[Surveillance of nosocomial infections in a cardiology hospital]. / Vigilancia de infecciones nosocomiales en un hospital de cardiología.

OBJECTIVE: To describe a nosocomial infection surveillance in a cardiology and thoracic diseases hospital in Northern Mexico during its initial months of activity. MATERIAL AND METHODS: Retrospective report of nosocomial infection surveillance performed by the infection control team with patient visits, clinical records, microbiology reports, and direct information obtained by health personnel. General and specific infection rates were analyzed by the chi 2 test and the Student t test. RESULTS: General infection rate was 4.99 during the 17 month period. Monthly infection rates changed significantly during infection control team vacations. By site of infection, surgical wound infection were the most common (1.14), followed by urinary tract infections (1.08), and bacteremia (0.72). However, when we grouped bacteremias and other catheter related infections, this intravascular access related infections were the most frequent (1.73). Gram positive organisms were most commonly isolated in nosocomial infections, except in pneumonias (gram negative organism) and urinary tract infections (Candida spp.). Mortality related to nosocomial infection was high (21.7%), and at least one third of those deaths was directly related to the infection. In hospital stay and cost were also increase with nosocomial infections. CONCLUSIONS: Nosocomial infection is common in interventional cardiology hospitals, producing high mortality and rising costs and length of stay. We need a more effective nosocomial infections surveillance system to eliminate changes in monthly case detection produced by the infection control team vacations, since this under-registry could result in lost of control of nosocomial infections and epidemic outbreak emergence.

Salmonella gastroenteritis outbreak among workers from a tertiary care hospital in Mexico City.

OBJECTIVE: To describe an outbreak of Salmonella gastroenteritis among employees of the National Institute of Nutrition (INNSZ) of Mexico City during July, 1994. METHODS: Employees who developed diarrhea or fever associated with gastrointestinal symptoms starting on July 14th were included for study as well as 50 healthy controls. A questionnaire was applied to all, and they also provided a stool sample, along with other 80 asymptomatic people (included the kitchen workers) in whom only stool culture was done. RESULTS: Ninety-seven employees that ate regularly at the Hospital's cafeteria were affected by the outbreak, and 67 of them (69%) could be evaluated. Most of them were nurses (34%), and handymen (27%). Most common symptoms were abdominal pain (97%), diarrhea (95%), nausea (91%), and fever (89%). Cultures from suspicious food items were all negative, but stool cultures from 10/70 cases were positive for Salmonella enteritidis vs. 0/133 in the controls. The ten S. enteritidis isolates resulted identical either by serotyping and by rapid amplified polymorphic DNA (RAPD) analysis. Cultures from all kitchen employees were negative for S. enteritidis. Breakfast meal on July 14th was associated with the development of gastroenteritis (61/67 cases vs 26/50 controls, p < 0.001), and particularly with an egg-covered meat plate (61/62 vs 13/26 controls, p < 0.0001). CONCLUSIONS: This outbreak was probably caused by eggs contaminated with Salmonella, since no one of the kitchen personnel was found to be an asymptomatic carrier, and the implicated recipe allows for inappropriate cooking. Recommendations to improve cooking procedures must be added to the usual regulations to diminish the frequency of foodborne disease outbreaks in hospitals.

Mycobacterial infection in Mexican AIDS patients.

To describe the characteristics of mycobacterial infection in Mexico, we reviewed records from patients who were seen at the AIDS Clinic of the National Institute of Nutrition in Mexico City from 1983 to 1992. Of 460 AIDS patients, 118 (25.6%) were found to have mycobacterial infections by positive Ziehl-Neelsen stain, culture, or both. Cultures were completed for 66 of the 118 stain-positive specimens. Mycobacterium tuberculosis was the most common species found (n = 13), followed by M. avium complex (n = 12); 21 infections were identified a nonspecific mycobacteria other than tuberculosis (MOTT) and 20 infections were from species other than tuberculosis. Susceptibility testing was performed in only two tuberculosis cases, with one strain showing multidrug resistance. We conclude that mycobacterial infection is common among our AIDS population, and MOTT may be at least as common as M. tuberculosis. Previous reports of the rarity of MOTT could be related to the lack of adequate diagnostic methods in developing countries.

Evaluation of the sensitivity of RT-23 purified protein derivative for determining tuberculin reactivity in a group of health care workers.

Since 1958 the World Health Organization (WHO) has recommended the use of RT-23 (2 tuberculin unit [TU] purified protein derivative [PPD]) instead of the standard 5 TU PPD to establish the prevalence of tuberculosis (TB) around the world. Before starting a hospital program to control the transmission of TB among health care workers at the National Institute of Nutrition (Mexico City), we compared the tuberculin product commonly used in Mexico (RT-23) with that used in the United States (5 TU PPD). In this trial the Mantoux test was performed on 80 health care workers from various areas of the hospital: 5 TU PPD and RT-23 were simultaneously inoculated in the left forearm and in the right forearm, respectively. The test results for both tuberculin products were read 72 hours later. When a 10-mm induration was used as the cutoff for reactivity, the specificity with use of RT-23 was 100%, but the sensitivity was only 57%. However, when a 5-mm induration was used as the cutoff, the sensitivity with use of RT-23 was 90%. Because of these findings, we advise that every country using RT-23 either consider changing the cutoff for reactivity or consider using 5 TU PPD for reassessment of the data on the incidence and prevalence of TB that were obtained with use of RT-23 (the product recommended by WHO).