Timing of Reoperation for Patients With Severe Peritonitis

Operations for severe intra-abdominal infection are intended to resolve the problem by dealing with the underlying pathology that caused the infection, or by addressing the intraperitoneal pathology caused by the disease (such as abscess), or both. This approach is called source control. Some of these initial surgical interventions predictably fail to achieve that goal and another intervention is required. Review of clinical series suggests that this occurs in 15% to 30% of unselected cases.^{1 - 8} Thirty years ago such reinterventions meant that another laparotomy was required, but in the interim, surgeons have learned that many cases, especially those involving intra-abdominal abscesses that follow a prior operation, can be managed with percutaneous drainage. This has been aided by increasingly available and increasingly accurate imaging technology. However, there remains a minority of cases that after failure of the initial operation require another laparotomy to resolve the intra-abdominal pathology.

Whether the reintervention is performed percutaneously or via open operation, most surgeons expect that the ultimate outcome will be more favorable if the reintervention occurs sooner rather than later. However, this proposition has no solid data support, other than the observation that patients who are ill with intra-abdominal infection tend to become more ill as time passes. The more serious the illness at the original intervention, the more serious the penalty for delayed treatment is assumed to be. In addition, the more serious the illness at the original intervention, the more likely it is that another intervention for source control will be required.⁹ However, in this setting, accurately recognizing the need for reintervention becomes more difficult.

Against this background, the concept of mandatory or planned relaparotomy for patients with initially severe peritonitis gained support in the 1980s and was supported by retrospective reviews suggesting improved outcomes.¹⁰ Until the study reported in this issue of JAMA by van Ruler and colleagues,¹¹ all evidence has been from retrospective series. These authors posed the question in a prospective, carefully designed and executed study that randomized patients with severe peritonitis to be managed with either planned relaparotomy or by relaparotomy only when found to be necessary by predefined clinical criteria (on-demand). The patients were severely ill at the time of initial operation and were enrolled in both university and regional medical centers. The groups were comparable, and the original surgeon thought that the focus of infection had been successfully surgically resolved in more than 90% of the patients in both groups. Prior observational studies have reported that the need for additional intervention is substantially more likely if the surgeon does not believe that the original attempt at source control was definitive.^{7 - 8 ,12}

In the primary comparison of mortality and major predefined morbidity, there was no statistically significant difference between the groups, although the differences that did occur favored the on-demand group. In the secondary end points of total number of reoperations, proportion of patients requiring 3 or more reoperations, number of percutaneous drainages required, days receiving ventilatory support, duration of stay in the intensive care unit and in the hospital, and cost, all differences were statistically significant and favored the on-demand group. In the on-demand group, 58% of the patients had no additional operations and of the 42% who had reoperations, 32% had no findings mandating operation. Thus, only 29% of the on-demand group had persistent or recurrent infection requiring operation after the first laparotomy for peritonitis. This compares with 34% of the mandatory relaparotomy group who had positive findings on relaparotomy. The rate of 32% negative relaparotomy in the on-demand group suggests that even in this group surgeons may be a little too quick to operate, and better tools are needed to make this decision.

The trial by van Ruler et al¹¹ is the best evidence yet that mandatory or scheduled relaparotomy for peritonitis is not helpful except in the obvious settings of patients whose first procedure has resulted in retained surgical packing or because the pathology could not be dealt with completely at the first operation. What surgeons should focus on now is the search for more accurate and sensitive methods to recognize in as timely a manner as possible when a patient will need another intervention. This may include improved understanding of clinical patterns, novel imaging techniques, and possibly new biomarkers. Ultimately, though, the diligent attention of the surgical team to the clinical progress of the patient after laparotomy for peritonitis is currently the most effective management technique.