Stool Consistency, but Not Frequency, Correlates with Total Gastrointestinal Transit Time in Children

Marina Russo, MD1, Massimo Martinelli, MD1, Elisa Sciorio, MD1, Carmine Botta2, Erasmo Miele, MD1, Gianfranco Vallone, MD2, and Annamaria Staiano, MD1

Objectives To evaluate the correlation between stool characteristics (consistency and frequency) and gut transit time in children and to determine whether the Bristol Stool Form Scale is a reliable method of assessing intestinal transit rate in children. Study design From March 2011 to March 2012, 44 children (25 boys and 19 girls, mean age 7.8 years) with a diagnosis of functional constipation and 36 healthy, nonconstipated children (17 boys and 19 girls, mean age 7.6 years) were enrolled. All participants maintained a 1-week stool diary, recording the time and date of every bowel movement and stool form, and then completed a validated questionnaire on functional constipation according to Rome III criteria. Whole gut transit time (WGTT) was then assessed using the radiopaque markers test. Results There was a significant correlation between stool form and WGTT in both constipated and nonconstipated children (correlation coefficient À0.84, P < .001). By contrast, there was no correlation between either stool frequency and WGTT or stool frequency and stool form. Multivariate logistic regression analysis, using WGTT as a dependent variable, showed that the sole variable significantly associated with WGTT was stool form (regression coefficient 2.9, OR 18.4, 95% CI 5.4-62.5, P < .001). Conclusion In this prospective, observational, case-control study, we show that stool form, as measured by the Bristol Stool Form Scale, rather than stool frequency, correlates with WGTT in both constipated and nonconstipated children. (J Pediatr 2013;162:1188-92).

unctional constipation is a common disorder in childhood, with a prevalence ranging from 7% to 30% in both West- Fern and non-Western countries.1 No organic cause is found in >90% of affected children.2 Delayed colonic transit is thought to be one of the main mechanisms underlying this disorder. Although most patients with functional constipation generally do not require diagnostic tests, the use of radiopaque markers (ROMs) is a validated method that can be used for measuring total gastrointestinal transit time.3 However, this technique, which relies on stool collection and/or radiation exposure, is rarely used to determine colonic transit time either in clinical practice or in epidemiologic studies but is reserved for use in evaluating patients with refractory constipation, defined by the lack of response to conventional constipation treatments. Stool form scales represent a useful tool for measuring colonic transit time in adults.4 The use of standardized and validated scales such as the Bristol Stool Form Scale (BSFS) has the advantage of avoiding exposure to radiation, stool handling, and discomfort apart from the mild unpleasantness of having to inspect feces to be able to decide which descriptor fits best. The BSFS has been used to evaluate stool form in a variety of clinical studies. The Rome Foun- dation recommends its use to assess stool form in adults with functional gastrointestinal disorders.5 Despite the im- portance of accurately assessing stool form changes, a stool form scale has not been validated for this use in pediatric patients. Lane et al6 proposed the use of the Modified BSFS for Children from 3 to 18 years of age, and Bekkali et al7 proposed an infant stool form scale and tested its usefulness by assessing defecation patterns in both premature and term infants. There is no reported correlation between stool consistency and whole gut transit time (WGTT) in children. Nevertheless, both stool form and stool frequency continue to be used as surrogate markers of bowel transit, both in research settings and in daily clinical practice.8,9 The aims of this study were, therefore, to evaluate the correlation between stool characteristics (consistency and frequency) and WGTT in children and to determine whether the BSFS is a reliable method of assessing intestinal transit rate in children.

From the 1Department of Pediatrics, University of Naples BM Bowel movement “Federico II”, Naples, Italy; and 2Department of BSFS Bristol Stool Form Scale Biomorphological Science and Functional Sciences, University of Naples “Federico II”, Naples, Italy BSSS Bristol Stool Scale Score The authors declare no conﬂicts of interest. ROMs Radiopaque markers WGTT Whole gut transit time 0022-3476/$ - see front matter. Copyright ª 2013 Mosby Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2012.11.082

