Rectal diameter measured by transabdominal ultrasonography for monitoring treatment response in children with functional constipation

Abstract

BACKGROUND

Transabdominal ultrasonography (USG) is a noninvasive tool for detecting fecal impaction in children with functional constipation (FC) and may predict treatment response. However, supporting evidence remains limited and controversial.

AIM

To evaluate changes in rectal diameter (RD) and anterior rectal wall thickness (AWT) measured by USG in children with FC after treatment with polyethylene glycol (PEG) for 8 weeks.

METHODS

This substudy of a randomized controlled trial included children aged 6 months to 18 years diagnosed with FC according to the Rome IV criteria at King Chulalongkorn Memorial Hospital (May 2024-February 2025). Participants received either commercial PEG or PEG-Chula as laxatives for FC. USG was performed at baseline and at 1, 4, and 8 weeks by the primary investigator. RD and AWT were measured following Klijn’s technique.

RESULTS

Of the 52 enrolled children, 44 completed all four visits. The median age was 50.5 [interquartile range (IQR): 28-92.5] months, and 65.9% were female. The median FC duration was 19.5 (IQR: 12-36) months. Stool frequency and consistency improved significantly over 8 weeks, with no difference between the commercial PEG and PEG-Chula groups. The mean RD decreased significantly from 2.70 cm (95%CI: 2.51-2.88) at baseline to 2.46 cm (95%CI: 2.30-2.62) at week 1, 2.45 cm (95%CI: 2.30–2.60) at week 4, and 2.33 cm (95%CI: 2.18-2.48) at week 8. AWT remained 0.29 cm (95%CI: 0.27–0.32) at baseline and 0.29 cm (95%CI: 0.26–0.30) at week 8 (P = 0.997). The time since the last defecation to USG significantly influenced RD (P = 0.030) but not AWT (P = 0.382).

CONCLUSION

USG-measured RD decreases significantly with effective PEG treatment and may serve as a useful noninvasive marker for monitoring treatment response in children with FC.

Key Words: Constipation; Rectal diameter; Ultrasonography; Children; Anterior wall thickness

Core Tip: Although transabdominal ultrasonography (USG) has been increasingly employed to assess fecal impaction in children with functional constipation (FC), evidence supporting its role in monitoring treatment responses remains limited. In this prospective substudy of a randomized controlled trial evaluating the efficacy of commercial polyethylene glycol (PEG) and PEG-Chula, USG-measured rectal diameter (RD) exhibited a significant decrease after 8 weeks of treatment, paralleling clinical improvement. In contrast, anterior rectal wall thickness did not change significantly during treatment. These findings indicate that USG-measured RD may be a practical, noninvasive parameter for monitoring treatment response in children with FC.

INTRODUCTION

Functional constipation (FC) is a common disorder of gut–brain interaction in children, affecting approximately 14% of children worldwide[1] and 12% in Asia[2]. The Rome IV criteria defined FC by the presence of symptoms such as infrequent defecation, stool retention behaviors, hard stools, painful defecation, and the passage of large-diameter stools[3]. Withholding behaviors are considered the main cause of childhood FC[1,4]. This retention leads to fecal impaction in the rectum and ultimately megarectum[1,4].

Digital rectal examination (DRE), often in combination with a general physical examination, is performed to detect fecal masses[4,5]. However, up to 85% of physicians refrained from performing DRE because of its invasiveness and uncomfortable nature[6], which may contribute to fear of defecation and unsuccessful toilet training. Other modalities, such as a palpable fecal mass via abdominal examination or abdominal X-ray imaging, have limitations, and findings are often misinterpreted. Thus, a patient-friendly, noninvasive test that can accurately assess fecal impaction and predict treatment success is beneficial.

In recent years, transabdominal ultrasonography (USG) has emerged as a novel, valuable, and noninvasive tool for detecting fecal impaction and monitoring treatment response in children with FC[7-10]. Thus, it has been proposed as a replacement for DRE in detecting fecal mass, as it strongly correlates with the presence of large stools[11-13]. The rectal diameter (RD) and anterior wall thickness (AWT) are parameters proposed for use to predict treatment response; however, results remain limited and controversial[8,9,14-17].

