Background and objective: The use of risk indexes, originally developed in the US for the assessment of SSI risk, is an useful instrument that must be analyzed according to each specific procedure. The addition of other possible SSI risk factors, like the use of perioperative antibiotic prophylaxis, could improve the predictive value of these indexes. The aim of this study was to determine the SSI incidence rate for craniotomy in patients admitted to the Neurosurgical Unit of the Hospital Clinic of Barcelona (Spain), to assess the use of standard NNIS and SENIC indexes, and to assess the possible effect of the addition of a new risk factor (adequate or inadequate use of perioperative antibiotic prophylaxis) to these indexes.
Patients and method: Risk factors for SSI were assessed following common standard definitions and procedures (CDC-NNIS) over a three-year period (1999-2001). NNIS and SENIC risk indexes were calculated. The effect of the addition of a new variable, namely perioperative antibiotic prophylaxis adequate (0 points) or inappropriate/no prophylaxis (1 point) on these indexes (modified indexes NNISa and SENICa) was also assessed. Statistical analysis included both parametric and non-parametric standard tests.
Results: The study included a total of 203 patients undergoing a craniotomy procedure (40% of all neurosurgical procedures). The overall SSI incidence rate was 6.8% (14 patients developed SSI). The cut-off point (75 percentile) for the duration of the procedure was 180 minutes instead of the commonly US reported 240 minutes. Patients who develop SSI had a trend towards having shorter operation times. For those patients in the lower risk groups, the SSI incidence rate was: NNIS (0, 1): 6.9%; SENIC (0, 1): 6.2%. If the modified indexes were used, the SSI incidence rate was: NNISa (0, 1): 4.2%; SENICa (0, 1): 4.9%. When NNIS and SENIC indexes, both standard and modified (NNISa and SENICa), were compared, no statistically significant differences between infected and non-infected patients were observed.
Conclusions: When applied to a health system other than the US, SENIC and NNIS indexes could be useful if adapted to each specific situation and procedure. The added value of a new risk factor (perioperative antibiotic prophylaxis) on standard NNIS and SENIC indexes shows a slight improvement in their prediction rate for SSI in patients undergoing craniotomy, mainly in those patients at lower risk for developing superficial SSI.