This Editorial View accompanies the following article: Capdevila X, Barthelet Y, Biboulet P, Ryckwaert Y, Rubenovitch J, d'Athis F: Effects of perioperative analgesic technique on the surgical outcome and duration of rehabilitation after major knee surgery. Anesthesiology 1999; 91:8–15.
THERE has been unquestionably renewed interest in regional anesthesia, or perhaps more correctly, "regional analgesia," which helps build interest in perioperative pain medicine. With the current improvements in both intraoperative general and regional anesthesia, it is unlikely that for unselected groups of patients, differences in outcome will be found for the intraoperative interval. Nevertheless, if advances in regional anesthesia are to continue to add value to our surgical patients, and if regional methods are to become more widely accepted by anesthesiologists, surgeons, patients, and administrators, more work is needed to prove that such techniques really offer measurable advantages over other methods that are perceived to be "easier," faster, and less dependent on specific technical skills, such as general anesthesia or postoperative patient-controlled analgesia. Furthermore, despite humanitarian concerns about minimizing pain in all of our patients-and we realize that this statement may be viewed as offensive to some-in this era of managed care and cost containment, and demonstrated "advantages" probably need to go beyond simply showing that pain assessment scores are improved or that less morphine is needed by patients. Without such concrete information on benefits, it becomes more difficult to introduce effectively the analgesic methods, particularly in the face of concerns about rare but severe complications (e.g., neuraxial hematomas associated with enoxaparin). 
What would constitute a "relevant" advantage of regional anesthesia and analgesia? Traditionally, studies have focused on "big" outcomes: death, blood loss, major complications (e.g., pulmonary emboli). For example, Davis et al.  demonstrated that spinal anesthesia reduced blood loss during hip replacement surgery, Christopherson et al.  showed an improved surgical outcome among patients undergoing peripheral vascular surgery with epidural anesthesia and analgesia, and Lui et al.  showed that for relatively healthy patients undergoing colon surgery, epidural anesthesia and analgesia provided the best balance of analgesia and side-effect profile along with improved gastrointestinal function and time-to-reach-discharge criteria. Unfortunately, as noted, many of these studies have focused on big outcomes: transfusion requirements, death, and reoperation rates. Such outcomes are critically important to the populations studied but are not easily transferable to the great majority of patients in whom regional anesthesia might be applicable. Is reduced mortality really a good reason to choose regional anesthesia over general anesthesia in a healthy 25-yr-old patient undergoing shoulder surgery? Does anyone seriously argue that regional anesthesia will significantly reduce mortality or blood loss in a group of reasonably healthy 50-yr-old patients undergoing peripheral orthopedic procedures accompanied by minimal blood loss? Conversely, advocates of regional anesthesia counter that, to date, no one has shown that choosing general anesthesia over regional anesthesia produces meaningful outcome improvements.
Despite these concepts, there is no question that the field of regional anesthesia and perioperative pain medicine needs more outcome studies. These need not always focus on death and disability (which are not often relevant issues or, at least, would be statistically impossible to study in groups with very low baseline mortality and morbidity rates), but they must examine the impact of therapy on some relevant variable that goes beyond simply assessing pain by visual analog scales or monitoring opioid use in the first 24–48 h postoperatively. In this issue of Anesthesiology, the article by Capdevila et al.  is an example of the type of work that is needed. Patients undergoing extensive knee surgery received a standardized general anesthetic and were randomly assigned to receive continuous epidural anesthesia, continuous femoral nerve block, or patient-controlled analgesia (note that both of the catheter techniques are best categorized as multimodal analgesia techniques  because significant concentrations of local anesthetic, as well as morphine and clonidine, were used). This was maintained for 72 h, during which time mechanically assisted continuous passive mobility exercises were used to achieve maximally tolerated ranges of motion. Seven days after surgery, patients were transferred to a rehabilitation facility for continued therapy and assessment. Although this rehabilitation arrangement is unusual for patients in the United States, it provided a superb opportunity for further controlled assessment in these French patients.
There are a number of important findings here, but perhaps the most relevant is the substantially shortened time until discharge from rehabilitation for patients in both of the regional analgesia groups (median, 37–40 days) compared with those in the group that received patient-controlled analgesia (50 days). Discharge criteria were objective and relatively rigid and included the requirement that a patient be able to walk uphill without assistance. Again, such prolonged rehabilitation stays would not be common in the United States. Its also possible that less intense rehabilitation efforts would result in longer recovery times in all groups (and a consequent "smearing" of the intergroup differences). However, this should not detract from the overall finding of the study: long-term recovery was facilitated by short-term postoperative regional analgesia maintained during the initial and intensely painful mobilization of the repaired/replaced knee joint. Many regional anesthesia/analgesia enthusiasts might suggest that these results are expected; furthermore, many orthopedic teams already use regional techniques. So what is new in these data? What seems to be new is that these results support the concept that postoperative analgesia is moving toward peripheral nerve techniques.  Although the epidural group in the study by Capdevila et al. had rehabilitative outcome that matched that of the femoral nerve block group, the femoral group had the most favorable side-effect profile of the two regional analgesia groups. An unanswered question from this investigation is whether our own (United States) cultural fixation in the last decade on length of stay as the prime measure of cost containment in medical practice contributes to poorer outcomes in selected patients, such as those needing multimodal rehabilitation.
No matter which "anesthetic world view" you hold-regional or general-it is clear that we all need to work together to further our understanding of outcome from our analgesic techniques. Data from these studies, coupled with further clinical experience, will help us understand how to best prescribe analgesia regimens for our patients.
Michael M. Todd, M.D.
Professor of Anesthesia;firstname.lastname@example.org
David L. Brown, M.D.
Professor and Head of Anesthesia; Department of Anesthesia; The University of Iowa; Iowa City, Iowa 52242–1009