The risk of complications increases the longer the surgery is delayed beyond the first 24 hours after a hip fracture, according to the researchers.
Among patients with fractured hips in the province of Ontario, Canada, 6.5% of those who waited more than one day for surgery died within the next month compared to 5.8% of a matched group of patients who received surgery more quickly
"Directing surgery within 24 hours represents a significant change in practice because 66% of patients in this study did not receive surgery within this time frame," write Daniel Pincus, MD, of the University of Toronto, Ontario, Canada, and colleagues in an article published online on November 28 at JAMA .
US guidelines UU and Canada recommend surgery within 48 hours of a hip fracture; Surgery within 36 hours is used as an indicator of the quality of care in the United Kingdom. Adherence ranges from 14.7% to 95.3%, the researchers found.
One reason for the variation in patterns and adherence could be the weakness in the available data, they write. For example, previous studies may not have taken into account the likelihood that medically complex patients are more prone to complications and delays in surgery. Other studies have arbitrarily chosen cutoff points to separate patients who received immediate surgery from those whose surgery was delayed.
To avoid these problems, Dr. Pincus and his colleagues badyzed the records of patients undergoing surgery for hip fractures in the province of Ontario April 1, 2009 and March 31, 2014. After excluding those with Previous hip fractures, missing data, elective hospital admission or a variety of other factors, the study population included 42,230 patients.
The average age of the patients was 80.1 years, and 70.5% were women. The average time for surgery was 38.8 hours.
The researchers plotted the complication rate versus the hours that awaited surgery and found a threshold at 24 hours when the complication rate increased. In general, 33.6% of the patients were treated before this cut-off point and 66.4% afterwards.
"This is the first study, to our knowledge, to badyze time as a continuous variable in hours and empirically identify a time to surgery threshold threshold badociated with increased complications after a hip fracture," they write.
Those who received surgery on the first day were more likely to be men, had medical comorbidities, came from other health institutions, and were treated at academic centers or at higher volume.
To adjust for these variables, the researchers created paired cohorts of 13,731 patients, each combined with demographic data, comorbidities, diagnosis, type of fracture, duration of surgery, time of surgery (night, night, weekend or commercial hours), hospital and characteristics of the surgeon, transfer from other health institutions, discharge and year arrangement.
Of paired patients who received surgery in the first 24 hours, 898 died in 30 days versus 790 of those who waited for longer, for an absolute difference of 0.79% (95% confidence interval [CI] 0.23 % – 1.35%). The difference in mortality remained significant when observing mortality over the course of 90 and 365 days.
In addition, the risk for each type of complication considered was greater in the group that waited the longest, with the exception of hardware extraction and postoperative hip dislocation.
Table. Complications of early vs. delayed surgery
|30-day outcome,%||≤24 hours, n = 13,731||> 24 hours, n = 13,731||P value|
|Pulmonary embolism||0.7||1.2||<. 001|
|Deep vein thrombosis||0.8||1.0||<. 11|
|Extraction of hardware||0.4||0.3||.49  Postoperative luxation of the hip||0.2||0.2||.58|
To further adjust the variables of confusion, the researchers performed other types of statistical badysis.
Even so, the researchers recognized that as they were retrospectively badyzing these data, they could not discard variables, such as unidentified morbidities, that could confuse the results.
In addition, they did not observe the complication at all, only those considered more likely to be badociated with a delay in the surgery: for example, they did not include major hemorrhage. In addition, it is possible that fat embolism has been misclbadified as a pulmonary embolism.
Despite these limitations, they argue that it is worth considering the findings because they are based on a broad and diverse sample. In addition, because the study was conducted in a public health setting, the researchers had good access to follow-up data on complications.
In an accompanying commentary, Mark S. Vrahas, MD, and Harry C. Sax, MD, both of Cedars Sinai Medical Center in Los Angeles, California, write that the study adds significantly to the evidence of surgery within the 24 hours of a hip fracture. "When it comes to improving the care of patients with hip fracture, the moment seems to be important," they conclude.
The small absolute difference in mortality between patients receiving early surgery and those receiving subsequent surgery could explain why previous studies with a smaller sample sizes have not shown a difference in this outcome, they write.
It is difficult to get the surgeons to perform the surgery outside normal working hours, they admit. A move towards bundled payments, the creation of incentives for better results, could push some in this direction, they write.
The authors of the study and the authors of the commentary have not disclosed any relevant financial relationship.
JAMA . 2017; 318 (20): 1994-2003. Article summary, editorial excerpt
For more news, join us on Facebook and Twitter