Assessment and reporting of perioperative cardiac risk by Canadian general internists: art or science?

Objective: Physicians may use several validated risk indices to estimate perioperative cardiac risk, but there is little evidence for interventions to reduce this risk. We were interested in evaluating how general internists assess, define, communicate, and attempt to modify perioperative cardiac risk.

Design: Cross-sectional survey of all 312 general internists in the Canadian Society of Internal Medicine with Canadian mailing addresses; 117 (38%) responded.

Results: Respondents' mean age was 46 years, 79% were male, and on average they did 17 preoperative consults per month. Of the 104 respondents who routinely performed preoperative assessments, 96% (100/104) informed patients of their perioperative cardiac risk, but 77% did so only subjectively (i.e., stating risk as low, moderate, or high). Respondents provided 8, 27, and 12 different definitions for low, moderate, and high risk, respectively, with marked variability in the range of definitions they provided: from <1% to < 20% for "low risk," from 1% to 2% to 20% to 50% for "moderate risk," and from >2% to >50% for "high risk." The 67% of respondents who reported using a perioperative cardiac risk index used a variety of indices and exhibited just as much variability in their risk estimates and definitions as those who didn't use risk indices. While virtually all advised perioperative beta blockade in patients with known coronary artery disease, they varied substantially in the recommended agent or dose; further, these internists were evenly split on whether antiplatelet agents should be held or continued perioperatively.

Conclusions: These physicians differed widely in their assessment of perioperative cardiac risk and their definitions of low, moderate, or high risk. This raises concerns about whether patients (and surgeons) are provided with adequate information to make fully informed decisions about the potential risks of elective surgical operations.