Complications in Dermatologic Surgery: Title and subTitle BreakHow Safe Is Safe?

THE TREND TOWARD outpatient surgery continues to rise. Medicare is encouraging more procedures in the nonhospital environment. Today surgery is performed in the surgical office that only 5 years ago would have been considered more appropriate for hospital operating rooms. These include orthopedic, obstetric, gynecologic, and cosmetic procedures. While many specialties that were reared in the traditional hospital operating room culture are now venturing into the office practice environment, dermatology by its very nature has lived in that world from the outset. Our specialty training is geared toward developing expertise in both outpatient medical and surgical treatment. A recent survey by the American Society for Dermatologic Surgery found that its members performed about 3.9 million procedures. Of these, skin cancer surgery was the most common with 1.4 million operations, suggesting a broad range in the nature of surgery performed by dermatologists in their offices.¹

Most likely the transition from hospital-based to office-based procedures has progressed because the technical interventions are safer than they have been in the past and do not require the extensive hospital infrastructure that supports conventional operating rooms. With respect to dermatology and its universe of operations, there is a general sense that the safety record of procedures performed by dermatologists is excellent. This view is supported by the malpractice rating that is assigned most dermatologic surgery procedures. It is interesting, though, that there is an actual dearth of data regarding patient safety and the complication rate for the bulk of cutaneous surgery performed in the office setting. Such information is important because dermatologic surgeons differ from traditional surgeons in an important way: we tend to operate alone, away from the oversight or observation of other colleagues. Surgeons who operate in an operating room are accustomed to colleagues who come by to observe, learn new techniques, and discuss cases. In addition, morbidity and mortality conferences are an essential quality-control device that all surgeons participate in as part of the requirement to maintain hospital credentials and operating privileges. These mechanisms have not become part of the standard of care in dermatologic surgery, perhaps for good reason: in the vast majority of cases, the rate of serious complications is extremely low.

In lieu of the quality and oversight mechanisms described above, it is important to understand the nature of risk in dermatologic surgery. Most of the studies in the literature that provide information on complication rates for outpatient procedures relate to anesthesia, general or laparoscopic surgery, or orthopedic and head and neck procedures. With the recent frenzy regarding the unsubstantiated risks involved in outpatient surgery performed by "non–plastic surgeon" physicians, it seems that an objective analysis of these and related claims is important—not because these false claims may cast dermatologists in an unfair harsh light, but because they mislead and frighten patients who might otherwise be able to obtain optimal care in the office-based dermatologic surgery environment.

Coldiron² recently presented data generated in a state that had mandatory reporting of office incidents. He found that dermatologists reported 2 incidents (gauze ignition burn under intravenous sedation and temporal nerve palsy under local anesthesia). Liposuction under general anesthesia resulted in the greatest number of reported incidents and none of the complications from liposuction was reported by dermatologists.² A recent article in Cosmetic Dermatology analyzed liposuction-related complications and mortality and found that dermatologists were not responsible for any liposuction-related deaths and the majority of such deaths occurred from liposuction procedures performed by plastic surgeons.³ These 2 approaches sought to determine whether risk in a given environment was specialty specific; they did not address the safety issue per se.

An alternative approach to understanding medical safety in the office setting is to look within our own specialty and provide outcomes-based data on the risks of complications in this increasingly popular environment. Inevitably, it is through data gleaned from evidence-based study that medical and surgical standards of care are articulated and confirmed. While it could be argued that the standard of care is that at which the majority of physicians practice, it is important to test those standards and confirm their validity.

Cook and Perone⁴ are to be commended for providing the first prospective assessment of the incidence of complications in outpatient dermatologic surgery. This thorough yearlong study performed on patients undergoing Mohs micrographic surgery and reconstruction included a substantial number of cases (N = 1358). Complete follow-up was obtained for more than 98% of the patients. The study was not performed in a blinded fashion: the surgeon performing the procedure evaluated each case in the immediate and delayed postoperative period. The 5 types of complications that were studied included wound infection, which was defined by culture confirmation of a clinically suspected infection; postoperative hemorrhage; hematoma; wound dehiscence; and flap or graft necrosis. The setting and conditions in which the tumor was removed and reconstruction was performed were also included in the study. For each complication event, factors that may have helped precipitate it were noted. The overall complication rate was 1.64%. For purposes of comparison, one study that looked at reconstruction performed by plastic surgeons after Mohs micrographic surgery had a complication rate of 3.3% when only criteria such as partial flap or graft loss and wound dehiscence were studied.⁵ However, this study did not specify whether the procedures were performed in an outpatient or hospital setting. The data from Cook and Perone confirm that office-based dermatologic surgery involving Mohs surgery and reconstruction is very safe when performed by properly trained physicians.

While the data provided by Cook and Perone are comforting, it should be noted that despite the meticulous care and detail that was taken with preoperative considerations, intraoperative technique, and postoperative wound care instructions, certain practices used by the investigators are not necessarily considered standard of care by all dermatologic surgeons. For example, all patients who underwent a flap or graft repair were given postoperative antibiotics. This approach to the use of antibiotics in dermatologic surgery raises long-standing unresolved questions and highlights how any assessment of clinical outcomes must be understood in the context of the actual clinical interventions.

Most dermatologic surgical procedures are considered "clean-contaminated," where there may be minor breaks in sterile technique. The infection rate in these types of procedures is thought to range between 5% and 10%. Studies that have looked at postoperative wound infection rates in dermatologic surgery describe an even lower rate of infection. Further, there is no standard for antibiotic prophylaxis or its use postoperatively. Use of antibiotics in such procedures is reasonable in selected cases in which factors such as surgical site, nature and length of procedure, level of contamination, and overall health of the patient are considered. While the incidence of wound infection in dermatologic surgery is low,^{6 - 7} it would certainly be beneficial to have well-defined guidelines regarding the use of antibiotics in dermatologic surgery. The devil is in the details: the studies necessary to arrive at statistically valid conclusions would require very large numbers of patients because the naturally occurring incidence of infection is so low. Indeed, in a study by Classen et al,⁸ nearly 3000 patients were needed to detect a statistically significant difference in the infection rate among groups of patients receiving antibiotics at different times. Such a study would be helpful to clarify the observation that prophylactic antibiotics are most effective when given before wounding and that a single preoperative dose provides optimal prophylaxis and postoperative use yields no further benefit.^{9 - 11}

Still, the question of whether the reduction of infection rates from a very low percentage to an even smaller percentage is sufficiently significant to justify antibiotic administration to all patients, including the overwhelming majority that are not at risk for developing infection should be answered. This effort should include an appraisal of allergic reaction, sensitization risk, and even cost. To complicate matters, the practice of "clean, aseptic" technique is also variable among dermatologic surgeons and any properly designed study would have to standardize all practices.

Another area of complication in dermatologic surgery that can be impacted upon by individual physician practice is that of hemorrhage and hematoma formation. Cook and Perone⁴ chose to limit the use of platelet inhibitors and blood thinners preoperatively. Otley et al¹² have demonstrated that risks for continuing therapy with warfarin and platelet inhibitors are minimal for most patients. In fact, the risks may be less than the inconvenience, costs, and risks of discontinuing the medications postoperatively. Otley et al found no statistically significant reduction in the rates of severe complications in patients who had their medications held preoperatively.¹² Regardless of these potentially significant variations in surgical practice, Cook and Perone provide excellent outcomes-based prospective data that substantiate the safety of dermatologic outpatient surgery, and their study is an important step to better defining the standard of care.