The abdominal wall of most patients with ECFs or EAFs is hostile; the surgeon might find that even entering the cavity itself presents a significant challenge. When possible, the surgeon should avoid going through the same incision used in prior operations, instead attempting to enter from non-violated areas of the abdominal wall such as the superior epigastric region or just over the pubic region and making your way in under direct vision from the inferior or superior aspect of the wound. It is really of great significance to avoid cutting on top your finger blindly, as the finger can easily be lifting small or large intestines for that matter that have been adhered to the abdominal wall, rather, this needs to be done under direct vision. An alternative method of entering the abdomen through a transverse incision has been advocated [35, 36], although I have not used that method in my practice.

A large number of patients cared for with an open abdomen have a split-thickness skin graft (STSG) (Fig. 7.10a, b). Such patients require special attention to ensure the success of their completion of surgery. Before the skin graft is excised, the neoskin, when pinched between the surgeon’s thumb and forefinger, should be easily elevated from the underlying tissue. Some surgeons do not attempt to excise the skin graft at all, but close the abdomen over it. When excision is attempted while the skin graft is adherent, dissection is very difficult and likely results in enterotomies and risks recurrent fistula formation [35].

Once the abdominal cavity is entered, the surgeon often faces a large ball of intestines wrapped by adhesions. Should these adhesions be separated? That question is as old as the surgery itself [36]. In my opinion, the surgeon should mobilize the entire segment of intestines, from the ligament of Treitz to the rectosigmoid. Doing so is tedious and time consuming, given previous abdominal surgeries and intra-abdominal inflammatory processes, and it is often complicated by new iatrogenic enterotomies. However, this is a must and when one does not do it, often patients will develop small bowel obstruction, as the symbiosis has been lost, when one releases only partially the adhesions. Other surgeons do not agree entirely with this approach; they suggest something in-between complete lysis, perhaps partial lysis of adhesions [37].

In patients with multiple ECFs or EAFs, resecting all of the fistulas may be challenging, but all of them must be resected [38–40]. The best scenario is when multiple fistulas are in close proximity to each other, so that the surgeon can excise the segment of fistulous tract “en masse.” Yet, if the fistulas are a large distance apart, more than one resection—and subsequently more than one anastomosis may be required; all are technically challenging. Because such patients are at high risk for developing short gut syndrome (see Chap. 25 on Short Gut Syndrome), adjunct procedures, such as strictureplasty, should be used in an attempt to avoid removing large segments of bowel. Intraoperatively, it is important for the surgeon to identify all fistulas. Care should be taken to avoid enterotomies, but if they do occur, any inadvertent injury to the bowel must be either repaired immediately or tagged with a suture so that it can be easily identified later during the course of the operation.

For reestablishing intestinal continuity, the hand-sewn, double-layer technique, not staplers, should be used [39]. In my practice, I prefer using continuous Vicryl™ (Ethicon, Somerville, NJ) sutures (Connell Technique) (Fig. 7.11). During this technique, the sutures go through the wall from the serosa to the mucosa, then from the mucosa to the serosa on the same side. The sutures then cross the incision to the serosa on the other side, and the pattern is repeated until suturing is completed. If the integrity of the anastomosis is questionable, it is reasonable to revise it or to create a proximal diverting ostomy. Excessive trimming of the mesentery, tension on the anastomosis, and inclusion of diseased bowel in the anastomosis must all be avoided [17, 41]. Operative treatment with takedown of ECFs is successful in 80–90% of patients, although the presence of an open abdomen lowers the success rate to 77.3% [4].

Once the continuity of the GI tract has been established, as described previously, creating a new abdominal wall may represent a serious surgical challenge. Multidisciplinary approaches and advanced surgical techniques may be necessary [35]. Whatever approach, whether single surgeon (general surgeon) or general surgeon with a plastic surgeon, the goal is to create functional and durable coverage of the abdominal cavity and to improve the patient’s quality of life. Native abdominal wall should be used; if that is not possible, biologic or prosthetic mesh can be used instead. In most patients, some sort of combination of reconstruction techniques will be needed. If native tissue can be used without undue tension, then it should be used. But, if midline tissue cannot be easily approximated or if mesh reinforcement is needed (as it is in almost all abdominal wall defects larger than 6 cm), then other techniques must be considered. For example, if midline tissue cannot be easily approximated, then lateral components need to be released to create a neoabdominal wall.

