Implementing an NT-proBNP-guided preoperative cardiac risk assessment algorithm in noncardiac surgery: a before-and-after cohort study

Several algorithms guide perioperative cardiovascular (CV) evaluation, and recent evidence supports more objective functional-capacity assessment and selective biomarker testing. We developed and implemented an updated evidence-based preoperative CV algorithm for patients undergoing noncardiac surgery. The goal was to identify patients at elevated risk for major adverse cardiovascular events and guide appropriate management without overutilizing cardiac testing or specialist consultation. We conducted a retrospective before-and-after cohort study at a university-affiliated hospital, comparing baseline care (March 2017-February 2020) with post-implementation period 1 (PI1; January-December 2021; NT-proBNP threshold 300 pg/mL) and post-implementation period 2 (PI2; January-August 2022; threshold 200 pg/mL). The algorithm combined clinical risk calculation, the Duke Activity Status Index, and selective NT-proBNP measurement or stress testing for asymptomatic adults with estimated perioperative CV event risk greater than 1%. Outcomes included NT-proBNP testing, cardiology referral, postoperative troponin measurement, elevated postoperative troponin/perioperative myocardial injury (PMI), 30-day myocardial infarction (MI), and mortality. Among 66,695 patients (34,990 baseline, 11,606 PI1, and 20,099 PI2), NT-proBNP testing increased from 220 (0.7%) at baseline to 174 (1.5%) in PI1 and 301 (1.6%) in PI2 ( p < 0.001 for each comparison). Among tested patients, the proportion with NT-proBNP greater than 300 pg/mL did not differ between baseline and PI1 (42.7% vs. 34.5%, p = 0.096); in PI2, 31.9% exceeded the protocol threshold of 200 pg/mL. Cardiology referrals decreased from 0.7% to 0.2% in PI1 and 0.5% in PI2 ( p < 0.001 for each; relative reductions 76.4% and 34.7%, respectively). Postoperative troponin measurement decreased from 1,343 (3.8%) to 322 (2.8%) and 550 (2.7%) ( p < 0.001 for each), whereas the proportion of monitored patients with elevated troponin/PMI increased from 13.7% to 25.5% and 19.8% ( p < 0.001 for each). Overall PMI rates were 0.5%, 0.7%, and 0.5%, respectively. Thirty-day MI and mortality did not differ. The algorithm increased selective NT-proBNP testing and reduced cardiology referrals and postoperative troponin testing without increasing stress testing or short-term adverse outcomes.