1188

Methods sessed in each child using the ROMs test. All children in- gested 20 ROMs at the same time.11 For the 3 days of transit testing, study children were asked to record the date A prospective, observational, case-control study was con- and time of every movement in a written diary and to ducted in 80 consecutive children (42 boys and 38 girls, mark corresponding stool form on the BSFS. They were mean Æ SD age 7.7 Æ 3.3 years, age range 4-15.4 years) re- also instructed to record any additional gastrointestinal ferred to the Department of Pediatrics of the University of symptom, such as abdominal pain, vomiting, and nausea, Naples “Federico II” from March 2011 to March 2012. as well as timing of meals, physical activity, and sleep. To de- The case group was composed of 44 consecutive patients termine WGTT, fecal radiographs were obtained of collected affected by functional constipation (25 boys and 19 girls, stools. No dietary restriction was required during the transit mean Æ SD age 7.8 Æ 3.3 years, age range 4-15.4 years), de- 10 testing, but children were asked to avoid strenuous and pro- fined using Rome III criteria. The control group consisted longed aerobic activities. of 36 healthy children with regular bowel movements (BMs) Informed consent for participation in this study was ob- (17 boys and 19 girls, mean Æ SD age 7.6 Æ 3.1 years, age tained from parents of all patients and the experimental de- range 4-14 years). All enrolled subjects and/or their parents, sign was approved by the Independent Ethics Committee depending on age, completed a validated questionnaire re- of the University of Naples, “Federico II.” garding functional constipation, according to Rome III criteria.10 Subjects were classified as having functional Statistical Analyses constipation on the basis of their responses to the question- All data were stored in a common database and statistically naire. The clinical characteristics of patients with functional analyzed using the SPSS version 8.0 (SPSS Inc, Chicago, constipation are reported in Table I. In cases of intractable Illinois). To determine stool form, data reported using the constipation resistant to conventional therapy, the Bristol Stool Scale Score (BSSS) were obtained from the following examinations were performed to exclude an 1-week diary recorded before the WGTT test. A numeric organic cause of constipation: blood cell count, celiac value ranging from 1 to 7 was used to express the average disease serology, thyroid hormone levels and antibodies, BSSS. Stool frequency was defined as the number of BMs in serum electrolytes, serum calcium, and allergy testing. 1 week and was obtained from the BMs diary. Data for the Subjects with symptoms or findings suggestive of organic whole gut transit with ROMs test were collected for each disease (eg, abnormal laboratory findings, persistent vomiting, gastrointestinal bleeding, and constitutional symptoms such as fever or weight loss), illnesses that may cause gastrointestinal symptoms, or a history of major abdominal surgery or development disability, were excluded. All participants maintained a 1-week stool diary, recording the time and the date of every BM and the stool form, using the BSFS. The BSFS is a 7-point scale describing and classify- ing stool form using different images and descriptions (Figure 1). For example, score 1 describes stools that are hard lumps, like nuts (hard to pass), whereas score 4 describes stools like a sausage or snake, smooth and soft in consistency. The BSFS is approved for use in epidemiologic surveys and in research to discriminate between patients with functional defecation disorders such as irritable bowel syndrome, diarrhea, and constipation.8 During the study period, the use of fecal softeners was not allowed. At the end of this 1-week period, WGTT was as-

Table I. Study population baseline characteristics P Characteristics Patients Control subjects value Sex .5 Boys, n 25 (56.8%) 17 (62%) Girls, n 19 (47.2%) 19 (53.8%) Age, mean Æ SD y 7.8 Æ 3.3 (4-15.4) 7.6 Æ 3.1 (4-14) .7 (range) BM (mean Æ SD n/wk) 4.8 Æ 1.1 5.7 Æ 1.1 .001 Æ Æ Æ WGTT (mean SD h 63.4 12.7 (24-96) 30.1 10.1 (20-61) .001 13 (range) Figure 1. BSFS (developed by Lewis and Heaton ). Repro- BSSS (mean Æ SD) 3.1 Æ 0.4 4.2 Æ 0.5 .001 duced with permission.

