Ultrasound-guided reduction of intussusception in infants in a developing world: saline hydrostatic or pneumatic technique?

Non-operative reduction has emerged as first line in the management of uncomplicated intussusception. The aim of this study was to compare the outcome of ultrasound-guided saline hydrostatic reduction and ultrasound-guided pneumatic reduction of intussusception in infants. This is a prospective study of infants with uncomplicated intussusception confirmed by ultrasound over a period of 21 months from December 2018 to August 2020. Fifty-two (69.3%) out of seventy-five infants were eligible and randomized based on simple random sampling technique into two groups: Group A included patients who had ultrasound-guided hydrostatic (saline) reduction; Group B included patients who had ultrasound-guided pneumatic (air) reduction. The success rates, time to reduction and complication rates were assessed. The success rates, between the saline hydrostatic reduction group and pneumatic reduction group, were comparable [17 (65.4%) versus 19 (73.1%); relative risk (RR) 0.8; 95% confidence interval (CI) 0.6–1.2; p = 0.54]. The mean time to reduction was higher in the saline hydrostatic reduction group (15.4 ± 5.1 min versus 10.8 ± 4.1 min; p = 0.003). There was no statistically significant difference in the perforation and recurrence rates between the two groups. Conclusion: Saline hydrostatic reduction and pneumatic reduction of uncomplicated intussusception under ultrasound guidance in infants might have comparable outcomes. However, pneumatic reduction may be faster. What is Known: • Ileocolic intussusception is the most common cause of intestinal obstruction in infants. • Ultrasonography is useful in the diagnosis and non-operative treatment of ileocolic intussusception. What is New: • Ultrasound-guided hydrostatic enema and ultrasound-guided pneumatic enema are similarly effective and safe techniques in the reduction of ileocolic intussusception. • Reduction of ileocolic intussusception under ultrasound guidance is a great technique that may prove useful in the developing world due to lower cost of required equipment. What is Known: • Ileocolic intussusception is the most common cause of intestinal obstruction in infants. • Ultrasonography is useful in the diagnosis and non-operative treatment of ileocolic intussusception. What is New: • Ultrasound-guided hydrostatic enema and ultrasound-guided pneumatic enema are similarly effective and safe techniques in the reduction of ileocolic intussusception. • Reduction of ileocolic intussusception under ultrasound guidance is a great technique that may prove useful in the developing world due to lower cost of required equipment.


Introduction
Ileocolic intussusception is the invagination of the terminal ileum into the colon [1][2][3]. It is the commonest cause of intestinal obstruction in infants, occurring mostly between three and nine months of age [2][3][4]. It often progresses to bowel ischemia and necrosis when left untreated [5,6]. When intussusception is not associated with shock, bowel gangrene, perforation or peritonitis, it is said to be uncomplicated.
The management of intussusception has evolved from operative to non-operative methods [7][8][9]. Non-operative Communicated by Piet Leroy * Isaac S. Chukwu Isaacchukwu8@gmail.com 1 reduction has become the standard of management with laparotomy reserved for complicated cases at presentation and failed reduction [10]. These non-operative methods are reported to be cheaper and safe in comparison to operative treatment [8,11]. The shift to non-operative management has been noted to be slow in some developing world probably as a result of substantial proportion of cases with late presentation and attendant complications and also the lack of facilities [7,9]. Despite these, current evidence indicate that non-operative reduction should be the initial management of uncomplicated intussusception despite delay in presentation [7][8][9][10]. Non-operative reduction of intussusception may be by hydrostatic or pneumatic enema [12,13]. The former may be with saline, barium or water-soluble contrast while the latter may be by air, oxygen or carbon (iv) oxide [14,15]. Non-operative reduction is undertaken under fluoroscopic or ultrasound guidance. Due to concerns of exposure to radiation and availability of the former in our environment, ultrasound guidance is commonly used [16,17]. Reports of success with hydrostatic reduction abound in literature with increasing use of pneumatic reduction [15]. There is presently no consensus, on which reduction technique has the highest reduction rates with lowest rate of complications [18].
Studies comparing these non-operative reduction methods under ultrasound guidance are scant in our setting. This study compared the outcome (success rate, time to reduction and complication rates) of saline hydrostatic reduction and pneumatic reduction of intussusception under ultrasound guidance in infants with a view to guide clinical decision making.