Thus, this study aimed to evaluate changes in RD and AWT, measured by transabdominal USG, in children with FC after an 8-week treatment with a polyethylene glycol (PEG)-based regimen.

MATERIALS AND METHODS

Study design and population

This substudy of a double-blinded, randomized controlled trial (RCT) was approved by the Institutional Review Board of Chulalongkorn University (No. 0093/67) and registered at ClinicalTrials.gov (NCT06357897). This study was conducted at the Clinical Research Center, Faculty of Medicine, Chulalongkorn University, from May 2024 to February 2025.

Children aged 6 months to 18 years were enrolled in this study if they met the Rome IV criteria for FC[18,19]. Conversely, children were excluded if they were diagnosed with organic causes of constipation (e.g., Hirschsprung’s disease, spinal cord anomalies, hypothyroidism, anal stenosis), suspected gastrointestinal obstruction, took medicines affecting bowel movement, had acute cardiac, liver, or pulmonary disorder, or had a history of PEG allergy.

Eligible children were randomly assigned to receive either the newly developed PEG-based formula (PEG-Chula) or a commercial PEG formula (Forlax^®) for 8 weeks. Both formulations contain PEG 4000 as the main active ingredient and differ in their flavoring agents and sweeteners. PEG 4000 doses were given from 0.5 to 1 g/kg/day. Children and guardians also received educational books and explanations on diet, physical activity, and toilet training (Supplementary material appendix 1).

Study procedure and outcome measurements

As described in the RCT study[20], the present study involved four visits. During the first visit, physicians collected medical history from caregivers. The Rome IV diagnostic criteria and FC-related symptoms were also recorded. Painful defecation was assessed using the facial pain scale (score 0-5). Stool consistency was evaluated using the Bristol stool scale (score 1-7)[21]. Then, children underwent physical examinations and USG and classified into group A (baseline RD < 3 cm) and group B (baseline RD ≥ 3 cm) based on their RD.

Caregivers were given diary books to record their children’s stool habits, dietary and water intake, adverse events, and compliance. Children’s diets were assessed based on 3-day food records, which included 2 weekdays and 1 weekend day at weeks 1, 4 and 8[22]. Then, dietary data were converted into nutrient intake using INMUCAL-Nutrients software (version 4.0), a Thai food composition database developed by the Institute of Nutrition, Mahidol University (Bangkok, Thailand)[23]. Adequate fiber intake was defined as consumption exceeding age plus 5 g/day, in accordance with the recommendations of the American Academy of Pediatrics[24]. Water intake was recorded daily during the 8 weeks and included all water types, such as fruit juice and beverages. Adequate water intake was defined according to the guidelines of the National Institute for Health and Care Excellence for children with constipation[25].

After treatment with PEG 4000 for 1, 4, and 8 weeks, stool output and related symptoms and signs, adverse events, and compliance were assessed. Children also underwent repeated USG at each visit to evaluate changes in RD and AWT. The time of the last defecation was also recorded before undergoing USG.

Transabdominal USG technique

USG was performed on all children at all four visits by the same investigator (Tran DL) under supervision (Sintusek P). The procedure was modified and followed the method described by Klijn et al[17], using a Samsung HS40 Ultrasound machine (ParagonCare Thailand Co., Ltd). In this approach, USG was performed with the child in the supine position, and the curved 2-8 MHz probe was placed approximately 2-4 cm above the pubic symphysis. Measurements were taken when the bladder was partially filled, and the transducer was angled about 15°-30° caudally from the transverse axis. The RD was measured in the transverse plane, and the AWT was measured from the outer wall to the inner wall (Figure 1). If necessary, the children were offered water, and measurements were repeated. The time of the last defecation recorded in the case record form for each participant. The procedure took approximately 10-15 minutes.

Figure 1  Measurement of rectal diameter and anterior wall thickness by transabdominal ultrasonography.

