Journal of Pediatric Surgery 50 (2015) 428–430

Contents lists available at ScienceDirect

Journal of Pediatric Surgery

j ourna l homepage: www.e lsev ie r .com/ locate / jpedsurg

Infectious complications following serial transverse enteroplasty in

infants and children with short bowel syndrome

Wendy K. Fujioka a, Robert A. Cowles a,b,⁎a Division of Pediatric Surgery, Department of Surgery, Columbia University College of Physicians and Surgeons, New York, NY, 10032, USAb Section of Pediatric Surgery, Department of Surgery, Yale School of Medicine, New Haven, CT 06520, USA

a b s t r a c ta r t i c l e i n f o

⁎ Corresponding author at: Section of Pediatric SurgerSchool of Medicine, 333 Cedar Street, FMB 131, New Hav2701; fax: +1 203 785 3820.

E-mail address: robert.cowles@yale.edu (R.A. Cowle

http://dx.doi.org/10.1016/j.jpedsurg.2014.07.0090022-3468/© 2015 Elsevier Inc. All rights reserved.

Article history:

Received 5 February 2014Received in revised form 19 June 2014Accepted 19 July 2014

Key words:EnteroplastyShort bowel syndromeBacteremiaFungemiaInfectionComplication

Background: Serial transverse enteroplasty (STEP) lengthens and tapers dilated small bowel in patients withshort bowel syndrome (SBS). Previous reports document encouraging outcomes with regard to tolerance forenteral nutrition (EN) and complications appear related to the re-operative nature of many cases and to thepresence of multiple staple lines. However, infectious complications following STEP have not been examined.Since infections, especially catheter-related blood stream infections (CRBSI), are considered detrimental ininfants and children with SBS, we sought to define the frequency and outcomes of peri-operative infectionsassociated with STEP.Methods: All children with SBS who underwent a STEP between 2004 and 2012 were indentified and theirmedical records were reviewed. Patients were considered candidates for a STEP if they had dilated smallbowel and failure to advance enteral nutrition. For the purpose of this study, infections occurring within a 14-day period after STEP were considered procedure-related and were the focus of the study.

Results: A total of 18 patients underwent 23 STEP procedures. Primary diagnoses included intestinal atresia,gastroschisis, necrotizing enterocolitis, and midgut volvulus. After the STEP, eight patients (35%) developedCRBSI, three developed wound infections, and two had urinary tract infections. Organisms isolated fromeither blood, wound or urine cultures included gram-positive cocci, gram-negative rods, and yeast.Perioperative antibiotics were administered in all cases with cefoxitin (43%) and piperacillin/tazobactam(30%) being most common. Neither antibiotic appeared superior in reducing the incidence of CRBSI. In threepatients with persistent bacteremia despite adequate antibiotic therapy, a 74% ethanol lock resulted innegative blood cultures in all cases. Only one central venous catheter required replacement acutely forpersistent fungemia.Conclusion: STEP can improve enteral tolerance. In this fragile patient population, however, STEP carries adocumented infectious burden. The optimal antibiotic prophylaxis and the role of ethanol locking in patientsundergoing STEP require further study.

© 2015 Elsevier Inc. All rights reserved.

Pediatric short bowel syndrome (SBS) is a form of intestinalfailure (IF) caused by a variety of prenatal and postnatal eventsincluding intestinal atresia, gastroschisis, necrotizing enterocolitis,and intestinal malrotation [1–3]. The loss of small intestinal surfacearea leaves affected infants and children with insufficient absorptivecapacity to sustain growth and development. While parenteralnutrition (PN) can be life-sustaining as the remnant intestine adapts,IF-associated complications such as cholestasis, sepsis, and venousthrombosis complicate management [2,4,5]. During this process ofintestinal adaptation, dilation of the adapting small bowel can

y, Department of Surgery, Yaleen, CT 06520. Tel.: +1 203 785

s).

occur leading to stasis, bacterial overgrowth and deterioration inabsorptive function.

Serial transverse enteroplasty (STEP) is an intestinal lengtheningprocedure that both lengthens and tapers the dilated small bowel inpatients with SBS [6,7]. Previous reports on STEP documentencouraging outcomes with regard to tolerance for enteral nutrition(EN) [5,8,9] and most complications appear related to the re-operative nature of many cases and to the presence of multiple staplelines resulting in leak, perforation, ulcer, or obstruction [1,4,10,11].Patients with SBS are known to be at high risk for catheter-relatedblood stream infections (CRBSIs) and these infections have beenproven to increase length of stay and health care costs [12–16]. Theinfectious burden associated with the STEP operation has not beenpreviously examined. Since any infection, especially a CRBSI, isconsidered so detrimental to infants and children with SBS, wechose to evaluate the frequency and outcomes of perioperativeinfections associated with performance of STEP operations.

