[Recent changes in the incidence of life-threatening events in the operating room: JSA surveys between 2001 and 2005]

In Japan, the number of medical staff charged with criminal liability has been increasing since 2000, and this medico-legal trend seems to be promoting topics of medical risk management in government, academic meetings and individual hospital. A survey conducted by the Japanese Society of Anesthesiologists (JSA) has been widely accepted among JSA-certified training hospitals, and its denominator has exceeded one million since 2001. The purpose of this investigation is to examine changes in the incidence of life-threatening events in the operating theater between 2001 and 2005 based on the data of the surveys.

JSA has conducted annual surveys of life-threatening and neurological events in the operating theater by sending and collecting confidential questionnaires to all JSA certified training hospitals. Cases of life-threatening events between 2001 and 2005 were analyzed. The recovery rates ranged from 76.2% (in 2005) to 91.6% (in 2002), and the annual patient numbers available for analysis ranged from 1,051,245 (in 2005) to 1,367,790 (in 2003) during the study period. The patients with ASA PS 1 or 2 were classified as having good physical status, and those with ASA PS 3 or 4 were classified as having poor physical status. Because mortalities (within 7 postoperative days) are more common in patients with poor physical status, in emergency patients, in neonate, in the elderly, and in patients undergoing cardiovascular surgery, the mortality rate in these patients were investigated. The recovery rate from cardiac arrest without any sequelae was also investigated. The causes of events were classified as follows: totally attributable to anesthetic management (AM), mainly to intraoperative pathological events (IP), to preoperative co-morbidity (PC), and to surgical management (SM). IP consists of pulmonary thromboembolism, acute coronary syndrome, anaphylaxis and so on. The incidence of cardiac arrest and mortality are indicated per 10,000 patients. Odds ratio and 95% confidential interval are shown in comparison with the incidence in 2001 to that in 2005.

The incidences of cardiac arrest were 6.12 in 2001, 5.79 in 2002, 5.89 in 2003, 5.09 in 2004, and 4.24/10,000 patients in 2005, respectively (odds ratio 0.69; CI 0.62-0.78). The incidences of death within 7 postoperative days due to intraoperative life-threatening events were 6.41 in 2001, 6.31 in 2002, 6.61 in 2003, 5.88 in 2004, and 4.91/10,000 patients in 2005, respectively (OR 0.77; CI 0.69-0.85). The incidences of death in patients with poor physical status (from 35.48 to 26.87/10,000 patients; OR 0.76; CI 0.66-0.86), in emergency patients (from 37.25 to 30.55/10,000 patients; OR 0.82; CI 0.72-0.93), in neonates (from 70.09 to 31.70/10,000 patients; OR 0.45; CI 0.22-0.91) and in the elderly (from 11.03 to 8.75/10,000 patients; OR 0.79; CI 0.68 to 0.92) decreased. The incidence of death in patients undergoing cardiovascular surgery ranged between 61.22 and 76.88/10,000 patients, and has not shown any significant decline. The incidences of death due to IP (from 0.65 to 0.42/10,000 patients; OR 0.64; CI 0.44-0.92), PC (from 4.14 to 3.30/10,000 patients; OR 0.80; CI 0.70-0.91) and SM (from 1.49 to 1.02/10,000 patients; OR 0.68; CI 0.54-0.87) decreased. However, the incidence of death due to AM ranged between 0.07 and 0.11/10,000 patients, and has not shown any significant decline partly because of the small number of deaths from this cause. Although recent trends in life-threatening events seemed to be favorable, the recovery rate from cardiac arrest decreased from 40.3% in 2001 to 30.7% in 2005 (OR 0.66; CI 0.51-0.84).

The incidence of life-threatening events in the operating room and mortality due to these events seemed to have decreased during the recent five years, probably because of progress in risk management in JSA-certified training hospitals. The decrease was obvious in the recent two years. However, the results should be interpreted cautiously, because the response rate to the questionnaire in 2005 was the lowest. To confirm this trend, we should perform a follow-up survey for 2006 and continue the survey. The reasons for the deterioration in the recovery rate from cardiac arrest should also be examined.