Outcome of Tapering Enteroplasty in Managing Jejunoileal Atresia

Abstract

Background:

Jejunoileal atresia (JIA) represents a common cause of neonatal bowel obstruction. There is a discrepancy between the diameters of the proximal and the distal bowel loops and this is managed with excision or tapering of the dilated proximal bowel loop. We aim primarily to evaluate the outcome of tapering enteroplasty (TE) in managing JIA and secondarily to compare the outcome of TE to non-TE.

Materials and Methods:

A retrospective analysis of records of all neonatal admissions with JIA from January 2017 to December 2018 at a tertiary university children's hospital. Type and location of atresia, time to full enteral feeds, length of stay (LOS), complications, and outcome were assessed in TE and non-TE groups.

Results:

Forty-one patients were included in the study; 29 (70.7%) cases had jejunal atresia and 12 (29.3%) had ileal atresia. Seventeen (41.4%) patients had TE. The median days to full feeds was 19; 28 in the TE group versus 16 in the non-TE group (P = 0.022). Four (9.7%) cases needed a re-exploration because of failure to start feeds; all in the non-TE group. The median LOS was 33 days for TE versus 22 days for non-TE (P = 0.101). Twenty-one cases (51.2%) developed a wound infection and showed a significantly longer median LOS of 29.5 versus 19 days (P = 0.019). Mortality was 7 (17.1%).

Conclusions:

TE did not show a superior outcome when compared to resection of the dilated bowel. It was associated with longer time to reach full enteral nutrition and longer LOS.

INTRODUCTION

Jejunoileal atresia (JIA) is one of the major causes of neonatal bowel obstruction[1] with an incidence of 3.4/10,000 live births.[2] The optimal surgical management of JIA is to establish bowel continuity and to avoid anastomotic complications and short bowel syndrome.[3] The distended proximal bowel represents a major technical challenge for the surgeons to perform an anastomosis.[4] As this dilated part shows alterations of neural and muscular elements,[5] the gold standard management is, therefore, resection of the dysfunctional bowel and primary anastomosis.[6,7]

Experimental studies have shown that a chronically obstructed small intestine has a low contractile pressure with poor propulsion and that it improves when the diameter of the dilated intestine is reduced.[8,9] Tapering was proposed because it reduced the diameter of the proximal dilated bowel, made it feasible to perform an end-to-end anastomosis, and preserved the bowel length.[10] Thomas first described tapering enteroplasty (TE) in 1969 for managing infants with short gut as an alternative to bowel resection.[11] Weber et al. subsequently published a case series of managing selected types of small bowel atresia with TE which showed a promising outcome.[10]

The aim of this study is to evaluate the outcome of primary TE in managing JIA in a single institute over a 2-year period and to compare this outcome with patients who did not have TE.

MATERIALS AND METHODS

A retrospective study of all neonatal admissions with JIA, who underwent surgery from January 2017 to December 2018 at Cairo University Specialized Pediatric Hospital, was conducted. Patients with associated gastroschisis and primary stoma were excluded from the study. The type of atresia was determined according to the Grosfeld classification.[12] The medical records of the selected patients were reviewed for gestational age, birth weight, age at surgery, associated anomalies, type and location of atresia, type of surgery, starting and reaching full enteral nutrition, complications (wound infection and the need for re-exploration), and outcome (length of stay [LOS] and mortality). The length of the remaining bowel was also reviewed, and congenital short bowel was diagnosed when the length of the intestine from the ligament of Treitz to the ileocecal valve was <60 cm.[13]

The outcome of the patients who underwent TE with minimal resection as an initial procedure was then compared to the outcome of the patients who underwent resection of the dilated bowel and primary anastomosis (non-TE group). The choice of surgical procedure was at the operating surgeon's discretion.

Data were coded and entered using the Statistical Package for the Social Sciences (SPSS) version 25 (IBM Corp., Armonk, NY, USA). Data were summarized using mean, standard deviation, median, minimum, and maximum in quantitative data and using frequency (count) and relative frequency (percentage) for categorical data. Comparisons between quantitative variables were made using the nonparametric Kruskal–Wallis and Mann–Whitney tests. For comparing categorical data, the Chi-square test was performed. The exact test was used instead when the expected frequency was <5. Correlations between quantitative variables were done using the Spearman correlation coefficient. Linear regression analysis was performed to predict days until full enteral nutrition. Values of P < 0.05 were considered statistically significant.

