Doi:10.1016/s1084-2756(03)00025-3

Seminars in Neonatology (2003) 8, 223–232
Surgical treatment of infants with necrotizing
enterocolitis
Agostino Pierro*, Nigel Hall
Department of Paediatric Surgery, The Institute of Child Health and Great Ormond Street Hospital forChildren NHS Trust, University College London, London, UK Summary With the improvements in neonatal intensive care, necrotizing enterocolitis
KEYWORDS ;
(NEC) has become the most common gastrointestinal emergency amongst infants in neonatal intensive care units. The incidence of NEC varies between 1 and 8% of neonatal intensive care unit admissions and the disease has a mortality rate between 20 and 40%. There are a number of surgical options available to the paediatric surgeon depending on the clinical condition of the infant and the extent of the disease.
However owing to a paucity of prospective data in this field and a lack of randomizedcontrolled trials there is little consensus as to which is the most appropriate. Primaryperitoneal drainage has become very popular in North America and Europe for thetreatment of perforated NEC in very low-birthweight infants. It is a useful manoeuvrein the resuscitation of critically ill infants and in some of these infants, furtheroperation may be avoided completely by inserting a peritoneal drain. Others howeverremain too unwell to undergo laparotomy and may die. Two randomized controlledtrials are currently underway to determine the real benefit of peritoneal drainage.
Laparotomy in very small neonates has become safer with improvements in anaesthe-sia and intensive care management. Resection and primary anastomosis has beenproposed as a valid treatment modality in neonates with both focal and multifocaldisease. The advantage of resection and primary anastomosis over stoma formation isstill controversial. Different surgical techniques such as diverting jejunostomy or ‘clipand drop’ have been described to deal with extensive disease and avoid massive smallbowel resection. Prospective studies and randomized controlled trials are needed todefine the best operative treatment for neonates with severe NEC.
2003 Published by Elsevier Ltd.
Introduction
nized risk factors such as congenital heart disease.
The majority of cases respond to intensive medical Necrotizing enterocolitis (NEC) is the most common treatment, which involves treatment of sepsis with surgical emergency in the neonatal period. With parenteral broad-spectrum antibiotics, correction advances in neonatal intensive care, the incidence of hypovolaemia and acidosis, ventilatory sup- is increasing and the disease affects up to 0.5% of port of respiratory failure and other supportive all live births and 3–5% of low-birthweight live measures. The intestine is decompressed by means births. Whilst primarily a disease of pre-term of a nasogastric tube and rested for a period of at infants, it is also observed in term infants, particu- least 7–10 days during which time nutritional intake larly those with co-existing morbidities and recog- is maintained by means of total parenteral nutrition(TPN). Despite excellent medical attention, a * Corresponding author: Department of Paediatric Surgery, number of infants develop more severe NEC requir- ing surgical intervention. This group of infants rep- WC1N 1EH, UK. Tel.: +44-20-7905-2175/2641; fax: +44-20- resents a major challenge for paediatric surgeons.
7404-6181E-mail address: a.pierro@ich.ucl.ac.uk (A. Pierro).
Despite increasing experience of this disease, there 1084-2756/03/$ - see front matter 2003 Published by Elsevier Ltd.
doi:10.1016/S1084-2756(03)00025-3 remains little consensus as to the most appropriate Indications for surgery in infants with NEC timing and nature of surgery. In addition to surgeryin the acute phase, a number of infants develop late complications of NEC, either following medi- PneumoperitoneumClinical deterioration despite maximal medical treatment cal or previous surgical treatment, and these Abdominal mass with persistent intestinal obstruction or complications require surgical intervention.
Paucity of evidence
Increased abdominal tenderness, distension and/or Whilst there are a number of surgical techniques in use for the management of NEC, there is little agreement amongst paediatric surgeons as to the approach most beneficial to the child. This stems primarily from the lack of quality prospective dataor randomized controlled trials concerning this con-dition. Although the disease is not rare, the share ofcases seen in each individual centre is sufficientlysmall to preclude large-scale studies on a local perforation associated with persistent signs of basis. The majority of published data comparing treatment modalities are retrospective reviews of Various authors have suggested other indications relatively small series of patients. There is inevi- for surgery, some of which remain controversial.
table bias within these studies and the numbers Occasionally, a laparotomy is performed because of involved are too small to draw reliable conclusions.