Statistical analysis

Continuous data are presented as mean ± SD or median [interquartile range (IQR)]. Categorical data are presented as n (%). Comparisons between continuous variables were conducted using Student’s t-test or Wilcoxon rank-sum test, whereas categorical variables were analyzed with the χ² test or Fisher’s exact test. Spearman’s rank correlation coefficient was used to assess the correlation between RD, AWT, and other variables. A mixed-effects model for repeated measures was employed to evaluate RD and AWT changes over time. Visits were included as a fixed effect. Clinically relevant covariates, such as age, sex, weight, height, FC duration, time since the last defecation before USG, and fiber and water intake, were evaluated for inclusion in the model. Covariates with a P < 0.25 were initially included in the analysis; in the subsequent step, only covariates with a P value < 0.05 were included in the final model. Subject-specific random effects were included, and an unstructured covariance matrix was used to account for within-subject correlations across repeated measurements. The results were illustrated in graphs to show the trends in outcomes. STATA version 18 was used for statistical analysis.

RESULTS

Of the 52 children enrolled in this study, 44 completed all four USG evaluations (Supplementary material appendix 2). The median age was 50.50 (28.00-92.50) months, and 65.91% were female. The median FC duration was 19.50 (12-36) months.

At baseline, the commercial PEG and PEG-Chula groups had mean stool frequencies of 2.02 ± 0.54 and 2.22 ± 0.78 times per week and mean stool consistency of 2.24 ± 0.83 and 2.15 ± 0.83, respectively. Comparing with baseline, stool frequency and consistency improved significantly over 8 weeks of treatment (P < 0.001), with no difference found between commercial PEG and PEG-Chula groups (P > 0.05).

The percentage of children consuming adequate fiber changed from 11.36% (5/44) at week 1 to 9.09% (4/44) at week 4 and to 22.73% (10/44) at week 8. Similarly, 34.09% (15/44) of children had adequate water intake at weeks 1 and 4, rising to 45.45% (20/44) at week 8.

Abdominal pain was reported significantly more often in group A (baseline RD < 3 cm) than in group B (baseline RD ≥ 3 cm) (P = 0.035). No significant differences were observed in demographic characteristics and other related symptoms between the two groups (Table 1).

Table 1 Demographic data and characteristics of children with functional constipation (n = 44), n (%)/median ± SD/median (interquartile range).

Variables	Total (n = 44)	Group A (RD < 3 cm) (n = 29)	Group B (RD ≥ 3 cm) (n = 15)	P value
Age (months)	50.50 (28.00-92.50)	47 (24.00-93.00)	54 (35.00-60.00)	0.699
Age 6 months to 4 years	21 (47.73)	15 (51.72)	6 (40.00)	0.460
Sex (female)	29 (65.91)	19 (65.52)	10 (66.67)	0.939
Weight (kg)	15.35 (11.60-25.90)	14.60 (11.50-26.30)	16 (12.20-21.40)	0.736
Height (cm)	99.50 (88.50-125.25)	97 (88.00-125.50)	102 (90.00-110.00)	0.755
FC in family	26 (59.09)	18 (62.07)	8 (53.33)	0.576
Previous treatment for FC	20 (45.45)	11 (37.93)	9 (60.00)	0.163
Duration of FC (months)	19.50 (12.00-36.00)	12.00 (12.00-30.00)	24.00 (12.00-40.00)	0.356
Stool frequency at baseline (times per week)	2.13 ± 0.67	2.21 ± 0.73	1.97 ± 0.55	0.268
Stool consistency at baseline (times per week)	2.19 ± 0.82	2.19 ± 0.79	2.20 ± 0.90	0.969
Painful defecation (facial pain score 0-5)	3.61 ± 1.08	3.45 ± 1.07	3.93 ± 1.05	0.159
Bloating or flatulence	11 (25.00)	6 (20.69)	5 (33.33)	0.843
Abdominal pain	12 (27.27)	11 (37.93)	1 (6.67)	0.035
Anal fissure	18 (40.91)	12 (41.38)	6 (40.00)	0.930
Last time of defecation before USG (hour)
Baseline (n = 34)	25.50 (19.00-50.00)	24.00 (19.00-48.00)	30.00 (22.00-54.00)	0.413
Week 1 (n = 43)	20.00 (5.00-30.00)	20.00 (6.00-23.50)	18.50 (4.00-46.00)	0.728
Week 4	18.50 (4.50-23.50)	17.00 (5.00-24.00)	21.00 (4.00-24.00)	0.293
Week 8	19.00 (5.50-23.50)	19.00 (5.00-21.00)	19.00 (6.00-24.00)	0.717
Adequate fiber intake at week 1	5 (11.36)	3 (10.34)	2 (13.33)	0.767
Adequate water intake at week 1	15 (34.09)	9 (31.03)	6 (40.00)	0.552

FC: Functional constipation; USG: Transabdominal ultrasonography; RD: Rectal diameter.

