewed for patient data, dates of positive and negative cultures for Candida from blood and other sites, dates of removal of the intravascular catheters, duration and dosage of amphotericin B administration, and outcome. Eight patients had persistent candidemia and died. Five patients were treated not in accordance with the short-course recommendations. Two had relapses; 1 was cured with catheter removal alone, and 2 were successfully treated with 26 and 30 days of amphotericin B therapy. Eighteen episodes (two episodes in 1 patient) of candidemia were cured using 7 to 14 days of amphotericin B therapy after the last positive blood culture. CONCLUSIONS: Once the bloodstream is sterilized, and there is no other evidence of invasive fungal disease, 7 to 14 additional days of amphotericin B at a dose of 0.5 mg/kg per day seems adequate for treatment of candidemia in children.

Hospital-acquired urinary tract infection.

From 16,534 admissions, 60 patients, 4 days to 15 years of age, with one or more hospital-acquired urinary tract infections were identified during a 5-year period by a prospective surveillance system. The patient charts were subsequently reviewed to characterize the population at risk for such infections and to describe the course and consequences of these infections. Infections in individual patients ranged from one to greater than 50. The hospital-acquired urinary tract infection rate for the study period was 14.2 infections per 1,000 admissions. In the patients in whom all urinary tract infections were well documented, the following characteristics were defined: (1) 92% (97 of 105) of the infections occurred in catheterized patients; (2) almost half (49 of 105) of the infections occurred in patients exposed to only intermittent catheterization; (3) 28% (29 of 105) of the infections were asymptomatic; (4) fever was the most frequent finding in the symptomatic patients and occurred in 66% (60 of 105); (5) pyuria was found in only 51% (35 of 69) of the urinalyses performed at diagnosis; (6) 85% (89 of 105) of the infections were single-organism infections; (7) 82% (101 of 123) of the causative organisms were Escherichia coli, Pseudomonas sp, coagulase-negative staphylococci, Enterococcus spp, Klebsiella spp, or Enterobacter sp. The urinary tract infections in the 60 patients were not complicated by bacteremia, and no direct relationship between the infections and the minimal mortality in our patients could be established.

Hospital-acquired urinary tract infections in the pediatric patient: a prospective study.

To determine through a prospective study the characteristics of hospital-acquired urinary tract infections (HAUTI) in children, 525 children subjected to bladder catheterization during a hospital admission were identified through surveillance of 12,316 admissions during a 24-month period. Urine culture results were available for 296 (56.4%) of the catheterized patients. In addition 12 noncatheterized children with a documented HAUTI were identified. The clinical courses of all patients with a HAUTI were followed for at least 6 months after their last HAUTI during the study period. Forty-four patients, 1 week to 17 years of age, with 1 or more HAUTI during a hospital unit admission were identified. A total of 51 HAUTI occurred. Thirty-nine (76.5%) of the infections occurred in patients subjected to catheterization. Thirty-two (10.8%) of 296 catheterized patients developed a HAUTI. Forty-three (84.3%) of the 51 infections were single organism infections. One HAUTI was associated with a wound infection with the same organism and one with a concurrent bacteremia with the same organism. Relapses were seen after 4 HAUTI. One reinfection was identified. There were no deaths directly associated with a HAUTI. Hospitalized children subjected to urinary tract catheterization are at significant risk for HAUTI. Complications are infrequent and not life-threatening.

Intravascular catheter colonization and related bloodstream infection in critically ill neonates.

Intravascular catheter tip colonization was prospectively evaluated in critically ill neonates to determine its relationship to the type of device used, duration of catheterization, insertion site and nosocomial bloodstream infection. Sixty-one percent (376 of 621) of all intravascular catheter tips were retrieved from 91 infants. Thirteen percent (41 of 310) of peripheral intravenous, 14% (6 of 42) of umbilical, 21% (3 of 11) of central venous, 36% (4 of 11) of peripheral arterial and 100% (2 of 2) of femoral catheters were colonized. Duration of catheterization was significantly longer for colonized lines (p less than .001). Eight of 26 (30.8%) peripheral intravenous catheters remaining in place for more than three days were colonized, compared with 33 of 284 (11.6%) at three days or less (p = 0.012). Coagulase-negative staphylococcus was the organism most frequently isolated from catheter tips and bloodstream infections. Catheter colonization rates in this population were higher than those found in adults. Heavily manipulated devices and those in place for longer periods of time were the most frequently colonized.

