Wounds may fail in one of two ways. Wound dehiscence describes partial or complete disruption of an abdominal wound closure with protrusion or evisceration of the abdominal contents. Incisional hernia is defined as an abnormal protrusion of a viscus through the musculoaponeurotic layers of a surgical scar. Since the time at which a scar may be said to have formed is open to debate, the full healing of the skin incision is used to make a convenient distinction between wound dehiscence and incisional hernia. Dehiscence of the wound occurs prior to cutaneous healing, while incisional hernias lie under a well healed skin incision. A parastomal hernia is the result of an incisional hernia occurring adjacent to a surgically created stoma. In this section the epidemiology and management of abdominal wound disruption and incisional hernias, including parastomal hernias, will be discussed together. The factors contributing to these complications show similarities and wound dehiscence may be associated with the subsequent development of an incisional hernia.

Disruption of an abdominal wound occurs in less than 1 per cent of patients undergoing major abdominal surgery. Incisional hernia is more common, occurring in approximately 10 per cent of patients. The aetiology of wound failure is related to the pre-operative condition of the patient, the technique of wound closure and postoperative complications.

The preoperative factors implicated in the aetiology of wound dehiscence and incisional hernia are age, male sex, previous irradiation, jaundice, uraemia, anaemia, diabetes, malnutrition, malignant disease, vitamin C depletion, obesity, and administration of steroids or cytotoxic drugs. Clinical studies of these variables in the aetiology of wound failure are sparse since they rarely occur in isolation. Experimental studies have, however, documented delayed healing of laparotomy wounds in uraemic and anaemic animals supporting a role for these factors in the aetiology of wound dehiscence. Some authors have reported malignant disease to be more common in patients with wound dehiscence (36 per cent) than in control patients (23 per cent), though others report no difference. There may also be an increased incidence of incisional hernia in patients with malignant disease. Jaundice is presumed to predispose to wound dehiscence because of its adverse effect on wound healing, but documentation of this is lacking. However, jaundice may be important in the subsequent development of incisional hernia. Malnutrition, as assessed by anaemia and protein depletion, increases the chance of wound dehiscence but probably not of incisional hernia. The role of obesity as a factor predisposing to wound dehiscence is supported by some but not others, in contrast to incisional hernia which is more common following laparotomy in the obese.

Norris wrote that ‘the elimination of postoperative wound dehiscence is entirely within the jurisdiction of the operating surgeon’. The type of incision, the choice of suture material, and the method of wound closure are of major importance in the aetiology of wound dehiscence but less important in the development of incisional hernia. However, the exact role of each is obscure since the majority of studies have compared different suture techniques using different materials.

Wound dehiscence of appendicectomy incisions is less common than that of midline laparotomy incisions. There is a very low incidence of incisional hernia (0.37 per cent) and no wound dehiscence associated with the use of a lateral paramedian incision, but this approach is time-consuming to perform and not suitable for emergency situations. Midline, standard paramedian, and transverse incisions all have a similar incidence of incisional hernia. Upper abdominal incisions may be more prone to dehiscence than lower abdominal incisions. However, incisional hernia is considerably more common through incisions more than 18 cm in length and when a stoma is created through the wound.

The theoretical advantages of a mass closure of abdominal incisions have found support in many clinical studies. For example, a 10.3 per cent wound dehiscence rate with layered closure with catgut has been reported and is to be compared with a rate of 0.93 per cent using a mass closure with interrupted steel sutures. The use of deep tension sutures to support a layered catgut closure was reported to prevent wound dehiscence but not subsequent incisional hernia. A reduction in the rate of wound dehiscence from 3.8 per cent to 0.8 per cent was reported on changing the technique of wound closure from layered catgut to mass closure with nylon or Dexon®. However, mass closure and layered closure of abdominal incisions are associated with similar rates of incisional hernia.

