The impact of a bedside hand sanitizer program on surgical site infections in neurosurgical patients in Vietnam

BACKGROUND: There are no data regarding hand hygiene practices in hospitalized patients in Vietnam. We conducted an intervention program to assess the impact of an alcohol-chlorhexidine-based hand sanitizer on surgical site infection (SSI) rates among neurosurgical (NS) patients in Ho Chi Minh City. METHODS: We conducted a prospective before and after study in NS patients admitted to Cho Ray Hospital between July 11 and August 15, 2000 (before), and July 14 and August 18, 2001 (after). The same two NS wards (A and B) were included in each study period. An ethanol (70%) and chlorhexidine gluconate (0.5%) hand sanitizer (Microshield® Handrub, Johnson and Johnson Medical, Arlington, Texas) was introduced on Ward A in September 2000. Training on proper use and installation of bedside sanitizer bottle holders was provided to Ward A. No intervention was made on Ward B. CDC surveillance definitions were used. Patient data on SSIs were collected on standardized forms and entered into EpiInfo version 6.04 (CDC) and analyzed using Stata (Stata Corp LP, College Station, Texas). RESULTS: Of the 786 total patients enrolled, 377 were enrolled during the before period (Ward A = 156, Ward B = 221), with 409 during the after period (Ward A = 159, Ward B = 250). The cohorts were similar except for a higher median ASA score during the after period (1 to 2, p < 0.001). On Ward A, the SSI rate between the two periods was reduced by 54% (8.3% to 3.8%; p = 0.09). Superficial SSIs were eliminated after the intervention (p = 0.007). On Ward B, the SSI rate increased by 28% (7.2% to 9.2%; p = 0.4). In patients without SSI, median post-operative length of stay and length of antimicrobial use were reduced for Ward A (both 8 to 6 days, p < 0.001), but not for Ward B; there were no differences among patients with SSI in either period. Mortality in patients without SSI on Ward A increased from 1.4% to 7.8% (p = 0.009), but was unchanged on Ward B. However, mortality in SSI patients on Ward B increased from 0 to 17% (p = 0.08), whereas there was no difference on Ward A. CONCLUSIONS: This study demonstrates that hand sanitizer use can 1) reduce SSI rates in Vietnamese NS patients, with particular impact on the incidence of superficial SSIs, and 2) reduce overall post-operative length of stay and antimicrobial use. Increases in mortality in patients without SSI may be related to pre-operative severity of illness. This study illustrates the impact alcohol-based hand hygiene products can have on reducing SSIs in developing countries.