Thirty-day mortality after noncardiac surgery is reported to be as high as 1.9%.1,2 Nonfatal perioperative cardiac complications occur in 5% of noncardiac surgery patients and contribute to length of stay and health care costs.2 Current models for predicting cardiac risk have limitations. The commonly used Revised Cardiac Risk Index (RCRI) does not predict cardiac events after major vascular surgery and underestimates vascular events.3,4 Diagnostic tests including electrocardiograms (EKGs) and stress tests weakly predict perioperative cardiac risk.4 Cardiopulmonary exercise testing has been recommended; however, its use is resource prohibitive and metabolic equivalents (METs) are subjective.4,5
The 6MWT is an easy to administer test of functional exercise capacity that can discriminate between low and high anaerobic thresholds in patients undergoing major noncardiac surgery (eg, a short distance predicts a lower threshold).6 It measures the distance in meters a person can walk in 6 minutes along a flat surface7 and has been shown to have predictive value for mortality of patients with idiopathic pulmonary fibrosis waiting for lung transplantation.8 The objective of our preliminary study was to investigate the ability of the 6MWT to predict perioperative myocardial injury by examining the association between 6MWT distance and postoperative troponin I concentrations in patients undergoing intermediate and high-risk major vascular or renal transplant surgery.
Our prospective study was approved by the University of California, Los Angeles institutional review board. Written informed consent was obtained from all subjects. Adult patients undergoing nonemergent, intermediate and high-risk major vascular surgery or renal transplantation from June 2014 to April 2016 at University of California, Los Angeles Medical Center were considered for inclusion. Major vascular surgeries included procedures done on the aorta and/or large branches (eg, aortic dissection repair). Renal transplantation included both living related and cadaveric grafts. We selected these populations because they were likely to have cardiac risk factors, which would aid our assessment of the 6MWT.
Subjects were recruited in the preoperative holding area on the morning of surgery. Patients were not incentivized and no patients declined to participate. 6MWTs were administered in the holding area by study personnel trained on the ATS protocol.7 We calculated the RCRI, which includes cardiac risk factors such as coronary artery disease and recorded METs (measure of subjective exercise tolerance).3 We monitored for 6MWT complications (eg, exercise-induced symptoms) and recorded posttest vital signs.
Troponin I testing using a Siemens CTNI immunoassay was performed on postoperative day 1 (POD1) morning blood draw samples. Values exceeding the 99th percentile of normal controls were considered “elevated” (>0.04 ng/mL). Values <0.04 were considered “negative.” Positive troponin cutoff values have not been defined for renal transplant patients; however, troponins >0.04 ng/mL are shown to be prognostic in this population,1 so this was the cutoff used. We performed chart review for the occurrence of intraoperative (eg, blood loss, arrhythmias) and postoperative (eg, myocardial infarction, death) major adverse cardiac events until 30 days after surgery.
We compared patients in the elevated and negative postoperative troponin I concentration groups with respect to variables known to be associated with cardiac risk and thus have the potential to influence troponin I concentrations (eg, 6MWT distance, RCRI, β blockers; Table). We used t tests for continuous variables, χ2 or Fisher exact tests for categorical variables, and the Wilcoxon test for ordinal variables. Univariable logistic regression models were constructed for 6MWT, RCRI, and METs.
Classification ability of the models was assessed using the area under the receiving operator characteristic curve (AUC). Assuming an AUC of 0.5 represents a test with no discriminatory capability, our goal was to have a 6MWT AUC of 0.7 because it is at least moderately predictive9 and better than previously described prediction models such as RCRI.3 We calculated the AUC for the 6MWT, RCRI, and METs and used the DeLong method for comparing correlated AUCs to detect differences between the 6MWT and the other tests. We generated an AUC for a combination of the 6MWT and RCRI. Finally, we calculated the 6MWT distance that maximizes sensitivity and specificity, as well as positive and negative predictive values.
