The timing of the second-look laparotomy has traditionally been advocated as between 24 and 48 h after the initial surgery, a period advocated from the first description of the procedure [1]. Whatever the length, the goal of this interval is to allow the second look to be performed when the overall physiology of the patient has been optimized. This will ensure that any stunned bowel has been reperfused to the extent that it no longer exhibits ambiguous vitality, in addition to increasing the likelihood of anastomotic healing and decreasing the chance of fistula formation. Moreover, adequate resuscitation and physiological support increase the patient’s tolerance of the second procedure.

Evidence supports the assertion that the 1- to 2-day intermission is sufficient to allow this optimization. In animal models, both serum and histopathological markers of intestinal ischemic injury are well past their peak by 24 h after perfusion is reestablished; by 48 h, they have returned to baseline [10]. This timeline is, of course, a guideline, and the clinical status of the patient should otherwise determine the schedule of treatment. After initiation of treatment and the primary laparotomy, an improvement in clinical parameters indicates progress toward optimization and resolution of any reperfusion injury. A lack of improvement or deteriorating clinical status, however, may indicate a missed bowel injury or the evolution of metabolically significant bowel necrosis and should compel re-exploration regardless of the length of time since the original surgery.

The technique of second-look laparotomy in AMI is entirely concerned with the inspection of the bowel. While ultrasonography, fluorescein, and angiography all have roles in the diagnosis of acute ischemia, their reliability in the prediction of intestinal viability is unacceptably low [11]. Clinical signs, such as absence of peristalsis, bowel wall edema, mucosal hemorrhage, and a lack of bleeding at cut bowel edges, are similarly poor predictors if employed at the time of the initial laparotomy. With a correction of the underlying vascular deficit, the patient’s physiological parameters and time for ischemia–reperfusion damage to declare itself; however, the utility of these clinical indicators is reasonable in determining the need for additional resection.

In recent years, the advent of minimally invasive techniques in general surgery has prompted consideration of laparoscopy in place of second-look laparotomy. Its applicability in this context has yet to be defined, but the authors believe that laparoscopy is inadvisable for several reasons [6, 12–15]. The benefit of laparoscopy over second-look laparotomy is minimal. The patient already bears a midline incision and is not spared another. The procedure requires both general anesthesia and paralysis to a degree indistinguishable from that required for laparotomy. The limitations of laparoscopy over laparotomy are significant. Laparoscopy provides only visual assessment of the bowel and does not allow any tactile examination of signs of viability. That visual assessment is itself limited to the small field of vision of the laparoscope. Running the bowel laparoscopy is a technically challenging procedure that, in the best of cases, cannot visualize more than a few centimeters of bowel at any time.

The most compelling reason to avoid laparoscopy, however, is that the pneumoperitoneum that is the sine qua non of laparoscopy is physiologically inadvisable in the context of poor visceral perfusion. Physiological intra-abdominal pressure in the human being is approximately 10 mmHg. If this pressure is the context for acute mesenteric ischemia, raising it by 150–200 % cannot reasonably be expected to benefit the patient. Pneumoperitoneum requires intra-abdominal pressures that have been shown to impair visceral blood flow. In a rat model, pneumoperitoneum at pressures required for laparoscopy results in decreased intestinal perfusion and bacterial translocation [16]. Avoidance of intra-abdominal hypertension, in fact, not only makes second-look laparoscopy inadvisable but makes a planned open-abdomen strategy preferable.

The physiological effects of increased intra-abdominal pressure are well documented. The term intra-abdominal hypertension (IAH) is used to denote pressures higher than 10 mmHg [17]. In a dog model, IAH was shown to decrease splanchnic perfusion by a factor of 2 at intra-abdominal pressures of only 20 mmHg [18]. An identical pressure increase in a pig model resulted in a drop in intestinal perfusion of almost 40 % [19]. A pressure increase to only 15 mmHg – less than standard pneumoperitoneum, far less than would be required for the diagnosis of abdominal compartment syndrome, and only 5 mmHg above healthy controls – caused a drop in bowel tissue oxygenation of more than a quarter. A pressure of 25 mmHg, still less than generally regarded as abdominal compartment syndrome, dropped oxygenation to 50 % of the normal. In a rat model, a moderate degree of IAH caused not only a drop in mesenteric and intestinal perfusion but also resulted in bacterial translocation [20]. Increasing intra-abdominal pressure was directly related to markers of gut ischemia [19]. These findings of decreased perfusion, intestinal ischemia, and bacterial translocation occurred in the context of normal cardiac output and mean arterial pressure. The patient with acute mesenteric ischemia, already at a physiological disadvantage, is particularly vulnerable to the deleterious effects of increased abdominal pressure. The burgeoning ischemia of IAH, when relieved with second-look laparotomy, could lead to another cycle of ischemia–reperfusion injury that will not complete manifestation until 24–48 h after the relaparotomy, decreasing the utility of the second look and worsening bowel injury.

