Management and outcomes of paediatric ileocolic intussusception at a paediatric tertiary care hospital: A retrospective cohort study

Abstract

Background

Rapid reduction of ileocolic intussusception is important to minimize the compromise in blood flow to the affected bowel segment. This study aimed to quantify the potentially modifiable time between diagnosis and initiation of pneumatic reduction, identify factors associated with delays, and characterize the outcomes of pneumatic reduction in a recent cohort.

Methods

This retrospective observational study occurred at a tertiary care paediatric hospital with a consecutive sample of all children with ileocolic intussusception September 2015 through September 2018. The primary outcome was the time between ultrasound diagnosis of intussusception and the beginning of pneumatic reduction. Independent variables were age of the patient, time of day of arrival, transfer from another facility, and intravenous access prior to ultrasound. Outcomes of pneumatic reduction were expressed as proportions.

Results

There were 103 cases of ileocolic intussusception (among 257,282 visits) during the study period. The median time between diagnostic confirmation and initiation of reduction was 36 minutes. This was shorter for transferred patients and children with intravenous access prior to ultrasound. One perforation was identified at the beginning of reduction, without hemodynamic instability. Six children (5.8%) underwent either open (n=4) or laparoscopic surgery (n=2) for reduction failure.

Conclusion

The median delay between diagnosis and initiation of reduction at this paediatric hospital was short, especially among patients transferred with a suspicion of intussusception and children with intravenous access prior to diagnosis. Complications from pneumatic reduction were infrequent.

Intussusception occurs when a portion of the intestine is introduced into a more distal segment, potentially causing intestinal ischemia (1). It is the most common cause of intestinal obstruction in children under 2 years of age (2). During ileocolic intussusception, prolonged ischemia can lead to perforation and death, hence the importance of timely diagnosis and treatment (3). Diagnosis is most commonly achieved by ultrasound imaging (4). In the absence of clear indications for immediate surgical intervention (e.g., identified perforation and hemodynamic instability), intussusception is usually reduced with a pneumatic or hydrostatic enema performed under fluoroscopy (5,6).

Intestinal perforation is an uncommon (<2% [6,7]) but potentially life-threatening complication of fluoroscopic reduction. The surgical team must therefore be notified prior to a reduction attempt and ready to intervene if necessary. Moreover, the American College of Radiology and the Society for Pediatric Radiology recommend that patients should ideally have an intravenous line before a nonsurgical reduction (8). A survey performed among 12 paediatric emergency departments across Canada in May 2018 reported that 75% of the centres required an intravenous access and 42% required the presence of a surgeon in the room before reduction (personal data from an informal survey). Thus, variable delays can occur depending on local processes.

Time to reduction is typically measured from the onset of symptoms. Some factors contributing to delays in reduction are difficult to control, such as delays in presentation and in clinical recognition. However, once diagnosed, a streamlined process can prevent further delays in management, which have been associated with worse outcomes (9–11). In this context, our study aimed (1) to quantify the potentially modifiable time lapse between diagnosis to initiation of reduction, which provides a performance comparator for other sites, (2) to identify factors associated with longer delays, and (3) to describe outcomes from pneumatic reduction in a recent paediatric cohort.

METHODS

Design

This was a retrospective observational study in a consecutive sample of children visiting a single Emergency Department (ED) with a final diagnosis of ileocolic intussusception.

Setting

This study was conducted at CHU Sainte-Justine, a tertiary care paediatric hospital with an annual ED volume of approximately 80,000 visits. The ED is staffed by emergency physicians and paediatricians. The modality of choice for the diagnosis of intussusception is ultrasonography. Ultrasounds for potential intussusception are available 24/7, and are prioritized by the radiology department. Weekdays from 08:00 to 16:00, they are performed by a technician or resident supervised by a paediatric radiologist; at other times, they are performed by a resident or a paediatric radiologist. Patients are usually not accompanied by a nurse. A final diagnosis is made once images are read by a staff radiologist. Results are then immediately communicated verbally to the treating physician. A staff radiologist is always present during reduction, and the surgery team is notified but does not have to be in the room. Typically, radiologists are in-house from 07:30 to 17:30 on weekdays, and 07:00 to 17:00 on weekends; residents are in-house until midnight on weekdays, and 08:00 to midnight on weekends. When not in-house, on-call staff radiologists and residents are available within 30 minutes.