The inanimate environment of an intensive care unit as a potential source of nosocomial bacteria: evidence for long survival of Acinetobacter calcoaceticus.

Environmental surface and personnel hand impression cultures were obtained during 13 sampling periods in the University of Virginia Pediatric Intensive Care Unit to document potential reservoirs of nosocomial pathogens. In 78 environmental cultures Staphylococcus aureus was found eight times and gram-negative bacilli ten times. The patient chart cover was the most commonly contaminated surface. Acinetobacter calcoaceticus was found in five of ten cultures positive for gram-negative bacilli. Thirty of 59 hand cultures were positive for S aureus and gram-negative bacilli; nurses and residents had both, respiratory therapists only gram-negative bacilli, and A calcoaceticus was the most commonly isolated bacterium of potentially nosocomial significance (14/30). Laboratory investigation of bacterial survival revealed that gram-negative bacilli survived on a dry formica surface from a few hours up to three days but Acinetobacter survived up to 13 days. Since A calcoaceticus has been implicated in many nosocomial infections, its long survival on a dry surface may be an additional factor in its transmission in hospitals and suggests that more attention be paid to environmental surfaces as a source of significant nosocomial pathogens.

The high rate of blood donor exposure for critically ill neonates.

The purpose of this study was to determine the number and volume of red blood cell (RBC) transfusions and the number of donors a newborn is exposed to during his or her newborn intensive care unit (NICU) stay. On one day at the Medical University of South Carolina (MUSC) and two days at the University of Virginia Hospital (UVH) all babies who had or were receiving RBCs comprised the study group. Patient records were reviewed at discharge. Fifty-two (70%) of the 75 NICU babies had or were receiving RBCs and were enrolled. The average number of RBC transfusions was nine (range 1 to 28, median 7) and the average transfusion volume was 16.5 ml (range 5 to 60) for a total volume of 148 ml transfused during a NICU stay. Each baby was exposed to an average of 6.9 donors (range 1 to 25, median 6.5). The practice of splitting RBC packs to share among different infants and of giving multiple small volume transfusions maximizes donor exposure and transfusion-related infectious risks in this population.

Intravascular catheter colonization in critically ill children.

In a prospective study, cultures were obtained of all intravascular catheters removed from children in an intensive care unit. Of 366 catheters removed from 217 children, 110 (30%) were found to be colonized, most commonly with coagulase-negative staphylococci. Despite the high rate of colonization, there were only nine instances (2%) of catheter-related bacteremia.

Nosocomial infection in neonatal intensive care units.

Nosocomial infection rates in neonatal intensive care units range from 5% to 25%. Both endemic and epidemic infections have been documented, with causative agents including gram-positive cocci, gram-negative bacilli, and viruses. This paper reviews the host, maternal, and environmental factors that influence susceptibility of neonates. Usual preventive measures, such as nursery design, staff apparel, handwashing, employee health, and neonatal care, as well as epidemic control measures to minimize infection risks and limit transmission of infectious organisms, are discussed.

Neonatal intensive care unit bacteremia: emergence of gram-positive bacteria as major pathogens.