429W.K. Fujioka, R.A. Cowles / Journal of Pediatric Surgery 50 (2015) 428–430

1. Methods

A retrospective chart review of all infants and children whounderwent a STEP procedure between 2004 and 2012 was performed.STEP operations were performed by a single surgeon with experiencewith the procedure. Medical records were reviewed from birth untilthe date of discharge following STEP. Patients were consideredcandidates for a STEP if they had dilated small bowel and failure toadvance enteral nutrition. One infant underwent STEP at the time ofintestinal atresia repair and therefore did not meet both of thesecriteria. Blood and urine were collected for culture and wounds andwere assessed based on clinical signs suspicious for infection. Thesesigns included fever, hemodynamic instability, unexplained leukocy-tosis or a concern regarding the healingwound. For the purpose of thisstudy, infections occurring within a 14-day period after STEP wereconsidered procedure-related and were the focus of the study. Dataincluding sex, gestational age at birth, age at time of surgery, primarydiagnosis, indication for surgery, and pre and post-STEP bowel lengthwere obtained. Perioperative and postoperative antibiotic coveragewere assessed and any positive blood, wound, sputum, and urineculture results along with their associated antimicrobial susceptibilityprofiles were examined. The use of other infection control measuressuch as antibiotic or ethanol lock therapywas recorded. The studywasapproved by the Institutional Review Board (Protocol # IRB-AAAD6042) of Columbia University. Analysis of variance was usedfor statistical comparison of antibiotic therapy in patients with orwithout infection. Statistical results were considered significant whenthe p-value was b0.05. Analysis was performed in Statview.

2. Results

Of the 18 patients (7 male) who underwent the STEP procedure,five patients had repeated STEP procedures, for an overall total of 23separate STEP operations. The median age at the time of surgery was16 months (range, 13 days–6 years) and primary diagnoses includedintestinal atresia, gastroschisis, necrotizing enterocolitis, midgutvolvulus, or a combination of these disorders (Table 1). The mostcommon indications for the STEP procedure were dilated small boweland failure to advance enteral feeds. Overall, the median increase inbowel length was 40% (range 21–63%). On long-term follow-up at amedian of 2.6 years (range 0.8–8.4 years), PN independence wasachieved in 8 of 18 (44%) patients and there was one death 4 yearsafter STEP resulting in an overall survival of 94% for the entire cohort.One patient required an intestinal transplant after the STEP andanother underwent a liver transplant because of IF-associated liverdisease prior to the STEP procedure.

Table 1Demographics of patient population.

Age at surgeryRange 13days–6 yearsMedian, months 16.5Interquartile range, months 8.73–28.4

GenderMales, n (%) 7 (39)Females, n (%) 11 (61)

Gestational ageRange, weeks 29-37.5Median, weeks 34

Length of stayRange, days 6–67Median, days 11.00Interquartile range, days 9–18

Primary diagnosisIntestinal atresia, n (%) 2 (11)Gastroschisis, n (%) 3 (17)Necrotizing enterocolitis, n (%) 6 (33)Combination, n (%) 7 (39)

Following the 23 STEP procedures, 14 (61%) had no documentedperioperative infections whereas nine (39%) had documentedinfections. Of these, eight (35%) were complicated by CRBSI, three(13%) by wound infection, and two (9%) by urinary tract infection.Organisms isolated from the blood, wound, or urine cultures wereidentified and underwent antibiotic susceptibility testing. The isolatesincluded gram-positive cocci, gram-negative rods, and yeast (Table 2).In three patients with CRBSI, more than one organism was culturedfrom the blood.

While perioperative care varied depending on the patient, certainprocedures were routinely followed. A bowel preparation was notperformed because most patients were felt to have dysmotility thatwould preclude the administration of a cathartic. Vascular catheterswere cared for by the nursing staff according to hospital protocolalthough the anesthesia service was given the responsibility for theintraoperative care of these catheters. A 10% povidone–iodinesolution was used for skin preparation prior to surgery because allpatients either had a gastrostomy present or were felt to be too youngfor other skin antiseptics. Finally, a single dose of perioperativeintravenous antibiotics was administered prior to the start of all cases,with cefoxitin being the most commonly used (n = 10) followed bypiperacillin/tazobactam (n = 5). Additional doses were administeredin the operating room if the case extended beyond 4 hours. Neitherantibiotic appeared superior in reducing the incidence of CRBSI (p =0.23 for cefoxitin and p = 0.47 for piperacillin/tazobactam). In threepatients with persistent bacteremia despite adequate antibiotictherapy, a 74% ethanol lock of the indwelling Broviac catheterresulted in negative blood cultures in all cases. Only one centralvenous catheter was replaced acutely for persistent fungemia.

