Hospital volume and outcomes of cardiothoracic surgery in Japan: 2005-2009 national survey.

OBJECTIVES

To elucidate the relationship between hospital volume and cardiothoracic surgical outcomes in Japan using the annual survey data, obtained between 2005 and 2009, collected by the Committee for Scientific Affairs of the Japanese Association for Thoracic Surgery.

METHODS

The relationship between hospital volume and 30-day mortality was analyzed using a logistic regression model. The empirical Bayes (EB) method was also used to stabilize any large variation resulting from a small sample size. Hospitals, whose lower limit of the EB mortality 95 % confidence interval was above the mean EB mortality of all hospitals, were defined as those with "inferior outcomes". The surgical procedures analyzed were coronary artery bypass grafting (CABG: elective + emergency), elective CABG, emergency CABG, single-valve surgery, surgery for acute type A dissection, open heart surgery for newborns, open heart surgery for infants, surgery for lung cancer, and surgery for esophageal cancer.

RESULTS

There were large variations in 30-day mortality for all procedures, particularly in the lower-volume hospitals. There was a significant but weak inverse correlation between the hospital volume and the 30-day mortality rate for elective CABG, emergency CABG, single valve surgery, surgery for acute type A dissection, and lung cancer surgery. There was no correlation between hospital volume and the 30-day morality for open heart surgery for newborns and infants, and esophageal cancer surgery. After EB method adjustment, there was no hospital with inferior outcomes for conventional operations such as elective CABG, single-valve surgery and lung cancer surgery. The ratio of hospitals with inferior outcomes in more complex procedures was 1.8 % for open heart surgery for newborns, 0.8 % for open heart surgery for infants, and 0.2 % for esophageal cancer surgery.

CONCLUSION

There is a weak or no inverse correlation between the hospital volume and the mortality in cardiothoracic surgery in Japan. Most of the low-volume hospitals are not associated with inferior outcomes. The performance of the lower-volume hospitals should be carefully scrutinized using risk adjustment.