An outbreak of nosocomial Staphylococcus epidermidis bacteremia in a neonatal intensive care unit in 1981 was epidemiologically linked to use of Broviac central venous catheters (p = 0.05) and prompted a review of nosocomial infection surveillance data for the previous 6 years, which indicated a striking shift from gram-negative to gram-positive bacteria, especially coagulase-negative staphylococci, as major bloodstream pathogens. In a case-control study with 68 patients with bacteremia between 1979 and September 1981 and 64 control patients, stratified by birth weight, three risk factors were identified: use of Broviac central venous catheters, surgical procedures, and intravenous hyperalimentation. The risk factors varied with the birth weight category. In the group of cases and controls with birth weights less than 1500 gm, only the use of Broviac catheters, which are usually left in the superior vena cava for several weeks, was associated with subsequent bacteremia (odds ratio = 7.1, p = 0.001). These observations, as well as the temporal association of the introduction of Broviac catheters with the subsequent shifts in causes to gram-positive organisms, suggest that the long-term indwelling vascular access devices contributed greatly to the changing pattern.

Paecilomyces lilacinus infection in a child with chronic granulomatous disease.

Chronic granulomatous disease (CGD) is a rare inherited disorder in which neutrophils do not appropriately generate cytotoxic superoxide anion, the respiratory burst, in response to invading bacteria or fungi as a part of normal host defence. We report the case of a child with CGD who had two abdominal wall abscesses caused by Paecilomyces lilacinus, an organism not previously known to cause infections in patients with CGD. The abdominal wall is a location that is rarely associated with Paecilomyces infections. Parenteral amphotericin B eradicated the infection in an immunocompromised child whereas this regimen has heretofore largely been unsuccessful in the treatment of this infection. Paecilomyces species and other fungi from immunocompromised hosts and thought to be laboratory contaminants, need to be carefully investigated for they may become pathogens in this clinical setting.

Alteration of normal gastric flora in neonates receiving ranitidine.

OBJECTIVES: To determine if the administration of ranitidine to neonates leads to an increase in gastric pH to > or = 4 and if this increase in gastric pH correlates with gastric colonization. STUDY DESIGN: 628 pH measurements and 276 gastric cultures were obtained from 86 neonates. Twenty-three patients received ranitidine and 63 patients served as controls. RESULTS: Treated patients had a mean gastric pH of 5.6 compared with a control mean pH of 4.4 (p < 0.0001). Gastric pH was significantly affected by feeding and postnatal age. 54 patients were colonized with pathogenic bacteria and/or yeast (n = 20 treated, n = 34 control). Length of hospitalization (p < 0.0001), increase in gastric pH (p < 0.01), days of antibiotics before culture (p < 0.0001), and ranitidine use (p < 0.0001) were associated with an increased rate of colonization. CONCLUSIONS: The use of ranitidine did lead to a significant increase in gastric pH and with this increase in gastric pH gastric colonization rates increased. No increased frequency of infection was found in ranitidine-treated infants.

Defining unnecessary disinfection procedures for single-dose and multiple-dose vials.

BACKGROUND: Recommendations in the literature conflict on the necessity of disinfecting single-use vials prior to aspiration of fluid. Interventions to disinfect the stopper surface on multiple-dose vials vary considerably. OBJECTIVES: To determine the necessity of alcohol disinfection of the stopper on single-dose vials and to compare povidone-iodine and alcohol versus alcohol-only disinfection of the stopper prior to each needle penetration on multiple-dose vials. METHODS: The rubber stopper surfaces of 100 single-dose vials were cultured for the presence of bacteria. To determine the efficacy of two procedures for disinfection of multiple-dose vials, 87 stopper surfaces routinely disinfected with both povidone-iodine and alcohol were cultured for bacteria. After a change in practice, 100 multiple-dose vials routinely disinfected with alcohol only were cultured for the presence of bacteria. RESULTS: Of the cultures done on single-dose vial stoppers, 99% were sterile. A comparison of the two disinfection techniques for multiple-dose vials revealed that 83 (95%) of the 87 vials prepped with both povidone-iodine and alcohol were sterile, compared with all stoppers disinfected with alcohol only. CONCLUSIONS: This study shows the lack of necessity of any disinfection procedure on the rubber stopper of single-dose vials and the efficacy of alcohol only for disinfecting the stopper of multiple-dose vials.