Methods
This is a prospective, randomized study that was carried out over a period of 21 months from December 2018 to August 2020 at the University of Nigeria Teaching Hospital (UNTH), Enugu, Southeast Nigeria. The study was approved by the UNTH Health Research and Ethics committee.
Inclusion criteria included children 1 to 12 months of age and presentation with uncomplicated intussusception confirmed by abdominal ultrasonography. Exclusion criteria were peritonitis determined by clinical features, bowel perforation sign on abdominal radiograph or abdominal ultrasound, nonresponsive shock and recurrent intussusception.
The patients were assigned to two groups. Group A included patients who had ultrasound-guided hydrostatic (saline) reduction. Group B included patients who had ultrasound-guided pneumatic (air) reduction. Eligible patients were randomized based on a simple random sampling technique. A total of 52 folded pieces of paper were prepared and mixed in a bag by a trained research assistant who was unaware of the demographic and clinical characteristics of the patients. Twenty-six folded pieces of paper had "A" printed on them and 26 folded pieces of paper had "B" printed on them. The bag was entrusted to the research assistant who picked a folded piece of paper from the bag, without looking into the bag (at the radiological suite just before the saline hydrostatic or pneumatic reduction) for each consecutive patient without replacement and assigned them to interventions. The folded pieces of papers were re-mixed in the bag each time. The letter "A" represented group A, those who had ultrasound-guided hydrostatic (saline) reduction, while letter "B" represented group B, those who had ultrasound-guided pneumatic (air) reduction. The sample population was attained when all the folded papers were picked. Written informed consent was obtained from the parents or legal guardians.
On presentation at the children emergency room, the patients were screened for suitability by clinical assessment of their history and physical examination and diagnosis of intussusception confirmed by abdominal ultrasound. The children were placed on nothing by mouth, and resuscitation was commenced with intravenous fluids, intravenous antibiotics (ceftriaxone 50 mg/kilogramme body weight/ day in 2 divided doses; metronidazole 7.5 mg/kilogramme body weight/dose 8 hourly), passage of nasogastric tube and urethral catheter. Serum electrolytes, urea and creatinine; and full blood count were done. Electrolyte imbalance when present was corrected and haemogram optimized accordingly to ensure a minimum of 10 g/dl. When the patient was fully resuscitated and making up to 1-2 ml/kilogramme body weight/hour of urine, the diagnosis was confirmed by abdominal ultrasonography at the radiological suite. Adequate preparation was made for immediate surgery in the event of a failed reduction and/or perforation of bowel. Eligible patients proceeded to have enema reduction.

Ultrasound-guided saline hydrostatic reduction
At the radiological suite, the child was placed supine on a couch. An ultrasound machine (GE Versana Essential S/W of Medical Systems (China) Co. LTD) with a linear transducer of 7.5 MHz was used by a radiologist to localize the mass. A size 22FG 2-way Foley catheter was inserted into the rectum and balloon inflated with 30 ml of sterile water. An adhesive tape was firmly applied (extending from one greater trochanter to the other, holding the buttocks together) to ensure a good anal seal. A 1 L infusion bag of normal saline (warmed to body temperature by placing it in a bowl of warm water) was suspended on a drip stand, 100 cm from the level of the patient. It was connected to the drainage channel of the Foley catheter with an infusion giving set. The fluid was allowed to run while monitoring the mass with 1 3 ultrasound. The pulse rate and oxygen saturation were monitored and recorded during the procedure. Complete reduction was confirmed by the disappearance of the intussusception mass and reflux of saline into the small bowel. Not more than three attempts at reduction were made. Each attempt was continued as far as the mass was progressively reducing as monitored with ultrasound. However, each attempt was aborted if after 3 min the mass was no longer reducing as monitored with ultrasound. Each attempt was aborted by disconnecting the infusion giving set from the Foley catheter; hence, reducing the pressure before another attempt was started again. Time to reduction was monitored with a stop watch. At the end of the procedure, the balloon was deflated and the catheter removed and the patient monitored.

