Clinical Scenarios in Surgery: Decision Making and Operative Technique (Clinical Scenarios in Surgery Series), 1 Ed.

Chapter 93. Necrotizing Enterocolitis

Richard Herman

Daniel H. Teitelbaum

Presentation

Patient A: This infant is a former 32-week premature infant who is now 3 weeks old weighing 1,800 g. He presents with increased residuals after enteral feedings, emesis, and abdominal distension. He has had no fevers. Abdominal exam is soft, distended without signs of erythema.

Patient B: This infant is a former 24-week premature infant now 10 days old weighing 900 g who presents with fever and abdominal distension. Abdominal exam is distended, tender, and discolored.

Differential Diagnosis

Acute neonatal abdominal distension is concerning for necrotizing enterocolitis (NEC), especially in the premature infant. However, there are many other causes for neonatal abdominal distension. These can vary from other surgical emergencies to relative benign entities. Potential surgical diagnoses/emergencies begin by distinguishing between those infants presenting acutely with abdominal distension in the newborn period and those presenting a few days to weeks later. The former would include neonatal ascites, distal intestinal obstruction, or obstruction of the genitourinary system. The latter would include intestinal volvulus, isolated intestinal perforation, Hirschsprung’s disease, or a partial obstruction of the gastrointestinal tract, such as an ileal stenosis. Nonsurgical pathologies include a functional ileus due to sepsis, bacterial or fungal, pseudomembranous colitis, or primary ascites or secondary ascites due to cardiac anomalies or other medical conditions.

Workup

After a full history and physical examination, both patients in Cases A and B will require a full set of laboratory values including a complete blood cell count, comprehensive metabolic panel, a blood gas, and routine abdominal radiographs (typically supine, and left and right decubiti films to assess for free air). Contrast enemas are contraindicated when considering a diagnosis of NEC; as such studies may result in colonic perforation. An upper gastrointestinal series is occasionally performed in a child where the diagnosis of a malrotation with volvulus is entertained.

Patient A has a slightly elevated white blood cell count of 27,000 cells/mm3 and a mildly depressed platelet count of 90,000/mm3. The abdominal radiograph shows pneumatosis intestinalis without evidence of pneumoperitoneum or portal venous gas (Figure 1).

images

FIGURE 1 • Abdominal x-ray showing pneumatosis (Red arrows).

Patient B has a low white blood cell count of 3,000 cells/mm3 and is also thrombocytopenic with a platelet count of 60,000/mm3. The abdominal radiograph is significant for pneumoperitoneum (Figure 2).

images

FIGURE 2 • Cross table lateral film showing pneumoperitoneum (Green arrow).

Discussion

NEC accounts for approximately 1% to 7% of all admissions to neonatal intensive care units, with infants weighing <1,500 g having an estimated incidence between 10% and 12%. Aside from prematurity and low birth weight, other risk factors include rapid initiation of feedings and certain medications, including indomethacin, theophylline, and aminophylline. About 90% of all cases of NEC occur after the initiation of enteral feedings; therefore, if a neonate has been NPO, development of NEC is less likely. The rate of advancing feeds has also been closely linked with the development of NEC. Certain nonspecific physical examination findings include lethargy, temperature instability, apnea, bradycardia, hypoglycemia, and shock. More specific abdominal findings include feeding intolerance (increased gastric residuals), abdominal distension, and blood per rectum (although not the most common cause for neonatal hematochezia). A fixed mass and palpable bowel loops are more concerning for necrotic intestine. Color change of the abdominal wall (darkened or erythematous) and scrotal color change to a reddish or a bluish hue are quite suggestive of advanced NEC. In the latter case, this may denote intestinal perforation with passage of meconium into a patent processus vaginalis. Vomiting and diarrhea can also been seen.

Though laboratory values are also nonspecific in the diagnosis of NEC, the following changes can be seen: an elevated or a depressed white blood cell count, thrombocytopenia, metabolic acidosis, and an elevated C-reactive protein. A depressed platelet count is quite common in neonates with septicemia of any cause, but a decline by more than 50% may suggest advanced NEC. Radiologic signs include pneumoperitoneum, portal venous gas, pneumatosis, intraperitoneal fluid, and fixed bowel loops on serial radiographs. With good ultrasonographic imaging, NEC can be detected with sensitivity near 100% in some series. Ultrasonographic findings include stagnant, nonmobile bowel loops, target sign (signifying the abnormal loops), portal venous air, and abdominal fluid. Stagnant, nonmobile loops also suggest necrotic bowel.

Diagnosis and Treatment

The treatment of NEC can vary from nonoperative management to surgical intervention. Even among each of these pathways, variability exists. In the patients with a mild course, or low suspicion, a 1-week course of antibiotics is recommended. This can be prolonged up to 2 weeks in some of the more complicated cases. Patient A would fall into this category and would likely receive a 2-week course of antibiotics while being maintained NPO. Surgical intervention can also vary from peritoneal drain placement to laparotomy with bowel resection, which would be indicated for patient B. The indications for each of these approaches will be discussed later on in this chapter (see Figure 3).

images

FIGURE 3 • Suggested management of NEC. Note, decision management may vary depending on the clinical presentation, and the weight threshold of 1,500 g is only a guideline. NEC, Necrotizing enterocolitis; NGT, nasogastric tube.

