Clinical Profile, Outcomes and Predictors of Mortality in Neonates Operated for Gastrointestinal Anomalies in a Tertiary Neonatal Care Unit- An Observational Study.

Background: Gastrointestinal (GI) malformations have varied short-term and long-term outcomes reported across various neonatal units in India.
Methods: This descriptive study was done to study the clinical profile, outcomes and predictors of mortality in neonates operated for congenital GI malformations in a tertiary neonatal care unit in South India between years 2011 and 2020. Details were collected by retrospective review of the case sheets.
Results: Total of 68 neonates were included with esophageal atresia (EA) in 10, infantile hypertrophic pyloric stenosis (IHPS) in 9, duodenal atresia (DA) in 10, ileal atresia in 8, jejunal atresia in 5, anorectal malformations (ARM) in 11, meconium ileus/peritonitis in 9, malrotation in 2, and Hirschsprung's disease (HD) in 4. Antenatal diagnosis was highest in DA (80%). Associated anomalies were maximum in EA (50%), the most common being vertebral, anal atresia, cardiac defects, tracheoesophageal fistula, renal and radial abnormalities, and limb abnormalities association (VACTERL). Overall mortality was 15%. IHPS, DA, Malrotation, HD and ARM had 100 % survival while ileal atresia had the least survival (38%). Gestational age <32 weeks (odds ratio [OR] 12.77 [1.96, 82.89]) and outborn babies (OR 5.55 [1.01, 30.33]) were significant predictors of mortality in babies operated for small intestinal anomalies. None of the surviving infants were moderately or severely underweight at follow-up.
Conclusion: Overall survival of surgically correctable GI anomalies is good. Among the predictors for mortality, modifiable factors such as in-utero referral of antenatally diagnosed congenital anomalies need attention. One-fifth had associated anomalies highlighting the need to actively look for the same. Although these neonates are vulnerable for growth failure, they had optimal growth on follow-up possibly due to standardized total parenteral nutritional policy during neonatal intensive care unit stay. Copyright:

INTRODUCTION

Gastrointestinal (GI) malformations are one of the most frequently encountered surgical pathologies contributing to about 15%–20% of the total congenital anomalies and are estimated to affect one in 1000–2000 live births.[12] The associated mortality is estimated to be in excess of 50% in many low- and middle-income countries (LMICs).[3] Advances in surgical management, intensive care medicine, multidisciplinary management, and postoperative nutritional support over the past decade have contributed greatly to improved survival of these neonates. However, there is a lot of disparity in the outcomes between LMICs and high-income countries.

Our study aims at presenting the spectrum of various GI malformations requiring surgery, their clinical profile, and outcomes in the past 10 years from our tertiary care neonatal unit.

Methods

This was a retrospective observational study done in level III neonatal intensive care unit (NICU), Chennai, India after the Institutional ethics committee approval (IEC-NI/20OCT/76/103). We included neonates born or admitted between January 2010 and September 2020 with congenital GI malformations requiring surgical intervention. Acquired causes of the surgical abdomen such as necrotizing enterocolitis were excluded from this study. Case sheets of eligible neonates were retrieved from medical records department using International Classification of Diseases (ICD) -10 codes (esophageal atresia [EA] Q39.1, infantile hypertrophic pyloric stenosis [IHPS] Q40.0, duodenal atresia [DA] Q41.0, jejunal/ileal atresia [JIA] Q41.2/Q41.4, meconium ileus/peritonitis P76.0/P78.0/E84.11, gastroschisis Q79.3, malrotation Q43.3, volvulus K56.2, anorectal malformations [ARM] Q42.3, Hirschsprung's disease [HD] Q43.1). The demographic data were collected by retrospective review of the case sheets. The children were followed up to find the current weight and weight for age (WFA) was plotted in the WHO Z-score charts. Children with WFA between-2 SD and-3 SD were taken as moderately underweight and those below-3 SD as severely underweight.[4] We used descriptive statistics to describe baseline variables.

We compared categorical outcome variables by Chi-square test or Fisher's exact test; normally distributed numerical variables by unpaired t-test, variables with skewed distribution by the Mann–Whitney U-test. Individual predictors for mortality were determined by univariate analysis. P <0.05 was considered statistically significant. We used SPSS version 21.0 (SPSS Inc., Chicago, IL).

RESULTS

The proportion of GI malformations needing surgical interventions among inborn neonates was 2/10,000 live births. Twenty-five (37%) were referred for management from other centers, out of which EAs were six (9%). Baseline and outcome characteristics are depicted in tables [Tables 1 and 2].

