To develop a new, simple, and practical risk index for patients undergoing coronary artery bypass graft (CABG) surgery, to develop a preoperative risk index that is predictive of surgical-site infection (SSI), and to compare the new risk indices with the National Nosocomial Infections Surveillance (NNIS) System risk index.

Potential risk factor and infection data were collected prospectively and analyzed by multivariate analysis. Two new risk indices were constructed and then compared with the NNIS System risk index for predictive power for SSI.

Alfred Hospital is a 350-bed, university-affiliated, tertiary-care referral center. The cardiothoracic unit performs approximately 650 CABG procedures per year.

All patients undergoing CABG surgery within the cardiothoracic unit at Alfred Hospital between December 1, 1996, and September 29, 2000, were included.

Potential risk factor data were complete for 2,345 patients. There were 199 SSIs. Obesity (odds ratio [OR], 1.78; 95% confidence interval [CI95], 1.24 to 2.55), peripheral or cerebrovascular disease (OR, 1.64; CI95, 1.16 to 2.33), insulin-dependent diabetes mellitus (OR, 2.29; CI95, 1.15 to 4.54), and a procedure lasting longer than 5 hours (OR, 1.75; CI95,1.18 to 2.58) were identified as independent risk factors for SSI. With the use of a different combination of these risk factors, two risk indices were constructed and compared using the Goodman-Kruskal nonparametric correlation coefficient (G). Risk index B had the highest G value (0.3405; CI95, 0.2245 to 0.4565), compared with the NNIS System risk index G value (0.3142; CI95, 0.1462 to 0.4822). The G value for risk index A constructed from preoperative variables only, was 0.3299 (CI95 0.2039 to 0.4559).

Two new risk indices have been developed. Both indices are as predictive as the NNIS System risk index. One of the new risk indices can also be applied preoperatively.

Respiratory isolation for 90% of individuals with acid-fast bacillus (AFB)-smear–positive tuberculosis (TB) is a recommended performance indicator in recent Infectious Diseases Society of America and Centers for Disease Control and Prevention guidelines. However, compliance with respiratory isolation reported from multiple centers in the United States and Europe falls short of that goal.

To identify missed clues in TB patients who are not appropriately isolated.

Retrospective survey.

A 900-bed voluntary hospital.

All patients with AFB-smear–positive TB admitted between January 1995 and December 1999 who were not appropriately isolated.

There were 173 TB cases admitted, including 106 with pulmonary TB. AFB smears were positive in 82 cases; 24 (29%) of these were not appropriately isolated. During the study period, the number of TB cases declined, but the proportion of appropriately isolated patients did not change. Most isolation failure cases were men (median age, 45.5 years); 21 of these patients were black, 2 were Hispanic white, and 1 was Asian, but none was non-Hispanic white. All isolation failure cases had at least one characteristic predictive of TB that could have been elicited at admission (eg, abnormal chest radiograph findings consistent with TB, fever, weight loss, a history of TB, a positive result on tuberculin skin test, hemoptysis, and human immunodeficiency virus infection).

Consistent with experiences at other hospitals, we found that the rate of isolation failure remained unchanged despite an overall decline in TB cases. In our experience, almost all isolation failures could be avoided by careful review of the history, physical examination, and chest radiograph for characteristics classically considered predictive of TB. (Infect Control Hosp Epidemiol 2002;23:595-599).

To study the role of amoeba-associated alpha Proteobacteria as infecting agents in intensive care units (ICUs).

Amoeba-associated alpha Proteobacteria were isolated from water samples taken from ICU taps and general hospital reservoir tanks using an amoebal co-culture procedure. Isolates were identified by 16S rDNA gene sequence comparison, and one isolate of each species was used as an antigen in a microimmunofluorescence assay to test the sera of the patients in the ICUs and compare them with those of control subjects.

The four university hospitals in Marseilles, France.

The sera of 85 patients in the ICUs with nosocomial pneumonia were tested.

We isolated 64 bacterial strains that were identified as Afipia species or close relatives within the Rhizobiaceae subgroup of alpha Proteobacteria. These bacteria were assigned to 8 different species. Eleven patients were found to have an elevated antibody titer to either Afipia genospecies 1, or 3 still unnamed bacteria. No specific antibodies were detected in 100 control subjects (P<.01).