1189

Downloaded from ClinicalKey.com at The Children's Mercy Hospital - Kansas City September 07, 2016. For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved. THE JOURNAL OF PEDIATRICS www.jpeds.com Vol. 162, No. 6 enrolled child. Delayed WGTT was defined as retention of stool form (regression coefficient 2.9, OR 18.4, 95% CI >2% of the ROMs at 33 hours after ingestion.11 5.4-62.5, P < .001). Stool frequency, age, and sex did not cor- Variables were screened for their distribution, and appro- relate with the WGTT (Table II). priate parametric or nonparametric tests were adopted as required. Cross-tabulations were evaluated by using the Discussion Spearman test. Statistical significance was predetermined as P < .05. The Spearman correlation along with 95% CIs was Stool form, rather than stool frequency, correlates with used to establish the relationship between stool form and WGTT in children. The BSFS should be considered as a valu- stool frequency with WGTT. Multivariate conditional logis- able tool for evaluating stool form in children. Bekkali et al7 tic regression analysis was used to determine which factors tried to develop a scale useful to describe and differentiate were independently associated with WGTT. The dependent physiologic and pathologic stool appearances in infants, variable was normal or delayed WGTT, and the effects of but they were not able to correlate this stool form scale all the above-mentioned variables were analyzed by using with colonic transit time. Lane et al6 proposed that a modified a stepwise procedure. BSFS, decreasing the number of stool categories from 7 to 5, should be used in pediatrics. However, the modified scale has Results yet to be validated for use in children <6 years of age, and no other study confirmed its test reliability.6 All enrolled children completed the study. Study groups were One previous study in children correlated colonic transit 12 well matched with respect to age and sex (Table II). Mean time with patient-reported symptoms. The presence of WGTT, measured by ROMs, was 63.4 hours for a high number of fecal incontinence episodes, low stool fre- constipated children compared with 30.1 hours for healthy quency, and palpable stools in the rectum were each variables children (P < .001). Constipated children presented a mean associated with a prolonged gut transit time. Otherwise, in- weekly BM frequency of 4.8 compared with 5.7 for formation questioning the correlation between WGTT and nonconstipated children (P < .001). The average BSFS stool clinical variables (including both form and frequency) 4 score was 3.1 for constipated children compared with derives from studies previously undertaken in adults. Using a score of 4.2 for nonconstipated children (P < .001). There wireless motility capsule and ROMs, there is a significant cor- was a significant correlation between stool form and relation between stool form and both colonic transit time and 4 WGTT in both constipated and nonconstipated children, WGTT in constipated adults. Other studies report a signifi- with a correlation coefficient of À0.85 (P < .001, 95% CI cant correlation between stool form and WGTT in healthy 13,14 À0.88 to À0.80) (Figure 2). By contrast, there was no adults. correlation between stool frequency and WGTT The validity of the BSFS in predicting colonic transit time (correlation coefficient À0.17, P = .1, 95% CI À0.38 to in adults with irritable bowel syndrome has been demon- 8 0.04). There was no correlation between stool frequency strated previously. In the present study, we confirm these and stool form (correlation coefficient 0.1, P = .1, 95% CI observations in children by showing a significant correlation À0.09 to 0.4). Chronologic age of children was not between stool form and WGTT in both healthy children and significantly correlated with any of the 3 variables those with functional constipation. Although many clinicians (coefficient correlation À0.02, À0.04, and À0.03, still ask about stool frequency and consider it to be a valid respectively; P = .8, .7, and .1, respectively). Sex did not marker for the indirect determination of colonic transit significantly influence WGTT, stool form, and stool frequency (coefficient correlation À0.02, À0.04, and 0.1, respectively; P = .7, .1, and .7, respectively). Forty-nine of 80 children (61%) had delayed WGTT. Thirty-nine of 44 constipated children (89%) presented with delayed WGTT compared with 28% of the healthy children. Multivariate logistic regression analysis, using normal or delayed WGTT as a dependent variable, indicated that the sole variable independently associated with WGTT was

Table II. Potential predictors of WGTT at multivariate logistic regression analysis Variable Regression coefﬁcient OR 95% CI P value Sex 0.1 1.1 0.3-4.4 .8 Mean age 0.02 1.0 0.9-1.01 .8 Stool frequency 0.3 1.3 0.7-2.3 .2 Figure 2. Correlations between WGTT and average BSSS in Stool form 2.9 18.4 5.4-62.5 <.001 both constipated and nonconstipated children.