Table 1

Baseline patient characteristics

	Esophageal atresia± tracheoesophageal fistula (n=10), n (%)	Infantile hypertrophic pyloric stenosis (n=9), n (%)	Duodenal atresia (n=10), n (%)	Ileal atresia (n=8), n (%)	Jejunal atresia (n=5), n (%)	Meconium ileus/meconium peritonitis (n=9), n (%)	Malrotation (n=2), n (%)	Hirschsprung disease (n=4), n (%)	Anorectal malformations (n=11), n (%)
Male	7 (70)	8 (89)	9 (90)	2 (25)	1 (20)	2 (22)	2 (100)	2 (50)	7 (63)
Gestation age at birth* (weeks)	37 (35-39)	39 (36-39)	39 (37-39)	35 (31-36.5)	34 (29-36)	31 (26-36)	35.5 (35-36)	37 (29-38)	36.5 (35-38)
Day of presentation*	1 (1-1)	21 (17-28)	1 (1-1)	1 (1-3)	1 (1-2)	5 (3-8)	5 (4-6)	2 (1-21)	2 (1-2)
IUGR	2 (20)	1 (11)	3 (30)	0	0	0	0	0	0
Birth weight* (kg)	2.5 (2-2.8)	3.1 (2.9-3.5)	3.1 (2.7-3.4)	2.34 (1.8-2.5)	1.8 (1.2-2.1)	1.8 (0.9-2.3)	2.1 (2-2.2)	3.1 (1.5-3.4)	2.76 (1.9-2.9)
Antenatal diagnosis	6 (60)	0	8 (80)	7 (87)	3 (60)	0	0	0	0
Polyhydramnios	5 (50)	0	8 (80)	6 (75)	4 (80)	6 (66)	0	0	0
Associated anomalies	5 (50)	0	7 (70)	1 (12)	0	0	0	0	2 (18)
Day of diagnosis*	1 (0-1)	29 (25-31)	1 (1-2)	2 (1-3)	1 (1.3)	4 (2-4)	6 (6-7)	1 (1-21)	2 (1-21)
Day of surgery*	2 (1-3)	31 (29-32)	2 (2-3)	2 (1.5-3)	2 (1.5-5)	5 (3-7.5)	8 (7-9)	7.5 (7-25)	2 (1-2)

Median (IQR). IQR: Interquartile range, IUGR: IUGR: Intrauterine growth restriction

Table 2

Outcome characteristics

	Esophageal atresia± tracheoesophageal fistula (n=10), n (%)	Infantile hypertrophic Pyloric stenosis (n=9), n (%)	Duodenal atresia (n=10), n (%)	Ileal atresia (n=8), n (%)	Jejunal atresia (n=5), n (%)	Meconium ileus/meconium peritonitis (n=9), n (%)	Malrotation (n=2), n (%)	Hirschsprung disease (n=4), n (%)	Anorectal malformations (n=11), n (%)
Primary anastomosis	10 (100)	9 (100) (ramset’s)	10 (100)	5 (62)	4 (80)	0	2 (100)	1 (25)	5 (45) (rectal pull through-2)
Placement of Stoma	0	0	0	3 (38)	1 (20)	4 (44)	0	3 (75)	6 (54)
Glove drain alone	0	0	0	0	0	4 (44)	0	0	0
Glove drain followed by resection anastomosis	-	-	z	-	-	1 (12)	-	-	-
Length of remaining bowel* (cm)				74 (49-165)	135 (78-146)	86 (60-106)	130 (120-140)
Duration of TPN* (days)	9 (8-12)	2 (2-3)	7 (2-9)	20.5 (16-26.5)	7 (6-15)	9 (5-9.5)	6.5 (6-7)	5 (3-10)	2 (1.5-4)
Duration of PICC line* (days)	12 (6-17)	2 (2-3)	7 (3-10)	19 (16-23)	9 (6.5-14.5)	8 (5-10)	5 (4-6)	5 (1-5)	3 (0-5.5)
Day of reaching full feeds* (days)	14 (3-17)	3 (2-4)	5 (4-10)	21 (16-27.5)	9 (7-17)	9 (7-10)	4.5 (4-5)	8 (8-14)	2 (1.3-4.8)
Duration of hospital stay* (days)	26 (13-29)	8 (5-18.5)	8 (5-16)	25 (16-43)	22 (11-33)	14 (10-18)	11 (10-12)	18 (11-50)	6 (5-12)
Duration of respiratory support* (days)	7 (4-10)	1 (0-1)	2 (1-3)	15 (8.5-20)	5 (4-7)	5 (3-10)	2.5 (2-3)	1 (0.5-1.5)	1 (0-2)
Survived	8 (80)	9 (100)	10 (100)	3 (38%)	4 (80)	7 (77)	2 (100)	4 (100)	11 (100)
Nature of feeds at discharge
MOM	6	7	6	2	3	7	2	11	4
Donor milk	0	0	1	0	0	0	0	0	0
Elemental formula	0	0	0	1	0	0	0	0	0
Formula	2	2	3	0	1	0	0	0	0
Need for re-surgery#	2 (20)	0	0	2 (22)	2 (40)	1 (10)	0	0	1 (8)
Readmissions needed	1 (10)	0	1 (10)	2 (22)	2 (40)	0	0	1 (25)	0
Present age of survivors in years*	6 (2-8)	4 (0.9-6.5)	4 (0.9 - 6.5)	4 (3-9)	7.5 (6.3-8.8)	3.5 (1-4.8)	2.5 (2-3)	6.5 (4-9)	7 (2-7)
Weight Z score at present*	0.5 (−1.2-1.2)	1 (−0.5-−2.5)	1 (−0.5-−2.5)	−1.5 (−0.5-−2.5)	−0.5 (−1.4-0.6)	0.6 (−1-+1)	0 (−1-+1)	0.7 (−0.5-1)	1.3 (0.2-2)