increased abdominal tenderness and distension In order for one surgical method to be proven and/or discolouration of the abdominal wall. The superior to another, large-scale, multicentre, differential diagnosis in these cases include midgut prospective, randomized controlled trials are volvulus, Hirschsprung's enterocolitis and, rarely, required. To date, these have been particularly intussusception.6 Similarly, a fixed dilated intesti- sparse in the field of paediatric surgery. This article nal loop defined by persistent location and con- focuses on the operative management of infants figuration for more than 24 h has been a proposed with severe NEC and on the recent advances in this indication for surgery. However, approximately half of the patients with this finding recover with-out undergoing an operation.5,7 The presence of gas Indications for surgery
in the portal vein on abdominal radiograph has beenproposed as an indication for surgery, although this Severity of NEC is classified according to the Bell is not accepted universally it does appear to carry a criteria which have been modified by Walsh and relatively poor prognosis.8 Rowe et al.5,9 suggested Kliegman.1 Whilst this system may aid some thera- that more than 90% of infants with portal vein gas peutic decisions and provide limited prognostic will develop intestinal necrosis, and approximately information, it does not take into account the half will have panintestinal involvement. These extent of disease on an anatomical basis which authors suggest operative intervention on appear- ance of portal vein gas on plain abdominal radio- There is great controversy surrounding the indi- graphs. Finally, according to Kosloske et al.,3 a cations for surgery in infants with NEC. Table 1 positive paracentesis (defined as aspiration of more summarizes the indications reported in the litera- than 0.5 ml of brown or yellow-brown fluid with ture. The indications for operation in a recent study visible bacteria on Gram stain) is suggestive of of neonates with NEC included pneumoperitoneum gangrenous bowel thereby indicating surgical inter- in 45% of the neonates, clinical deterioration in 37% and intestinal obstruction in 18%.2 The most widely have been reported on the presence of sealed accepted indication for surgery is the presence of pneumoperitoneum. Unfortunately, pneumoperi- toneum is not always demonstrable in neonates infants with severe NEC11 and its prevalence is with intestinal perforations.3–5 Other absolute indications for surgery in acute NEC include con- arotomy.12 The value of thrombocytopenia in tinued deterioration of an infant that is refrac- defining the extent of the disease and the need for tory to maximal medical treatment, and an an operation is controversial. The trend of the abdominal mass or abscess secondary to intestinal platelet count during the course of the disease is Surgical treatment of infants with necrotizing enterocolitis considered particularly important. A sudden, pro- found drop in the platelet count appears to predictthe presence of gangrenous bowel13 and may indi- cate the need for operative intervention.14 Platelet consumption in combination with one or more other factors, such as abdominal mass or radiographic evidence of fixed abnormal bowel loops, is con- Proximal jejunostomy‘Clip and drop’ technique sidered an important criterion to determine the need for surgery.15 Persistent thrombocytopenia16or decreasing platelet count17 is considered import-ant in diagnosing clinical deterioration and theneed for surgery. A study of risk factors and severityindices in NEC showed that infants that were number of options exist, including resection with operated had significantly lower platelet counts enterostomy, resection with primary anastomosis, pre-operatively than the ones managed conserva- proximal jejunostomy, ‘clip and drop’ technique tively.18 Ververidis et al.2 showed that the greater the extent of the disease, the lower the platelet The patient's weight and clinical status as well as count. The nadir platelet count was significantly the extent of the disease influence the choice of lower in infants who died than in survivors. None of surgical intervention. At laparotomy, the extent of the patients with platelet count >100×109/l died. In the disease can be classified as focal when it is predicting intestinal gangrene, severe thrombo- limited to a single intestinal segment; multifocal if cytopaenia has a sensitivity of 69%, specificity of it includes two or more intestinal segments with 60% and positive predictive value of 89%, whilst a more than 50% of the small intestine viable and rapid fall in platelet count has a sensitivity of 32%, panintestinal when the majority of small and large specificity of 89% and positive predictive value of bowels are involved with less than 25% viable bowel 92%. Therefore, the presence of thrombocyto- paenia cannot reasonably predict the extent of thedisease or indicate the need for laparotomy; how- Primary peritoneal drainage
ever, a very low platelet count or a rapid fall in The management of intestinal perforation related platelet count is a negative prognostic sign.
to NEC in extremely low-birthweight infants(ELBW), defined as <1000 g, remains controversial.
Surgical treatment
In 1975, Marshall and Ein presented at the meetingof the Canadian Association of Paediatric Surgeons, Up to 50% of neonates with NEC develop advanced Winnipeg, Manitoba, Canada, the use of percu- disease that requires operative treatment.19 Butter taneous peritoneal drainage before laparotomy, as et al.20 reported an increase in operative rate from a method of stabilizing and improving the systemic 46% between 1990 and 1994 to 69% between 1995 status of premature infants with intestinal perfora- and 1999. According to this study, the increase was tion secondary to NEC. Initially, it was hoped that due to an increase in the percentage of stage III the drainage of air and stools from a child too patients and an increase in referrals of post-NEC unstable for a laparotomy would relieve symptoms strictures. To some it may be surprising that small, of abdominal compartment syndrome and infec- unstable, critically ill infants are able to tolerate tion, and subsequently improve tolerance of a major surgical interventions. In a recent study laparotomy. Two years later, Ein et al.24 from the investigating safety of the operative procedure Hospital for Sick Children in Toronto reported the itself, investigators found that even very low- use of peritoneal drainage without laparotomy (PPD) in the management of newborn infants with complicated NEC. Since then, there have been There is no general consensus concerning the several reports, which have suggested that PPD may ideal surgical management of NEC. Current surgical serve as a definitive therapy rather than an options include primary peritoneal drainage (PPD) intermediary for laparotomy. In 1990, the same authors25 presented their 13-year experience with The principal surgical objectives of laparotomy PPD in 37 patients with bowel perforation second- in acute NEC are to control sepsis, remove gan- ary to NEC. Sixty-five percent of these patients grenous bowel and to preserve as much bowel weighed less than 1000 g and 88% less than 1500 g.
length as possible.5,22,23 Within these objectives, a These patients were all septic, acidotic and too unstable to undergo a primary laparotomy. Of the same treatment modality (PPD/laparotomy). The 37 reported patients, one-third completely recov- authors concluded that, using the current available ered with PPD and did not require any further data, it is not possible to determine whether PPD is operative intervention. In the remaining two-thirds associated with improved survival, and that only a (26 patients), nine (24%) died rapidly before randomized controlled trial could identify the best laparotomy (within 24 h) for continued clinical In light of relatively poor quality evidence from deterioration, and seven (22%) underwent a retrospectively collected data, it is not surprising delayed laparotomy for bowel occlusion or fistula that there is little consensus amongst paediatric formation. The authors concluded that laparotomy surgeons. In an attempt to resolve this issue, there with resection and bowel anastomosis or stoma are currently two multicentre, prospective, ran- could be the preferred treatment for NEC and per- domized controlled trials in progress comparing foration, but they recommended PPD for ELBW peritoneal drainage with primary laparotomy in infants less than 1000 g (Pierro: NET trial) or 1500 g In 1988, Cheu et al.26 reported their experience with PPD in 51 of 92 infants with perforated NEC.