3. Discussion

The STEP procedure has gained popularity as an adjunct in the careof children with short bowel syndrome and dilated bowel. It canlengthen and taper the bowel and has been shown to improve enteraltolerance. The current study provides additional evidence of its utilitybecause almost half of the patients who underwent STEP wereeventually weaned from PN.

Despite these encouraging results of STEP on nutritional tolerancein this fragile patient population, this study has identified aconcerning and previously unrecognized infectious burden associatedwith this operation. Of particular concern is the 35% incidence ofCRBSI because these not only expose the patient to the detrimentaleffects of sepsis but alsomay require catheter removal with associatedloss of vascular access sites.

Table 2Infectious complications after serial transverse enteroplasty.

Blood-stream infections Wound infections UTI

Total patients, n (%) 8 (35) 3 (13) 2 (9)Total infections, n 11 6 2Appearance

Range, days 0-13 5-14 2-11Median, days 3 4 6.5Gram-positive cocci 5 4 1

E. faecalis E. faecalis S. aureusE. faecium E. faeciumS. epidermidis S. coag negS. hominis (2) S. viridans

Gram-negative rod 5 2 1K. pneumoniae (2) E. coli E. ClocaeK. oxytoca P. aeruginosaS. marcesansE. coli

Fungi 1 0 0Candida albicans

430 W.K. Fujioka, R.A. Cowles / Journal of Pediatric Surgery 50 (2015) 428–430

This study represents an initial attempt to characterize the burdenof infection associated with complex gastrointestinal procedures onthis particularly vulnerable group of patients with SBS. The lack of acontrol group that did not undergo a STEP procedure for comparisonis unavoidable and therefore it is impossible to determine withcertainty that the STEP itself was independently responsible for all thedocumented infections. In addition, the study was not able to identifyspecific factors that may prevent CRBSI or other infections in patientsundergoing STEP and this is likely a result of the relatively smallsample size.

Classically, the perioperative antibiotic dose is administered in anattempt to prevent perioperative infections during gastrointestinalsurgery. In this study, it appears that the perioperative antibiotic wasinsufficient to prevent infections even when broad spectrum agentssuch as piperacillin/tazobactam were used. One could postulatethat poor catheter care during the perioperative period could beresponsible for the high incidence of CRBSI. While this may indeedbe a factor, other theories may also adequately explain the findings ofthe study. It is accepted that dilated segments of small bowel promotebacterial overgrowth and that bacterial translocation occurs in thispatient population. While the current study cannot prove thatbacterial translocation with seeding of vascular catheters, theoperativewound, and the urinary tract occurs, this remains a plausibleexplanation for the findings presented here. For comparison, and incontrast, two recent studies documenting blood and surgical siteinfections after cardiac surgery reported infection rates of b10%[17,18]. These data suggest that there is a distinct aspect of the STEP,apart from the perioperative catheter care, that may result in higherrates of blood, surgical site, and urinary infections.

Many studies have examined the efficacy of ethanol [19–25],heparin [24], or antibiotic [20,21] lock therapies, as well as alcohol-swabbed caps [26], to prevent CRBSI in patients with indwellingcentral-venous catheters receiving PN. In this study, the use of 74%ethanol locking successfully resulted in negative blood culturesin all cases with persistent bacteremia despite appropriateantibiotic treatment.

In summary, the data presented here suggest that the infectiousburden associated with the STEP operation is significant and thatthe causes are likely multifactorial. Although no specific causes forinfections were identified, the results of this study underscore theimportance of meticulous central venous catheter care both in andout of the OR, timely administration of perioperative antibiotics andpossibly the expansion of the use of ethanol locking protocols inthis patient population. Since the perioperative antibiotic dose doesnot seem to reliably prevent the occurrence of blood, wound, andurinary tract infections, this patient population must be carefullymonitored for an infectious complication after the STEP procedure.While not formally evaluated in this study, it is possible that otheradjuncts such as preoperative gut decontamination in patientsundergoing the STEP procedure would be beneficial and should befurther examined.

Acknowledgment

This study was funded by a grant from the Charles Edison Fund(Newark, NJ).

References

[1] Modi BP, Javid PJ, Jaksic T, et al. First report of the international serial transverseenteroplasty data registry: indications, efficacy, and complications. J Am Coll Surg2007;204(3):365–71.

[2] Modi BP, Jaksic T. Pediatric intestinal failure and vascular access. Surg Clin NorthAm 2012;92(3):729–43.

[3] Wales PW, de Silva N, Kim J, et al. Neonatal short bowel syndrome: population-based estimates of incidence andmortality rates. J Pediatr Surg 2004;39(5):690–5.

[4] Ching YA, Fitzgibbons S, Valim C, et al. Long-term nutritional and clinical outcomesafter serial transverse enteroplasty at a single institution. J Pediatr Surg 2009;44(5):939–43.

