Factors Associated with a Failed Nonoperative Reduction of Intussusception in Children.

AIMS: The aim of this study was to evaluate the factors associated with the failure of nonoperative reduction of intussusception in children.
METHODS: A retrospective study was conducted in a tertiary care pediatric surgery hospital. The children admitted to the department of pediatric surgery between November 2013 and February 2020 with the diagnosis of Intussusception were included.
RESULTS: A total of 106 (67%) children underwent pneumatic reduction. Eighty-nine (84%) children had a successful reduction. A higher rate of failed reduction was found in children who presented at or after 48 h of the onset of symptoms (P = 0.03) and abdominal distension at presentation (P < 0.002). On multiple logistic regression analysis, the children presenting at or after 48 h of the onset of symptoms (odds ratio [OR] = 11.3; P = 0.039) and abdominal distension at presentation (OR = 4.46; P = 0.021) were found to be associated with increased risk of failure of nonoperative reduction. The variables age <1 year, weight <10 kg, pain abdomen, vomiting, bilious vomiting, fever, bleeding per rectum, and palpable mass were not associated with the failed nonoperative reduction. The variables, presentation at or after 48 h of the onset of symptoms (OR = 2.812; P = 0.045) and abdominal distension at presentation (OR = 8.758; P = 0.000) were found to be associated with an increased need for surgery.
CONCLUSION: The risk factors for failed nonoperative reduction of intussusception include a presentation at or after 48 h of the onset of symptoms and the presence of abdominal distension at presentation. The delayed presentation was associated with the increased need for surgery and increased chances of intestinal nonviability. Copyright:

INTRODUCTION

Intussusception is the most common cause of intestinal obstruction in infants and toddlers with an incidence of one to four every 2000 children.[12] The nonoperative method of reduction is the gold standard treatment for intussusception.[2] The rate of successful reduction of intussusception is as high as 95% and as low as 46%.[34] The British Society of Pediatric Radiologist guidelines recommends at least 75% successful nonoperative reduction rate.[5] The successful nonoperative reduction of intussusception carries less morbidity and mortality.[2] The better we understand the risk factors for failed reduction, the better we can prognosticate the condition and can prepare for the possible interventions. Hence, a study was conducted to evaluate the factors associated with the failure of nonoperative reduction of intussusception in children.

METHODS

A retrospective study was conducted in a tertiary care pediatric surgery hospital. The children admitted to the department of pediatric surgery between November 2013 and February 2020 with the diagnosis of intussusception were included. The data were collected from electronic medical records and patient charts. The data collected included demographic parameters (age, gender, and weight), presentation (duration of symptoms, previous history of intussusception, pain abdomen, abdominal distension, vomiting, fever, bleeding per rectum, palpable mass, and peritonitis), ultrasonography findings, and the method of intervention (nonoperative reduction or surgery). The type of nonoperative reduction (pneumatic or hydrostatic) and its outcome, recurrence, nature of the surgery, and intraoperative findings (perforation, gangrene, and lead point) were also documented.

The children admitted with intussusception were resuscitated and underwent pneumatic reduction if there were no contraindications (pneumoperitoneum, peritonitis, or persistent hypotension). The pneumatic reduction was performed by a pediatric surgeon under C-arm guidance. A Foley's catheter was placed into the rectum of the child and the buttock was strapped to prevent air leak. The air was insufflated up to a pressure of 80 to 120 mmHg. The success of the procedure was determined by air reflux into a few loops of the small bowel. If that did not happen, the procedure was stopped at 3 min and re-attempted after 5 min for a total of three attempts each lasting 3 min. If the procedure was unsuccessful, the entire procedure was retried after an hour break. If the procedure failed, the child was taken to the operation theatre for surgery. After the successful reduction, the oral feeds were reinitiated and the child was discharged on the next day. The child with contraindication for nonoperative reduction directly underwent surgery after resuscitation. The children were followed up at 15 days, 30 days, 3 months, 6 months, and then yearly.

Statistical analysis

The statistical analysis was performed using IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp. The descriptive variables were expressed as number and percentage, while the continuous data were expressed as mean ± standard deviation or median (range). Differences were evaluated using the Pearson Chi-squared test or Fisher's exact test for noncontinuous data and Student's t-test or Mann–Whitney U-test for continuous data. Multinomial logistic regression analysis was used to identify the risk factors. The difference between the groups was considered significant when the “P” value was <0.05 with a 95% confidence interval.

