New Orleans—Chest x-rays often are included in routine preoperative testing for patients undergoing low-risk elective surgery but are infrequently indicated. An educational intervention successfully lowered the frequency with which this “inappropriate” test is ordered, and in so doing helped decrease the utilization rates of several other types of preoperative testing as well.
“Like most institutions, we were looking at costs and recognized we might not be operating as efficiently as possible,” said Jonathan P. Wanderer, MD, assistant professor of anesthesiology at Vanderbilt University, Nashville, Tenn. “As part of that, we’ve been implementing changes when medical practice suggests we should be doing things differently than we currently are. And one area where that’s clear is chest x-rays, which used to be part and parcel of the preoperative workup, but from a screening standpoint don’t have that much value.”
An electronic learning module was created that specified indications for preoperative chest x-rays; all surgical providers were required to complete the module during August 2012. The researchers then identified and analyzed preoperative testing orders between February 2009 and March 2014, including chest x-ray, electrocardiogram (EKG), basic metabolic panel (BMP), packed cell volume (PVC), platelets, comprehensive metabolic panel, complete blood count (CBC), partial thromboplastin time/prothrombin time (PTT/PT), type and screen, urinalysis and “other labs.” The procedures were then stratified according to American Society of Anesthesiologists (ASA) physical status and surgical risk classifications.
In all, 43,320 anesthetic records were identified before the intervention date, compared with 24,013 after. The rate of requested chest x-rays fell from 33% to 20% (P<0.001). More telling was the fact that during the last six months of data, chest x-rays were requested in only 16% of surgical cases. In the subpopulation of low-risk patients (ASA physical status 1) undergoing low-risk surgery, chest x-ray rates fell from 8% to 2% (P<0.001).
“Subsequent analysis revealed that there were a handful of clinicians who were responsible for almost all the remaining orders,” Dr. Wanderer told Anesthesiology News. “So we’ve been working with them directly, and have effectively dropped the rate even further. So I think we’re now down to the point where we’re ordering them when there’s an acute disease process or surgical procedure where imaging will be helpful.”
Perhaps not surprisingly, EKG, PCV, BMP, PTT/PT, comprehensive metabolic panel, CBC, urinalysis and platelets all showed statistically significant reductions in relative frequencies after the intervention date. Interestingly, type and screens and “other labs” showed statistically significant increases in relative frequencies after the intervention date.
Dr. Wanderer, who is reporting his findings here at the ASA’s 2014 annual meeting (abstract A1140), had only good things to report about the program’s success, which he said demonstrates that changing physician behavior may not be as difficult as some believe. “Some of our practices get backed into our routine procedures and aren’t really second-guessed,” he said. “We learned that some of the check boxes that were being selected during the preoperative workup weren’t an intentional decision by the surgeon, but were almost filled out by default.”
And because these practice patterns are certainly not limited to Vanderbilt, employing similar steps may help other institutions achieve the same benefits. “We’re now taking a closer look at our rationale for ordering EKGs, another area where we found that we could change our practice,” he added. “We’ve again ended up having conversations with our colleagues and convinced them that we needed a reason to order the test rather than just doing it routinely. And that has resulted in practice changes by merely asking the question: If we do this test, what value are we likely to get from it?”
Charles B. Watson, MD, emeritus chair of anesthesia and deputy surgeon-in-chief at Bridgeport Hospital in Bridgeport, Conn., explained that because of the low incidence of positive findings from age-related or routine chest x-rays, his institution does not perform routine chest x-rays unless the patient has known active disease or new findings suggestive of pulmonary pathology. “The ASA has published guidelines on preoperative testing since the early 2000s and not identified routine testing of healthy individuals as useful [Anesthesiology 2012;116:522-538],” he said.
As Dr. Watson explained, educating medical practitioners on the benefits of avoiding routine chest x-rays can sometimes take years, as it did at his institution. “Information was circulated by email, policy documents to offices and physicians, and in meetings/surgical conferences over the years,” he said. “Now, our staff rarely order chest films, and only with positive clinical findings. The surgical motivation for screening patients for active symptoms of pulmonary disease is the understanding that last-minute chest films will likely delay their procedures.”