Two recent studies seem to indicate that PPD is They concluded that PPD was useful in resuscitating more advantageous in neonates with isolated per- ELBW infants with intestinal perforation secondary forations of the gastrointestinal tract not related to to NEC, but considered PPD a stabilizing procedure NEC.34,35 These perforations are usually isolated, and not an alternative to laparotomy. Other idiopathic or related to indomethacin treatment for authors have advocated the use of PPD as primary patent ductus arteriosus. The infants affected have treatment for NEC and perforation for all children neither pneumatosis intestinalis nor portal vein weighing less than 1500 g and for unstable babies more than 1500 g.27 In this study, 79% of patientssurvived after PPD and 17 out of 23 survivorsrequired no further operative intervention. In ad- Resection of affected bowel and
dition, the Toronto group28 reported a survival rate enterostomy formation
of 69% in neonates less than 1000 g treated withperitoneal drainage compared with a survival rate The resection of affected bowel and exteriorization of only 22% in neonates who underwent laparotomy of remaining segments was once thought to be the only safe surgical approach to the infant with gan- We recently conducted a systematic review of grenous bowel. This technique remains widely used the topic and failed to discover any randomized but has been superseded by more novel approaches trial on the usage of PPD in the management of in some centres. Resection of necrotic bowel in NEC. In spite of the lack of conclusive evidence, neonates with NEC has the theoretical advantage of peritoneal drainage has become a popular method reducing bacterial translocation and correcting the of treatment of ELBW infants with complicated septic state of the patients. The conventional view NEC, whilst other surgeons still prefer to perform a is that it is safer to exteriorize the bowel ends as laparotomy, resection of the diseased bowel, with the presence of peritonitis, inflammation of the either a primary anastomosis or stoma creation and bowel wall and the reduced intestinal blood supply peritoneal toilet for this disease.29–32 Resection of in patients with NEC are unfavourable factors for the necrotic bowel is considered by some authors to the healing of the anastomosis.36 In addition, the be the key factor in improving the condition of the stoma allows adequate healing and rests the distal neonate.32 Moss et al.33 recently performed a bowel prior to subsequent re-anastomosis.23 meta-analysis on this topic reviewing the results of There are, however, several disadvantages to 10 published studies. The combined probability of this approach. It is often difficult to re-establish survival was similar for PPD (55%) and laparotomy adequate enteral feeding sufficient for adequate (67%). A marked bias in treatment assignment was weight gain. High output stomas, in particular, found, with a greater proportion of smaller babies carry a risk of dehydration and electrolyte imbal- undergoing PPD compared with laparotomy (931 vs.
ance, and the importance of early closure to avoid chronic salt and water loss has been highlighted.
of birthweight on survival did not show a signifi- Enterostomies have been associated with signifi- cant difference in survival between PPD and cant morbidity whilst present, including stenosis, laparotomy. In addition, there were significant prolapse and excoriation of the surrounding skin.
variations in outcomes between institutions for the Stoma closure involves a second anaesthesia and is Surgical treatment of infants with necrotizing enterocolitis usually performed once the infant is thriving and et al.23 found that resection and primary anastomo- fully recovered from the acute stage of the illness.
sis was associated with a higher survival rate than Metabolic or physical problems, however, may de- enterostomy. In addition, resection and primary mand earlier surgery. In one series, the incidence anastomosis was not associated with higher inci- of complications in infants with enterostomies for dence of intestinal strictures and longer duration of NEC was 68%.37 This high rate has prompted the PN compared with resection and enterostomy.
