Bennett, William E. Jr*; Heuckeroth, Robert O.†
*Department of Pediatrics, Section of Pediatric Gastroenterology, Hepatology, and Nutrition, Indiana University School of Medicine, Indianapolis, IN
†Department of Pediatrics, Division of Pediatric Gastroenterology and Nutrition, Washington University School of Medicine, St Louis, MO.
Address correspondence and reprint requests to William E. Bennett Jr, MD, Department of Pediatrics, Section of Pediatric Gastroenterology, Hepatology, and Nutrition, Indiana University School of Medicine, 705 Riley Hospital Dr, ROC 4210, Indianapolis, IN 46202 (e-mail: email@example.com).
Received 10 June, 2011
Accepted 15 August, 2011
This research was funded in part by NIH grant NIDDK T32 DK077653-17 (W.E.B.).
The authors report no conflicts of interest.
The prevalence of constipation in the pediatric population is high, approaching 30% (1). Approximately 5% of general pediatric outpatient visits and up to 25% of pediatric gastroenterology outpatient visits are for constipation (2). Textbooks and clinical practice guidelines from the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition suggest that hypothyroidism should be considered in the differential diagnosis of constipation (2,3). For this reason, thyroid studies are frequently obtained as part of the standard evaluation for children with constipation, as well as for other problems in pediatric gastroenterology offices. Data to justify routine screening for hypothyroidism in constipated but otherwise healthy children, adolescents, and young adults are lacking.
Studies comparing bowel movement frequency or other bowel symptoms in adult patients with euthyroid, hypothyroid, and hyperthyroid have yielded mixed results (4,5). Physiologic studies show altered anorectal function in adult patients with both hypo- and hyperthyroidism, but equivocal findings in bowel transit times or symptoms of constipation (6,7). To our knowledge, there are no studies of the prevalence of hypothyroidism in pediatric patients referred for evaluation of constipation. In the present study, we performed a retrospective chart review to determine the prevalence of unrecognized hypothyroidism among pediatric patients referred to our gastroenterology office who had thyroid testing performed.
We reviewed records for all of the patients with free thyroxine (free T4) or thyroid-stimulating hormone (TSH) tests ordered by the pediatric gastroenterology division at St Louis Children's Hospital between January 1, 2003 and January 1, 2008. Hypothyroidism was defined by the following laboratory values: free T4 < 0.8 ng/μL (for all ages), TSH > 10 mIU/L (4–30 days of age), or TSH > 5.5 mIU/L (older than 30 days). A full chart review was performed for all patients with thyroid testing, which included review of clinical presentation (ie, indications for ordering thyroid tests, as enumerated in the pediatric gastroenterology consultation note), medical history, medication history, other laboratory tests, radiologic evaluation, and consultation with other subspecialty services. All of the medical contacts within a large, metropolitan hospital system using a single electronic medical record were ascertained, although no patient contact was initiated. The US Department of Health and Human Services Center for Medicare and Medicaid Services 2011 Clinical Laboratory Fee Schedule (8) quotes a median (mean) cost of $12.69 ($12.47) for a free T4 and $23.64 ($23.58) for a TSH. There is considerable variation in the cost in practice. For instance, the amount billed at our large referral children's hospital in 2011 was approximately $205 for TSH and $176 for free T4, whereas the amount collected ranged widely, from $41 to $102 for TSH and $35 to $88 for free T4. Similarly, an endocrinology consultation is listed by Health and Human Services as $47.18, although patients may be billed as much as $455, depending upon insurance coverage. To determine the total number of patients examined for constipation and slow growth during this time period, we counted all of the diagnostic codes consistent with these diagnoses. Individual chart reviews were performed only for patients who had thyroid testing.
The human research protection office and the institutional review board at Washington University School of Medicine reviewed and approved the present study.