To calculate power, we used methods described by Obuchowski and McClish10 and expected to observe 4 patients with negative postoperative troponin I concentrations for every 1 patient with elevated postoperative troponin I concentrations. We based this estimate on the VISION trial, in which 11.6% of all-comer surgical patients had a “prognostically relevant” elevated postoperative troponin.1 Using this expected rate, and an α level of .05, to have 80% power to detect an AUC difference of 0.70–0.50, we needed 80 negative and 20 elevated postoperative troponin I concentration patients. Statistical analyses were performed using IBM SPSS V23 (Armonk, NY). P values <.05 were considered statistically significant.
Among 51 major vascular and 49 renal transplant patients, the mean age was 61 ± 15.8 years, mean height was 1.7 m ± 0.10, mean weight was 75.8 kg ± 17.05, and 60% were men. Characteristics did not differ by postoperative troponin I concentrations (data not shown).
No patients experienced complications from the 6MWT. Nine patients had procedural blood loss requiring blood transfusion; there was no association with postoperative troponin I concentrations. Seventeen patients had elevated troponin I concentrations on POD1 (10 renal transplant, 7 major vascular). All 100 patients were asymptomatic for myocardial injury until discharge and did not have electrocardiogram changes. The median length of hospital stay postoperatively was 3 days (range, 1–21). In the 30-day period after surgery, no patients were readmitted with major adverse cardiac events.
Results of the univariate analysis yielded no difference between the elevated and negative postoperative troponin I concentration groups with respect to any variable except the 6MWT distance (P = .005; Table). The univariable model with only 6MWT yielded a significant association between every 10 m walked and postoperative troponin I concentration (OR, 0.89 [95% CI, 0.82–0.97]; P = .006)]. The AUCs of the 6MWT, METs, and RCRI were 0.71 (95% CI, 0.57–0.85), 0.51 (95% CI, 0.33–0.68), and 0.59 (95% CI, 0.44–0.75), respectively. Using the DeLong method, the AUC of 6MWT was not different from that of RCRI (P = .23 versus 6MWT) or that of METs (P = .14 versus 6MWT; Figure). The combined AUC using 6MWT and RCRI was 0.70 (95% CI, 0.56–0.85).
The 6MWT distance that maximizes sensitivity and specificity for predicting elevated postoperative troponin I concentrations was 307 m (sensitivity 53% [95% CI, 29.2%–76.7%], specificity 87% [95% CI, 79.5%–94.0%]). At this distance, the positive predictive value was 45% (95% CI, 23.2%–66.8%). The negative predictive value was 90% (95% CI, 83.4%–96.6%). Thus, 90% of patients who walked more than 307 m had negative postoperative troponin I concentrations. However, only 45% of patients who walked <307 m had elevated postoperative troponin I concentrations. From our univariable logistic regression model, we estimated that for every 50 m gained on the walk test, the odds of an elevated postoperative troponin I concentration decrease by about 43% (OR, 0.57 [95% CI, 0.38–0.86]).
Our preliminary study is the first to our knowledge to evaluate feasibility of the 6MWT in the perioperative environment. In our study, lower walk distances were associated with elevated troponins on POD1, which highlights the potential of the 6MWT as a screening tool in major noncardiac surgery patients. The 6MWT AUC was not significantly different from the RCRI or METs when compared via the DeLong method. Thus, the test may be most useful when combined with tools such as the RCRI, even though our data did not reveal a superior aggregate AUC.