The technique of mass closure depends on the placement of sutures through all layers of the abdominal wall except the skin at a distance of more than 1 cm from the edge of the musculoaponeurotic layers of the anterior abdominal wall. Such sutures should not be placed more than 1 cm apart, and the total length of suture used in a continuous mass closure should be more than four times the length of the wound. The sutures should not be placed under tension, to allow for postoperative swelling of 30 per cent. Tension may be associated with ischaemia of the tissue enclosed within the mass closure and the development of incisional hernias.

The use of catgut in the closure of all but appendicectomy wounds is now uncommon because of the high rate of wound dehiscence. Nylon suture is now most commonly used, since it is easier to handle than steel wire. The use of nylon sutures and a mass closure technique is associated with an incidence of wound dehiscence below 1 per cent. Jenkins reported only one dehiscence in 1505 abdominal wound closures. However, persistent wound sepsis and discharging sinuses have been associated with the use of non-absorbable sutures, and synthetic absorbable sutures that retain their strength far longer than catgut have certain theoretical advantages. Polyglycolic acid (Dexon® , polyglactin (Vicryl® ) and polydioxanone (PDS® ) have been used successfully for the closure of laparotomy wounds. The rate of wound dehiscence is comparable to that following mass closure with nylon, but there is no reduction in the rate of sinus formation and the incidence of incisional hernia is higher.

Increased intra-abdominal pressure due to inadequate postoperative analgesia, vomiting, the development of a postoperative chest infection resulting in coughing, and gross distension from paralytic ileus are important in the aetiology of both wound dehiscence and incisional hernia. Improvements in anaesthetic technique, postoperative care for the prevention of chest infection and good postoperative analgesia have contributed to the reduction in the incidence of wound disruption seen in recent years. Wound sepsis also predisposes to the development of wound dehiscence and incisional hernia.

Wound disruption may be occult or overt and partial or complete. Overt wound dehiscence follows removal of the skin sutures. The skin incision may partly or completely open to allow frank evisceration or show the presence of a herniation of bowel through a partial or complete defect in the musculoaponeurotic closure. Occult wound dehiscence occurs with disruption of the musculo-aponeurotic layers beneath intact skin sutures.

Wound dehiscence is at least twice as common in men as in women and is more common in patients over the age of 60. Wound disruption occurs on the sixth to ninth postoperative day in over 55 per cent of patients. Twenty-one per cent of the dehiscences occur on removal of the sutures. The patient may show signs of gross dehydration, a rise in temperature and pulse rate, and a peripheral leucocytosis, especially if an occult dehiscence has been overlooked for more than 24 h. Signs of bronchopneumonia and meteorism may also be present. A copious serosanguinous discharge presages dehiscence in about one-third of patients. Alternatively, a boggy swelling or frank sepsis of the wound may be the only signs of dehiscence of the fascial layer.

The first priority in the treatment of wound dehiscence is the correction of fluid depletion and electrolyte disturbance. Patients are often dehydrated, and cardiovascular collapse secondary to sepsis and shock from evisceration of the bowel must be treated prior to the administration of a general anaesthetic for repair of the dehiscence. If occult dehiscence is suspected, removal of skin sutures allows direct inspection of the fascial closure. While resuscitation is in progress, the dehiscence is covered by the liberal application of gauze swabs soaked in normal saline; a Velcro corset or many-tailed bandage may prevent further evisceration.

Confirmed dehiscence of an abdominal wound is best treated by resuture. At operation the skin sutures and the remnants of the previous fascial closure are removed. The edges of the wound are debrided of necrotic tissue and the wound resutured with a careful mass closure using No. 1 nylon. Gross abdominal distension may be due to an ileus or intestinal obstruction secondary to early formation of adhesions. Decompression of the bowel by retrograde milking of the intestinal contents into the stomach and nasogastric aspiration may considerably ease the process of closure of the wound; division of any obstructing adhesions is important.