search for alternative strategies which may avoid the need for repetitive surgery and complications associated with stomas, whilst ensuring that the domized controlled trials. It is possible that the underlying surgical principles of treating NEC by more critical patients with multifocal disease were allocated to enterostomy in preference over resec-tion with anastomosis, as indicated by the rela-tively high incidence of post-NEC strictures after Resection and primary anastomosis
enterostomy. It is notable, however, that in thestudy by Fasoli et al.,23 resection and primary Resection and primary anastomosis was once con-sidered a hazardous option because of the risk of anastomosis was associated with acceptable mor- anastomotic leakage due to poor healing of intra- bidity and mortality even in patients with NEC peritoneal anastomoses in the presence of peri- spread to multiple intestinal segments. The results tonitis, inflammation of the bowel wall and com- of intestinal resections and multiple primary anas- promised intestinal blood supply. In recent years, a tomoses for preservation of bowel length have not number of centres have published retrospective been reported in the literature. Hall and Pierro reviews of infants treated with intestinal resection (unpublished data) have treated seven neonates and primary anastomosis. Kieswetter et al.38 (weight 800–1800 g) with resection and multiple showed that in patients with localized NEC, an 89% intestinal anastomoses: four neonates, who had survival rate can be achieved with resection and two intestinal anastomoses, had no complications, primary anastomosis. Harberg et al.39 reported two, who had three and four intestinal anastomo- minimal morbidity and low mortality (11%) with ses, had anastomotic leaks and one neonate who resection and primary anastomosis in patients with had three anastomoses developed an anastomotic localized or multiple areas of the disease regardless stricture. The results of this small series are of the degree of peritoneal contamination. Sparnon encouraging, however, more than two intestinal and Kiely32 reported an uncomplicated and rapid anastomoses may carry an increased risk of recovery in seven of 10 neonates with primary anastomosis. Griffiths et al.40 reported 76% survival Most paediatric surgeons would agree that every after resection and primary anastomosis com- effort should be made to preserve the ileocaecal pared with 39% in patients who underwent resec- valve in neonates with NEC. Ladd et al.,47 in a tion and enterostomy. A similar favourable experi- retrospective review of infants with NEC, found ence with resection and primary anastomosis has no difference in duration of TPN, post-operative been reported by Pokorny et al.15 and Parigi stay and growth between infants with or without a et al.41. In contrast, Cooper et al.42 showed that the ileocaecal valve. Similarly, in a more recent study, survival rate of neonates that underwent primary there were no significant differences in duration anastomosis was lower (64%) than that of neonates of TPN, incidence of NEC recurrence and hospital that had an enterostomy (79%). Ade-Ajayi et al.43 stay between 40 neonates with NEC who had the reported 89% survival after resection and primary ileocaecal valve resected and 43 who did not.23 anastomosis for advanced NEC. Recurrence of However, the incidence of post-NEC strictures was NEC and strictures occurred in 22 and 17% of the significantly lower in neonates who had the valve neonates, respectively. Fasoli et al.23 reported that resected compared with neonates who retained in neonates with localized NEC, resection and pri- the ileocaecal valve, possibly indicating that not mary anastomosis had similar results to resection resecting bowel of questionable viability in order to spare the ileocaecal valve could be less advan- Whilst resection and primary anastomosis was tageous than resecting the valve itself. Survival initially described in selected patients with focal rate was not affected by the removal of the valve.
disease and good general condition, it is also now These results seem to indicate that neonates adapt gaining acceptance as a valid treatment option for well to the loss of the ileocaecal valve. The critical severe NEC and even for multifocal disease.22 In a issues in performing an intestinal resection are study of 46 neonates with multifocal disease, Fasoli the length of the retained small bowel and the gestational age of the neonate. It is well known that Clip and drop technique
pre-term neonates have potential for bowelgrowth48 and the earlier findings23,47 appear to This method complies with surgical principles and indicate that their bowel adaptive capacity to the also avoids stoma formation. For the infant with loss of the ileocaecal valve is remarkable.
extensive bowel necrosis, Vaughan et al.50 advo-cated the resection of all segments of grossly non-viable Operations for panintestinal disease
aspiration of peritoneal contamination, clippingthe ends of remaining bowel and returning them to The techniques described thus far are of particular the abdomen. This is followed by a second-look use for the infant with one or more short segments laparotomy with delayed anastomosis 48–72 h of NEC. Multiple resections and primary anasto- later. In their small series, all three infants with moses may be appropriate for some infants with NEC survived,50 and in a subsequent report of four more widespread disease, provided careful atten- infants in whom this technique was employed, one tion is paid to the viability of resection margins.
died and the remaining three required stoma Infants with NEC affecting a large proportion of the formation at the second look.8 This technique has gastrointestinal tract pose a particularly difficult the advantages of preserving bowel length and problem, and treatment of this group remains par- ticularly controversial. The surgical principles inthese children are difficult if not impossible to Patch, drain and wait
fulfil. Due to the length of bowel involved, it isoften not possible to fully remove all gangrenous The basic principle of this technique, reported by intestine whilst salvaging adequate length for sus- Moore,51 is to resect no bowel and do no entero- tainable life. It is for these reasons that in the infant with panintestinal NEC who is unstable and laparotomy, insertion of bilateral Penrose drains critically ill, some surgeons would forego further from the undersurface of the diaphragm into the treatment. However, when there is doubt, a pelvis with exit sites in both lower quadrants. Wait- number of techniques have recently been reported ing is important and post-drainage laparotomy with the aim of allowing time for stabilization of should not be performed before 14 days. The the infants' general condition and the possibility of bilateral Penrose drains capture faecal fistulas some healing of the gastrointestinal tract to occur.
and function as de-facto enterostomies. Moore's Due to the severity of the disease, the mortality 15-year personal experience indicates no mortality with this approach during the first 60 post-operative days, no major morbidity and no second Proximal jejunostomy
operations required in 70% of patients, possibly dueto spontaneous ‘auto-anastomosis’. The author51 Initially proposed by Martin and Neblett,30 surgical advocates this technique for infants with extensive creation of a high jejunostomy in the presence of gut necrosis (NEC totalis). There are no reports panintestinal disease has been reported in one from other centres employing the same technique.
series of 10 infants.49 This technique allowsdecompression and defunctioning of the diseased Authors' preferred surgical strategy
intestine but does not remove gangrenous segmentsand may permit continued bacterial translocation.