*Median (IQR), #Other than stoma closure, re-anastomosis. IQR: Interquartile range, TPN: Total parenteral nutrition, PICC: Peripherally inserted central catheter, : MOM: Mother’s Own Milk

ARMs were the most common anomaly, followed by EA and DA. Among ileal atresia, Type 3 was the most common type encountered (75%). Out of 11 neonates with ARM, 3 (27%) were high, while 7 (64%) were intermediate and 1 (9%), low. One baby had meconium ileus and eight had meconium peritonitis.

Out of the 10 EA cases, 50% had vertebral, anal atresia, cardiac defects, tracheoesophageal fistula, renal and radial abnormalities, and limb abnormalities (VACTERL) association. The most common anomaly associated with DA was trisomy 21 followed by structural heart defects. Two (18%) babies with ARM have associated anomalies, VACTERL in one baby and bilateral hydronephrosis in the other.

Duodeno-duodenostomy was performed in all babies with DA. Two neonates (25%) with ileal atresia developed short bowel syndrome, out of which 1 expired. One baby with meconium peritonitis developed short bowel syndrome but survived.

Among the total of 68 neonates with congenital GI anomalies, 10 (15%) expired, 4 (36%) due to surgery-related causes, and 6 (55%) due to sepsis. Anastomotic leak was present in 5 (7%) and re-surgery was needed in 12%. Ileal atresia had the highest mortality (62%) while IHPS, DA, Malrotation, HD and ARM had 100 % survival. Four neonates were operated for Hirschsprung's disease and all survived. Duration of both central line and TPN were highest in ileal atresia cases with a median of 20 days.

None of the surviving infants were moderately or severely underweight at follow-up [Table 2]. Median age of the three surviving infants who had ileal atresia was 4 (interquartile range 3–9) years and WFA Z-score was ‒1.5 (‒0.5 to ‒2.5) which was lower than that in other conditions.

Univariate analysis showed that mortality was significantly high in lower gestational age <32 weeks and outborn babies [Table 3]. Lower gestation age was found to be independent predictor of mortality on multivariate logistic regression analysis.

Table 3

Predictors of mortality in neonates operated for small intestinal anomalies (duodenal atresia, Jejunal atresia, ileal atresia, meconium peritonitis/ileus, malrotation) (n=34)

Baseline characteristics	Univariate analysis
	Survived (n=26), n (%)	Expired (n=8), n (%)	OR (95% CI)	P ¥
Gestational age (very preterm babies born at<32 weeks)	3 (11.54)	5 (62.50)	12.77 (1.96-82.89)	0.008
Birth weight (<2 kg)	7 (26.92)	5 (62.50)	4.52 (0.84-24.10)	0.077
Male	14 (54)	3 (38)	0.51 (0.1-2.61)	0.423
Outborn	20 (77)	3 (38)	5.55 (1.01-30.33)	0.048
Age at admission (days)*	1 (1-2)	1 (1-1)	0.64 (0.25-1.69)	0.372
Culture positive sepsis	8 (31)	4 (50)	2.25 (0.45-11.33)	0.326
Duration of long line (days)*	8 (5-11)	9.5 (7-20)	1.13 (0.99-1.3)	0.076
Time to reach full feeds	9 (5-12)	0 (0-19)	1 (0.9-1.11)	0.983
Shock	12 (46)	5 (63)	1.94 (0.38-9.88)	0.423
Short bowel syndrome	5 (19)	3 (38)	2.52 (0.45-14.24)	0.296
Primary anastomosis	19 (73)	4 (50)	0.37 (0.07-1.89)	0.231
Re-surgery	2 (8)	1 (13)	1.71 (0.14-21.82)	0.678
Anastomotic leak	3 (12)	2 (25)	2.56 (0.35-18.92)	0.358
	Multivariate analysis AOR (95% CI)			P €
Gestational age (very preterm babies born at<32 weeks)	11.87 (0.04-0.89)			0.048
Outborn	5.03 (0.78-35.81)			0.106