RESULTS

A total of 158 children were admitted with the diagnosis of intussusception during the study period. The median age of the study group was 13.5 months (1 month to 16 years). The follow-up ranged from 6 months to 5 years. Ninety-six (61%) were male children. Vomiting was the most common presentation seen in 109 (69%) children. The triad of symptoms (pain abdomen, vomiting, and currant jelly stools) was seen in 21 (13%) children [Table 1].

Table 1

Clinical profile of children with intussusception

Variables	n (%)
Total number of children	158
Median (range) age of the study group	13.5 months (1 month to 6 years)
Age <1 year, n (%)	69 (44)
Male: female ratio	96:62
Mean weight of study group (kg)	10.5±6.4
Previous history of intussusception	4
Clinical presentation, n (%)
Pain abdomen	88 (56)
Abdominal distension	52 (33)
Vomiting	109 (69)
Bilious vomiting	10 (6)
Fever	19 (12)
Bleeding per rectum	74 (47)
Palpable mass	43 (27)
Peritonitis	6 (4)

Out of 158, five (3%) children were diagnosed to have intussusception when they were explored for an acute abdomen. The rest 153 (97%) were diagnosed by ultrasonography. Transient intussusception was found in 26 (16%) children who were observed and discharged with uneventful follow-up. Fifteen (9%) children had either ileoileal or jejunojejunal intussusception which were persistent and hence were operated on. Six (4%) children with proven intussusception presented with peritonitis underwent surgery.

A total of 106 (67%) children underwent pneumatic reduction. Eighty-nine (84%) children had a successful reduction. Fourteen (13%) children with failed reduction underwent laparotomy. Twelve of these 14 children had to undergo bowel resection (seven gangrene and five lead points) and the rest two had a successful operative reduction. Two (2%) children with perforation following reduction, one child (1%) with suspected failed reduction, and one (1%) with suspected pneumoperitoneum underwent surgery. Seven (8%) of 89 successful reductions had a recurrence. The management of these children is depicted in Figure 1.

Figure 1

Presentation and management of intussusception

There were a total of eight (5%) children who had secondary intussusception. The lead points found were five cases of Meckel's diverticulum, a case of lymphoma, an ileal polyp, and an ileal stricture. A total of 48 (30%) children required surgical intervention which included 28 (58%) cases of bowel resection [Figure 1 and Table 2].

Table 2

Operative management of intussusception (n=48)

Surgical procedure	n (%)
Surgical reduction	13
Laparotomy	8
Ileo-ileal	1
Ileocolic	7
Laparoscopy	5
Ileo-ileal	2
Jejuno-jejunal	1
Ileocolic	1
Colocolic	1
Surgical reduction with ileo-cecopexy	2
Resection-anastomosis	28
Gangrene	16
Meckel’s diverticulum	5
Lymphoma	1
Polyp	1
Irreducible	5
Ileal stricturoplasty	1
Negative laparotomy	4
Doubtful pneumoperitoneum	1
Doubtful reduction	1
Spontaneous reduction	2

Four (3%) children had a previous history of nonoperative reduction of intussusception. Three of these four children were managed by pneumatic reduction with uneventful follow-up. The other child, an 8-year-old boy, underwent laparotomy for failed reduction and he was found to have a distal ileal mass (lymphoma) requiring ileocecal resection with ileo-ascending anastomosis. We had one death in our study. A male infant presented in shock and sepsis, underwent laparotomy, resection of gangrene, and ileo-ascending anastomosis. But he succumbed to sepsis on the postoperative day 1.

There were a total of 110 (69%) children who presented at or after 48 h of the onset of symptoms. These children had a higher incidence of abdominal distension at presentation (43/110 vs. 9/48; P = 0.01), a higher failed reductions (15/69 vs. 2/37; P = 0.03), and so a higher need for surgery (41/110 vs. 7/48; P = 0.004) compared to the children who presented within 48 h of the onset of symptoms. Currant jelly stools were found more in children presented within 48 h of the onset of symptoms [29/48 vs. 45/110; P = 0.02; Table 3].