Very rarely, the patient is unfit for surgery or the disrupted wound is too grossly contaminated to allow immediate surgery. Packing of the wound to return the bowel to the abdominal cavity followed by the application of strapping may allow the patient's condition to improve over a few days so that the wound may be closed as a secondary procedure. Wound dehiscence treated conservatively is inevitably followed by development of incisional hernia. Even after resuture, incisional hernia develops in almost 50 per cent of patients.

The prognosis of wound dehiscence is grave, and becomes worse with advancing age and with gross suppuration of the wound. Mortality rates of 24 and 15 per cent have been reported. The outcome is better if the wound disruption is recognized and treated early, in patients whose wounds are clean, and if prolapse of the intestine does not occur. Death is most commonly due to multisystem failure.

Incisional hernias can develop at any age, although the mean age of patients developing this complication is 58 years, compared with 46 years for patients with intact wounds. Although clinical evidence of incisional hernia may be delayed for more than 10 years after laparotomy and less than 50 per cent of incisional hernias are apparent at 1 year, the use of radio-opaque markers has shown separation of the musculoaponeurotic layers as early as 1 month postoperatively in patients who subsequently develop incisional hernias.

The presentation of incisional hernia depends on the site of the original wound, the size of the neck of the hernia, the size of the hernia, and the presence of complications. Small defects in the scar may result in large hernias, and this may predispose to incarceration and strangulation. Large hernias are unsightly and may give rise to abdominal discomfort (Fig. 1) 1444. Pressure necrosis and ulceration may occur in the skin overlying a large hernia.

The majority of incisional hernias are asymptomatic, and the majority of symptomatic hernias may be managed by conservative measures. Obese patients benefit from weight reduction, since this reduces intra-abdominal pressure and may render a symptomatic hernia asymptomatic. Some patients find benefit from support by an elastic corset. In the fit patient, the indications for elective repair of incisional hernia are discomfort, enlargement, unacceptable appearance, or significant risk of strangulation. The presence of incarceration or strangulation is an indication for emergency surgery.

Factors which predispose to the development of incisional hernias are relative contraindications to repair. Weight reduction improves the chance of successful repair in obese patients. Cessation of smoking and optimization of respiratory function reduce the chances of postoperative cough. In patients with large incisional hernias, the contents of the hernia may have lost the right of domain in the abdomen: return of the contents of the hernia to the abdominal cavity may result in increased intra-abdominal pressure, splinting of the diaphragm, and a significant reduction in pulmonary reserve in patients with chronic respiratory disease. Recurrence of malignant disease, cachexia, ascites, renal failure, and hepatic failure are also contraindications to repair of an incisional hernia. The preoperative assessment of patients undergoing repair of incisional hernias is therefore important. Prophylactic administration of antibiotics is indicated to reduce the incidence of wound sepsis and the chance of failure of the repair.

Many different surgical approaches have been described for the repair of incisional hernias: no single technique is satisfactory for all hernias and surgical treatment has a high failure rate. The types of repair may be divided into five basic categories.

1.The defect may be repaired in the same way as a laparotomy wound is repaired, with a mass closure of No. 1 nylon.

2.The rectus sheath may be overlapped as in the ‘Mayo’ double-breasted ‘vest over pants’ repair.

3.The defect may be repaired by the use of a darn of nylon or fascia lata.

4.Lower midline incisions may be amenable to closure by swinging muscle over to close the defect.

5.Large defects may be repaired by implanting a non-absorbable mesh of tantalum, Marlex, Mersilene, or polytetrafluoroethylene (PTFE).

The initial steps in the repair of an incisional hernia are the same, irrespective of the technique used. Generally, the original incision is reopened, and it is often necessary to excise an ellipse of redundant skin. The skin and subcutaneous tissues are dissected from the hernia sac back to the defect in the musculoaponeurotic layers and 3 to 4 cm of the fascia around the hernia are exposed.