¥Univariate logistic regression analysis, *Median (IQR), €Multivariate logistic regression analysis. AOR: Adjusted odds ratio, CI: Confidence interval, OR: Odds ratio, IQR: Interquartile range

DISCUSSION

In antenatally diagnosed malformations, multidisciplinary team helps in better planning, management, and optimal outcome in the neonate. The antenatal detection rate of EAs was low (9%–10%) in the study by Gupta[5] while in our study, it was higher (50%). Our study had an antenatal diagnosis rate of 80% for DA which is higher than other studies with the detection rate of 50%–70%.[67] As reported in the literature, antenatal diagnosis of ARM is usually difficult and mostly missed.[8] All neonates with meconium ileus/peritonitis in our study had polyhydramnios in antenatal scans. One of these neonates presented as nonimmune hydrops, which had been reported by the authors.[9] This higher rate of antenatal diagnosis of congenital GI malformations can be attributed to our perinatal center with a fetal medicine unit.

Inverted superior mesenteric artery (SMA)-superior mesenteric vein (SMV) relationship was seen in postnatal scans in both cases of malrotation. However, a normal SMA/SMV relationship can be seen in up to 29% of patients with surgically proven malrotation.[1]

The incidence of anomalies associated with EA ± TEF ranges between 30% and 60% in different studies.[51011] In ours, 50% had VACTERL association similar to Galarreta et al. study.[11] Half of the cases with DA were associated with Down syndrome as reported in previous studies.[6] Genetic predisposition along with environmental factors have been implicated in the pathogenesis of IHPS. Moreover, there is still ongoing debate whether the condition is congenital or acquired.[12] Hence, we also included neonates with IHPS in our study. The median postnatal age of these cases at presentation was around 3 weeks of life and hence were admitted and managed in our unit.

Among the postoperative complications of EA mentioned in the literature, anastomotic stricture is the most common (30%–40%).[5] It was present only in 10% of our cases. Surgical procedure for JIA is usually based on site and type of atresia, proximal segment dilatation, length of remaining bowel, and general condition of the neonate. Most of our babies were preterm babies. Based on the surgeon's discretion and condition of the bowel, ileostomy and re-anastomosis later was opted in 3 (38%) of the ileal atresia cases. Reports suggest that resection of dilated proximal bowel should be minimized, as TPN can take care of the period of dysmotility and impaired anastomotic function.[13] However, preservation of as much bowel length as possible at the risk of creating a poorly functioning anastomosis can produce significant morbidity and mortality.[14]

All our EAs were associated with TEF and the mortality was 20% which was lower than that reported by Gupta study (68%).[5] However, our number of EA cases was low when compared to their volumes and all our cases were either Type B, C, or D. The outcome of IHPS was excellent (100%) similar to that in literature.[15] Our mortality in JIA was 46% which was much higher compared to that in developed countries[16] but comparable with the Indian study by Gupta et al.[14] Around two-third of mortality in our JIAs were due to sepsis. Short bowel syndrome was present in 21% of our neonates with ileal/jejunal atresia comparable with the study by Burjonrappa et al.[16] Survival in malrotation was better with a lower incidence of short bowel syndrome than previous studies.[1718] Prognosis was good in ARM with 100% survival, more than the reported survival of 85%–95%.[19] This might be due to most of the ARM being intermediate type and staged surgery.

Weight on follow-up was good even for ileo-jejunal atresia (median Z-score being ‒0.5 in jejunal atresia and ‒1.5 in ileal atresia). Early diagnosis of the condition, proper surgical management, standardized TPN policy in postoperative period and regular follow-up probably helped in the prevention of growth failure.

Significant predictors of death in the study by Sarkar et al. were gestational age and nature of GI malformations.[1] In our study, predictors of mortality in neonates with small intestinal anomalies were lower gestational age and babies referred from outside. Longer duration of peripherally inserted central catheter, though not a significant predictor showed a trend toward increased mortality. Strategies to reduce the duration of long line and central-line associated bloodstream infection are important, as sepsis has contributed to significant number of deaths in our study. Poorer outcomes in outborn babies can be attributed to lesser pickup in the antenatal period, delayed postnatal diagnosis, and intervention.

CONCLUSION

Overall survival of GI malformations was 85%, with IHPS, DA, Malrotation, HD and ARM having 100% survival. Associated anomalies were present in 26% with half of the tracheoesophageal fistula cases having VACTERL association. This highlights the need to actively look for associated anomalies. Early referral or preferably in-utero to tertiary neonatal unit can result in a better outcome. Modifiable factors such as culture positive sepsis and longer duration of long line needs attention. Although neonates with GI anomalies are vulnerable for growth failure, our neonates had optimal growth on follow-up possibly due to standardized TPN policy during NICU stay.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.