Local practices mandate intravenous access prior to pneumatic reduction. Each pneumatic reduction includes one to three attempts, with up to three levels of increasing air pressure, if needed (to a maximum of 120 mmHg). An unsuccessful attempt is defined as the absence of any movement of the intussusception for a continuous period of 3 minutes at a given pressure. Following successful reduction, most patients are discharged home after an ED observation period of 4 to 6 hours.

Participants

All children with a final diagnosis of ileocolic intussusception presenting to the ED September 23, 2015 through September 21, 2018 were included. Those with a diagnosis of isolated ileoileal intussusception were excluded as they do not require pneumatic reduction.

Outcomes

The primary outcome was the time between the moment of official radiological diagnosis of intussusception (completion of the ultrasound) and the beginning of the first reduction attempt. Other outcomes included the success probability for pneumatic reduction, as well as complications associated with this procedure.

Independent variables

A series of independent demographic and clinical variables were included as part of the data collection process, including baseline demographics, duration of symptoms, whether the patient was transferred from a referring site, and presence of intravenous access prior to the ultrasound.

Procedure

Data for patients with a final diagnosis of intussusception were retrieved from the ED information management system database. Electronic medical charts of participants were then manually reviewed. Timed images from the hospital imaging management system were collected as follows: the time of the last ultrasound image was recorded as the time of diagnosis, whereas the start of nonsurgical reduction was recorded as the first fluoroscopic image.

Data were collected by two of the authors using a standardized case report form created for this study. To ensure data collection validity, 15% of eligible charts (12,13) were blindly evaluated in duplicate. Discrepancies in chart review were resolved by consensus among the two raters.

Data analysis

Data were entered in an Excel database (Microsoft Inc., Richmond, WA) and analyzed with SPSS v25 software (SPSS Inc., Chicago, IL).

The inter-rater reliability was measured for the charts evaluated by two raters using the Kappa score or intraclass correlation coefficient (ICC) (14). A priori, it was decided that variables having a Kappa score or an ICC lower than 0.6 would not be used in the analysis.

The primary analysis was the median time between intussusception diagnosis and reduction. Other median times assessed were time to diagnosis (from ED arrival to diagnosis at ultrasound) and time between arrival and final reduction, with corresponding first and third quartiles. The association between time to reduction and intravenous use were measured by comparing the median times for children who already had an intravenous line at the moment of first ultrasound exam to patients who required an intravenous insertion following ultrasonographic diagnosis. The association between time to reduction and independent variables was assessed using the Mann–Whitney test given non-normal distributions.

Sample size

This was an observational study aiming to report the median time to reduction, and it did not test a hypothesis. However, we aimed to evaluate at least 100 cases.

Ethics

Approval was obtained from the Sainte-Justine University Hospital Research Ethics Board. Given the design of the study, patient consent was not required.

RESULTS

During the study period, 257,282 children were evaluated at the ED including 103 cases of ileocolic intussusception (one case/2,500 ED visits). Baseline demographics of the participants are described in Table 1. Most patients were less than 3 years old (median of 18 months old) and 67% were male; 37 patients (36%) were transferred from another hospital of which 34 had the correct diagnosis made at the referring site. The duration of symptoms prior to the ED visit varied widely, as demonstrated by the first and third quartiles of 7 and 48 hours.

Table 1.

Baseline demographics of the participants (n=103)

	N (%)
Median age in months (first and third quartile)	18 (10 and 29)
Sex male	69 (67)
Median duration of symptoms (first and third quartile)	12 h (7 and 48)
Transfer from another hospital	37 (36)
Diagnosis proven at another facility before transfer	34 (33)
Intravenous access present at arrival in the emergency department	35 (34)
Intravenous access present during ultrasound*	59 (57)

*Two patients did not have an ultrasound.

Upon duplicate review of 18 (15%) eligible charts were evaluated in duplicate, all items except “presence of a surgeon during enema” demonstrated an excellent reliability with Kappa scores or ICC greater than 0.65.

