Background

The Gupta Perioperative Risk for Myocardial Infarction or Cardiac Arrest (MICA) scale is a common metric for evaluating which patients are at risk of perioperative cardiac complications. However, this score has not been externally validated in the bariatric population. The MBSAQIP database, the largest bariatric-specific clinical dataset, is an ideal tool for validating the predictive power of the MICA scale in this complex population.

Methods

From the 2015-2017 MBSAQIP database, the MICA scores of all patients undergoing metabolic and bariatric surgery were calculated. The discriminative ability and diagnostic accuracy of the MICA score in predicting cardiac morbidity was calculated.

Results

A total of 474,194 patients were included in this validation cohort. The MICA scale had a fair discriminative ability between low (<0.5% risk of cardiac complications), medium (0.5–1.0%), and high (>1.0%) risk patients, with complication rates of 0.083%, 0.434%, and 1.092%, respectively. On ROC analysis, the scale was found to have an area under the curve (c-statistic) of 0.752. While this denotes fair accuracy, it is significantly less so than other previously reported cohorts (0.874–0.884.)

Conclusions

While the MICA scale demonstrates moderate accuracy, validation in this external dataset shows the scale to be significantly less accurate than previously described. The MICA scale remains a valuable tool for assessing cardiac risk in bariatric patients; however, reliance on this scale alone may be insufficient to accurately predict the likelihood of cardiac complications. Findings should be supplemented with further diagnostic efforts in the setting of clinical suspicion for cardiac disease.