RESULTS

The study comprised a total of 41 cases: 17 males (41.4%) and 24 females (58.6%). Twenty-nine (70.7%) cases had jejunal atresia and 12 (29.3%) had ileal atresia. The median gestational age was 38 weeks (range 34–38 weeks), and the median weight was 2.5 kg (range 1.5–3.75 kg). Twelve cases (29.2%) had associated anomalies: volvulus (2), ARM (1), intestinal perforation (5), malrotation (3), and cardiac anomalies (1). The patients with associated anomalies had a median LOS of 24 days and 8.3% mortality versus 23.5 days and 20.7% for those without anomalies. Associated anomalies did not significantly affect the LOS (P = 0.369) or the outcome (P = 0.616). All five cases with bowel perforation had atresia located in the ileum.

The classification of JIA in our group is summarized in Table 1. Three cases underwent a Ladd's procedure in addition to primary anastomosis due to associated malrotation.

Table 1.

Operative data for the 41 patients operated on for jejunoileal atresia

Findings	Value
Median age at operation (days) (range)	1 (1-28)
Resection of the dilated bowel and primary anastomosis	24 (58.6)
Tapering enteroplasty with minimal resection	17 (41.4)

Values presented as absolute n (%) of patients

The postoperative data and the outcome are summarized in Table 2. All patients received total parenteral nutrition till they tolerated 120 ml/kg/day of enteral feed. Twenty-one cases (51.2%) developed a wound infection (includes wound erythema, edema, or discharge). The median LOS among patients with wound infection was 29.5 days in comparison to 19 days among patients without wound infection (P = 0.019). The mortality among patients with wound infection was 27.3% in comparison to 5.3% among patients without wound infection (P = 0.062). All mortality cases had evidence of sepsis. In the five cases with suspected leaks, intraoperative findings confirm the initial diagnosis. A redo anastomosis was done in all cases. The four cases that were explored due to failure to start enteral nutrition were found to have dense adhesions and some showed persistent caliber differences. The four cases underwent adhesiolysis whereas two cases had additional tapering. The decision to taper was based on surgeon preference.

Table 2.

Postoperative data and outcome in 41 patients with jejunoileal atresia

Findings	Value
Median days to start enteral nutrition (range)	8 (2-28)
Median days until full enteral nutrition (range)	19 (4-65)
Complication
Second abdominal operation	9 (21.9)
Leakage	5 (12.2)
Failure to start enteral nutrition	4 (9.7)
Wound infection	21 (51.2)
Median LOS (days) (range)	24 (5-74)
Mortality	7 (17.1)

Values presented as absolute n (%). LOS: Length of stay

Seventeen (41.4%) patients had TE with minimal resection [Table 3]. None of the patients with ileal atresia underwent TE. Patients who had TE reached full enteral nutrition after a median postoperative period of 28 days in comparison to 16 days with patients who did not have TE (P = 0.022). One patient had congenital short bowel. He had type IV jejunal atresia, underwent TE, started enteral nutrition on day 19 postoperative, reached full enteral nutrition on day 43, and was discharged alive after a LOS of 74 days.

Table 3.

Parameters associated with tapering enteroplasty in 41 patients with jejunoileal atresia

	Tapering enteroplasty		P
	Yes (n=17)	No (n=24)
Location of atresia
Jejunum	17 (100)	13 (54.2)
Ileum	0	11 (45.8)
Days until start of enteral nutrition, median (range)	9.2 (2-19)	7.5 (2-28)	0.191
Days until full enteral nutrition, median (range)	28 (12-46)	16 (4-65)	0.022
2nd abdominal operation due to failure to start enteral nutrition	0	4 (15.4)	0.089
Anastomotic leak	2 (11.8)	3 (11.5)	1
Wound infection	10 (58.8)	12 (50)	0.577
LOS (days), median (range)	33 (7-74)	22 (3-61)	0.101
Mortality	2 (11.8)	5 (20.8)	0.447

Values presented as absolute n (%). LOS: Length of stay

We compared the start and the reaching of full enteral nutrition in the different types of atresia [Table 4]. There is a statistical significance between the different types of atresia and the median time to reach full feeds (P = 0.040).