These preliminary results support the hypothesis that ICU patients are exposed to amoeba-associated alpha Proteobacteria.

When the incidence of methicillin-susceptible Staphylococcus aureus (MSSA) infection or colonization increased in our neonatal intensive care unit (NICU), we sought to further our understanding of the relationship among colonization with MSSA, endemic infection, and clonal spread.

A retrospective cohort study was used to determine risk factors for acquisition of a predominant clone of MSSA (clone “B”).

A 45-bed, university-affiliated, level III-IV NICU.

Infants hospitalized in the NICU from October 1999 to September 2000.

Infection control strategies included surveillance cultures of infants, cohorting infected or colonized infants, contact precautions, universal glove use, mupirocin treatment of the anterior nares of all infants in the NICU, and a hexachlorophene bath for infants weighing 1,500 g or more.

During the 1-year study period, three periods of increased incidence of MSSA colonization or infection, ranging from 6.4 to 13.5 cases per 1,000 patient-days per month, were observed. Molecular typing using pulsed-field gel electrophoresis demonstrated two predominant clones, clone “B” and clone “G,” corresponding to two periods of increased incidence. Multivariate analysis demonstrated that length of stay (OR, 1.035; 95% confidence interval [CI95], 1.008 to 1.062; P = .010) (increased risk per day) and the use of H2-blockers (OR, 20.44; CI95, 2.48 to 168.26; P = .005) were risk factors for either colonization or infection with clone “B,” and that the use of peripheral catheters was protective (OR, 0.06; CI95, 0.01 to 0.43; P = .005).

Control of MSSA represents unique challenges as colonization is expected, endemic infections are tolerated, and surveillance efforts generally focus on multidrug-resistant pathogens. Future studies should address cost-effective surveillance strategies for endemic infections.

To report a program of continuous surveillance of surgical-site infections (SSIs) using basic surveillance methods.

Analysis of routine prospective surveillance data.

Two hospitals in Ireland (300 and 350 beds) that merged and moved to a new 650-bed hospital in 1987.

59,335 surgical sites of postoperative patients.

Surgical sites were surveyed by one infection control nurse and SSI rates were produced for selected operations and surgical services. The program was conducted in general accordance with the 1999 HICPAC guidelines, but differed in surveillance strategy. Operations were limited to two to three risk classifications, assigned by the infection control nurse.

The overall SSI rate was 4.5%, with 2.4% in clean surgery. Apart from increases in the 3rd, 4th, 13th, and 14th years, rates remained relatively stable during the 16 years. Few significant decreases in SSI rates in surgical services or specific operations were shown, apart from the following: vascular surgery, 8.1% to 5% between the first 8 years and the last 8 years; general surgery services, 9% to 5%, and gynecology, 15.8% to 1.7%, both in the first year compared with in subsequent years; and gastric operations, 21% to 4.3% between the first year and the second year. Organ/space infection was identified in 0.5% of 17,804 operations, including 0.4% meningitis after neurosurgical procedures, 3% graft infections after vascular bypass operations, and 0.2% intra-abdominal infections after abdominal surgery.

With the use of basic principles of surveillance and modest resources, procedure-specific SSI rates were produced, with little significant change during the 16 years. Despite limitations in case-finding, risk stratification, feedback, and surveillance methods, the overall SSI rates were comparable with other published data.

To identify delayed prophylaxis from a preexisting database and strategies to improve performance.

A skilled nursing facility with 14 floors (4 buildings). The “outbreak unit” was a 50- to 60-bed floor.

We performed surveillance during six seasons using one protocol. Prophylaxis was started when influenza was cultured in the building and 10% of residents on the floor had a new respiratory illness within 7 days. We defined delayed prophylaxis as four or more residents on a floor with positive cultures whose specimens had been collected within 5 days before the application of prophylaxis.

We identified 14 examples of delayed prophylaxis. In three, delayed prophylaxis was related to the 3.9-day delay between culture collection and culture report. There was a high degree of commonality among building attack rates within a season. During six seasons, the first case in the last building occurred 27 to 64 days after the first case in the facility. The two seasons with the greatest activity (68 and 154 cases, respectively) began with explosive, multi-floor outbreaks in a single building. The match between the circulating strain and the vaccine was good, except in 1997-1998 when there were seven examples of delayed prophylaxis.

