To study the relationship between serum high-density lipoprotein cholesterol (HDL-C), total serum cholesterol, and nosocomial infection in patients undergoing general surgery.

Prospective cohort study, with an extended follow-up to 1 month after hospital discharge.

The general surgery service of a tertiary hospital.

Nosocomial infection, mainly surgical-site infection (SSI), urinary tract infection, respiratory tract infection (RTI), and bacteremia.

1,267 surgery patients aged 10 to 92 years.

182 subjects acquired 194 nosocomial infections, a cumulative incidence of 14.5%; most (116, 62.3%) were postoperative wound infections. There was an increase in infection risk at low levels of HDL-C, and both low and high total cholesterol levels. After adjusting simultaneously for several confounders, including total cholesterol, low levels of HDL-C (≤20 mg/dL) yielded an odds ratio (OR) of 2.2 (95% confidence interval [CI95], 0.6-7.9) for SSI and an OR of 10.3 (CI95, 0.7-151.5) for RTI. Otherwise, no trend was observed between HDL-C levels and infection risk, and no increased risk of nosocomial infection was observed for HDL-C values in the range of 21 to 49 mg/dL. Serum cholesterol showed a U-shaped relationship with nosocomial infection risk. Both low levels (below 102 mg/dL) and high levels (above 290 mg/dL) of total cholesterol were associated with a higher risk of SSI (mainly those caused by gram-negative bacteria) and RTI in comparison with the reference group (139-261 mg/dL).

Serum HDL-C and total cholesterol seem to be associated with the risk of nosocomial infection in surgical patients

To study the spread of extended spectrum ß-lactamase-producing, but aminoglycoside-susceptible, Klebsiella pneumoniae strains in our hospital over an 8-month period, by using two genotypic markers.

Ribotyping (using two endonucleases) and randomly amplified polymorphic DNA analysis (RAPD; using two different 10-mer primers) were applied to the epidemiological typing of clinical K pneumoniae isolates from stools, ileal fluid, or urine of hospitalized children.

The surgical intensive care ward (S1: 9 patients, 17 isolates), surgical unit (S2: 2 patients, 2 isolates), and gastroenterology ward (GE: 1 patient, 1 isolate) of the Robert Debre Hospital of Paris, France.

Ribotyping of the 20 clinical isolates, the type strain of the species, and two epidemiologically unrelated isolates with EcoRI and HindlII revealed 6 and 5 different patterns, respectively. Six ribotypes were identified by using these two enzymes. RAPD generated 6 distinct patterns, in complete agreement with ribotyping. Our geno-typic results showed that 11 patients from wards Sl, S2, and GE harbored genotypically related strains, suggesting nosocomial transmission and cross-colonization between and within the three wards.

Ribotyping and RAPD appear to be reliable methods for distinguishing K pneumoniae strains. The spread of one strain of K pneumoniae in different units of our hospital was demonstrated by both methods. However, RAPD has the advantage of simplicity and rapidity conferred by polymerase chain reaction.

To describe the characteristics of and trends in nosocomial infection among human immunodeficiency virus (HIV)-infected patients.

Multicenter prospective cohort study.

HIV-infected patients were enrolled at time of first inpatient admission at five Veterans' Administration Medical Centers (VAMCs).

As of March 1995, 2,541 patients with 6,625 inpatient admissions had been monitored in the five VAMCs. A total of 530 nosocomial infections were detected using standard Centers for Disease Control and Prevention definitions. Overall distribution by infection site was 31% for primary bloodstream infections (BSIs), 28% for urinary tract infections, 15% for pneumonia, and 26% for all other sites. Of BSIs, 63% were central line-associated bloodstream infections (CLABs). The rate of CLABs per 1,000 central line days was 6.5 (range, 2.3-8.3) for all patients from participating hospitals, similar to the median CLAB rate of 6.0 for patients in medical intensive-care units (ICUs) of National Nosocomial Infections Surveillance (NNIS) System hospitals from January 1990 through September 1994. For ICU-specific CLABs, the rate from hospitals reporting at least one ICU CLAB was 12.7 (range, 12.1-13.1), comparable to the 90th percentile of NNIS hospital medical ICUs (13.1). Staphylococcus aureus, associated with 35% of BSIs, was the most common nosocomial BSI pathogen. Our data demonstrated the following: 13 (10%) of 134 patients with CD4 counts ≥200 cells/mm3 had a CLAB, compared with 61 (6%) of 1,011 patients with CD4 counts <200 cells/mm3, P=.08; the per-day risk of CLABs did not change with increased duration of catheterization (P=.4); and the per-day risk of a temporary (ie, short-term) CLAB was greater than that of a permanent CLAB (P<.001).

The data suggest that HIV-infected patients were at higher risk of acquiring a BSI than were patients in the NNIS population; patients with CD4 counts ≥200 cell/mm3 and temporary central lines were at increased risk for BSI, perhaps reflecting widespread prophylaxis with trimethoprim-sulfamethoxazole among patients with CD4 counts <200 cells/mm3; and, in contrast to most studies, S aureus, not coagulase-negative Staphylococcus, was the most common BSI pathogen.

To show the value of mathematical modeling in simulating the spread of nosocomial pathogens in an intensive-care unit (ICU), to provide a framework for listing available knowledge; to predict the benefits of various control measures; and to supplement the epidemiological assessment of these measures.

Simulated outbreak of a nosocomial pathogen in an ICU, based on a deterministic compartmental model describing both person-to-person spread and indirect spread between patients through staff members.

Simulation of three typical colonization control measures: effective handwashing compliance among staff members, ICU antimicrobial policy, and curtailing ICU admission of colonized patients.

In controlling colonization, effective handwashing compliance reduced staff member colonization, but only moderately limited patient colonization unless the ICU was isolated strictly by curtailing the admission of colonized patients. The impact of antibiotic policy was very slight.

In the field of nosocomial infection, mathematical modeling appears to be a valuable tool that can be used to evaluate the magnitude of the expected effects of control strategies and to guide the selection of the best randomized clinical trials to pursue.

To determine policies at children's hospitals regarding immunizations, annual tuberculosis (TB) screening, and blood or body fluid exposure follow-up, particularly as they apply to physicians.

A three-page survey was sent to infection control practitioners (ICPs) in April 1994 at hospitals affiliated with the National Association of Children's Hospitals and Related Institutions. One follow-up mailing was sent to nonresponding ICPs.

Responses were received from 62 (67%) of 93 ICPs. Thirty-five (66%) of 53 children's hospitals had an immunity policy that applied to medical students, 42 (79%) of 53 to resident physicians, 32 (52%) of 62 to hospital-based physicians, and 18 (29%) of 62 to private or community physicians (who admit patients to one hospital). Physicians were required to show evidence of an annual TB screen at 36 hospitals (58%). Immunity policies or TB screening were provided for the following physician groups: medical students, 13 (21%); resident physicians, 43 (69%); hospital-based physicians, 50 (81%); and private or community physicians, 23 (37%). Infection control practitioners reported that the following diseases had been identified within the past 5 years at their hospitals: measles, 82%; mumps, 40%; rubella, 31%; TB, 94%; hepatitis B, 94%; pertussis, 90%; varicella, 98%; and influenza, 94%. Physicians in these institutions were reported to have contracted the following diseases from patient exposure: measles, hepatitis B, TB, pertussis, varicella, and influenza.

Children's hospitals vary widely in their policies regarding healthcare-worker immunity, and, in many cases, physicians may not be protected from nosocomial transmission of community infections.