The closure of complex abdominal wall defects can be challenging. Traumatic injuries, tumor resections, necrotizing infections, enterocutaneous fistulas, previous surgeries, damage control laparotomy, and congenital defects can result in large abdominal wall defects that make reconstruction difficult. Surgical techniques, autologous and exogenous grafts have been developed to aid in the closure of such complex ventral wall defects and improve outcomes in these patients. One of the approaches is the development of musculofascial flaps that can be mobilized and brought to the midline to allow closure. In cases when there is a need for tissue transposition in order to establish a no-tension fascial closure, I use myocutaneous flap transposition through lateral component separation, as described previously [42, 43] (Fig. 7.12 a–f), although in most recent years posterior component separation has gained popularity. The component separation technique is based on an enlargement of the abdominal wall surface by separating and advancing the muscular layers.

Component separation provides additional medial transposition of musculofascial flaps that allows for reconstruction of giant abdominal wall defects often without the additional need for mesh. Defects up to the size of 20 cm at the level of mid-abdomen can be closed by this technique.

The earliest description of these musculofascial flaps dates back to the nineteenth century. These techniques were described by Guillouid in 1892, Chrobak in 1892, Gersuny in 1893, and Noble in 1895 [44]. Alfonso Albanese [42] from Argentina is credited with the first ever description of the technique that involved dividing the external oblique muscle vertically to enable the closure at midline by suturing together the rectus abdominis muscles in 1951. However in 1990, the technique was modified and refined by Ramirez et al. from Johns Hopkins University Hospital and was described as “components separation” [43]. This technique utilizes the medial advancement of bilateral, innervated, bipedicled, rectus abdominis-transversus abdominis-internal oblique muscle flap complexes to close ventral abdominal wall defects. In their landmark study, Ramirez and coworkers utilized human cadavers to demonstrate that the external oblique muscle can be separated from the internal oblique in an avascular plane by incising the external oblique fascia with an incision just lateral to the linea semilunaris [43]. Similarly, the rectus abdominis muscle with its overlying fascia can be separated from the posterior rectus sheath. This separation of the anatomic components allows significant mobilization for approximately 10, 20, and 8 cm in the upper, middle, and lower thirds of the abdomen, respectively. They subsequently utilized these findings clinically to reconstruct abdominal wall defects in 11 patients successfully.

Several other authors have also developed and utilized other autologous tissue transfer techniques for the repair of large sized abdominal wall defects. Wangansteen used the tensor fasciae latae flap from the thigh to reconstruct lower abdominal wall defects [45]. Ger and Duboys used muscle flap transfer to reconstruct full thickness abdominal wall defects [46]. However, the use of such free muscle flap transfer results in denervation of the muscle flap, which over time results in muscular atrophy, abdominal wall laxity, protuberance, and predisposition to hernia recurrence. Therefore, such an approach often requires the need for additional reinforcement of the repair with synthetic mesh. Similarly, the transfer of large sized flaps results in donor site scarring and deformity [47]. The components separation technique provides the advantage of preserving the innervation to the muscle flaps, hence maintaining the dynamic support and integrity of the abdominal wall. The absence of free tissue transfer also prevents the development of donor site morbidity and provides a more aesthetically feasible repair.

The transection of the rectus abdominis has been suggested by some surgeons as a contraindication [48, 49] for component separation. However, there is a lack of substantial evidence to support the claim that the use of component separation in patients with rectus violation results in adverse surgical outcomes.