Influenza may involve buildings sequentially with a commonality of building attack rates. Explosive, multi-floor outbreaks early in the season could lead to a lower threshold for prophylaxis within a larger area when initial cases are encountered later in the season. This strategy could have prevented five examples of delayed prophylaxis. Rapid testing of multiple specimens while waiting for culture confirmation could have prevented three examples of delayed prophylaxis. (Infect Control Hosp Epidemiol 2002;23:600-603).

Although many catheter-related blood-stream infections (CRBSIs) are preventable, measures to reduce these infections are not uniformly implemented.

To update an existing evidenced-based guideline that promotes strategies to prevent CRBSIs.

The MEDLINE database, conference proceedings, and bibliographies of review articles and book chapters were searched for relevant articles.

Laboratory-based studies, controlled clinical trials, prospective interventional trials, and epidemiologic investigations.

Reduction in CRBSI, catheter colonization, or catheter-related infection.

The recommended preventive strategies with the strongest supportive evidence are education and training of healthcare providers who insert and maintain catheters; maximal sterile barrier precautions during central venous catheter insertion; use of a 2% chlorhexidine preparation for skin antisepsis; no routine replacement of central venous catheters for prevention of infection; and use of antiseptic/antibiotic-impregnated short-term central venous catheters if the rate of infection is high despite adherence to other strategies (ie, education and training, maximal sterile barrier precautions, and 2% chlorhexidine for skin antisepsis).

Successful implementation of these evidence-based interventions can reduce the risk for serious catheter-related infection.

To identify the routes of transmission during an outbreak of infection with hepatitis C virus (HCV) genotype 2a/2c in a hemodialysis unit.

A matched case-control study was conducted to identify risk factors for HCV seroconversion. Direct observation and staff interviews were conducted to assess infection control practices. Molecular methods were used in a comparison of HCV infecting isolates from the case-patients and from patients infected with the 2a/2c genotype before admission to the unit.

A hemodialysis unit treating an average of 90 patients.

A case-patient was defined as a patient receiving hemodialysis with a seroconversion for HCV genotype 2a/2c between January 1994 and July 1997 who had received dialysis in the unit during the 3 months before the onset of disease. For each case-patient, 3 control-patients were randomly selected among all susceptible patients treated in the unit during the presumed contamination period of the case-patient.

HCV seroconversion was associated with the number of hemodialysis sessions undergone on a machine shared with (odds ratio [OR] per additional session, 1.3; 95% confidence interval [CI95], 0.9 to 1.8) or in the same room as (OR per additional session, 1.1; CI95, 1.0 to 1.2) a patient who was anti-HCV (genotype 2a/2c) positive. We observed several breaches in infection control procedures. Wetting of transducer protectors in the external pressure tubing sets with patient blood reflux was observed, leading to a potential contamination by blood of the pressure-sensing port of the machine, which is not accessible to routine disinfection. The molecular analysis of HCV infecting isolates identified among the case-patients revealed two groups of identical isolates similar to those of two patients infected before admission to the unit.

The results suggest patient-to-patient transmission of HCV by breaches in infection control practices and possible contamination of the machine. No additional cases have occurred since the reinforcement of infection control procedures and the use of a second transducer protector.

To describe an outbreak of mediastinitis in heart transplant recipients.

Retrospective and contemporaneous cohort study.

Urban tertiary-care university hospital with a large cardiac transplantation program.

Heart transplant recipients.

Modifications of donor harvest technique; procedures aimed at decreasing skin and mucosal bacterial colonization; strict aseptic technique in the intensive care unit; and aggressive policing of established infection control practices.

In April 1999, mediastinitis rates among heart transplant recipients increased abruptly from a baseline of 6 cases per 100 procedures to sequential quarterly rates of 22, 31, and 50 cases per 100 procedures, whereas infection rates in other cardiac operations were unchanged. Bacteria causing these infections were multidrug-resistant "nosocomial" organisms. The epidemic occurred 2 months after a change in the United Network for Organ Sharing organ allocation algorithm. This change resulted in an increase in the duration of preoperative hospitalization from a median of 52 to 79 days (P = .008) and may have promoted prolonged hospitalization of patients with high illness severity. Aggressive multidisciplinary interventions were temporally associated with a return to preoperative mediastinitis rates without changing length of hospitalization prior to transplantation.