The AMI patient is not only particularly vulnerable to IAH but is particularly likely to develop the condition as is any patient undergoing emergency abdominal surgery – a third of whom will end up with the formal diagnosis of abdominal compartment syndrome [21]. The high prevalence of abdominal compartment syndrome (ACS) among acute care surgical patients is likely a consequence of the frequent need for fluid resuscitation in this population. Crystalloid volumes of 10 L or greater, rates of 250 mL/kg, or the administration of more than 6 L in 6 h have all been associated with a high risk of abdominal compartment syndrome [17]. In examining resuscitation to supranormal oxygen delivery parameters, Balogh et al. demonstrated a prevalence of either IAH or ACS of 58 % as well as an increase in visceral ischemia [22]. While the diagnosis of bowel ischemia predisposes the patient to IAH, intra-abdominal hypertension in the surgical patient has been associated with bowel necrosis at laparotomy [21, 23].

The clear benefits of second-look laparotomy, combined with the hazards posed by the development of intra-abdominal hypertension and the likelihood that such hypertension will develop, mandate careful consideration of maintenance of an open abdomen after laparotomy for acute mesenteric ischemia. Closure of the abdomen after laparotomy for AMI will increase the likelihood of intra-abdominal hypertension and worsened gut perfusion and possibly increase the subsequent necessity of additional bowel resection at second-look laparotomy. Delay of fascial closure until completion of the second-look laparotomy minimizes the probability of IAH-related gut ischemia, allows a more liberal resuscitation protocol, and facilitates second-look surgery.

The maintenance of an open abdomen after initial laparotomy for AMI is almost universally advantageous. Though the benefits of this strategy may be clear, the maintenance of the open abdomen and the eventual final closure are its attendant challenges. Despite the nomenclature, the open abdomen requires some method of temporary closure in the interval between the initial and final laparotomies, however separate in time those operations must be. Such temporary closure must accomplish several goals. First, it must provide coverage and protection of the viscera. Desiccation of the intra-abdominal contents must be prevented, particularly within the threatened bowel and anastomotic or staple lines. The obverse of this priority is the need to minimize insensible losses in the patient whose volume status is critical. Second, the temporary closure must allow for the maintenance of domain within the abdomen and prevent retraction of the fascial envelope. The preservation of domain as well as of the fascial edges is vital to eventual final closure. Retraction of fascial edges occurs rapidly. Without the use of fascial bridging, the chances of successful primary closure fall to zero after only 96 h [24].

Management of intra-abdominal fluid collection and stagnation with an eye to contamination control is an additional requirement and must be balanced with the need to avoid desiccation of the viscera. Finally, the temporary abdominal closure must avoid intra-abdominal hypertension and abdominal compartment syndrome. Though this risk seems obviated entirely by an open abdominal strategy, it is not [22].

In the initial damage control experience, simple skin approximation was used for temporary abdominal closure; the resulting incidence of ACS led to the abandonment of this practice [25]. Aside from this, skin approximation is a suboptimal method of interval closure for other reasons. It does not prevent fascial retraction, nor does it avoid early loss of domain. Its watertight nature, in addition to enabling the development of ACS, does not allow control or drainage of contaminated intra-abdominal fluid and increases the risk of evisceration and skin necrosis. The complication rate of skin closure, no matter what the method, can be as high as 36 %, and so this technique is generally no longer used [25]. Skin interposition methods, such as the mesh silo and the Bogota bag, are not watertight and thus avoid the problems associated with fluid accumulation and stagnation such as ACS and infection. As these methods involve the sewing of an interposition material to the skin, and not the fascia, they do not prevent the latter from retracting. As such, rates of successful primary closure are generally poor.