For the Rome IV criteria, caregivers or children with FC (88.64%) commonly reported painful defecations or hard bowel movement, followed by stool retention behaviors (77.27%), large-diameter stool (68.18%), and defecation < 2 times per week (65.91%) (Figure 2).

Figure 2  Rome IV diagnostic criteria in the past 1 month.

Correlation of RD and AWT to baseline characteristics

Spearman correlation analysis revealed that the FC duration was significantly associated with baseline RD, with a weak positive correlation (ρ = 0.35, P = 0.021). In contrast, FC duration was not significantly correlated with AWT (ρ = 0.25, P = 0.107). No other demographic or clinical variables, such as age, weight, height, stool frequency, or stool consistency, showed a significant correlation with either RD or AWT (Table 2).

Table 2 Correlations of rectal diameter and anterior wall thickness to baseline variables (n = 44).

Variables	Baseline RD (cm)		Baseline AWT (cm)
	Spearman’s rho	P value	Spearman’s rho	P value
Age (months)	0.22	0.148	0.15	0.337
Weight (cm)	0.19	0.212	0.17	0.263
Height (cm)	0.23	0.139	0.18	0.25
Duration of FC (months)	0.35	0.021	0.25	0.107
Stool frequency at baseline (times per week)	-0.22	0.149	-0.07	0.672
Stool consistency at baseline (BSS 1-7)	-0.09	0.569	-0.01	0.925
Time last defecation to USG (hour)	0.09	0.616	0.003	0.987

AWT: Anterior wall thickness; FC: Functional constipation; RD: Rectal diameter; USG: Transabdominal ultrasonography.

Changes in RD and AWT after 8 weeks of PEG treatment

USG-measured RD gradually decreased over time following the intervention, with mean values of 2.70 cm (95%CI: 2.52-2.89) at baseline, 2.46 cm (95%CI: 2.30-2.61) at week 1, 2.45 cm (95%CI: 2.30-2.60) at week 4, and 2.33 cm (95%CI: 2.18-2.48) at week 8. The reduction in RD was significant at week 8 (P = 0.001) compared with baseline (Figure 3A and Table 3).

Figure 3 Mean change from baseline in rectal diameter and anterior wall thickness after 8 weeks of treatment. A: Rectal diameter; B: Rectal anterior wall thickness.

Table 3 Mean of rectal diameter and anterior wall thickness after 8 weeks of treatment.

Variables	Mean (95%CI)	Mean change from baseline (95%CI)	P value
Rectal diameter (cm)
Baseline	2.70 (2.51-2.88)
Week 1	2.46 (2.30-2.62)	-0.24 (-0.45 to -0.03)	0.024
Week 4	2.45 (2.30-2.60)	-0.24 (-0.46 to -0.03)	0.025
Week 8	2.33 (2.18-2.48)	-0.37 (-0.59 to -0.14)	0.001
Anterior wall thickness (cm)
Baseline	0.29 (0.27-0.32)
Week 1	0.29 (0.27-0.31)	0.005 (-0.02-0.03)	0.740
Week 4	0.29 (0.27-0.30)	0.003 (-0.03-0.03)	0.865
Week 8	0.29 (0.26-0.30)	-0.004 (-0.03-0.02)	0.997

The mean AWT was 0.29 cm (95%CI: 0.27-0.32) at baseline and remained stable over time, reaching 0.29 cm (95%CI: 0.26-0.30) at week 8. However, this change was not significant (P = 0.997) (Figure 3B and Table 3).

The time of the last defecation before USG significantly affected the change in RD (P = 0.03) but did not significantly affected the change in AWT (P = 0.382). Age, sex, weight, height, and FC duration did not significantly affected the change in both RD and AWT (Supplementary material appendix 3).