The authors' preferred approach to the operative A second-look laparotomy and intestinal recon- management of infants with NEC is illustrated in structive surgery are performed after 6–8 weeks and the aim is once again to preserve as much In infants with focal disease involving a small bowel length as possible. In this series, eight infants length of small or large bowel, a resection and survived to undergo a second procedure; in these, primary anastomosis is usually performed, with the resection of necrotic segments and anastomosis exception of patients unstable during the operation was performed to restore intestinal continuity.
who are managed with a stoma at the level of the There was, however, a significant incidence of TPN- related complications and only a 50% long-term In infants with multifocal disease (>50% of the TPN-free survival rate. This procedure is useful in bowel assumed to be viable), various surgical neonates with NEC affecting the majority of the options are available. Resection and one or more intestine, but the high morbidity and mortality rate intestinal anastomoses (preferably not more than two) are performed when it is possible to ascertain Surgical treatment of infants with necrotizing enterocolitis Authors preferred operative management of advanced NEC.
the viability of the bowel distal to NEC without orogastric tube is inserted to decompress the causing significant bleeding. Stoma (with or with- out intestinal resection), proximal enterostomy or anaerobic coverage are given intravenously for at clip and drop technique are performed when: (a) it least 7 days. Ventilatory and circulatory support are is not possible to ascertain the status of the bowel distal to the NEC; (b) the distal bowel is of dubious Advances in neonatal intensive care have been sub- viability; (c) attempts to dissect the distal bowel stantial in recent years, and this article is not cause significant bleeding and (d) the patient is unstable peri-operatively. In these circumstances, The metabolic insult of the necrotic bowel and a stoma with or without intestinal resection(s) is sepsis contributes to stress the neonate's homeo- the preferred surgical option. However, if this static mechanisms. Sepsis is an intriguing pathologi- would lead to massive intestinal resection, the clip cal condition associated with many complex and drop technique described earlier will be metabolic and physiological alterations.52 Studies adopted in an attempt to salvage as much intestinal in adults have shown that the metabolic response to length as possible. If the mobilization of the bowel sepsis is characterized by hypermetabolism as loops affected by NEC causes significant bleeding, a documented by an increase in resting energy high diverting jejunostomy is preferred.
expenditure by up to 49%.53 In addition, increased In patients with panintestinal disease (>75% of tissue catabolism,53 gluconeogenesis and hepatic small and large bowel involved), two options are release of glucose54 have been described. The considered: (a) proximal diverting jejunostomy existing knowledge on the metabolic response to when the intestinal resection would cause signifi- sepsis in infants is limited. A recent metabolic study cant bleeding or loss of the majority of the small in septic neonates with NEC by Powis et al.55 failed bowel and (b) clip and drop technique in the at- to show any increase in whole body protein turn- tempt to salvage some of the affected bowel and over, synthesis and catabolism. This data suggests avoid a short bowel syndrome. In neonates with that the calorific requirement of infants with NEC is total intestinal gangrene, closure of the abdomen not increased. The metabolic rate and hormonal and treatment withdrawal are considered.
response to stress and sepsis in infants may bedifferent from that of adults, and therefore, it is Pre- and post-operative management
not possible to adapt recommendations made foradults to the neonatal population. It is possible that Infants with NEC are amongst the most critically ill neonates divert the products of protein synthesis and most difficult patients to care for in paediatric and breakdown from growth into tissue repair. This surgery. Commonly, they are small premature may explain the lack of growth commonly observed neonates who require intensive care support and a in infants with critical illness or sepsis. Further team approach combining the expertise of paedi- studies are needed in this field to delineate the atric surgeons, neonatologists, intensivists, car- metabolic response of newborn infants to NEC and diologists and microbiologists. A nasogastric or severe sepsis by exploring the relationship between nutrition and immunity. The standard practise in from 26 to 48 days. The usual management of most paediatric surgical centres is to avoid enteral post-NEC strictures is intestinal resection and pri- feeding during the acute stage of the disease, and NEC include intestinal malabsorption, short bowel In addition to requiring nutritional support, syndrome, hepatic cholestasis and recurrence of neonates with NEC may present with clotting NEC. Malabsorption can derive from various causes abnormalities and require repeated transfusions of including enzyme depletion, intestinal dysmotility, platelets and clotting factors. The Thomsen– hypersecretion of gastric acid and bacterial over- Freidenreich (T) cryptantigen is a naturally occur- growth.5 Short bowel syndrome is the most serious ring antigen found on the surface of all human long-term gastrointestinal complication of NEC erythrocytes, and is normally concealed by a with an incidence of up to 23% in NEC survivors.68 layer of N-acetylneuraminic acid. This antigen Hepatic cholestasis is a common complication of infants with NEC requiring long-term PN. The acetylneuraminic acid layer is cleaved from the factors contributing to the development of this cell surface by neuraminidase, thus exposing complication are likely to be multifactorial, and are the underlying antigen. T cryptantigen activation not yet fully determined. Neonates, especially has been reported in some cases of neonatal those born pre-term, with low-birthweight are par- NEC.56–61 Activation of the T cryptantigen renders ticularly at risk. Infection, intestinal bacterial over- erythrocytes susceptible to haemolysis following growth, and lack of enteral stimulation contribute the transfusion of blood products containing anti-T significantly to the development of the disease.5,69 antibodies. In a recent study of 104 neonates with Introduction of minimal bolus enteral feeding in the advanced NEC,62 the incidence of T cryptantigen recovery phase of the disease is important because activation was 22% and was significantly higher in it stimulates gallbladder contractility,5,70,71 pro- infants with stage III disease (30%) compared with motes bile flow and aids in bowel adaptation.5 The those with stage II disease (4%). Screening of recurrence rate of NEC varies between 4 and neonates with advanced NEC for T cryptantigen 6%.72,73 The presence of an intestinal stricture may activation is advised to identify those at risk of further promote the recurrence of NEC.