One child went directly to surgery because of hemodynamic instability at ED arrival, while 102 underwent pneumatic reduction. The median time (Q1, Q3) between diagnostic confirmation and initiation of pneumatic reduction was 36 (15, 60) minutes. This was shorter for patients transferred from another facility (24 minutes) than nontransferred patients (44 minutes). The median time from ED arrival to final diagnosis was of 124 (91, 159) minutes. Again, this was shorter for transferred patients (91 versus 160 minutes).

In a subanalysis of the 66 children seen initially at our hospital, having intravenous access prior to the ultrasound was associated with a shorter time to reduction (median time 39 versus 50 minutes). This was statistically significant using a Mann–Whitney test (P=0.005). Other factors evaluated as potential predictors of prolonged delays (age and time of arrival) failed to reach statistical significance (Table 2).

Table 2.

Predictors of prolonged delay for reduction

	Median delay in minutes first and third quartiles	P-value on Mann– Whitney test
Age group		0.407
0–12 months	27 (14 and 51)
13–24 months	44 (17 and 68)
>24 months	35 (14 and 61)
Time of arrival		0.884
0–7:59	30 (11 and 89)
8:00–16:59	36 (14 and 65)
17:00–24:00	35 (17 and 54)
Transfer from another facility		0.003
Yes	24 (11 and 48)
No	44 (20 and 69)
Intravenous access during ultrasound		0.005
Yes	39 (14 and 60)
No	50 (21 and 74)

Of the 102 patients who underwent a pneumatic reduction as the initial reduction procedure, the initial procedure was successful in 93 (92%) children (Table 3). Five children underwent a second series of pneumatic reduction after a variable period of time, which was successful in four of five children. The only complication during pneumatic reduction was a perforation identified 2 minutes following the onset of the procedure. That patient remained hemodynamically stable and was urgently transferred to the operating room, where an ileocecal perforation was successfully repaired. Intussusception recurred in three (3%) children during the 6-hour observation period following successful reduction.

Table 3.

Outcomes and procedure details among patients who received a pneumatic reduction (n=102)

	N (%)
Presence of surgeon during first reduction
Yes	48 (47)
No	12 (12)
Unknown	42 (41)
Number of reduction attempts*
1	73 (71)
2	17 (17)
3–4	11(10)
Pressure
80–89 mmHg	29 (28)
90–99 mmHg	29 (28)
100–119 mmHg	18 (18)
120 mmHg	10 (10)
Unknown	16 (16)
First pneumatic reduction successful	93 (90)
Delayed second pneumatic reduction (n=5)
Successful	4 (80)
Failure	1 (20)
Any complication
Desaturation	0
Perforation	1 (1)
Hypotension	0
Surgery (including one patient who went directly to surgery)
Open	4
Laparoscopic	2
Hospitalization	20 (19)
Recurrence during the 6 hours of observation	3 (3)

In total, six children (including the patient who went straight to surgery) underwent either open (n=4) or laparoscopic surgery (n=2), three of whom had an identified pathological lead point (Supplementary Appendix 1).

Discussion

This study identified a median time of 36 minutes between intussusception diagnosis and initial pneumatic reduction at a paediatric tertiary care hospital. This time was shorter among patients who already had an intravenous line during the ultrasound examination, and among transferred patients. Of note, the median time from ED arrival to diagnosis was approximately 2 hours, whereas the median time from arrival to reduction completion was slightly over 3 hours. Finally, pneumatic reduction was effective in most patients, with the only identified complication being a case of intestinal perforation.

To our knowledge, this is the first study to report the potentially modifiable delays between diagnosis and reduction for intussusception in children. Minimizing the time to reduction is paramount. A retrospective cohort study in two tertiary care centres found that patients who eventually required surgery had a higher median time to nonsurgical intervention. The overall median time between ED presentation and fluoroscopic reduction in this prior study was 7.3 hours. The authors reported a 5.2% increase in the probability of needing a surgical intervention for every additional hour of delay to reduction (15). Nonetheless, prior evidence suggests that nonsurgical reduction remains effective in the majority of affected children (16) and that short-term recurrence is infrequent (17), as observed in this cohort.

