To determine the attributable risk of death due to catheter-related septicemia (CRS) in critically ill patients when taking into account severity of illness during the intensive-care unit (ICU) stay but before CRS.

Pairwise-matched (1:2) exposed-unexposed study.

10-bed medical-surgical ICU and an 18-bed medical ICU.

Patients admitted to either ICU between January 1, 1990, and December 31, 1995, were eligible. Exposed patients were defined as patients with CRS; unexposed controls were selected according to matching variables.

Matching variables were diagnosis at ICU admission, length of central catheterization before the infection, McCabe Score, Simplified Acute Physiologic Score (SAPS) II at admission, age, and gender. Severity scores (SAPS II, Organ System Failure Score, Organ Dysfunction and Infection Score, and Logistic Organ Dysfunction System) were calculated four times for each patient: the day of ICU admission, the day of CRS onset, and 3 and 7 days before CRS. Matching was successful for 38 exposed patients. Statistical analysis was based on nonparametric tests for epidemiological data and on Cox's models for the exposed-unexposed study, with adjustment on matching variables and prognostic factors of mortality.

CRS complicated 1.17 per 100 ICU admissions during the study period. Twenty (53%) of the CRS cases were associated with septic shock. CRS was associated with a 28% increase in SAPS II. Crude ICU mortality rates from exposed and unexposed patients were 50% and 21%, respectively. CRS remained associated with mortality even when adjusted on other prognostic factors at ICU admission (relative risk [RR], 2.01; 95% confidence interval [CI95], 1.08-3.73; P=.03). However, after adjustment on severity scores calculated between ICU admission and 1 week before CRS, the increased mortality was no longer significant (RR, 1.41; CI95, 0.76-2.61; P=.27).

CRS is associated with subsequent morbidity and mortality in the ICU, even when adjusted on severity factors at ICU admission. However, after adjustment on severity factors during the ICU stay and before the event, there was only a trend toward CRS-attributable mortality. The evolution of patient severity should be taken into account when evaluating excess mortality induced by nosocomial events in ICU patients.

To assess the relative importance of risk factors for surgical-site infections (SSIs) in orthopedic patients and thereby determine which risk factors to monitor in the national surveillance of SSI in The Netherlands.

Reanalysis of data on SSI and associated risk factors from two surveillance projects on nosocomial infections, carried out in 1992 and 1993 in The Netherlands: Project Surveillance Nosocomial Infections in the region of Utrecht (PSZU) and the first Project Surveillance Surgical Wound Infections (SWIFT-1). Odds ratios (ORs) were calculated for age, gender, preoperative stay, and the number of operations. In addition, in PSZU, other nosocomial infections, and, in SWIFT-1, prophylactic antibiotics, acute surgery, and wound contamination were studied.

The study was confined to hospitalized orthopedic patients (PSZU, 4,872; SWIFT-1, 6,437).

In PSZU, the following ORs were significant in a multivariate model: age 0-44 years, 1.0; 45-64 years, 1.6; 65-74 years, 4.7; and 75-99 years, 6.0. For a preoperative stay over 4 days, the OR was 3.3 (95% confidence interval [CI95], 2.5-4.0), and for multiple surgery, 2.5 (CI95, 1.9-3.0). For females, the OR was 0.8 (not significant). The same model applied to SWIFT-1 gave similar ORs. Adjustment for additional nosocomial infections (PSZU) decreased the ORs for ages over 65 years remarkably. The OR for additional nosocomial infections in patients under 65 years of age was 15.6 (CI95, 4.3-57.4). Adjustment for prophylactic antibiotics, acute surgery, and wound-contamination class (SWIFT-1) did not influence the ORs of the original model, but showed that wound-contamination class was an important risk factor.

Age, additional nosocomial infections, wound-contamination class, preoperative stay, and the number of operations were identified as important risk factors for SSI in Dutch orthopedic patients.

To determine the attributable hospital stay and costs for nosocomial methicillin-sensitive Staphylococcus aureus (MSSA) and methicillin-resistant S aureus (MRSA) primary bloodstream infections (BSIs).

Pairwise-matched (1:1) nested case-control study.

University-based tertiary-care medical center.

Patients admitted between December 1993 and March 1995 were eligible. Cases were defined as patients with a primary nosocomial S aureus BSI; controls were selected according to a priori matching criteria.

Length of hospital stay and total and variable direct costs of hospitalization.

The median hospital stay attributable to primary nosocomial MSSA BSI was 4 days, compared with 12 days for MRSA (P=.023). Attributable median total cost for MSSA primary nosocomial BSIs was $9,661 versus $27,083 for MRSA nosocomial infections (P=.043).