Mass closure of an incisional hernia may be appropriate if the defect can be approximated without undue tension. The peritoneum is opened, adhesions to the undersurface of the scar lysed to clear 3 to 4 cm on the peritoneal surface and the hernia sac, and its covering of weak scar tissue is excised. The hernia is then closed with a mass closure of interrupted or continuous non-absorbable suture such as No. 1 nylon. The sutures are placed according to the principles of mass closure of a primary wound. The repair may be reinforced by the addition of interrupted far-and-near sutures or with an onlay graft of Marlex mesh.

Opening of the peritoneum may result in postoperative ileus, and abdominal distension may compromise the security of the repair. In addition, dissection of the peritoneal and fibrous sac of the hernia may prove difficult and result in inadvertent enterotomy of underlying bowel. The ‘keel operation’ is an extraperitoneal repair for midline incisional hernias which avoids opening the peritoneum, minimizes postoperative ileus, and allows early mobilization. The hernial sac and the neck of the hernia are cleaned of fibrofatty tissue and the hernia is inverted by the placement of interrupted mattress nylon sutures to close the defect in the musculoaponeurotic layer. The use of relaxing incisions in the anterior rectus sheath should be avoided since the repair may fail through these incisions (Fig. 2) 1445.

The Mayo ‘double-breasted’ repair described for the repair of paraumbilical and epigastric hernias may be used to close small incisional hernias, especially if the direction of the original incision was transverse (for example, a hernia through an incision previously used for a transverse colostomy). After excising the hernial sac, the anterior rectus sheath is overlapped by a double layer of interrupted nylon mattress sutures so that the upper layer of the rectus sheath overlies the lower layer (Fig. 3) 1446.

Fascia lata and nylon darn repairs have largely been replaced by the use of non-absorbable implants. However, midline incisional repairs may be repaired by incising the anterior rectus sheath 1 cm from the edge of the hernial defect on each side and suturing the medial edges of the anterior rectus sheath to invert the hernial sac. The repair is completed by placing a nylon darn between the lateral edges of the incision in the anterior rectus sheath.

Nuttall has described a repair of lower abdominal midline incisional hernias by overlapping the rectus abdominis muscle on each side. The anterior rectus sheath is incised on each side of the defect, and the rectus abdominis detached from its insertion as close as possible to the pubic symphysis. Each rectus abdominis muscle is then reattached to the opposite side of the pubic tubercle with nylon sutures, and the overlap of muscle loosely sutured together. The operation is completed by the closure of the anterior rectus sheath with non-absorbable sutures.

Large defects should be repaired by implanting a mesh of non-absorbable material: the recurrence rate after mesh repair may be as low as 11 per cent, compared with 44 per cent after direct suture. Tantalum gauze has now been replaced with materials such as Marlex® , Mersilene® , and PTFE, though the last is expensive. The gauze is sutured to the edges of the defect in the musculoaponeurotic layers with a non-absorbable nylon suture. Both intraperitoneal and extraperitoneal placement have been described (Fig. 4) 1447.

A closed suction drain is placed deep to the skin closure and left until drainage has ceased. Postoperative care is as for any patient undergoing abdominal surgery, except that the patient is taught to sit up in ways which do not place tension on the repair. The recurrence rate following repair of incisional hernias is between 25 and 44 per cent, and second repairs are equally unsuccessful.

Hernias develop alongside 5 to 10 per cent of colostomies and 3 to 10 per cent of ileostomies. Parastomal hernias are more common when the stoma is sited lateral to the rectus muscle than when the stoma is brought out through the rectus abdominis. The siting of a stoma in the incision used for laparotomy is associated with a very high incidence of incisional hernia. The extraperitoneal approach for the formation of a colostomy does not prevent the development of a parastomal hernia. A chronic cough, obesity, malnutrition, postoperative sepsis, and abdominal distension may predispose to the formation of a parastomal hernia.