Free T4, TSH, or a combination of these tests was ordered 1280 times during a 5-year period in a total of 873 unique patients. The mean age was 7.4 years, with a range from 10 days to 20.5 years, and 454 (52.0%) were male. A total of 443 children (50.7%) had constipation alone, 93 (10.7%) had slow growth alone, 85 (9.7%) had vomiting, 80 (9.2%) had cholestasis or hypertransaminasemia, 40 (4.6%) had both constipation and slow growth, 57 (6.5%) had another indication for the test, and 75 (8.6%) did not have an indication for the test clearly indicated in the consultation note. Of the 2911 unique patients with constipation (6197 total encounters), 483 (16.6%) had thyroid testing performed. Of the 420 children examined for slow growth (1215 encounters), 133 (31.7%) had thyroid testing performed. Because this is a retrospective analysis, the decision to test for hypothyroidism was at the discretion of the attending physician. We were unable to identify any change in practice patterns during the years evaluated.
A total of 817 of these children (93.6%) had normal thyroid tests. Fifty-six children (6.4%) had elevated TSH, decreased free T4, or a combination of the two. Of the 56 abnormal tests, 40 were normal when repeated. Eleven of these 40 had documentation, indicating that TSH levels were drawn during an acute illness, and elevated levels were thought to be a transient phenomenon. For the other 29 of 40 children whose thyroid tests normalized, no documentation for the likely cause for the transiently abnormal TSH was given. Of the remaining 16 children with abnormal testing who either did not normalize or were not retested, 7 had known hypothyroidism before testing in our gastroenterology office. Nine of the 56 abnormal thyroid tests (1.0% of all children evaluated) led to a new diagnosis of hypothyroidism. These results are summarized in Figure 1.
Of the 9 children with newly diagnosed hypothyroidism, only 7 had constipation. Two of the 9 children were being evaluated for slow growth without constipation. Three children referred for constipation had trisomy 21, so thyroid function testing should have been part of their routine health evaluation as recommended by the American Academy of Pediatrics (9). Two children with constipation and newly diagnosed hypothyroidism were taking medications (interferon-γ for active hepatitis C and carbamazepine for epilepsy) known to cause secondary hypothyroidism (10,11). In each case, hypothyroidism resolved after cessation of the medication. One patient with hypothyroidism was being evaluated for both slow growth and constipation. Thus, only 1 patient with newly diagnosed hypothyroidism presented to our office with constipation as their only recognized symptom of disease, as reported by parents and recorded in our office consultation.
Of those patients with normal screening results, it is unknown how many went on to develop hypothyroidism later in life. At the time of the review in 2009 to 2010, none of the patients with a normal TSH went on to have a repeat measurement performed by another service at our institution that was abnormal. It is unknown whether any of these children had thyroid testing performed at other institutions within this time period.
The costs for laboratory testing were at least $27,400 for the 1280 blood draws ordered in our gastroenterology office, although they may be as high as $213,120. Eighteen patients were referred for endocrinology consultations, adding at least an additional $849 in consultation fees, but this may be as high as $8190, depending upon insurance coverage. Eight of the referred children had hypothyroidism as a final diagnosis. Total costs billed for thyroid evaluation of children in our study were above and beyond the cost of the initial pediatric gastroenterology consultation, which was not included in this analysis. The cost of screening the normal patients was $18,653, but may be as high as $147,159, again depending on insurance coverage. Intangible costs include repeated painful procedures (venipuncture) performed on 56 children, indirect costs to the family, such as transportation, child care, and missed work for additional appointments, and were not included in the above calculations.
A medical position statement from the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition recommends thyroid hormone evaluation in children only in cases of severe, refractory constipation, and in most cases before consultation with a pediatric gastroenterologist (2). At our institution, the majority of patients referred to our pediatric gastroenterology clinic for constipation have already tried unsuccessfully to manage the condition in conjunction with their pediatrician before referral. These children often have had symptoms for months or years before referral and may therefore be considered refractory (although, admittedly, the degree and duration of constipation in all of the children referred to our clinic was not assessed in the present study). Even among this selected patient population, our data suggest that thyroid function tests are unlikely to be abnormal in the absence of other clinical indications or risk factors for hypothyroidism, such as concomitant growth failure, medications, or genetic predisposition (as in trisomy 21).
