[Surgical volume and mortality due to intraoperative critical incidents at Japanese Society of Anesthesiologists certified training hospitals: an analysis of the annual survey in 2002].

Background: We have previously showed that surgical volume affects mortality due to intraoperative critical incidents among patients undergoing cardiac surgery, the surgery with the highest risk, using data obtained by the annual survey in 2001 conducted by the Japanese Society of Anesthesiologists (JSA). In this study, we investigated whether surgical volume affects mortality due to intraoperative critical incidents independent of the surgical site.
Methods: We investigated this relationship using data obtained from the 2002 annual survey conducted by the Subcommittee on Surveillance of Anesthesia-related Critical Incidents, JSA. Between January 1, 2002 and December 31, 2002, 1,987,988 patients were registered from 704 training hospitals certified by the JSA. Intraoperative critical incidents occurred in 2,844 patients. Of these, 804 patients died within 7 postoperative days. The overall mortality was 4.61 per 10,000 anesthetics. Hospitals were divided into 5 groups according to their annual surgical cases: Group A, fewer than 1,000 (62 hospitals); Group B, 1,000-1,999 (204 hospitals); Group C, 2,000-3,999 (288 hospitals); Group D, 4,000-5,999 (110 hospitals); Group E, more than 6,000 (40 hospitals). Hospitals were also divided into 2 groups according to mortality: Group 1, under 20.00 per 10,000 anesthetics (672 hospitals); Group 2, equal to or higher than 20.00 per 10,000 anesthetics (32 hospitals). Total number of deaths in Group 2 was 158. Mortality was expressed as the mean (95% confidence interval). Statistical analysis was performed using chi-square test and Fisher test. A p value of <0.05 was considered significant.
Results: The mortality rates in Groups A-E were 14.89 (8.48-21.3), 3.86 (3.05-4.67), 3.88 (3.19-4.57), 4.04 (3.20-4.88), and 3.12 (2.19-4.05) per 10,000 anesthetics, respectively. Average surgical cases and mortality in Group 1 were 2,789 (2,775-3,002) and 3.24 (2.90-3.58), respectively, while those in Group 2 were 1,672 (1,243-2,101) and 22.18 (30.58-45.94), respectively. If all patients in Group 2 (n=53,509) had been treated in the hospitals of Group 1, 139-143 deaths might have been avoided.
Conclusions: Surgical volume was shown to affect mortality independent of the surgical site. Hospitals with low surgical volume should pay significant attention to improving surgical outcomes. These results also suggest that centralization or regionalization should be discussed from the perspective of socio-economical problems as well as patient safety.