Ultrasound-guided pneumatic reduction
At the radiological suite, the child was placed supine on a couch. An ultrasound machine (GE Versana Essential S/W of Medical Systems (China) Co. LTD) with a linear transducer of 7.5 MHz was used by a radiologist to localize the mass. A size 22FG 2-way Foley catheter was inserted into the rectum and balloon inflated with 30 ml of sterile water. An adhesive tape was firmly applied (extending from one greater trochanter to the other, holding the buttocks together) to ensure a good anal seal. The part of the aneroid sphygmomanometer containing the hand bulb, release valve and aneroid gauge was connected to the drainage channel of the Foley catheter via a short tube (cut from the tubing of a urine bag). The hand bulb was then squeezed to release air into the bowel while monitoring the mass with ultrasound. The pressure was carefully monitored using the aneroid gauge and not allowed to exceed 120 mmHg at any time. The pulse rate and oxygen saturation of the patient were monitored and recorded during the procedure. Complete reduction was confirmed by the disappearance of the intussusception mass and reflux of air into the small bowel. Not more than three attempts at reduction were made. Each attempt was continued as far as the mass was progressively reducing as monitored with ultrasound. However, each attempt was aborted if after 3 min the mass was no longer reducing as monitored with ultrasound. Each attempt was aborted by disconnecting the short tube from the Foley catheter; hence, reducing the pressure before another attempt was started again. The time to reduction was also monitored with a stop watch. At the end of the procedure, the adhesive tape was removed, the balloon was deflated, the catheter removed, and patient monitored.
Those who had a successful reduction of intussusception were taken back to the ward and monitoring continued. The nasogastric tube was removed when the drainage was minimal, no longer bilious and there was no abdominal distension. Oral feeds were commenced when patients were able to pass faeces and/or flatus. The intravenous fluids were stopped when oral intake was fully established; urethral catheter was then removed. Patients were discharged when recurrence was ruled out by the absence of symptoms of intussusception. The parents were warned of the possibility of recurrence of intussusception and the need to present early if symptoms such as abdominal pain, vomiting, passage of blood and mucus and abdominal distension recurred. Patients who did not have successful reduction proceeded to have surgery. Delayed repeat enema was not done. Abdominal ultrasonography was not routinely done to confirm recurrence before the patients were discharged from the hospital.
The patients were followed up at the Paediatric Surgery out-patient department and via phone calls. They were seen on the first, fourth and twenty-fourth week following discharge from the hospital. Each patient was followed up for a minimum of 6 months. The parents/legal guardian were called on phone and reminded of follow-up visits. The symptoms of recurrence of intussusception such as abdominal pain, vomiting, passage of blood and mucus and abdominal distension were assessed for during each follow-up visit, and the parents were warned of the possibility of recurrence of intussusception the need to present early if symptoms such as abdominal pain, vomiting, passage of blood and mucus and abdominal distension recurred.

Outcome measures
The primary outcome measure was the success rate of enema reduction. The secondary measures included the time to enema reduction and complication -recurrence and perforation rates.
Successful reduction was defined as the disappearance of the intussusception mass, reflux of saline/air into the small bowel and resolution of symptoms of intussusception.
The time to reduction was the time from the instillation of the normal saline or insufflation of air to when successful reduction occurred by ultrasound assessment.
Failed reduction was defined as presence of the intussusception mass after 3 attempts (or less than or equal to 3 attempts if perforation occurred) at reduction have been made.
Recurrence was defined as return of symptoms of intussusception 12 h or more after a successful reduction [19].
Successful reduction and recurrence were assessed by a senior registrar in Paediatric Surgery who was not involved in the enema reduction.

Data collection
A proforma was used for data collection. This included: patient's age, sex, weight, duration of symptoms prior to presentation, phone number of parents, date of admission, clinical features, type and outcome of non-operative reduction, details of operative treatment for failed reduction, duration of hospital stay and duration of follow-up.

Statistical analysis
Data were captured, stored, retrieved and analysed using Statistical Package for Social Sciences (SPSS) of IBM SPSS statistics for windows, version 20 (SPSS Inc., Chicago, Illinois). Descriptive data were expressed as percentages, mean ± standard deviation and presented in tables and charts. Shapiro-Wilk test was used to test for normality of the data. The independent samples t test was used for descriptive numerical data and chi-square was applied to categorical data. Independent samples t test was used to compare the mean time to reduction while chi-square was used to compare the success, recurrence and perforation rates of the two groups. Relative risk with 95% confidence interval was calculated for the success, recurrence and perforation rates. p value <0.05 was considered statistically significant.