Group B (RD ≥ 3 cm) exhibited a significant reduction in mean RD over time, decreasing from 3.33 cm (95%CI: 3.10-3.57) at baseline to 2.63 cm (95%CI: 2.39-2.87) at week 8 (P < 0.001). In contrast, RD reduction in group A (RD < 3 cm) was not significant, decreasing from 2.35 cm (95%CI: 2.17-2.54) at baseline to 2.17 cm (95%CI: 2.00-2.34) at week 8 (P = 0.177) (Table 4).

Table 4 Mean rectal diameter at each visit by baseline rectal diameter group.

Variables	Mean (95%CI)	Mean change from baseline (95%CI)	P value
Group A (RD < 3 cm) (n = 29)
Baseline	2.35 (2.17-2.54)
Week 1	2.32 (2.16-2.47)	-0.04 (-0.28-0.21)	0.774
Week 4	2.30 (2.15-2.46)	-0.05 (-0.30-0.19)	0.686
Week 8	2.17 (2.00-2.34)	-0.18 (-0.45-0.08)	0.177
Group B (RD ≥ 3 cm) (n = 15)
Baseline	3.33 (3.10-3.57)
Week 1	2.75 (2.53-2.96)	-0.59 (-0.90 to -0.28)	< 0.001
Week 4	2.73 (2.52-2.95)	-0.60 (-0.93 to -0.28)	< 0.001
Week 8	2.63 (2.39-2.87)	-0.70 (-1.05 to -0.35)	< 0.001

RD: Rectal diameter.

DISCUSSION

This study evaluated changes in RD and AWT using USG in children with FC after 8 weeks of treatment with PEG-based regimen. Along with significant improvements in stool frequency and stool consistency, a significant reduction in RD was observed in the group with RD ≥ 3 cm, whereas no significant change in AWT was noted from baseline.

Based on the Rome IV criteria, painful defecation or the passage of hard stools was the most commonly reported symptom, followed by withholding behaviors and the passage of large-diameter stools. This pattern is in line with findings from a national survey on FC in Thailand[26]. In contrast, infrequent defecation was observed in 65.91% of children, which is notably quite high. The observed differences may be attributable to variations in study populations, as this study included children with FC duration ranging from 6 months to 18 years, whereas the national survey in Thailand[26] focused on toddlers and young children (6 months to 3 years). In addition, this study revealed a correlation between enlarged RD and longer FC duration that aligns with the results of previous studies[10]. This finding might imply that the children with FC in the present study had more chronic disease, with enlarged RD potentially accounting for the infrequent defecation. A cutoff of ≥ 3 cm of RD has been proposed as an adjunct diagnostic criterion for pediatric FC[27,28]. Using this cutoff for RD, abdominal pain symptoms were reported significantly less frequently in the group with RD ≥ 3 cm (37.9% vs 6.7%). In the early stages of constipation, fecal retention leads to stool accumulation in the rectum, which increases intrarectal pressure and bowel movements, resulting in abdominal pain[4,5,29]. Children with FC tend to have poor appetite because of gastrocolic reflex after a meal, leading to abdominal pain. However, in chronic FC, abdominal pain might not be predominant as an enlarged rectum, which was confirmed in this study.

In contrast to results of previous studies, the present study did not find significant correlations between RD or AWT and other patient characteristics, such as age, weight, height, stool frequency, stool consistency, or last defecation to USG[7,15,30,31]. FC duration was significantly correlated with RD at baseline. However, determining the factors that influence RD and AWT in children with FC remains challenging, as current evidence is inconsistent and sometimes contradictory. Differences in methodology, USG machine resolution, operator experience, and patient preparation (such as defecation before USG, full bladder, and cooperation) may affect study results. Establishing the USG protocol and a multicenter setting to validate prior evidence can improve diagnostic accuracy and enhance the clinical utility of USG.

