Effects of Aprotinin During Cardiac Surgery/Long Term Death Rates

Cardiac Surgery

The dept. of Anesthesiology currently has a database of subjects whom had surgery and received either Aprotinin or Amicar in the OR. The current viewpoint is that Aprotinin is more harmful than Amicard. In an effort to see what the long term outcomes were for subjects whom had surgery here at Upstate, it was decided to look at long term death rates to see if any differences. A student t-test will be used to determine statistical significance where a p value of \<0.05 will be deemed significant. Using data from 462 subjects that had undergone cardiac surgery at SUNY Upstate Medical University, CABG only and the long term mortality rate from the Mangano, et.al. publications, the unadjusted mortality for the two drugs are Aprotinin 5.4% and Amicar 1.2%. A power analysis was performed using the hospital mortality rates of 5.4% and 1.2% with the sample size in the propensity data and a p-value of 0.05. The result was a power of 81.7%.

Charts for 462 subjects will be reviewed from 2002-2005 based on the surgery performed. From each chart, the subject name, date of birth,and social security number will be recorded along with the state of residence and year of surgery. This information will be transferred to a CD that is encrypted and password protected. The CD will be sent via FedEx overnight to the Center for Disease Control(CDC), National Death Registry. The CDC will use the information on the CD to extract death outcomes for each subject's information provided. Once all information is complete, the CDC will transfer on to a CD that is encrypted and password protected and return to our office via FexEx overnight. The information on the CD will then be used to determine what the long term death rates were for subjects here at SUNY Upstate Medical University and stored in an anonymous database. Once the information is verified as accurate, the CD will be destroyed and no identifying information will be kept. The CDC will also destroy the information after 60 days, with no links to the identifying information.

No related papers found yet.

United States