Abstract 1598 Clinical Research in Developing Countries Poster Symposium, Sunday, 5/2

Objective: compare the rates of central line-associated bloodstream infection (CL-BSI) and ventilator-associated pneumonia (V-PNEU) of an University NICU with the rates published in July 1998 by the National Nosocomial Infections Surveillance (NNIS) system of the CDC-Atlanta.

Method: from 1/97 to 6/98, CL-BSI and V-PNEU defined by NNIS system (CDC, 1994) were studied in a Brazilian University Hospital NICU. The following data was collected for 4 BW categories: number of patient-days, of central line-days, and of ventilator-days. The device utilization ratios - UR (for central line: CL-UR and for ventilator: V-UR) had the number of specific devices as numerator and of patient-days as denominator. CL-BSI was obtained by the number of blood stream infections divided by central line-days, and V-PNEU by the number of pneumonias divided by ventilator-days. NICU data was compared to July/1998 NNIS rates at percentile 50 by Z-Test (CDC, 1993). Z>1.6 or Z<-1.6 was considered significant (*).

Results: are shown in the table.

Table 1 No caption available
Full size table

Conclusion: although NICU device utilization ratios were similar to NNISS data in low birthweight newborn infants, the central line-associated bloodstream infection rates for all birthweight categories and the ventilator-associated pneumonia rates for infants > 1500g were higher than percentile 50 NNIS rates. The use of NNIS system allows an effective comparison between American and other countries neonatal nosocomial infection rates, and should help to outline strategies to reduce NICU infection rates in developing countries.