Changes in organ allocation for transplant that prolong waiting time in the hospital and alter illness acuity may lead to increased rates of postoperative infection. Measures to limit bacterial colonization may be a helpful countervailing strategy.

To describe and investigate the cause of an outbreak of 10 cases of nosocomial invasive infection with Aspergillus flavus in a hematologic oncology patient care unit.

A retrospective cohort study.

The hematologic oncology unit of a comprehensive cancer center.

Ninety-one patients admitted to the hematologic oncology service between January 1 and December 31,1992, for 4 or more consecutive days were included in the study.

Ten (18%) of 55 patients admitted from July to December 1992 were diagnosed as having invasive aspergillosis compared with 0 (0%) of 36 patients admitted from January to June 1992 to the same patient care units. Patient characteristics, mortality rate, autopsy rate, and admitting location did not change significantly during the course of the year to result in a sudden increase in the number of aspergillosis cases. The source of the outbreak was the high counts of Aspergillus conidia determined from air sampling in the non–bone marrow transplant wing during the outbreak. After high-efficiency particulate air (HEPA) filters were installed as an infection control measure, there were only two additional cases of nosocomial aspergillosis in the 2 years following the outbreak.

This outbreak occurred among hematologic oncology patients with prolonged granulocytopenia housed in an environment with neither HEPA filters nor laminar air flow units. Our data demonstrate that in the setting of an outbreak of aspergillosis, HEPA filters are protective for highly immunocompromised patients with hematologic malignancies and are effective at controlling outbreaks due to air contamination with Aspergillus conidia.

Occupational exposure to human immunodeficiency virus (HIV) is an important threat to healthcare workers. Centers for Disease Control and Prevention guidelines recommend prompt institution of prophylaxis. This requires (1) immediate prophylaxis after exposure, pending test results that may take more than 24 hours in many hospitals; or (2) performance of a rapid test. The Single Use Diagnostic System (SUDS)® HIV-1 Test is used to screen rapidly for antibodies to HIV type 1 in plasma or serum, with a reported sensitivity of more than 99.9%. We used this test from January 1999 until September 2000, when it was withdrawn from the market following reports claiming a high rate of false-positive results.

We reviewed the results of postexposure HIV testing during 21 months.

A total of 884 SUDS tests were performed on source patients after occupational exposures (883 negative results, 1 reactive result). The results of repeat SUDS testing on the reactive specimen were also reactive, but the results of enzyme immunoassay and Western blot testing were negative. A new specimen from the same patient showed a negative result on SUDS testing. This suggested a specificity of 99.9%. In the 4 months after SUDS testing was suspended, there was 1 false-positive result on enzyme immunoassay for 1 of 132 source patients (presumed specificity, 99.2%).

Use of the SUDS test facilitated rapid and accurate evaluation of source specimens, obviating unnecessary prophylaxis.

To describe compliance with and the safety and prophylactic efficacy of zanamivir among patients at risk of developing influenza-related complications after exposure to both influenza A and B viruses.

Nonrandomized trial using both historical and contemporaneous controls from ward populations within the same facility.

A 547-bed urban hospital providing complex continuing care and rehabilitation.

Fifty patients on a single ward concomitantly exposed to both influenza A and B during an influenza outbreak.

All patients were screened for symptoms of influenza and nasopharyngeal swabs were sent for viral culture prior to treatment (two oral inhalations [2 × 5 mg] twice daily for 5 days) or prophylaxis (two oral inhalations [2 × 5 mg] once daily for 14 days) with zanamivir. Patient symptoms, compliance, and drug tolerance were monitored daily. Demographic data, comorbidities, vaccination status, and functional status (Katz score) were collected for all patients.

The mean age of the patients was 70.6 years (standard deviation, ± 16.4 years). Ninety-four percent had two or more comorbidities, 71% were dependent in two or more activities of daily living, 63% had been vaccinated for influenza, and 82% had received amantadine. Zanamivir was well tolerated, with 93% of the patients completing their course. The efficacy for preventing symptomatic infection was 82% and 91% (95% confidence interval, 62% to 98%) based on historical and contemporaneous controls, respectively.

These data suggest that zanamivir can be used safely and effectively as a prophylactic agent in the management of influenza outbreaks in a complex continuing care population with multiple comorbidities. (Infect Control Hosp Epidem 2002;23:604-608).