Although prior studies identify important outcome differences from positive postoperative troponins,1 we cannot make such definitive conclusions. Small sample size and lack of poor outcomes limited testing of our model’s predictive capability. Also, the threshold used for calling a troponin “elevated” (0.04 ng/mL) is based on normal controls and its use in renal transplant patients may be problematic. However, the Vascular events In noncardiac Surgery patIents cOhort evaluatioN (VISION) study showed that even in patients with chronic kidney disease, positive postoperative troponin was predictive of 30-day mortality.1 Next, we did not have preoperative troponins and recognize postoperative concentrations could reflect preoperative concentrations in patients with diabetes, for example. Although in our study, there was no difference in diabetes prevalence among elevated and negative troponin groups. It is worth nothing that patients with intermittent claudication were included in our study but none reported symptoms during the 6MWT. So, we did not further examine the utility of the test in predicting risk in these patients. Other studies have demonstrated that the 6MWT is helpful when looking at outcomes in patients with peripheral arterial disease.11
Troponin was tested only in the morning on POD1, so the true incidence of myocardial injury may be underestimated. Finally, our best combined 6MWT sensitivity and specificity of 53% (95% CI, 29.2%–76.7%) and 87% (95% CI, 79.5%–94.0%), respectively, occurred at a walk distance of 307 m. A low sensitivity increases the false-negative rate.
Still, this preliminary data may have value for those who administer and interpret the test. The 6MWT is a useful and time-efficient test that has the potential to predict elevated troponins after major noncardiac surgery. Larger studies using a greater variety of surgical patients are needed to help elucidate the test’s true utility in this setting.
Name: Anahat K. Dhillon, MD.
Contribution: This author helped with conceptual development and article preparation.
Name: Andrew A. Disque, MD.
Contribution: This author helped with enrollment and study conduct.
Name: Christine T. Nguyen-Buckley, MD.
Contribution: This author helped with enrollment and study conduct.
Name: Tristan R. Grogan, MS.
Contribution: This author helped with statistical analysis and figure preparation.
Name: Dana L. Russell, MPH.
Contribution: This author helped in data collection, analysis, and article preparation.
Name: H. Albin Gritsch, MD.
Contribution: This author helped with identifying study patients.
Name: Jacques P. Neelankavil, MD.
Contribution: This author helped with conceptual development, enrollment and study conduct, and article preparation.
This manuscript was handled by: Tong J. Gan, MD.
1. Devereaux PJ, Chan MT, Alonso Coello P, et al. Association between postoperative troponin levels and 30-day mortality among patients undergoing noncardiac surgery. JAMA. 2012;307:2295–2304.
2. Devereaux PJ, Sessler DI. Cardiac complications and major noncardiac surgery. N Engl J Med. 2016;374:1394–1395.
3. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation. 1999;100:1043–1049.
4. Duncan D, Wijeysundera DN. Preoperative cardiac evaluation and management of the patient undergoing major vascular surgery. Int Anesthesiol Clin. 2016;54:1–32.
5. Kertai MD, Boersma E, Bax JJ, et al. A meta-analysis comparing the prognostic accuracy of six diagnostic tests for predicting perioperative cardiac risk in patients undergoing major vascular surgery. Heart. 2003;89:1327–1334.
6. Sinclair RC, Batterham AM, Davies S, Cawthorn L, Danjoux GR. Validity of the 6 min walk test in prediction of the anaerobic threshold before major non-cardiac surgery. Br J Anaesth. 2012;108:30–35.
7. ATS Committee on Proficiency Standards for Clinical Pulmonary Function Laboratories. ATS statement: guidelines for the six-minute walk test. Am J Respir Crit Care Med. 2002;166:111–117.
8. Lederer DJ, Arcasoy SM, Wilt JS, D’Ovidio F, Sonett JR, Kawut SM. Six-minute-walk distance predicts waiting list survival in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2006;174:659–664.
9. Vanagas G. Receiver operating characteristic curves and comparison of cardiac surgery risk stratification systems. Interact Cardiovasc Thorac Surg. 2004;3:319–322.
10. Obuchowski NA, McClish DK. Sample size determination for diagnostic accuracy studies involving binormal ROC curve indices. Stat Med. 1997;16:1529–1542.
11. McDermott MM, Ades PA, Dyer A, Guralnik JM, Kibbe M, Criqui MH. Corridor-based functional performance measures correlate better with physical activity during daily life than treadmill measures in persons with peripheral arterial disease. J Vasc Surg. 2008;48:1231–1237, 1237.e1.