Table 4.

Enteral nutrition correlated with location and type of atresia

	Type of atresiaa								P
	Jejunal atresia				Ileal atresia
	Type I	Type II	Type III	Type IV	Type I	Type II	Type III	Type IV
Days until start of enteral nutrition (median)	7.5	16	9	10	4.5	4	4.5	2	0.077
Days until full enteral nutrition (median)	16.5	19	27.5	28	23	27	10.5	7	0.040

^aLacking a case of ileal atresia of unknown subtype

A linear regression analysis was calculated to predict days until full enteral nutrition using the type of atresia and TE as independent variables. A significant regression equation was found (F (2,30) = 4.631, P < 0.018), with an R² 0.236, with the type of atresia being the significant predictor (P = 0.027).

DISCUSSION

The surgical management of JIA aims to establish bowel continuity and avoid anastomotic complications as well as short bowel syndrome.[3] However, the distended proximal bowel represents a major technical challenge.[4] To overcome this, surgical management includes resection of the dysfunctional bowel and primary anastomosis[6,7] or TE to preserve the bowel length.[10] In this study, we investigated the outcome of primary TE in managing JIA and compared this with the outcome of resection of the dysfunctional bowel and primary anastomosis.

Ileal atresia in our cohort was more likely to perforate compared to jejunal atresia. A similar observation was made by Heij et al.[14] This supports the assumption that a difference in compliance between the jejunal and ileal walls. When atresia occurs, the jejunal wall seems to be capable of massive distension, while the ileal wall is rarely very dilated and tends to perforate.[14] This may also explain why none of the ileal atresia patients underwent TE in our study.

Seventeen (41.4%) of our patients had a TE with minimal resection; only one of them had congenital short bowel. Our study showed a median of 19 days to full enteral nutrition, which is similar to the results of other studies.[15,16,17] The median days to full enteral nutrition was significantly longer in the TE group versus the nontapering group (28 vs. 16), P = 0.022. Dewberry et al. reported a similar finding of 25 days versus 21 days, but it was statistically insignificant.[18] In our study, the subtype of bowel atresia affected the median days to full enteral nutrition (P = 0.040), whereas jejunal atresia type IV showed the longest time 28 days.

A linear regression analysis was performed to predict days until full enteral nutrition using the type of atresia and TE as independent variables. The significant independent variable was the type of atresia (P = 0.027).

Five patients (12.2%) developed an anastomotic leak; however, this showed no statistically significant association with TE. A 7% leak rate was also one of the significant complications reported by Stollman et al.[19] Four (9.7%) patients needed a second abdominal operation because of failure to start enteral nutrition; all of them did not have TE. Dewberry et al. reported that 13% of a nontapered group underwent reoperation for bowel dysmotility versus 38% in a tapered group.[18] Although our finding contradicts this, our outcome may be attributed to the end-to-oblique anastomotic technique performed in the nontapering group due to a persisting discrepancy between the proximal and distal loops despite resection.

The median LOS in the tapering group was longer than the median LOS in the non-tapering group (33 vs 22 days) (P 0.101). These values are similar to what is reported by Dewberry et al. (59 vs 31 days, respectively) (P 0.25).[18]

The most significant complication in our cohort was wound infection, which developed in 51.2% of the patients. Expectedly, wound infection significantly prolonged the LOS. Mortality increased among patients with wound infection, but it was not statistically significant. Although mortality attributed to JIA has decreased steadily throughout the 20^th century,[19] our study showed a higher mortality of 17.1% when compared to the 11% published by Stollman et al. in a study of 114 patients.[19] In perspective, our mortality rate is lower than the average mortality of 45.6% attributed to intestinal atresia, reported in a systematic review of the outcome of neonatal surgery in Africa.[20]

The limitations of our study were its retrospective nature, the small sample size, and the group allocation which depended on the operating team's choice. Therefore, the results should be carefully interpreted.

CONCLUSIONS

TE did not show a superior outcome when compared to resection of the dilated bowel in our cohort. It was associated with a longer time to reach full enteral nutrition and a longer LOS. TE could be reserved for cases where resection of the dilated bowel would cause a significant shortening of the gut or for cases with very proximal atresia adjacent to the ligament of Treitz.

Declarations

Ethics approval and consent to participate retrospectively registered.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.