The neurodevelopmental implications of NEC The only direct implication of T cryptantigen have not been fully investigated. Approximately activation for the management of NEC is that a 50% of the neonates with NEC are neurodevelop- selective transfusion policy has been recommended mentally normal.5,74 The neurological sequelae in in order to reduce the risk of haemolysis. This has children with NEC seems to be correlated with been identified in a number of infants when NEC is complicated by T cryptantigen activation and car- medical diseases rather than with NEC itself.
ries a poor prognosis.63–65 In one series of 128 The mortality rate of neonates with NEC depends infants with NEC, the mortality rate was 40% follow- on the severity of the disease, associated anomalies ing post-transfusion haemolysis compared with 6% and gestational age. Over the past three decades, when post-transfusion haemolysis did not occur.63 the survival rate of neonates with NEC has improved When infants with T cryptantigen activation require progressively. This has been most apparent in very blood products, the use of packed red cells, washed low-birthweight infants.5 In a study of 83 neonates platelets and low titre anti-T FFP is advis- who required a laparotomy for advanced NEC (Bell's able.62,66,67 Should haemolysis occur, exchange stage II and III), the overall mortality rate was 30%.
transfusion may abolish the haemolytic process and Causes of death included multisystem organ failure (nϭ10), sepsis (nϭ14) and congenital cardiacabnormality occurred within 30 days from the diagnosis of NEC.
Outcome of surgical intervention
The mortality rate was higher (67%) in patients withpanintestinal involvement of the disease compared Approximately half of the neonates with NEC who with patients with focal (12%) or multifocal NEC clinical resolution of the disease. The most commoncomplication in this group of infants is post-NECintestinal stricture. Butter et al.20 reported an Conclusions
incidence of post-NEC strictures of 48% in neonatestreated medically. The time, from the onset of NEC The fact that so many infants with NEC survive is until the diagnosis of intestinal stricture, ranged testimonial to the advances in medical and surgical Surgical treatment of infants with necrotizing enterocolitis 9. Rowe MI, Reblock KK, Kurkchubasche AG et al. Necrotiz- ing enterocolitis in the extremely low birth weight infant.
J Pediatr Surg 1994;29:987–90.
10. Ricketts RR. The role of paracentesis in the management of infants with necrotizing enterocolitis. Am J Surg 1986;
52:61–5.
11. Hutter JJ Jr, Hathaway WE, Wayne ER. Hematologic ab- normalities in severe neonatal necrotizing enterocolitis.
J Pediatr 1976;88:1026–31.
12. O'Neill JAJ. Neonatal necrotizing enterocolitis. Surg Clin North Am 1981;61:1013.
13. O'Neill JAJ, Stalman TM, Meng HC. Nectrotizing entero- colitis in the newborn. Ann Surg 1975;182:280–6.
14. Foglia RP. Necrotizing enterocolitis. Curr Probl Surg 1995; 32:757–823.
15. Pokorny WJ, Garcia-Prats JA, Barry YN. Necrotizing entero- Mortality rate according to extent of the disease in 83 colitis: incidence, operative care, and outcome. J Pediatr neonates with NEC who required laparotomy (adapted from Surg 1986;21:1149–54.
16. Grosfeld JL, Chew H, Schlater M. Changing trends in necro- tizing enterocolitis. Ann Surg 1991;214:300–7.
17. Ross MN, Waine ER, Janic JS et al. A standard for comparison for acute surgical necrotizing enterocolitis. J Pediatr Surg
1989;24:938–1002.
care. Surgical intervention plays a vital role in the 18. Schrober PH, Nassiri J. Risk factors and severity indices in treatment of a proportion of infants with NEC, and necrotizing enterocolitis. Acta Paediatr 1994;396(Suppl):
is undoubtedly essential for their survival. How- ever, the selective patients who would benefit 19. Kosloske AM. Surgery of necrotizing enterocolitis. World J most from surgery and the precise nature of that Surg 1985;9:277–84.
surgery remain largely unclear. Despite the intro- 20. Butter A, Flageole H, Laberge JM. The changing face of surgical indications for necrotizing enterocolitis. J Pediatr duction of new surgical techniques in recent years Surg 2002;37:496–9.
and a general improvement in morbidity and mor- 21. Anveden-Hertzberg L, Gauderer MW. Surgery is safe in very tality rates, there remains little consensus amongst low birthweight infants with necrotizing enterocolitis. Acta paediatric surgeons as to the most appropriate Paediatr 2000;89:242–5.
way to treat many of these children. Prospective, 22. Pierro A. Necrotizing enterocolitis: pathogenesis and multicentre, randomized controlled trials are treatment. Br J Hosp Med 1997;58:126–8.
23. Fasoli L, Turi RA, Spitz L et al. Necrotizing enterocolitis: required to enable us to answer these questions extent of disease and surgical treatment. J Pediatr Surg 1999;34:1096–9.
24. Ein SH, Marshall DG, Girvan D. Peritoneal drainage under References
enterocolitis. J Pediatr Surg 1977;12:963–7.