Table 3

Comparison between children according to the time of presentation

Variables	Presentation <48 h (n=48)	Presentation ≥48 h (n=110)	P
Clinical profile
Median age (range)	12 months (2 months to 15 years)	14 months (1 month to 16 years)	0.86†
Age <1 year (n=69)	22 (46)	47 (43)	0.72*
Mean weight (kg)	10.4±6.4	10.5±6.3	0.85+
Pain abdomen	23 (48)	65 (59)	0.19*
Abdominal distension	9 (19)	43 (39)	0.01*
Vomiting	37 (77)	72 (65)	0.15*
Bilious vomiting	4 (8)	6 (5)	0.49*
Fever	3 (6)	16 (15)	0.14*
Bleeding per rectum	29 (60)	45 (41)	0.02*
Palpable mass	17 (35)	26 (24)	0.13*
Peritonitis	0	6 (5)	0.18#
Management of intussusception
Unsuccessful pneumatic reduction (n=17)	2/37 (3)	15/69 (23)	0.03*
Perforation following reduction (n=2)	1/37 (3)	1/69 (1)	0.65*
Recurrence (n=7)	4/37 (11)	3/69 (4)	0.3*
Need for surgery (n=48)	7 (15)	41 (37)	0.004*
Gangrene (n=16)	2 (4)	14 (13)	0.1*
Lead point (n=8)	0	8 (7)	0.1#

†Mann-Whitney U-test, *Chi-square test, +Independent sample t-test, #Fisher’s exact test

A higher rate of failed reduction was found in children who presented at or after 48 h of onset of symptoms (P = 0.03) and abdominal distension at presentation [P = 0.002; Table 4]. On multiple logistic regression analysis, the children presenting at or after 48 h onset of symptoms (odds ratio [OR] = 11.3; P = 0.039) and abdominal distension at presentation (OR = 4.46; P = 0.021) were found to be associated with increased risk of failure of nonoperative reduction. The variables age < 1 year (OR = 0.466; P = 0.165), weight < 10 kg (OR = 1.641; P = 0.468), pain abdomen (OR = 1; P = 0.99), vomiting (OR = 0.562; P = 0.39), bilious vomiting (OR = 6.75; P = 0.136), fever (OR = 2.23; P = 0.357), bleeding per rectum (OR = 2.55; P = 0.162), and palpable mass (OR = 2.74; P = 0.135) were not associated with the failed nonoperative reduction.

Table 4

Successful versus unsuccessful pneumatic reduction

Variables	Successful reduction (n=89)	Failed reduction (n=17)	P
Median age (range)	13 months (1 month to 16 years)	12 months (5 months to 7 years)	0.65†
Age <1 year (n=50)	42 (47)	8 (47)	0.99*
Mean weight (kg)	10.2±6.7	9.6±4.2	0.75+
Weight <10 kg	56 (63)	11 (65)	0.89*
Presentation within 48 h	35 (39)	2 (12)	0.03*
Pain abdomen	48 (54)	8 (47)	0.6*
Abdominal distension	20 (22)	10 (59)	0.002*
Vomiting	67 (75)	12 (71)	0.68*
Bilious vomiting	4 (4)	2 (12)	0.24*
Fever	6 (7)	3 (18)	0.14*
Bleeding per rectum	46 (52)	12 (71)	0.15*
Palpable mass	27 (30)	7 (41)	0.38*

†Mann-Whitney U-test, *Chi-square test, +Independent sample t-test

The variables presentation at or after 48 h of the onset of symptoms, abdominal distension at presentation, and bilious vomiting were associated with an increased need for surgery [Table 5]. On logistic regression analysis, only the factors, presentation at or after 48 h of the onset of symptoms (OR = 2.812; P = 0.045) and abdominal distension at presentation (OR = 8.758; P = 0.000) were found to be associated with an increased need for surgery. The bilious vomiting had no association with the need for surgery (OR = 2.425; P = 0.277).

Table 5

Surgery versus nonsurgery group

Variables	Surgery group (n=48)	Nonsurgery group (n=110)	P
Median age (range)	14 months (4 months to 12 years)	13.5 months (1 month to 16 years)	0.86†
Age <1 year (n=69)	19 (40)	50 (45)	0.49*
Mean weight (kg)	10.7±5.8	10.4±6.6	0.74+
Weight <10 kg	28 (58)	68 (62)	0.68*
Presentation within 48 h	7 (15)	41 (37)	0.004*
Pain abdomen	24 (50)	64 (58)	0.34*
Abdominal distension	33 (69)	19 (17)	0.000*
Vomiting	34 (71)	75 (68)	0.74*
Bilious vomiting	6 (13)	4 (4)	0.04*
Fever	8 (17)	11 (10)	0.23*
Bleeding per rectum	29 (60)	45 (41)	0.02*
Palpable mass	15 (31)	28 (25)	0.45*
Peritonitis	6 (13)	0	0.001#

†Mann-Whitney U-test, *Chi-square test, +Independent sample t-test, #Fisher’s exact test