Results
Seventy-five infants presented with intussusception within the study period (December 2018 to August 2020). Fiftytwo (69.3%) patients who met the inclusion criteria were enrolled; completed the study; and were included in the final data analysis with twenty-six patients in each group. Of the twenty-three (30.7%) infants that were excluded; 7 (9.3%) patients presented with shock, 8 (10.7%) patients had gross abdominal distension with respiratory distress, 3 (4.0%) patients had peritonitis, 1 (1.3%) patient had bowel perforation, and 4 (5.4%) patients were referred from another facility with failed enema reduction.
There was no statistically significant difference in the demographic and clinical characteristics of the patients in the two groups ( Table 1). The age range at presentation in Group A was 4 months to 12 months with a median age of 5 months while the age range at presentation in Group B was 4 months to 9 months with a median age of 6 months. The mean age of patients in Group A was 5.5 ± 1.8 months, while the mean age of patients in Group B was 6.1 ± 1.6 months (t = 1.27, p = 0.24). Group A had 16 males (61.5%) and 10 females (38.5%) while Group B had 19 males (73.1%) and 7 females (26.9%). The sex distribution was similar in the two groups (p=0.37). The mean duration of symptoms in Group A was 72.0 ± 48.9 h (range: 24-216 h), while the mean duration of symptoms in Group B was 76.8 ± 38.8 h (range: 24-168 h) (t = 0.39, p =0.69). Patients in Group A were followed up for a mean duration of 13.8 ± 3.4 months (range: 6-20 months), while patients in Group B were followed up for a mean duration of 14.4 ± 4.4 months (range: 7-20 months) (t = 0.55, p=0.55).
Seventeen patients in Group A had a successful reduction giving a success rate of 65.4%, while nineteen patients in Group B had a successful reduction giving a success rate of 73.1%. Statistical analysis of the success rates of the two groups gave a p value of 0.54 (relative risk (RR) 0.8; 95% confidence interval (CI) 0.6-1.2) which was not statistically significant ( Table 2). The mean time to reduction for Group A was 15.43 ± 5.12 min (range: 7-25 min), while the mean time to reduction for Group B was 10.86 ± 4.13 min (range: 3-16 min) (Fig. 1). Statistical analysis of the time to reduction for the two groups gave a p value of 0.003 (t = 3.54, p<0.05) which is statistically significant.
One patient in Group A had a perforation during reduction giving a perforation rate of 3.8%. There was no account of perforation in Group B. Statistical analysis of the perforation rates of the two groups gave a p value of 0.31 [RR 0.9 (CI 0.9-1.0)] which was not statistically significant.
Two patients in Group A had recurrence after reduction giving a recurrence rate of 7.7%, while one patient in Group B had recurrence giving a recurrence rate of 3.8%. Statistical analysis of the recurrence rates of the two groups gave a p-value of 0.55 [RR 0.9 (CI 0.8-1.0)] which was not statistically significant. For the two patients who had recurrence in Group A, the first patient had recurrence 1 week after reduction and the second patient presented 5 weeks after reduction. The second patient who died while being resuscitated was the only death recorded in this study-giving a fatality rate of 3.8% in Group A. The patient who had recurrence in Group B developed symptoms 36 h after an initial successful reduction. He had a repeat ultrasound-guided pneumatic (air) reduction which was successful. He was discharged after counselling on the risk of another recurrence.
All (34.6%) of the patients with a failed reduction in Group A had ileocolic intussusception. All (26.9%) of the patients with a failed reduction in Group B had ileocolic intussusception. No pathological lead point was seen at surgery in both groups. All (26.9%) of the patients in Group B with a failed reduction had bowel resection and anastomosis done for them. Eight patients in Group A (30.8%) had bowel resection and anastomosis done, while one patient (3.8%) had manual reduction done. No patient who had a failed reduction had any form of enterostomy done. Intraoperative findings in 7 patients in Group A with a failed reduction (26.9%) were oedematous irreducible intussusception, oedematous reducible intussusception in 1 patient (3.8%) and oedematous intussusception with a perforation on the colon adjacent to a gangrenous area in 1 patient (3.8%). The intraoperative finding in all the patients in Group B Fig. 1 A box plot showing time to reduction for the two groups with a failed reduction (26.9%) was oedematous irreducible intussusception.
Moreover, for the 23 patients who were excluded, 3 patients died while being resuscitated in lieu of surgery, 4 patients with oedematous reducible intussusception had manual reduction, and 16 patients with gangrenous intussusception had bowel resection and anastomosis.