The majority of parastomal hernias are asymptomatic, and only 10 to 20 per cent of patients require repair. Small hernias may be controlled by the use of a well-designed colostomy belt. The indications for surgical intervention are mainly related to difficulties in maintaining the stoma appliance. If the hernia is large, it may be difficult to apply a bag. The patient may be unable to see the stoma because of a large hernia (Fig. 5) 1448. Reduction of the hernia on lying down and its prolapse on standing may cause the appliance to dislodge. The development of strangulation of a parastomal hernia is an absolute indication for repair. Some patients find large parastomal hernias cosmetically unacceptable. The contraindications to repair are the same as for repair of an incisional hernia.

Two surgical options are available for the repair of a parastomal hernia. The stoma may be resited elsewhere and the original stoma site closed, repairing the hernia at the same time: this is appropriate for patients in whom the original stoma site is unsatisfactory. In some patients who have had multiple abdominal incisions, however, the choice of sites for a stoma may be limited. Under these circumstances, local repair of the hernia is indicated.

Local repair may be performed via a peristomal incision. The defect in the abdominal wall musculature is then repaired with interrupted nylon sutures. An alternative is to approach the stoma via an incision at least 10 cm from the stoma, using the original laparotomy incision. The stoma is approached by subcutaneous dissection and repaired by the placement of a collar of Marlex or other non-absorbable mesh. The technique has the advantage of avoiding a fresh surgical incision in the vicinity of the stoma, which may cause problems with the fitting of an appliance.

Bucknall TE, Cox PJ, Ellis H. Burst abdomen and incisional hernia: a prospective study of 1129 major laparotomies. Br Med J 1982; 284: 931–3

Donaldson DR, Hegarty JH, Brennan TG, Buillou PJ, Finan PJ, Hall TJ. The lateral paramedian incision-experience with 850 cases. Br J Surg 1982; 69: 630–2

Dudley HAF. Layered and mass closure of the abdominal wall: a theoretical and experimental analysis. Br J Surg 1970; 57: 664–7

Ellis H, Gajraj H, George CD. Incisional hernias: when do they occur? Br J Surg 1983; 70: 290–1

Goligher JC, Irvin TT, Johnston D, de Dombal FT, Hill GL, Horrocks JC. A controlled clinical trial of three methods of closure of laparotomy wounds. Br J Surg 1975; 62: 823–9

Grace RH, Cox S. Incidence of incisional hernia after dehiscence of the abdominal wound. Am J Surg 1976; 131: 210–2

Greenall MJ, Evans M, Pollock AV. Midline or transverse laparotomy? A random controlled clinical trial. Part 1: Influence on healing. Br J Surg 1980; 67: 188–90

Guiney EJ, Morris PJ, Donaldson GA. Wound dehiscence. Arch Surg 1966; 92: 47–51

Lamont PM, Ellis H. Incisional hernia in re-opened abdominal incisions: an overlooked risk factor. Br J Surg; 75: 374–6

Leaper DJ, Pollock AV, Evans M. Abdominal wound closure: a trial of nylon, polyglycolic acid and steel sutures. Br J Surg 1977; 64: 603–6

Leslie D. The parastomal hernia. Surg Clin N Am 1984; 64: 407–15

Maingot R. Umbilical and incisional hernia. In: Maingot R, ed. Abdominal Operations. New York: Appleton-Century-Crofts, 1980: 1618

Nuttall HCW. Rectus transplantation for midline incisional herniae. Br J Surg 1937; 25: 344–50

Pearl RK. Parastomal hernias. World J Surg 1989; 13: 569–72

Pollock AV, Evans M. Early prediction of late incisional hernias. Br J Surg 1989; 76:953–4

Read RC, Yoder G. Recent trends in the management of incisional herniation. Arch Surg 1989; 124: 485–8

Sjödahl R, Anderberg B, Bolin T. Parastomal hernia in relation to the site of the abdominal stoma. Br J Surg 1988; 75: 339–41

Zrukowski ZH, Matheson NA. ‘Button hole’ incisional hernia: a late complication of abdominal wound closure with continuous non-absorbable sutures. Br J Surg 1987; 74: 824–5