[5] Javid PJ, Kim HB, Duggan CP, et al. Serial transverse enteroplasty is associated withsuccessful short-term outcomes in infants with short bowel syndrome. J PediatrSurg 2005;40(6):1019–23 [discussion 1023–1024].

[6] Kim HB, Fauza D, Garza J, et al. Serial transverse enteroplasty (STEP): a novelbowel lengthening procedure. J Pediatr Surg 2003;38(3):425–9.

[7] Suri M, Dicken B, Nation PN, et al. The next step? Use of tissue fusion technology toperform the serial transverse enteroplasty–proof of principle. J Pediatr Surg 2012;47(5):938–43.

[8] Kaji T, Tanaka H, Wallace LE, et al. Nutritional effects of the serial transverseenteroplasty procedure in experimental short bowel syndrome. J Pediatr Surg2009;44(8):1552–9.

[9] Oliveira C, de Silva N, Wales PW. Five-year outcomes after serial transverseenteroplasty in children with short bowel syndrome. J Pediatr Surg 2012;47(5):931–7.

[10] King B, Carlson G, Khalil BA, et al. Intestinal bowel lengthening in children withshort bowel syndrome: systematic review of the Bianchi and STEP procedures.World J Surg 2013;37(3):694–704.

[11] Wales PW, de Silva N, Langer JC, et al. Intermediate outcomes after serialtransverse enteroplasty in children with short bowel syndrome. J Pediatr Surg2007;42(11):1804–10.

[12] Cole CR, Frem JC, Schmotzer B, et al. The rate of bloodstream infection is high ininfants with short bowel syndrome: relationship with small bowel bacterialovergrowth, enteral feeding, and inflammatory and immune responses. J Pediatr2010;156(6):941–7.

[13] Cole CR, Hansen NI, Higgins RD, et al. Bloodstream infections in very low birthweight infants with intestinal failure. J Pediatr 2012;160(1):54–9.

[14] Gutierrez IM, Kang KH, Calvert CE, et al. Risk factors for small bowel bacterialovergrowth and diagnostic yield of duodenal aspirates in children with intestinalfailure: a retrospective review. J Pediatr Surg 2012;47(6):1150–4.

[15] Mohammed A, Grant FK, Zhao VM, et al. Characterization of posthospitalbloodstream infections in children requiring home parenteral nutrition. JPEN JParenter Enteral Nutr 2011;35(5):581–7.

[16] Goudie A, Dynan L, Brady PW, et al. Attributable cost and length of stay for centralline-associated bloodstream infections. Pediatrics 2014;133(6):e1525–32.

[17] Abou Elella R, Najm HK, Balkhy H, et al. Impact of bloodstream infection on theoutcome of children undergoing cardiac surgery. Pediatr Cardiol 2010;31(4):483–9.

[18] Murray MT, Krishnamurthy G, Corda R, et al. Surgical site infections andbloodstream infections in infants after cardiac surgery. J Thorac Cardiovasc Surg2014;148(1):259–65.

[19] Chambers ST, Peddie B, Pithie A. Ethanol disinfection of plastic-adherentmicroorganisms. J Hosp Infect 2006;63(2):193–6.

[20] Chu HP, Brind J, Tomar R, et al. Significant reduction in central venous catheter-related bloodstream infections in children on HPN after starting treatment withtaurolidine line lock. J Pediatr Gastroenterol Nutr 2012;55(4):403–7.

[21] Huang EY, Chen C, Abdullah F, et al. Strategies for the prevention of central venouscatheter infections: an American Pediatric Surgical Association Outcomes andClinical Trials Committee systematic review. J Pediatr Surg 2011;46(10):2000–11.

[22] Jones BA, Hull MA, Richardson DS, et al. Efficacy of ethanol locks in reducingcentral venous catheter infections in pediatric patients with intestinal failure.J Pediatr Surg 2010;45(6):1287–93.

[23] Oliveira C, Nasr A, Brindle M, et al. Ethanol locks to prevent catheter-relatedbloodstream infections in parenteral nutrition: a meta-analysis. J Pediatr 2012;129(2):318–29.

[24] Wales PW, Kosar C, Carricato M, et al. Ethanol lock therapy to reduce the incidenceof catheter-related bloodstream infections in home parenteral nutrition patientswith intestinal failure: preliminary experience. J Pediatr Surg 2011;46(5):951–6.

[25] Cober MP, Kovacevich DS, Teitelbaum DH. Ethanol-lock therapy for the preventionof central venous access device infections in pediatric patients with intestinalfailure. JPEN J Parenter Enteral Nutr 2011;35(1):67–73.

[26] Castello FV, Maher A, Cable G. Reducing bloodstream infections in pediatricrehabilitation patients receiving parenteral nutrition. J Pediatr 2011;128(5):1273–8.