As indicated in Step 7.1 the procedure begins with a midline exploratory laparotomy to enter the abdominal cavity. Care should be exercised to dissect the entire abdominal wall free from any adhesions to the bowel and omentum under direct visualization. Following this, flaps containing skin and subcutaneous tissue are lifted off of the underlying anterior rectus sheath and the external oblique fascia. These flaps should extend caudally to the inguinal ligament and cranially beyond the costal margin. The lateral extent of the flaps should be at least 2–3 cm lateral to the linea semilunaris in order to allow adequate exposure of the external oblique fascia. The blood supply of the skin comes from perforators arising from the deep epigastric and superficial inferior epigastric arteries. Extensive dissection of the skin flaps can disrupt these perforators, predisposing the overlying skin flaps to surgical site infections, skin necrosis, and wound dehiscence. Next the semilunar line and insertion of the external oblique fascia are identified. A vertical incision is made 1–2 cm lateral to the semilunar line extending from the inguinal ligament to at least 5 cm cranial to the costal margin. The cranio-caudal extent of this incision is important to allow adequate release along the entire length of the abdominal wall. A plane is developed deeply to the external oblique but superficially to the internal oblique. The dissection should be continued laterally in this avascular intermuscular plane extending up to at least the midaxillary line. If midline approximation cannot be achieved and additional mobilization is required, the dissection should be extended to the posterior axillary line to allow additional release. Component release should be performed bilaterally. It is important to avoid dissection deep to the internal oblique as the neurovascular plane exists between the internal oblique and the transversus abdominis muscles where blood vessels and nerves supplying the obliques and the rectus abdominis muscle traverse. Dissection in this plane may damage this neurovascular bundle or the Spigelian fascia predisposing the patient to a Spigelian hernia [47].

If midline approximation cannot be achieved, further medial advancement of the rectus abdominis-internal oblique complex of up to 2 cm can be achieved with posterior component separation. This is performed by longitudinally incising the posterior rectus sheath 1–2 cm from the midline and separating it from the rectus abdominis muscle. Care should be exercised to avoid injury to the inferior epigastric vessels running deep to the rectus muscle. The anterior fascia is debrided and the healthy tissue is approximated in the midline. The muscles are approximated using non-absorbable interrupted sutures.

I prefer, as do most surgeons, to reinforce the hernia repair with a mesh graft following component separation to help reduce recurrence; however, a randomized controlled trial comparing component separation with and without mesh repair found similar recurrence rates between the two approaches [49]. Redundant skin flaps are excised and finally, the undermined skin edges are approximated in layers in a standard fashion. Drains are placed using separate skin stab incisions between the skin flaps and the external oblique aponeurosis to reduce dead space and fluid collection [50].

There are various modifications of the component separation procedure . Some authors perform this procedure using minimally invasive surgical techniques (see Chap. 14), but the rates of recurrence of hernia are similar [51].Although many surgeons are familiar with anterior component separation (ACS), in recent years posterior component separation (PCS) with transversus abdominus release (TAR) has become popular. Detailed technical aspects of this procedure paying particular attention to the surgical anatomy have been reported [52]. The main principle of PCS is that the perforating vessels are spared, and the mesh is placed between rectus muscle anteriorly and posterior rectus fascia/peritoneum/preperitoneum posteriorly. Once you have dealt with all adhesions and other concomitant procedure, such as reconstitution of GI tract or other procedures as described above, the posterior approach to the retrorectus space is performed by incising the medial edge of the posterior rectus sheath at the medial edge of the rectus abdominis muscle. The edge of the transected posterior rectus sheath is grasped with clamps and retracted medially and posteriorly, allowing easy lateral dissection of the retrorectus space. During this stage of the operation one has to be cognizant not to injure intercostal nerves that perforate rectus muscle. The posterior lamina of the internal oblique aponeurosis is incised just medial to the entry of the intercostal nerves as they enter the rectus muscle posteriorly [52].

The dissection of this segment should start as cranially as you can. At the point of transition of posterior lamina of the internal oblique fascia you will be able to see the medial aspect of the transversus abdominus muscle (TAM). The muscle fibers and fascia of TAM can be separated from the underlying thin posterior transversus abdominis fascia and peritoneum with a right angle clamp. But, this separation requires a careful dissection under the muscle fibers of TAM . One has to be careful not to enter peritoneum, but if this does occur, one must make sure to identify and close with absorbable suture. Transection of TAM can be done in a number of ways but I agree with these authors the transection of the TAM should start as far cranially as possible where these muscle fibers are prominent and progressing caudally aids markedly this part of the component separation [53]. This extraperitoneal space now can be extended laterally and caudally in order to make space for the prosthesis. This dissection is facilitated greatly with a sweeping move of your hand. I prefer that this space extend to the costal margin and join the central tendon of the diaphragm in the midline. Once the space is created to your satisfaction, the posterior rectus sheaths are approximated with running absorbable suture. Fixation of the mesh superiorly, inferiorly and laterally with sutures, will help you with positioning the mesh appropriately. A number of techniques can be used to place the rest of the sutures. I prefer to use a Carter-Thomason suture passer, but other suture passers are just as good to fix the mesh to the anterior abdominal wall.