1. Walsh MC, Kliegman RM. Necrotizing enterocolitis: treat- 25. Ein SH, Shandling B, Wesson D et al. A 13-year experience ment based on staging criteria. Pediatr Clin North Am 1986; with peritoneal drainage under local anesthesia for necro- 33:179–201.
tizing enterocolitis perforation. J Pediatr Surg 1990; 2. Ververidis M, Kiely EM, Spitz L et al. The clinical significance 25:1034–6.
26. Cheu HW, Sukarochana K, Lloyd DA. Peritoneal drainage for enterocolitis. J Pediatr Surg 2001;36:799–803.
necrotizing enterocolitis. J Pediatr Surg 1988;23:557–61.
3. Kosloske AM. Indications for operation in necrotizing 27. Morgan LJ, Shochat SJ, Hartman GE. Peritoneal drainage as enterocolitis revisited. J Pediatr Surg 1994;29:663–6.
primary management of perforated NEC in the very low 4. Puntis J, McNeish AS, Allan RN. Long term prognosis of birth weight infant. J Pediatr Surg 1994;29:30–4.
Crohn's disease with onset in childhood and adolescence.
28. Azarow KS, Ein SH, Shandling B et al. Laparotomy or drain Gut 1984;25:329–36.
for perforated necrotizing enterocolitis: who gets what and 5. Albanese CT, Rowe MI. Necrotizing enterocolitis. In: O'Neill why? Pediatr Surg Int 1997;12:137–9.
JAJ, Rowe MI, Grosfeld JL, Fonkalsrud EW, Coran AG, 29. Kosloske AM, Papile LA, Burstein J. Indications for operation editors. Pediatric surgery. St Louis, MO: Mosby 1998;vol. 2:
in acute necrotizing enterocolitis of the neonate. Surgery 1980;87:502–8.
6. Stringer MD, Spitz L. Surgical management of neonatal 30. Martin LW, Neblett WW. Early operation with intestinal necrotising enterocolitis. Arch Dis Child 1993;69:269–701.
diversion for necrotizing enterocolitis. J Pediatr Surg 1981; 7. Caty MG, Azizkhan RG. Nectrotizing enterocolitis. In: 16:252–5.
Ashcraft KW, Murphy JP, Sharp RJ, Sigalet DL, Snyder CL, 31. Musemeche CA, Kosloske AM, Ricketts RR. Enterostomy in editors. Pediatric surgery. Philadelphia, PA: W.B. Saunders necrotizing enterocolitis: an analysis of techniques and timing of closure. J Pediatr Surg 1987;22:479–83.
8. Molik KA, West KW, Rescorla FJ et al. Portal venous air: the 32. Sparnon AL, Kiely EM. Resection and primary anastomosis poor prognosis persists. J Pediatr Surg 2001;36:1143–5.
for necrotizing enterocolitis. Pediatr Surg Int 1987;2:101–4.
33. Moss RL, Dimmitt RA, Henry MC et al. A meta-analysis cial nutrition in support in clinical practice. London: Edward of peritoneal drainage versus laparotomy for perforated necrotizing enterocolitis. J Pediatr Surg 2001;36:1210–3.
55. Powis MR, Smith K, Rennie M et al. Characteristics of protein 34. Rovin JD, Rodgers BM, Burns RC et al. The role of peritoneal and energy metabolism in neonates with necrotizing drainage for intestinal perforation in infants with and with- enterocolitis—a pilot study. J Pediatr Surg 1999;34:5–10.
out necrotizing enterocolitis. J Pediatr Surg 1999;34:143–7.
56. Kirsten GF, Smith J, Pieper C et al. The necessity for 35. Cass DL, Brandt ML, Patel DL et al. Peritoneal drainage as T-cryptantigen activation screening in babies with necrotis- definitive treatment for neonates with isolated intestinal ing enterocolitis. S Afr Med J 1996;86:546–8.
perforation. J Pediatr Surg 2000;35:1531–6.
57. Klein RL, Novak RW, Novak PE. T-cryptantigen exposure in 36. Tam PKH. Necrotizing enterocolitis—surgical management.
neonatal necrotizing enterocolitis. J Pediatr Surg 1986; Semin Neonatol 1997;2:297–305.
21:1155–8.
37. O'Connor A, Sawin RS. High morbidity of enterostomy and its 58. Mupanemunda RH, Kenyon CF, Inwood MJ et al. Bacterial- closure in premature infants with necrotizing enterocolitis.
induced activation of erythrocyte T-antigen complicating Arch Surg 1998;133:875–80.
necrotising enterocolitis: a case report. Eur J Pediatr 1993; 38. Kiesewetter WB, Taghizadeh F, Bower RJ. Necrotizing 152:325–6.
enterocolitis: is there a place for resection and primary 59. Seges RA, Kenny A, Bird GW et al. Pediatric surgical patients anastomosis? J Pediatr Surg 1979;14:360–2.
with severe anaerobic infection: report of 16 T-antigen 39. Harberg FJ, McGill CW, Saleem MM et al. Resection with positive cases and possible hazards of blood transfusion.
primary anastomosis for necrotizing enterocolitis. J Pediatr J Pediatr Surg 1981;16:905–10.
Surg 1983;18:743–6.
60. Squire R, Kiely E, Drake D et al. Intravascular haemolysis in 40. Griffiths DM, Forbes DA, Pemberton PJ et al. Primary association with necrotising enterocolitis. J Pediatr Surg 1992;27:808–10.
experience. J Pediatr Surg 1989;24:515–8.