Nosocomial primary BSI due to S aureus significantly prolongs the hospital stay. Primary nosocomial BSIs due to MRSA result in an approximate threefold increase in direct cost, compared with those due to MSSA.

To help define the scope of nosocomial legionnaire's disease (LD) and to assess use of recommended diagnostic methods and transmission control practices.

We surveyed 253 hospitals participating in the National Nosocomial Infections Surveillance (NNIS) System. The anonymous survey included questions about episodes of nosocomial LD, environmental sampling practices, maintenance of hospital water systems, and diagnostic techniques.

Of 192 hospitals that responded, 29% reported at least one episode of nosocomial ID from 1990 through 1996, and 61% of these reported at least two episodes. Of 79 hospitals with transplant programs, 42% reported nosocomial LD, compared with 20% of hospitals without transplant programs. Environmental sampling had been conducted by 55% of hospitals, including 79% of those reporting nosocomial LD. Legionella were isolated in 34% that sampled potable water and 19% that sampled cooling system reservoirs. Supplemental potable-water decontamination systems were installed in 20% of hospitals. Only 19% routinely performed testing for legionellosis among patients at high risk for nosocomial LD.

Nosocomial LD is relatively common among NNIS hospitals, especially those performing organ transplants. Environmental sampling for Legionella is a common practice among NNIS hospitals, and Legionella often are isolated from sampled hospital cooling towers and hospital potable-water systems. Hospitals have responded to suspected nosocomial LD infection with a variety of water sampling and control strategies; some have not attempted to sample or decontaminate water systems despite identified transmission.

To examine the impact of a new policy to ensure appropriate use of vancomycin in a 461-bed tertiary-care hospital.

We instituted a policy that allowed physicians to prescribe vancomycin but that required them to complete a vancomycin continuation form and document that use conformed to Hospital Infection Control Practices Advisory Committee (HICPAC) guidelines if they wished to continue the drug beyond 72 hours. Vancomycin was stopped automatically at 72 hours if use was not consistent with guidelines, if an infectious diseases consultant did not approve the drug, or if the form was not completed. A pharmacist and infectious diseases specialist monitored use of vancomycin prospectively and interacted with prescribers when indicated. Educational efforts were limited to printing the HICPAC guidelines on the form and providing information about the policy in a newsletter. Patterns of prescribing and the economic impact of the form were evaluated over a 6-month period.

Only 29% to 48% of vancomycin orders initially met HICPAC guidelines, but 77% to 96% of use was appropriate after 72 hours when the form was used. Inappropriate surgical prophylaxis, empirical therapy of intensive-care unit and transplant patients, and therapy for inadequately documented coagulase-negative staphylococcal infections remained problems. Vancomycin use fell from a mean of 136 (±52) g/1,000 patient days in the 12 months before the form to 78 (±22) g/1,000 patient days in the 9 months after institution of the form (P<.05). Net vancomycin acquisition costs and costs of ordering vancomycin serum levels fell by $357 and $19 per 1,000 patient days, respectively (P<.05). This represented annualized saving of approximately $47,000 in drug and monitoring costs. No adverse patient outcomes were seen as a result of the program.

A vancomycin continuation form can decrease inappropriate vancomycin use and may save money. Additional educational efforts may be required to increase compliance with HICPAC guidelines during initial prescribing.

To evaluate the risk of tuberculosis (TB) transmission to patients potentially exposed to two healthcare providers who worked in outpatient settings for several weeks prior to being diagnosed with acid-fast bacilli smear-positive pulmonary TB.

Potentially exposed patients were notified by letter and television reports of the recommended evaluation for TB infection or disease and availability of free screening at the hospital. Prevalence of infection in the screened patients and the incidence rate of TB over the subsequent 2 years were compared to those of a control group of unexposed outpatients.

An urban inner-city hospital.

1,905 patients with potential exposure to the ill healthcare workers; 487 (25%) presented for evaluation. Controls consisted of 951 unexposed patients.

361 potentially exposed patients had their tuberculin test read; 97 (27%) had a purified protein derivative ≥10 mm. In the comparison group, 148 (25%) of 600 with test readings had a ≥10-mm reaction (risk ratio, 1.18; 95% confidence interval, 0.86-1.60). In multivariate analysis, male gender, non-white race, and older age were significantly associated with a positive tuberculin test; exposure was not. No TB cases were identified during screening. Two years after the exposure, 7 TB cases had been reported to the state registry among 1,905 potentially exposed patients (184 cases/100,000 person-years) versus 4 cases in the comparison group of 951 (210 cases/100,000 person-years).

Evaluation of patients exposed to health-care workers with TB disease in ambulatory settings of an inner-city hospital revealed no evidence of transmission of Mycobacterium tuberculosis due to the exposure.