Appendicitis in Sweden: quality results.

Appendicitis and appendectomy continue to be among the most common diagnoses and operations performed. It is remarkable that after all these years there is still the potential for controversy and publication on this subject. In this issue of Annals of Surgery, Andersson and colleagues review the mortality in all patients in Sweden having an appendectomy from January 1987 to November 1996, revealing interesting data that can be interpreted in several ways. The authors are concerned (and have been in the past) that some appendectomies may be performed unnecessarily and that patients have an excess mortality from a surgically managed disease that could be treated nonoperatively with antibiotics and adequate resuscitation. They describe a seven-fold increase in expected mortality in those patients having appendectomy compared to the normal population and attribute this to the trauma of surgery and anesthesia. In their introduction and in the concluding sentence it is stated that surgical dogma accepts an error rate of 20-25% in diagnosis, and that we have the right to be wrong in this disease as it is benign and not harmful to operate for the wrong diagnosis. Andersson in his previous writings has indicated that he does not accept this premise. Their data indicate some striking points. They rightly point out the case fatality and mortality rates are amongst the lowest reported. This is quite an achievement when one considers that the data base is for all appendectomies for the entire country for almost a 10-year period. They further point out that in those patients with a perforated appendix or another diagnosis (25% of the population), the mortality seems reasonable (71% of the total deaths) given the nature of the primary clinical problem. The remaining deaths occurred in those patients with a nonperforated appendicitis (19.5%) or nonsurgical abdominal pain (9.4%). It is certainly possible that some of these were preventable, although the data provided do not refer to co-morbidities or in any real way help to determine the actual cause of death. Table 4 lists the attributed causes taken from the registry but death certificate causes of death are notoriously inaccurate. The low rate of autopsies, with no information from those done, further puts into question the real causes of death. When mortality from nonsurgical abdominal pain or other diagnoses is attributed to appendicitis, as it is in 13 cases here, the underlying causes of death in table 4 can be called into question (as they point out) based on misclassification. The data are remarkable for their accurate description of the mortality cost to Sweden of this very common disease. It is striking that 253 (88%) of the deaths occurred in the 8.9% of the study group, 60 years of age and over. If there is an area for improvement in results, surely it is in the aged population. It is noted that there are three important trends in appendicitis in modern times: first, a decreasing incidence of appendectomies; second, a shift in the proportion of operations in the elderly; and lastly, improved diagnostic accuracy. The first and third are probably related and strongly suggest that surgeons with the improved diagnostic tools of ultrasound and computed tomography are making the correct diagnosis more frequently, therefore not relying on a ”harmless” laparotomy to resolve the diagnosis. The increased use of laparoscopy may also contribute to the increased diagnostic accuracy, and one expects that it will do so in the future. In the present study, the area where improved diagnosis would be most likely to improve outcome is in the patients who do not have appendicitis and who would profit either from no operation or one directed with appropriate timing to their basic pathologic condition. Before suggesting that the increased standardized mortality rate seen in those patients with a nonperforated appendix would have been less with a different therapeutic approach, we must know diagnostic procedures, how they were prepared for the operation, and their co-morbid conditions, as 88% of the deaths were in patients over 60. While retrospective review of charts is appropriately out of fashion, the most information about the causes of death and whether they were preventable would come from such a study. It is hard to be persuaded that their data do suggest “surgical trauma per se” is the cause of their proposed excess mortality rates. One could only do that having been reassured that patients had been correctly resuscitated and received appropriate antibiotics and diagnostic investigations. The postoperative care and all its ramifications would also have to be evaluated. Is that necessary, however, to use the information provided by the manuscript? Probably not, because the areas for further improvements in care aresuggested and should address the disease in the elderly and the nonappendiceal based diagnostic categories. They suggest nonoperative therapy but do not provide a frame work in which to do that. A prospective randomized trial would be difficult and would require so many patients that it would not be feasible. A prospective observational study of outcomes in appendectomies would answer many of the questions posed by this study. Already defined are the case fatality rates and standardized mortality ratios against which prospective data could be compared. Data forms could be designed to determine accurately the causes of death and whether or not they were preventable. Indeed the sub-rosa hypothesis of role of surgery, nonoperative therapy, and even interval appendectomy could be examined. One might even hypothesize that the management of this entity might be a surrogate for the management of acute intra-abdominal disease. The data, however, do not in any way indicate that surgical trauma has anything to do with, if it really is in excess, the seven-fold increase in standardized mortality rates. The Swedes as a nation are to be congratulated for having the ability to get these data and for the excellent results demonstrated for their entire country in the management of this common but often difficult clinical condition.