1190 Russo et al

Downloaded from ClinicalKey.com at The Children's Mercy Hospital - Kansas City September 07, 2016. For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved. June 2013 ORIGINAL ARTICLES time, we show that stool frequency is a poor surrogate mea- 2. Loening-Baucke V. Chronic constipation in children. Gastroenterology sure of gut transit time. Glia et al9 reported a moderate cor- 1993;105:1557-64. relation between stool frequency and colonic transit time but 3. O’Donnell LJ, Virjee J, Heaton KW. Detection of pseudodiarrhoea by simple clinical assessment of intestinal transit rate. BMJ 1990;300: exclusively in subjects with extremely delayed transit time 439-40. 9 4,8,15,16 (#2 BMs/wk). Findings from other studies, however, 4. Saad RJ, Rao SS, Koch KL, Kuo B, Parkman HP, McCallum RW, et al. Do do not support the contention of using stool frequency as stool form and frequency correlate with whole-gut and colonic transit? a surrogate for determining gut transit time. Consistent Results from a multicenter study in constipated individuals and healthy with this literature, our data in children confirm that stool controls. Am J Gastroenterol 2010;105:403-11. 5. Longstreth GF, Thompson WG, Chey WD, Houghton LA, Mearin F, frequency should not be recommended as an indicator of Spiller RC. Functional bowel disorders. Gastroenterology 2006;130: gut transit time. 1480-91. Colonic transit time has been previously measured by us- 6. Lane MM, Czyzewski DI, Chumpitazi BP, Shulman RJ. Reliability and ing both scintigraphy17-21 and ROMs.22-25 A significant pro- validity of a modified Bristol Stool Form Scale for children. J Pediatr longation of both total and segmental transit times is 2011;159:437-41.e1. 7. Bekkali N, Hamers SL, Reitsma JB, Van Toledo L, Benninga MA. Infant recorded in adolescents with chronic constipation compared stool form scale: development and results. J Pediatr 2009;154:521-6.e1. 26 27 with healthy adolescents. Gutierrez et al observed that 8. Heaton KW, O’Donnell LJ. An office guide to whole-gut transit time. one-half of 38 children with chronic constipation had normal Patients’ recollection of their stool form. J Clin Gastroenterol 1994;19: colonic transit, 37% prolonged transit time in the rectosig- 28-30. moid and left colon segments, and 13% had slowed transit 9. Glia A, Lindberg G, Nilsson LH, Mihocsa L, Akerlund JE. Clinical value 27 of symptom assessment in patients with constipation. Dis Colon Rectum times in all colonic segments. The reasons why subgroups 1999;42:1401-10. of constipated children have delays in either total or segmen- 10. Rasquin A, Di Lorenzo C, Forbes D, Guiraldes E, Hyams JS, Staiano A, tal colonic transit times are not yet completely understood et al. Childhood functional gastrointestinal disorders: child/adolescent. but could include either dysfunction of muscle layers in the Gastroenterology 2006;130:1527-37. colonic wall or an alteration of the enteric nervous system.28 11. Corazziari E, Cucchiara S, Staiano A, Romaniello G, Tamburrini O, Torsoli A, et al. Gastrointestinal transit time, frequency of defecation, Testing of WGTT can prove useful in the diagnostic and anorectal manometry in healthy and constipated children. J Pediatr workup of chronic constipation, particularly in those chil- 1985;106:379-82. dren who prove refractory to conventional therapies. How- 12. De Lorijn F, van Wijk MP, Reitsma JB, van Ginkel R, Taminiau JAJM, ever, due to the inconvenience and cost of tests that allow Benniga MA. Prognosis of constipation: clinical factors and colonic tran- the formal measurement of WGTT, health care providers sit time. Arch Dis Child 2004;89:723-7. 13. Lewis SJ, Heaton KW. Stool form scale as a useful guide to intestinal require objective clinical measures to assess for delayed transit time. Scand J Gastroenterol 1997;32:920-4. gut transit. Based on the results of this prospective study, 14. Degen LP, Philips SF. How well does stool form reflect colonic transit? stool form should be considered as a valuable indicator of Gut 1996;39:109-13. WGTT. 15. Wyman JB, Heaton KW, Manning AP, Wicks AC. Variability of colonic This study does have some limitations. First, the number function in healthy subjects. Gut 1978;19:146-50. 