DISCUSSION

Intussusception is one of the most common abdominal emergencies in toddlers with an incidence of 0.33–2.4 per 1000 live births.[6] Early intervention reduces the risks of complications (bowel gangrene, perforation, peritonitis, shock, and death) and increases the chances of successful reduction.[5]

Ileocolic intussusceptions contribute to 90% of intussusceptions while ileoileal, jejunojejunal, and colocolic intussusceptions contribute to the rest 10% of intussusceptions.[2] The nonoperative reduction is the procedure of choice for ileocolic and colocolic intussusceptions.[2] The various nonoperative methods include pneumatic reduction, saline reduction, and barium reduction.[2] The pneumatic reduction is found to be less messy, easy, faster, safer, and more successful than hydrostatic reduction.[16]

The success rate of nonoperative reduction varies from 44% to 95%.[345] The success rate of nonoperative reduction in our study (84%) was similar to Talabi et al. and Xie et al.[27] This finding was superior to that of other studies.[48910111213]

Our study revealed that the factor, presentation at or after 48 h of the onset of symptoms, had an increased risk of failed nonoperative reduction. This is similar to other studies such as Reijnen et al. and Xiaolong et al.[1415] Kaiser et al., Khorana et al., and Chung et al. found that the children who presented after 24 h had an increased risk of failed nonoperative reduction.[41016] To contradict these findings, Talabi et al., Binkovitz et al., Curtis et al., He et al., and Steadman et al. found that the duration of symptoms had no association with the nonoperative reduction of intussusception.[2111317] In our study, we found that 50% of children with failed reduction had gangrene and two had a perforation. This suggests that the prolonged duration of symptoms is associated with intestinal nonviability.

In our study, the infants (age <1 year) were not found to have an increased risk of failed nonoperative reduction. This finding is supported by Talabi et al., Kaiser et al., Khorana et al., Binkovitz et al., Curtis et al., He et al., and Reijnen et al.[2410111213141617] Even though the majority of infants (<1 year) in our study group presented late, they had no significant association with increased failed nonoperative reduction. The infants in the studies by Xialong et al. and Fallon et al. had increased chances of a failed nonoperative reduction.[1518] It could be because of the small-caliber lumen and highly competent ileocecal valve in infants leading to a higher failure rate.[2]

Our study found that the abdominal distension at presentation had an increased risk of failed nonoperative reduction. This is similar to other studies such as Kaiser et al. and Khorana et al.[410] Abdominal distension usually occurs late and its indirect evidence of late presentation. This is useful in prognosticating children presenting with a nonreliable history.

Rectal bleeding had no association with a failed nonoperative reduction, which was similar to Curtis et al.[12] A few studies contradict this finding of our study.[10131415] A study by Fallon et al. found that the rectal bleeding had an association with a failed reduction on univariate analysis but logistic regression analysis ruled out the association.[18]

Khorana et al. found that the variables vomiting, fever, and palpable mass had a positive association with a failed nonoperative reduction.[10] But in our study, none of these factors had an association with a failed nonoperative reduction. Vomiting was the most common presentation in most of the studies including ours.[2412] Despite this, many studies have shown no association between the vomiting and the failed reduction.[4121415] The studies Xiaolong et al. and Fallon et al. ruled out an association between fever and failed reduction.[1518] The disappearance of palpable mass has been a criterion to define a successful reduction in many studies, hence considering this to be a prognostic factor for failed reduction may not be appropriate.[1920]

Certain ultrasonography findings such as the presence of peritoneal fluid, trapped fluid between the bowel loops, and thickened bowel loops were associated with the increased risk of failed nonoperative reduction.[10131518]

As per Steadman et al., fluid resuscitation was associated with an increased risk of failed pneumatic reduction. They proposed that fluid resuscitation will increase the edema of the intussusceptum.[17] But further studies may require evaluating this factor.

Feldman et al. found out that the chances of failed nonoperative reduction significantly reduce with sedation.[21] The drugs used in sedation such as propofol, ketamine, and midazolam have smooth muscle relaxant action which helps in increased chances of reduction. But there is also a danger of increased chances of perforation.[21]

CONCLUSION

The nonoperative reduction is the treatment of choice for children with intussusception. The risk factors for failed nonoperative reduction of intussusception include a presentation at or after 48 h of the onset of symptoms and the presence of abdominal distension at presentation. These two factors were also found to be associated with an increased need for surgery. The variables age, weight, vomiting, fever, rectal bleeding, and palpable mass were not associated with failed nonoperative reduction.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.