In children, stomas should be rare and reversible. Although the decision to create a stoma is to some degree always a matter of judgment, there are some indications that are largely technical (high imperforate anus, long-segment Hirschsprung disease, ileal pouch-anal anastomosis), some that are somewhat obvious because an anastomosis would clearly be unsafe (NEC with necrotic bowel, fecal peritonitis with sepsis), and finally some that are more or less a matter of style or personal preference on the basis of experience and training. These decisions often serve to define a surgeon as “conservative” or “aggressive.” Regardless, because stomas are not without complications and make life difficult for the patients and their parents, the decision to create a stoma should not be taken lightly, the surgeon should be able to justify the decision in that particular patient in that particular circumstance, and, to the extent possible, the parents’ opinion should be sought ahead of time. It is clear that many of the traditional indications for creation of a stoma are historically based and, although we will likely never see a randomized trial to prove it, there are many surgeons with the courage to challenge surgical dogma and who have generated a great deal of experience that supports the informed and thoughtful decision to avoid a stoma under certain circumstances.
The Bishop–Koop ileostomy was a major advance in the surgical treatment of meconium ileus when it was first introduced several decades ago and many pediatric surgeons still routinely use it in these patients. Many surgeons, however, have found that evacuation of the inspissated meconium by irrigating through an enterotomy made in the dilated portion of the ileum that is then closed primarily works well and avoids the issues related to the care and closure of a stoma. Likewise, stomas should almost never be necessary in healthy children after ileocecal resection for intussusception, severe appendicitis, or even Crohn’s disease. Leveling colostomy in healthy infants with short-segment Hirschsprung disease should rarely, if ever, be necessary as primary repair in the newborn period has proven to be safe and effective. Bowel perforations after blunt or penetrating trauma can usually be safely repaired primarily without creation of a stoma unless there are clear extenuating circumstances – significant delay in diagnosis, severe chemical peritonitis, profound sepsis. In the end, although it is clearly important to learn from historical precedent, the surgeon should be able to justify the creation of a stoma with more than “because this is how it has always been done.”
Despite the historical variation in types of stomas, stomas are essentially either end-stomas, in which case they should be matured in the manner of a Brooke ileostomy, or some modification of the loop ileostomy. End-colostomies can be flush but function better and are easier to care for when they are matured somewhat. There is no need to routinely use a rod or tube to secure a loop stoma – proper suture placement and a small but adequate fascial opening should be enough to prevent retraction of the stoma. In the case of Hirschsprung disease or imperforate anus, the mucous fistula should be brought up separately so as to avoid spillover and subsequent impaction of stool. To prevent prolapse, the mucous fistula should not be matured and, in fact, is probably best made flush and rather small, perhaps by using just the corner of a stapled bowel end closure. In most other cases, a double-barrel or Turnbull stoma should be used. The latter involves “Brooking” the proximal end and leaving the distal end flush with the skin. Most of the other stomas described were designed to allow closure at the bedside because a trip to the operation room was so dangerous. There is little need to use these old-fashioned constructs today.
Complications of stomas are not uncommon but are largely preventable. Parastomal hernias are often due to an excessively large fascial opening and improper suture fixation of the bowel. The same factors predispose to prolapse, which is one of the more frustrating complications of stomas. Prolapse is more common in mucous fistulas, probably due to the effect of peristalsis in bowel that has been fixed in this way. Maturing the mucous fistula makes matter worse because the everted segment of bowel peristalses in the direction of prolapse. Prolapse of the proximal segment usually occurs in stomas created when the bowel was extremely dilated and has now decompressed to a more normal size. In these cases, it is probably best to either find a way to remove the stoma or to completely re-site it. Proximal jejunal stomas should be avoided if at all possible because of the fluid and electrolyte and skin care problems that arise from the high output. Whenever possible, all stomas should be placed in the lower abdomen at a site that has been carefully marked by an enterostomal therapist. Transverse colostomies should also be avoided as they are difficult for patients to manage and accept. Bowel obstruction frequently occurs at or near ostomies, sometimes simply due to adhesions and other times due to volvulus around the stoma. Operation should not be delayed simply because the surgeon is convinced that “the stoma is patent” based on digital exam or intubation of the stoma. Finally, an examining finger should rarely if ever be placed in a child’s ileostomy due to the significant risk of circumferential injury to the bowel or its blood supply.