Hold Off a Bit Before Operating on Broken Hips, Study Suggests

SAN FRANCISCO — Hip fracture repair is best started about 20 hours after presentation in elderly patients on blood thinners, not immediately and not more than a day later either, a researcher said here.

Mortality and complications such as venous thromboembolism (VTE) and pneumonia were minimized when surgery began at about the 20-hour mark after patients using warfarin or other anticoagulants arrived at the hospital, according to Chunyuan Qiu, MD, MS, of Kaiser Permanente in Baldwin Park, California.

In an analysis of more than 35,000 patients treated in the Kaiser Permanente system from 2009 to 2020, there was a clear U-shaped curve for mortality and complications when plotted against time from admission for patients on anticoagulants. The nadir for each curve fell right around 20 hours, said Qiu, speaking at the American Society of Anesthesiologists annual meeting, who called it “the sweet spot.”

No such pattern was seen for patients not on anticoagulants at admission. In fact, the data in this group suggested “the earlier the better” was optimal for surgery. Mortality was low to start and climbed steadily with increasing delay.

It therefore appeared that the benefit of delay in the anticoagulated patients likely stemmed from letting their medications wash out enough to restore some degree of normal clotting.

Study Details

For the study, Qiu and colleagues looked to a hip fracture registry for patients in the Kaiser Permanente system. Because the researchers suspected that anti-clotting medications would be an important factor, they stratified patients according to their use of anticoagulants within 100 days prior to surgery. About 15% of the cohort had done so.

Outcomes of interest were mortality at 30 and 90 days and at 1 year, plus 90-day incidence of pneumonia, VTE, and myocardial infarction. Some surgeries were performed within 10 hours, while others were delayed as long as 48-72 hours.

In the anticoagulated patients, mortality stood at 8.3%, 16.3%, and 30.5% at 30 days, 90 days, and 1 year, respectively. These rates were roughly 50% higher than in the non-anticoagulated patients, further suggesting that having anticoagulants on board drove poor outcomes in hip fracture surgery.

While 30-day mortality approached 9% for anticoagulated patients who underwent surgery relatively early — within about 10 hours — it was less than 6% for those whose surgeries were around Qiu’s “sweet spot.” Likewise, 90-day mortality was lower by about 5 percentage points with this timing; 1-year mortality was also lower but to a smaller degree.

Similar patterns were seen for 90-day incidence of pneumonia, VTE, and myocardial infarction in the anticoagulated patients.

For those not on these drugs, the U-shaped curves were replaced with upward-trending straight lines. This was also the case for pneumonia. VTE did show a relatively high 90-day rate when surgery was performed within about 10 hours, with a nadir close to 16 hours, followed by a gradual increase as delay extended past 30 hours. Timing of surgery did not seem to matter for myocardial infarction risk in the non-anticoagulated patients.

Please enable JavaScript to view the

comments powered by Disqus.