How the Evidence Has Changed My Practice

Evidence-based medicine has changed my practice in the past 5 years, as it should. We generally all start out doing things as we learned them in our residency. After all, at least some of our staff surgeons are our mentors in addition to being our teachers, and we have a great deal of respect for both. Those ladies and gentlemen performed the operations they did because they judged that the procedures provided some type of benefit to the patients, and we trusted and respected their judgment. For many of the operations that we perform, we will continue to follow the easy path of our apprentice module and continue to do what we learned to do in our residency. However, I often tell my residents that I will be very disappointed if they are still doing what I taught them 15 years from now. The only thing that is constant in surgery is change. If they are not changing for the better, they may well be getting worse. The day they quit changing for the better is the day they should consider quitting surgery. I know that I am doing fewer than 20 percent of my procedures the way I learned to do them in my residency, and I feel and hope that I am continuing to improve. I have found higher levels of evidence very helpful in this quest for improvement in my practice. I used to improve my practice by adapting new methods that achieved the same or better goals of treatment with less invasiveness, fewer complications, shorter recovery times, and which made sense from a physiologic and wound healing point of view. Evidence-based medicine is a newer tool that helps me do this more objectively with the benefit of the experience of thousands of other patients and surgeons to help me make the decision to change my practice. I will point out two common operations where level I evidence has helped me change my practice. THUMB BASAL JOINT ARTHROPLASTY VERSUS LIGAMENT RECONSTRUCTION AND TENDON INTERPOSITION I finished training at McGill in 1984. This was where I learned silicone implant replacement of the trapezium, which was the standard of care for thumb trapeziometacarpal arthritis. This operation had stood the test of time and worked well for many patients. In 1986, two things occurred almost at the same time. First, silicone synovitis became recognized as an entity.1 Disintegrating bits of solid silicone were blamed for degenerative changes in the bones and joints around which they were inserted. Second, Burton and Pelegrini2,3 published landmark level III evidence, retrospective comparative studies that showed that ligament reconstruction and tendon interposition had results that were as good as or better than the embattled silicone trapezial replacements that were being blamed for destroying wrist bones. This is when North America fell in love with ligament reconstruction and tendon interposition. I changed my practice from silicone to ligament reconstruction and tendon interposition, and stopped using silicone trapezial replacement, as did many others. Most American hand surgeons still perform ligament reconstruction and tendon interposition today, although the variations of full flexor carpi radialis, split flexor carpi radialis, abductor pollicis longus, and others have tried to capture their market share. I performed ligament reconstruction and tendon interposition or its variations along with the majority until the new best-evidence articles were published. In 2005, a Cochrane review4 of randomized, quasi-randomized, or controlled trials in which pain, physical function, patient global assessment, range of motion, or strength was measured as an outcome of trapeziectomy versus ligament reconstruction and tendon interposition was published. It provided level I evidence that showed the following: No one procedure produced greater strength than any other. Although this also appears to be the case for pain and physical function, there was insufficient evidence to be conclusive. Trapeziectomy is safer and has fewer complications than the other procedures studied in this review; conversely, trapeziectomy with ligament reconstruction and tendon interposition has more. In 2009, a second Cochrane review5 on trapeziectomy versus ligament reconstruction and tendon interposition came to similar conclusions, namely, that although it appears that no one procedure produces greater benefit in terms of pain and physical function, there was insufficient evidence to be conclusive. Trapeziectomy has fewer complications than trapeziectomy with ligament reconstruction and tendon interposition (level I evidence). As a result of the 2005 Cochrane study, I changed my practice in 2006. I stopped routinely performing ligament reconstruction and tendon interpositions and started performing simple trapeziectomies. It has been my impression that, as the level I evidence suggests, the outcomes are the same but the complications are fewer, and that trapeziectomy is better tolerated by patients, with a more rapid recovery, as it is simpler surgery. In addition, it is easier to perform using the wide-awake approach (i.e., no tourniquet and no sedation, just local infiltration with lidocaine and epinephrine). This is also easier and more convenient for the patients than general anesthesia or sedation with blocks and a tourniquet. I am pleased with this change in my practice. COMPRESSION OF THE ULNAR NERVE AT THE ELBOW: TRANSPOSITION VERSUS DECOMPRESSION A 2007 meta-analysis of randomized controlled trials (level I evidence) in the Journal of Bone and Joint Surgery6 suggested the following: There is no difference in motor nerve-conduction velocities or clinical outcome scores between simple decompression and ulnar nerve transposition for the treatment of ulnar nerve compression at the elbow in patients with no prior traumatic injuries or surgical procedures involving the affected elbow. Confidence intervals around the points of estimate were narrow, which probably exclude the possibility of clinically meaningful differences. These data suggested that simple decompression of the ulnar nerve is a reasonable alternative to anterior transposition for the surgical management of ulnar nerve compression at the elbow. A 2008 meta-analysis of randomized controlled trials7 (level I evidence) found no statistically significant difference. Instead, there was a trend toward an improved clinical outcome with transposition of the ulnar nerve as opposed to simple decompression. At the end of my training in 1984, my teachers at McGill were performing subcutaneous transposition, which I performed until the first meta-analysis of 2007, at which time I changed my practice to simple decompression. As the level I evidence suggested, my outcomes seem to be no worse in the past 4 years, but my complications are fewer, and simple decompression seems to be better tolerated by the patients, with a more rapid recovery, as it is simpler surgery. This operation is also simpler to perform with the wide-awake approach under pure local anesthesia with no sedation and no tourniquet. This is also easier and more convenient for the patients than general anesthesia or sedation with blocks and a tourniquet. I am also pleased with this change in my practice. THE COURAGE TO CHANGE PRACTICE When we as surgeons have been performing an operation that we learned from trusted individuals, and that operation seems to be benefiting most of the patients most of the time, it is difficult for us to change. In contrast, if an alternative procedure is providing high levels of evidence that its outcomes are no worse or perhaps even better with fewer complications, perhaps we should consider having the courage to change. After all, this is not about us. This is about the patients. For their sake, we need to make it a point to become aware of new articles with a high level of evidence and at least consider having the courage to change practice. Donald H. Lalonde, M.D. Dalhousie University Hilyard Place, Suite C204 600 Main Street Saint John, New Brunswick E2K 1J5, Canada [email protected]

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.