Surgical Approach

One of the foremost decisions to make in an infant with NEC is determining if a surgical intervention is required. For all of the above-mentioned tests, the only absolute indication for surgery, whether it is a peritoneal drain placement or laparotomy, is pneumoperitoneum. Kosloske et al. described 12 criteria for operating in NEC and classified them into categories of best, good, fair and poor indicators for the need of surgical intervention (Table 1).

TABLE 1. Indications for Operative Intervention

images

Kosloske AM. Indication for operation in necrotizing enterocolitis revisited. J Pediatr Surg. 1994;29:663.

Once the decision to operate is made, one must decide the surgical approach. Over the past decade, tremendous controversy has arisen regarding the most optimal intervention. While an exploratory laparotomy has typically been viewed as the gold standard of care, many surgeons have found that primary peritoneal drainage was an effective treatment of NEC. The earliest suggestion of the effective use of primary peritoneal drainage was first published by Ein et al. in 1977 as a means of stabilizing premature neonates with intestinal perforations. In 1994, Morgan et al. reported a 79% survival rate for infants <1,500 g treated with primary peritoneal drainage. These earlier studies led to a prospective, randomized, controlled trial by Moss et al. In this study, primary peritoneal drain placement in preterm infants <1,500 g demonstrated no difference in survival, or other clinically important early outcomes, compared to an open laparotomy. Another recent prospective randomized controlled trial by Rees et al., from 13 countries, demonstrated conflicting results from Moss’s trial, where peritoneal drainage was found to be relatively ineffective. However, in the Rees trial, the study was limited to infants <1,000 g versus 1,500 g in the Moss study. However, differences are striking between the studies, as Rees et al. had a 74% conversion rate of peritoneal drain placement to laparotomy after a median of 2.5 days, compared to a 9% conversion in the Moss study. Reasons for conversion were increasing inotrope and ventilatory support, increasing or persistent pneumoperitoneum, palpable mass with persistent pneumoperitoneum, drainage of stools from drain site. A prospective cohort study by Blakely et al. in 2006 showed concurring results to Rees; however, the patients were not randomized, and the patients in the peritoneal drainage group were, younger, smaller, and sicker; more were on vasopressors and high-frequency ventilation. However, only 23% of the peritoneal drain placement patients ultimately received a laparotomy. In their conclusions, they state that drain placement is not an effective treatment for advanced NEC. Thus, the precise approach remains controversial. A suggested approach is shown in Figure 3.

Surgical Approach

Peritoneal Drainage: Drains are placed by making a small incision, usually in the right lower quadrant and dissecting into the peritoneum. Enteric contents are allowed to escape and be cultured. One often can do a limited inspection of the intestine in this area and should also consider performing an irrigation of the abdominal cavity. Afterward, either one- or two ¼ -in. Penrose drains are inserted into the peritoneum. The first is placed transversely and the second more superiorly. Often, additional holes may be cut into the Penrose drains to increase drainage. Drains are sutured in place and placement confirmed with radiographs (Figure 4A,B). After 1 week and the fluid draining has decreased and cleared, the drain is slowly backed out.

images

FIGURE 4 • A: Abdominal x-ray post peritoneal drain placement. (Drain marked with arrow.) B: View of peritoneal drain placement in right lower quadrant.

Open Laparotomy: Laparotomy is typically performed on all infants larger than 1,500 g, those <1,500 g based on surgeon preference, or clinical failure of drain placement in the first 48 hours. An infraumbilical transverse incision is made and the peritoneal cavity is entered. The bowel is carefully run and the areas of obvious necrosis are resected. Often, it may not be completely apparent during the initial exploration as to the extent of necrosis. In these cases only the grossly necrotic portions are removed, and the ends of the resected bowel are tied off and dropped into the peritoneal cavity. A second-look laparotomy is then performed between 24 and 48 hours to examine questionable areas. The most proximal part of the bowel is brought up as an ostomy and the remaining viable portions of bowel re-anastomosed. Some authors have reported successful management by avoiding an ostomy and performing a primary anastomosis. While this may work with very limited disease, more diffuse NEC will almost always result in a diverting ostomy.

Each of the above procedures carries its own risk. For peritoneal drain placement, visibility is poor and hepatic injury, or small intestinal injury can occur during placement. Laparotomy complications include hepatic injury, intestinal leak, sepsis, bleeding due to a typical coagulopathic state, stoma necrosis, and recurrent NEC. Multiple complications can be seen with the creation of an ostomy in the neonate; infections, dehiscence, stenosis, hernias, prolapse, retraction, and bowel obstructions. In particular, most infants are coagulopathic and the liver lacks normal fibrose supportive structures. In these cases, even minor trauma to the liver may result in virtually noncontrollable hemorrhage. Correction of coagulopathy and even injection of fibrin glue directly into the liver fracture may be helpful; however, such injuries often may lead to significant morbidity or death.

