High sensitive troponin T predicts 30-day mortality after non-cardiac surgery
Relationship Between High Sensitive Troponin T Measurements and 30-Day Mortality After Noncardiac Surgery
Presented at ACC.17 by PJ DevereauxMar. 19, 2017 - news
Myocardial injury after non-cardiac surgery (MINS), defined as ischaemic injury of the myocardium during or within 30 days after surgery, is common and independently associated with increased 30-day mortality. The use of hsTnT assays was recently approved and these assays are used in clinical practice, however, it is not known whether hsTnT levels after non-cardiac surgery can predict 30-day mortality.
In this prospective, international cohort study (VISION), the association between the peak hsTnT levels during the first 3 postoperative days and the 30-day mortality after surgery was evaluated, in 21 790 patients ≥45 years of age, who stayed at least one night in hospital after non-cardiac surgery. Patients had hsTnT measurements 6-12 hours after surgery and daily for 3 days. 40.4% had preoperative hsTnT measurements. Moreover, hsTnt thresholds that independently altered patients’ risk of death and had an absolute risk of 30-day mortality ≥3% and an adjusted hazard ratio (aHR) ≥3.0 were determined.
- The most common types of surgery were major orthopaedic surgery (16%), major general surgery (20%) and low-risk surgery (35%).
- The 30-day mortality was 1.2% (95% CI: 1.1-1.4%).
- According to multivariable analysis, peak hsTnT values ≥20 ng/L, occurring in 24.0% of patients, were associated with higher 30-day mortality and had an aHR ≥3.0 and an absolute risk of 30-day mortality ≥3%, compared to the reference group (peak hsTnT ≤5 ng/L, n=5218). Peak hsTnT of ≥20 to <65 ng/L (n=4049): aHR: 23.63 (95% CI: 10.32-54.09), peak hsTnT ≥65 to <1000 ng/L (n=1118): aHR: 70.34 (95% CI: 30.60-161.71), peak hsTnT ≥1000 ng/L (n=54): aHR: 227.01 (95% CI: 87.35-589.92).
- Patients with a peak hsTnT value of ≤5, ≥20 to <65, ≥65 to <1000, and ≥1000 ng/L had 30-day mortality rates of 0.1%, 3.0%, 9.1%, and 29.6%, respectively.
- In a continuous analysis, an absolute hsTnT change of ≥5 ng/L independently increased patients’ risk of 30-day mortality (aHR: 4.70; 3.53-6.27).
- 13.8% of patients with elevated perioperative hsTnT had their peak value before surgery.
- 93.1% of patients with myocardial injury after non-cardiac surgery did not have ischaemia-related symptoms, and would not have been detected without hsTnT measurements.
- The OR of the composite endpoint of non-fatal cardiac arrest, CHF, coronary revascularisation, and mortality in patients with myocardial injury after non-cardiac surgery was 8.47 (95% CI: 6.94-10.34).
In patients undergoing non-cardiac surgery, a peak postoperative hsTnT measurement of ≥20 ng/L and an absolute change of ≥5 ng/L during the first 3 days after surgery was significantly associated with 30-day mortality. These data suggest that unrecognized heart injuries may account for about 25% of the deaths within the first 30 days after non-cardiac surgery, and that a better management of ischemic heart injury detected by hsTnT has the potential to prevent many of the deaths after non-cardiac surgery.
During the press conference, the VISION study was applauded for informing physicians about what are normal and what are higher values obtained with the newer hsTNT tests. For patients it is important to better know their risks.
In current clinical practice, hsTNT is not commonly measured, which means ‘low-hanging fruit that we’re ignoring’. Dr. Devereaux suggests that hsTNT is measured, and he advises that patients with elevated hsTNT values be seen by a specialist. Moreover, statin and aspirin therapy may be considered. He noted that the VISION study concerns observational data, but the study population had coronary disease, and evidence exists that statins and aspirin provide benefit. Dr. Devereaux proposes a pragmatic approach to measure hsTNT in anyone over 65 years old, and in those over 45 with known CVD.
Our coverage of ACC.17 is based on the information provided during the congress.