In the 102 children with attempted pneumatic reduction in the current study, the only complication was an intestinal perforation noted 2 minutes following the onset of the procedure. A retrospective study conducted in a tertiary care setting evaluated the risk of serious complications associated with reduction (pneumatic or hydrostatic). Among 433 children undergoing this procedure, 5 (1.2%) had a perforation during the procedure, 4 of whom had hemodynamic instability (7). Another meta-analysis reported a perforation rate of 0.39% (6). Of note, recent meta-analyses suggest better outcomes with pneumatic as compared to hydrostatic reduction (18,19). Intestinal perforation can be further complicated by tension pneumoperitoneum, which requires urgent intervention (initial needle decompression followed by surgery) to relieve hemodynamic and ventilatory compromise (20). Although our sample size may have been too small to capture this complication, a retrospective study reported a 4% rate of tension pneumoperitoneum during pneumatic reduction at another paediatric centre (21).

Our findings demonstrate that the intussusception management process described above is associated with relatively short times to reduction and overall favourable clinical outcomes. The slightly longer time to reduction in patients who did not have intravenous access during ultrasonography is unlikely to be of clinical significance in a setting with efficient interdepartmental patient transfers and a high expertise in paediatric vascular access. However, this expertise is not uniform. Although intravenous access is a common procedure in the paediatric ED, it can be difficult to perform in children. The first-time failure rate of intravenous cannulation in children varies between 31 and 45% according to prior reports (22–24). Moreover, venipuncture is a leading cause of procedural pain and distress in paediatric patients (25–27). In addition, complications related to the presence of an intravenous line constituted a leading cause of adverse events in a retrospective surveillance study of randomly selected American paediatric inpatient units (28). Indeed, the need for intravenous access therefore constitutes an important clinical decision.

In this context, a wider discussion regarding the universal requirement of intravenous access for hemodynamically stable patients prior to reduction seems warranted. This discussion must include paediatric emergency physicians, radiologists, surgeons, hospitalists, allied healthcare professionals, and families. Any potential recommendations must be individualized to local centre characteristics and patient demographics. Depending on local practices and the expertise of health professionals, it might be hypothesized that patients who look ill prior to the diagnosis of intussusception (e.g., severely dehydrated, peritoneal signs on exam) will already have an intravenous line inserted prior to ultrasonography. Future studies evaluating the utilization of intravenous analgesic or antiemetic medications among patients with ileocolic intussusception would further inform practice guidelines for patients with less concerning presentations but who nonetheless are found to require nonsurgical reduction.

Similarly, requiring the presence of a staff surgeon on site before the onset of pneumatic reduction may also cause delays to the procedure, depending on time of presentation and in-house coverage. Given the overall low complication rate associated with pneumatic reduction and relative simplicity of percutaneous needle decompression, notifying the surgical team that a patient will undergo pneumatic reduction without requiring a physical presence on site may be a reasonable alternative. A more precise identification of clinical and radiological factors (e.g., free fluid in abdomen, fluid in intussusceptions, absence of Doppler flow, and pathological lead point) associated with an increased risk of nonsurgical reduction failure or complication will allow the incorporation of risk stratification into clinical practice guidelines.

The primary study limitation is the retrospective design. Being a single-centre study, the generalizability of our results is limited by local system factors that may differ from other centres. Moreover, one third of patients were transferred to the ED with a suspected diagnosis of intussusception. Though a subgroup analysis was performed on patients who were not transferred, the significant number of transferred patients in this cohort must be considered when interpreting the study findings.

CONCLUSION

The median delay from diagnosis to initiation of reduction at this paediatric hospital was short, especially among patients transferred with a suspicion of intussusception and children with intravenous access. Complications from pneumatic reduction were infrequent. Pneumatic reduction was successful on first attempts in a majority of patients, and complications were infrequent. Future studies should focus on methods to identify children at risk of complications from pneumatic reduction of intussusception.

Funding: There are no funders to report for this submission.

Potential Conflicts of Interest: All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.