A decreasing tendency in RD was observed during the 8-week PEG 4000 therapy. The time of the last defecation before USG significantly affected the change in RD. This finding is consistent with the results of previous studies[8,9,32]. After 4 weeks of treatment with PEG, Joensson et al[8] reported a reduction in mean RD from baseline to the end of the study (3.96 ± 8.2 cm vs 2.69 ± 5.6 cm). RD was significantly reduced in children who responded to treatment (P < 0.001), whereas no significant change was observed in nonresponders (P = 0.700). Similarly, Karaman et al[7] found that post-treatment RD was smaller pre-treatment, although the reduction was not significant. Notably, RD in children with FC before and after treatment remained greater than that in healthy children. Conversely, Gulcin et al[16] reported an increase in RD following treatment from 1.80 ± 1.55 cm at baseline to 2.35 ± 1.42 cm after 12 weeks of treatment (P = 0.008), likely because of rectal evacuation before USG in children with FC. This finding proved that the time of the last defecation to USG was significantly related to the change in RD. Other factors, including age, sex, weight, height, and duration of FC, were included in the analytical model and showed no significant effect on the study results. It was acknowledged that colonic gas also affected measurements in some cases[28,33,34], although only images with clear visualization of the rectum were recorded. Further studies should be conducted under more standardized conditions, and operator experience remains essential to ensure measurement reliability. Again, it emphasizes the need for carefully interpreting isolated RD values and highlights the importance of standardizing USG timing in both clinical practice and research settings.

In the subgroup analysis, a significant reduction in RD was observed only in participants with RD ≥ 3 cm. In the study by Di Pace et al[32], RD significantly reduced in both groups after 3 months of treatment. Among children with megarectum, those who received enemas during the first week showed a significantly smaller RD reduction at each visit than those treated without enemas during the initial week. This finding suggests that management strategies may need to differ between children with and without megarectum.

An expert panel has noted that RD is associated with fecal loading in both children and adults[28]. In addition, USG has been recommended as a useful tool for assessing the pathophysiology of constipation in adults[35]. However, important differences should be considered. In adults and the elderly, RD may be influenced by age-related changes in pelvic floor function, colonic motility, and underlying conditions, which differ from pediatric FC[36].

While a RD cutoff of 3 cm has been established for predicting fecal loading in children, no corresponding cutoff has been defined for adults because of limited available evidence[28]. Further prospective studies are needed to characterize rectal luminal contents and to support the broader application of USG in the management of constipation.

AWT is another potential parameter measured by USG. No significant change in AWT was noted during 8 weeks of treatment. Similarly, Karaman et al[7] also reported no significant difference in AWT before and after treatment (0.28 ± 0.07 cm pretreatment vs 0.29 ± 0.09 cm posttreatment). Treatment duration of 8 weeks may not be sufficient to cause detectable changes in AWT. Thus, further studies with longer follow-up periods are warranted to explore this potential relationship.

This study has several limitations. First, USG was performed by an unblinded but well-trained pediatrician; therefore, measurement bias and limited validation cannot be excluded. Second, the small sample size and short follow-up duration may limit the generalizability of the findings. Future studies involving larger cohorts, standardized and consensus-based imaging protocols, and longer follow-up periods are warranted to further validate USG as a reliable assessment tool when performed by blinded and experienced operators in this clinical context.

CONCLUSION

This study demonstrates RD reduction measured by USG following effective PEG treatment for 8 weeks. RD may serve as a useful noninvasive marker for monitoring treatment response in children with FC. This finding supports the role of USG as an adjunctive tool in the clinical follow-up, particularly in children with an enlarged rectum. Further studies on standardized assessment protocols for RD using USG are needed to confirm its clinical applicability.

ACKNOWLEDGEMENTS

We would like to express our gratitude to the Maha Chakri Sirindhorn Clinical Research Center under the Royal Patronage, Department of Pediatrics, Clinical Sciences program, Faculty of Medicine, Chulalongkorn University, for their invaluable support. We are also deeply thankful to Dr. Teerasak Phewplung from the Department of Radiology for expert consultation, and Paragon Care Thailand Co., Ltd. for generously providing the ultrasound machine used in this study. In addition, we gratefully acknowledge the contributions of our research team-Mr. Santirat Sopee, Ms. Arisa Ama, and Ms. Pakpine Phunnoi for their dedication and assistance in data collection. We would like to thank Miss Alisara Pittiyayon at Electricity Generating Authority of Thailand for the image Figure 1. Finally, we are grateful to all the participating children and their caregivers for their involvement in this study.