INDIANAPOLIS – Slower walking speed on the timed-up-and-go test in elderly patients scheduled for surgery is a significantly better forecaster of postoperative complications and 1-year mortality than are the considerably more complex patient risk calculators currently considered standard of care, Dr. Thomas N. Robinson reported at the annual meeting of the American Surgical Association.
"I think what walking speed reflects is global reduced physiologic reserve. It’s frailty. And by definition, an individual who’s frail will have adverse health care outcomes," explained Dr. Robinson, a general surgeon at the University of Colorado, Denver.
Dr. Thomas N. Robinson
Use of preoperative walking speed to assess postoperative risk is a paradigm shift, he noted. Current surgical risk assessment strategies rely upon math-heavy patient risk calculators which evaluate single end-organ dysfunction, in some cases summing up the individual scores for heart, lung, liver, and other organ dysfunction in an attempt to define chronic disease burden. But in older patients, this approach is less effective than a simple frailty assessment based upon mobility: that is, walking speed, he continued.
Surgeons in the Veterans Affairs health care system use a risk calculator that involves input of 24 variables. Hitting the "compute risk" button then produces the patient’s estimated 30-day morbidity and mortality risks.
Dr. Robinson presented a prospective study in which the VA tool was compared to the timed-up-and-go (TUG) test in 272 patients aged older than 65 years who were followed for a minimum of 1 year after elective surgery. To see how the two tests performed across surgical specialties, the investigators included 174 patients with a cardiac operation and 98 who had colorectal surgery.
In the TUG test, a clinician starts a stopwatch as the patient rises from a chair, walks 10 feet, returns, and sits back down. The patient is instructed to walk at his or her normal pace and is free to use a walking aid. Dr. Robinson chose to study the TUG rather than a simple 5-meter gait speed test because he considers TUG more relevant to surgical patients.
"TUG combines lower extremity strength to stand up in addition to walking speed. And if you think about somebody who needs to make the transition from hospital to home, lower extremity strength is important," he said.
The investigators categorized a TUG time of 10 seconds or less as fast, 11-14 seconds as intermediate, and 15 seconds or longer as slow. Of note, the subjects’ TUG times were unrelated to common comorbid conditions in the elderly, including stroke, diabetes, heart failure, and hypertension.
In contrast, walking speed was strongly associated with classic indicators of frailty. For example, impaired cognition was present in 3% of subjects with a fast TUG time, 41% of those with an intermediate time, and 92% of slow performers. Another frailty indicator – a history of falling within the past 6 months – was present in 7% of the fast group, 21% of intermediate TUG walkers, and 85% of those with a TUG speed of 15 seconds or more.
In the cardiac surgery group, one or more postsurgical complications occurred in 11% of the 53 patients in the fast group, 25% of 88 patients with an intermediate TUG time, and 52% of 33 individuals in the slow group. The 1-year mortality rates were 2%, 3%, and 12%, respectively.
Similarly, in the colorectal surgery group, the complication rate was 12% among 30 fast walkers, 29% of 42 patients in the intermediate group, and 77% of 26 patients in the slow group. The 1-year mortality rates were 3%, 10%, and 31%, respectively.
The investigators judged comparative test performance in predicting postoperative morbidity and mortality on the basis of the receiver operating characteristic area under the curve, which was 77% with the TUG test compared to 55% with the VA risk calculator in the colorectal surgery patients. In the cardiac surgery group, the figures were 68% for TUG and 55% with the risk calculator.
Geriatricians typically measure TUG in seconds as a continuous variable. Dr. Robinson and coworkers decided the test would be more useful for surgeons if they created the three discrete categories of fast, intermediate, and slow.
Discussant Dr. Michael E. Zenilman praised the investigators for what he called "an outstanding study," and one that’s particularly welcome right now, as the wave of aging baby boomers swells.
"As we take care of more elderly patients, it’s important that we develop tools like this to quickly and objectively assess risk. The tools that we have now, such as the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) models and the VA risk calculator, are for me just too complicated," said Dr. Zenilman of Johns Hopkins University, Baltimore. Dr. Zenilman is the university’s vice chair and regional director of surgery for the Washington area. Noting that TUG, Mini-Mental Status scores, history of falling, and serum albumin levels all have been shown to serve as proxies for frailty, he asked Dr. Robinson to predict which one he thinks will win out as a postoperative risk predictor.