61. Williams RA, Brown EF, Hurst D et al. Transfusion of infants 41. Parigi GB, Bragheri R, Minniti S et al. Surgical treatment with activation of erythrocyte T antigen. J Pediatr 1989; of necrotizing enterocolitis: when? how? Acta Paediatr 115:949–53.
1994;396(Suppl):58–61.
62. Hall N, Ong EG, Ade-Ajayi N et al. T cryptantigen activa- 42. Cooper A, Ross A III, O'Neill JA et al. Resolution of intract- tion is associated with advanced necrotizing enterocolitis.
able cholestasis associated with total parenteral nutrition J Pediatr Surg 2002;37:791–3.
following biliary irrigation. J Pediatr Surg 1985;20:772–4.
63. Novak RW, Abbott AE Jr, Klein RL. T-cryptantigen determi- 43. Ade-Ajayi N, Kiely E, Drake D et al. Resection and primary nation affects mortality in necrotizing enterocolitis. Surg anastomosis in necrotizing enterocolitis. J R Soc Med 1996; Gynecol Obstet 1993;176:368–70.
89:385–8.
64. Anand KJ, Hickey PR. Halothane–morphine compared with 44. Parigi GB, Bragheri R, Minniti S et al. Surgical treatment high-dose sufentanil for anesthesia and postoperative anal- of necrotizing enterocolitis: when? how? Acta Paediatr gesia in neonatal cardiac surgery see comments. N Engl J 1994;396(Suppl):58–61.
Med 1992;326:1–9.
45. Cooper A, Ross AJ III, O'Neill JA Jr et al. Resection 65. Cruickshank AM, Fraser WD, Burns HJ et al. Response of with primary anastomosis for necrotizing enterocolitis: a serum interleukin-6 in patients undergoing elective surgery contrasting view. J Pediatr Surg 1988;23:64–8.
of varying severity. Clin Sci (Colch) 1990;79:161–5.
46. Ade-Ajayi N, Kiely E, Drake D et al. Resection and primary 66. Facchinetti F, Bagnoli F, Bracci R et al. Plasma opioids in the anastomosis in necrotizing enterocolitis. J R Soc Med 1996; first hours of life. Pediatr Res 1982;16:95–8.
89:385–8.
67. Chwals WJ. The newborn as a surgical patient. Metabolic 47. Ladd AP, Rescorla FJ, West KW et al. Long-term follow-up considerations. In: Rowe MI, Grosfeld JL, Fonkalsrud EW, after bowel resection for necrotizing enterocolitis: factors Coran AG, editors. Pediatric surgery. St Louis, MO: Mosby affecting outcome. J Pediatr Surg 1998;33:967–72.
48. Touloukian RJ, Smith GJ. Normal intestinal length in 68. Ricketts RR. Surgical treatment of necrotizing enterocolitis preterm infants. J Pediatr Surg 1983;18:720–3.
and the short bowel syndrome. Clin Perinatol 1994; 49. Sugarman ID, Kiely EM. Is there a role for high jejunostomy 21:365–87.
in the management of severe necrotising enterocolitis? 69. Pierro A. Cholestatic jaundice in newborn infants receiving Pediatr Surg Int 2001;17:122–4.
parenteral nutrition. Semin Neonatol 1996;1:1–4.
50. Vaughan WG, Grosfeld JL, West K et al. Avoidance of stomas 70. Jawaheer G, Pierro A, Lloyd DA et al. Gall bladder contrac- and delayed anastomosis for bowel necrosis: the ‘clip and tility in neonates: effects of parenteral and enteral feeding drop-back’ technique. J Pediatr Surg 1996;31:542–5.
published erratum appears in Arch Dis Child Fetal Neonatal 51. Moore TC. Successful use of the ‘patch, drain, and wait’ Ed 1995; 73:F198. Arch Dis Child Fetal Neonatal Ed 1995; laparotomy approach to perforated necrotizing entero- 72:F200–2.
colitis: is hypoxia-triggered ‘good angiogenesis’ involved? 71. Jawaheer G, Shaw NJ, Pierro A. Continuous enteral feeding Pediatr Surg Int 2000;16:356–63.
impairs gallbladder emptying in infants. J Pediatr 2001; 52. Vlessis AA, Goldman RK, Trunkey DD. New concepts in the 138:822–5.
pathophysiology of oxygen metabolism during sepsis see 72. Ricketts RR, Jerles ML. Neonatal necrotizing enterocolitis: comments. Br J Surg 1995;82:870–6.
experience with 100 consecutive surgical patients. World J 53. Plank LD, Connolly AB, Hill GL. Sequential changes in the Surg 1990;14:600–5.
metabolic response in severely septic patients during the 73. Stringer MD, Brereton RJ, Drake DP et al. Recurrent necro- first 23 days after the onset of peritonitis see comments.
tizing enterocolitis. J Pediatr Surg 1993;28:979–81.
Ann Surg 1998;228:146–58.
74. Stevenson DK, Kerner JA, Malachowski N et al. Late mor- 54. Takala J, Pitkanen O. Nutrition support in trauma and sepsis. In: Payne-James J, Grimble G, Silk D, editors. Artifi- Pediatrics 1980;66:925–7.

Source: http://pedsurg.com.pe/Tratamiento%20Quirurgico%20de%20infantes%20con%20NEC.pdf