Discussion
Ultrasonography is useful in enema reduction of intussusception. This is more so in resource-poor setting like ours where fluoroscopy is not routinely available. We compared the success rate, time to reduction and complication rate of ultrasound-guided hydrostatic enema and ultrasound-guided pneumatic enema in the reduction of intussusception. Our study made use of simple, cheap and easily assembled equipment for enema reduction of intussusception.
We found that there is no significant difference in the success rate between the two techniques which suggests that ultrasound-guided hydrostatic enema and ultrasoundguided pneumatic enema may be equally effective. This was similar to the findings of Huai et al. [20]. The success rate of hydrostatic reduction in this study was in keeping with success rates noted by other authors [8,9,21]. The success rate of hydrostatic reduction reported by Elebute et al. [22] was lower than that found in our study. This may be because routine hydrostatic reduction in their facility started during the period of the study. Ekenze et al. [7] and Ahmad et al. [12] noted higher success rates in their patients who had hydrostatic reduction. This variation may be because hydrostatic reduction was only done in patients who presented within 72 h in the study by Ekenze et al. [7] and 48 h in the study by Ahmad et al. [12]. Duration of symptoms was not a consideration in this study. Nonetheless, the success rate of pneumatic reduction in this study is in consonance with the findings of Joda et al. [23]. Lee et al. [13] noted a higher success rate in their patients who had pneumatic reduction compared to the success rate in this study. The success rate correlated with the pneumatic pressures used in their study.
There was a significant difference in the time to reduction between the two groups. The time to reduction in patients who had hydrostatic reduction agreed with the findings of Mensah et al. [24]. Ali et al. [25] reported a time to reduction of 12 to 35 minutes which is not consistent with the finding in this study. This may be because of the low hydrostatic pressures (60 to 100 mmHg) applied during the study. However, Beger et al. [26] recorded a mean reduction time of 12 minutes which is lower than the mean reduction time noted in this study and attributed the lower mean reduction time to the olive oil added to the normal saline which was used for hydrostatic reduction in their study. Moreover, Singh et al. [27] noted a time to reduction of 3 to 10 minutes in their patients who had pneumatic reduction which is at variance with the finding in this study. This may be explained by improved technique and experience with pneumatic reduction which is routinely done in their institution.
Time to reduction is remarkably shorter in pneumatic reduction when compared to hydrostatic reduction. Ali et al. [25] explained that reduction is easier with pneumatic enema as the friction between the intussusceptum and intussuscipiens decreases following air insufflation under pressure. Dung et al. [28] explained further that air has inherent compressibility that enables it to quickly sweep and dissect between the intussusceptum and intussuscipiens leading to rapid reduction. Moreover, Zambuto et al. [29] clarified that intracolonic pressures during hydrostatic reduction remain low compared to pneumatic reduction where they noted intermittent, high, sharp pressure peaks.
There was no perforation noted in patients who had pneumatic reduction in this study. This may be because the reduction pressure was closely monitored (not exceeding 120 mmHg) [13]. Moreover, Joda et al. [23] documented one perforation (at a pressure of 90 mmHg) in their patients who had pneumatic reduction. This affirmed that perforation can occur at reduction pressures below 120 mmHg.
Meanwhile, there was no significant difference in the recurrence rate between the two groups. Mensah et al. [24] reported a recurrence rate of 11.1%, in their patients who had hydrostatic reduction, which is higher than the recurrence rate in this study. This may be as a result of sample size differences. The time to recurrence in patients who had hydrostatic reduction in this study is in contrast to the findings of Elebute et al. [22] and Ahmad et al. [12] as their patients developed symptoms within 48 h of the procedure. This may be due to differences in the study population characteristics. The fatality rate of 3.8% noted among the patients who had hydrostatic reduction is in keeping with the findings of Ogundoyin et al. [8] whose patient also developed peritonitis prior to demise. The recurrence rate among patients who had pneumatic reduction is in consonance with the finding of Joda et al. [23]; their patient also presented within 48 h.

Limitations
This study is limited by the small sample size of the two groups which were compared. Also, the two methods of non-operative reduction were carried out by different members of the team which might have affected outcome although they were trained and expertise ensured before embarking on the study. However, to the best of the authors' knowledge, this is the first study in Nigeria to compare the outcomes of ultrasound-guided saline hydrostatic reduction and ultrasound-guided pneumatic reduction of intussusception in infants. The results obtained in this study could serve as a roadmap for other studies comparing the outcomes of ultrasound-guided saline hydrostatic reduction and ultrasound-guided pneumatic reduction of intussusception in infants. The choice of ultrasound guidance will help in resource-poor settings like ours where fluoroscopy may not be readily available. Delayed repeat enema can also be done without a fear of exposure to radiation.

Conclusion
This study has shown that the success rates of saline hydrostatic reduction and pneumatic reduction are comparable. It is faster to achieve reduction using pneumatic reduction compared with saline hydrostatic reduction. The complication rate of saline hydrostatic reduction and pneumatic reduction are similar with respect to the perforation and recurrence rates.
These findings should be considered in clinical decision making and parental counselling in infants, with uncomplicated intussusception, undergoing ultrasound-guided enema reduction. In clinical settings, we propose that ultrasoundguided hydrostatic or ultrasound-guided pneumatic enema may be used in the reduction of ileocolic intussusception in children as they are similarly effective and safe. Use may be based experience and choice of the surgeon.