16. Chaussade S, Khyari A, Roche H, Garret M, Gaudric M, Couturier D, of control subjects enrolled into the study may be considered et al. Determination of total and segmental colonic transit time in con- too small to establish definitive conclusions about WGTT stipated patients. Results in 91 patients with a new simplified method. in healthy subjects. Second, this study did not include an Dig Dis Sci 1989;34:1168-72. evaluation of segmental colon transit times, required to be 17. Krevsky B, Malmud LS, D’Ercole F, Maurer AH, Fisher RS. Colonic able to determine just where the delayed transit time was transit scintigraphy. A physiologic approach to the quantitative measurement of colonic transit in humans. Gastroenterology 1986;91: occurring. 1102-12. This prospective, observational, case-control study shows 18. Stanghellini V, Tosetti C, Corinaldesi R. Standards for non-invasive that stool form, as measured by BSFS, rather than stool methods for gastrointestinal motility: scintigraphy. A position statement frequency correlates with WGTT in both constipated and from the Gruppo Italiano di Studio Motilita Apparato Digerente (GIS- nonconstipated children. Therefore, the BSFS should be MAD). Dig Liver Dis 2000;32:447-52. 19. Notghi A, Harding LK, Kumar D. Radioisotope determination of re- considered for use as a qualitative assessment of delayed gional colonic transit in severe constipation. Gut 1993;34:1289. transit accounting for cases of functional constipation in 20. Notghi A, Kumar D, Panagamuwa B, Tulley NJ, Hesslewood SR, children. n Harding LK. Measurement of colonic transit time using radionuclide imaging: analysis by condensed images. Nucl Med Commun 1993;14: 204-11. Submitted for publication Jul 17, 2012; last revision received Oct 23, 2012; accepted Nov 29, 2012. 21. Maurer AH, Krevsky B. Whole-gut transit scintigraphy in the evaluation of small-bowel and colon transit disorders. Semin Nucl Med 1995;25: Reprint requests: Annamaria Staiano, MD, Department of Pediatrics, 326-38. University of Naples “Federico II,” Via S Pansini, 5, 80131 Naples, Italy. E-mail: [email protected] 22. Hutson JM, Chase JW, Clarke MC, King SK, Sutcliffe J, Gibb S, et al. Slow-transit constipation in children: our experience. Pediatr Surg Int 2009;25:403-6. References 23. Southwell BR, Clarke MC, Sutcliffe J, Hutson JM. Colonic transit studies: normal values for adults and children with comparison of radiolog- 1. Van den Berg MM, Benninga MA, Di Lorenzo C. Epidemiology of child- ical and scintigraphic methods. Pediatr Surg Int 2009;25:559-72. hood constipation: a systematic review. Am J Gastroenterol 2006;101: 24. vd Baan-Slootweg OH, Liem O, Bekkali N, van Aalderen WM, 2401-9. Rijcken TH, Di Lorenzo C, et al. Constipation and colonic transit times

Stool Consistency, but Not Frequency, Correlates with Total Gastrointestinal Transit Time in Children 1191

in children with morbid obesity. J Pediatr Gastroenterol Nutr 2011;52: 27. Gutierrez C, Marco A, Nogales A, Tebar R. Total and segmental colonic 442-5. transit time and anorectal manometry in children with chronic idio- 25. Bautista Casasnovas A, Varela Cives R, Villanueva Jeremias A, Castro- pathic constipation. J Pediatr Gastroenterol Nutr 2002;35:31-8. Gago M, Cadranel S, Tojo Sierra R. Measurement of colonic transit 28. van den Berg MM, Di Lorenzo C, Mousa HM, Benninga MA, time in children. J Pediatr Gastroenterol Nutr 1991;13:42-5. Boeckxstaens GE, Luquette M. Morphological changes of the 26. Zaslavsky C, da Silveira TR, Maguilnik I. Total and segmental colonic enteric nervous system, interstitial cells of Cajal, and smooth muscle in transit time with radio-opaque markers in adolescents with functional children with colonic motility disorders. J Pediatr Gastroenterol Nutr constipation. J Pediatr Gastroenterol Nutr 1998;27:138-42. 2009;48:22-9.

1192 Russo et al