Special Intraoperative Considerations

There are two main special intraoperative considerations in NEC. The first is that one must always consider bowel preservation and the prevention of short bowel syndrome if possible. Often, multiple blind loops or additional ostomies can be created, to facilitate saving as much bowel as possible. At times, a single proximal stoma may be performed, and the more distal bowel segments are joined together and allowed to heal. One may stent open these distal anastomoses over a silastic catheter.

The second consideration is to decide if one is dealing with NEC-totalis, the most severe case of NEC. This diagnosis is made when <10% of the bowel is viable. Most of these infants are <1,000 g and mortality has ranged from 42% to 100% depending on the series. If these infants survive, they will require long-term TPN, management of short bowel syndrome, and many ultimately need a small bowel transplant. When NEC totalis is encountered, immediate and very direct discussion with the parents is needed to decide whether ongoing treatment is desired. If there is minimal or no viable bowel the surgeon should seriously consider giving the parents an option not to continue care. This is important, particularly in a child who has other comorbidities that may make their quality of life substandard (e.g., severe intracranial hemorrhage).

Postoperative Management

For postoperative drain placement, the infant is kept NPO on total parenteral nutrition and antibiotics for approximately 14 days. The characteristics of the drain output are observed. After the output is minimal and no longer feculent in nature, the drain is slowing backed out starting 7 days after insertion. Removal is usually at a rate of 1 to 2 cm per day; the last part of the drain will usually fall out on its own. After 2 weeks have passed, feeds are slowly restarted. Feedings are generally started at a continuous infusion, and the hourly volume increases 1 mL daily. NEC strictures are common and occur in 9% to 36% of cases. In general, these strictures are found within the colon, most commonly at the splenic flexure, but as well on the left side of the colon. A contrast enema should be considered prior to the initiation of feeds and if the infant experiences feeding intolerance.

Any patient who has undergone drain placement needs to be closely monitored and evaluated for potential conversion to laparotomy. Conversion should be considered in any patient who fails to improve, or whose condition worsens. Deterioration of the clinical condition includes increasing vasopressor or ventilator requirements, organ failure including renal failure, persistent pneumoperitoneum, worsening coagulopathy, or a worsening lactic acidosis.

Postlaparotomy care is similar with respect to 2 weeks of NPO, TPN, and antibiotics. Feeds are then usually slowly restarted after gastrointestinal function has returned. If the proximal ostomy has a high output, then one should make an attempt to “refeed” the output from the ostomy into the mucous fistula. The reestablishment of gastrointestinal continuity is done approximately 8 weeks postoperatively. Prior to reestablishing continuity, a contrast enema study should be performed to check for strictures or other anatomic problems.

Case Conclusion

Patient A: Undergoes nonoperative management, is followed with routine x-rays, placed on antibiotics and NPO for 14 days. Feeds are slowly restarted and the patient is monitored for feeding tolerance.

Patient B: Undergoes placement of peritoneal drain, made NPO and given intravenous antibiotics. Starting postoperative day 7 the drain is slowly backed out. After a completed 2-week course of antibiotics, and return of bowel function, enteral feeds are slowly started after a contrast enema demonstrates no colonic stricture.

TAKE HOME POINTS

·        Confirm diagnosis of NEC versus other pathology.

·        Obtain appropriate imaging and laboratory values.

·        Decide operative versus nonoperative management.

·        For nonoperative management, decide duration.

·        If operative management, decide on drain placement versus Laparotomy.

·        Preserve as much bowel as possible.

·        Be aware of post-NEC strictures.

·        When reestablishing enteral feedings, begin a slow feeding course, advance of one ml of hourly feeds daily.

SUGGESTED READINGS

Blakely ML, Tyson JE, Lally KP, et al. Laparotomy versus peritoneal drainage for necrotizing enterocolitis or isolated intestinal perforation in extremely low birth weight infants: outcomes through 18 months adjusted age. Pediatrics. 2006;117(4):680–687.

Kim SS, Albanese CT. Necrotizing enterocolitis. In: Grosfeld J, et al., ed. Pediatric Surgery. 6th ed. Philadelphia, PA: Mosby Elsevier, 2006:1427–1453.

Kosloske AM. Indication for operation in necrotizing enterocolitis revisited. J Pediatr Surg. 1994;29:663–666.

Morgan LJ, Shochat SJ, Hartman, GE. Peritoneal drainage as primary management of perforated NEC in the very low birth weight infant. J Ped Surg. 1994;29:310–314.

Moss RL, Dimmitt RA, Barnhart DC, et al. Laparotomy versus peritoneal drainage for necrotizing enterocolitis and perforation. N Engl J Med. 2006;354(21):2225–2234.

Rees CM, Eaton S, Kiely EM, et al. Peritoneal drainage or laparotomy for neonatal bowel perforation? A randomized controlled trial. Ann Surg. 2008;248(1):44–51.



If you find an error or have any questions, please email us at admin@doctorlib.info. Thank you!