This retrospective review is limited by the fact that only a small percentage of children with constipation who were examined in our office were tested for hypothyroidism (16.6%) and it is not clear whether there is a difference between the tested and untested populations. For this analysis, we decided to assess all of the patients with thyroid testing, rather than patients with a specific diagnosis (such as constipation), to better understand the utility and practice patterns for thyroid hormone testing in a pediatric gastroenterology referral practice. We also did not make a prospective effort to identify children with recognized hypothyroidism or a family history of hypothyroidism, although it is anticipated that this would have been discussed as a routine part of our clinical care. Furthermore, it is possible that children who presented with constipation were identified as having hypothyroidism in their pediatrician's office and were referred to an endocrinologist instead of to our gastroenterology office. This is likely to be uncommon, however, because few of the children referred to our office for constipation have any thyroid testing before the office visit, although this was not assessed systematically in all of the patients. Despite these concerns, the extremely low incidence of hypothyroidism in our referral population suggests that testing for hypothyroidism in children who have constipation as their only symptom is unlikely to be helpful.
Large-scale epidemiologic studies estimate that although 4.6% of the US population has elevated TSH, only 0.3% of the population (including children) has clinically significant hypothyroidism (12). Assuming that our results are representative, the likelihood of identifying a new case of hypothyroidism among constipated children without other clinical indicators (in our population, approximately 0.2% or 1/443) is no greater than the likelihood of finding children with hypothyroidism by random population screening (0.3%). The yield of thyroid testing improves when the denominator is changed: in patients with both constipation and slow growth, 2.5% (1/40) were hypothyroid; in patients with slow growth alone, 2.2% (2/93) were hypothyroid. Thus, our data suggest that although slow growth increased the likelihood that a child has hypothyroidism, the presence of constipation did not affect the likelihood that a child had hypothyroidism (P = 0.6, z test, SigmaPlot 11). Furthermore, the financial cost of finding a single case of unrecognized hypothyroidism is high. Therefore, we question the wisdom of routine thyroid function testing in children with constipation unless there are other compelling clinical data to suggest the diagnosis.
In conclusion, we have reviewed all of the thyroid tests ordered by our pediatric gastroenterology office during a 5-year period and found that only a small proportion of those we tested whose symptom was isolated constipation had hypothyroidism (0.2%), whereas those evaluated for constipation with slow growth or slow growth alone were much more likely to be hypothyroid (2.5% and 2.2%, respectively). Identifying a single case of hypothyroidism during this time period in a child who presented with constipation alone costs at least $18,653. Although it may be argued that this is a reasonable price to pay, constipation alone did not increase the likelihood of hypothyroidism above the population prevalence for this disease.
Although there are some limitations to our analysis, our data suggest that careful consideration should be given when ordering free T4 and TSH concentrations in children with constipation without other clinical indicators of hypothyroidism. This question needs further investigation before these results can be applied, ideally with prospective observation of a large pediatric cohort.
The authors thank Michael Daft and Dennis Dietzen for their assistance in data acquisition. The authors thank Phillip Tarr for valuable insights.
1. Ip KS, Lee WT, Chan JS, et al. A community-based study of the prevalence of constipation in young children and the role of dietary fibre. Hong Kong Med J 2005;11:431–6.
2. Constipation Guideline Committee of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition. Evaluation and treatment of constipation in infants and children: recommendations of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr 2006;43:e1–13.
3. Wyllie R. Functional constipation. In: Kliegman R, ed. Nelson Textbook of Pediatrics, 18th ed. Philadelphia: Saunders; 2007.
4. Baker JT Harvey RF. Bowel habit in thyrotoxicosis and hypothyroidism. Br Med J 1971;1:322–3.
5. Crooks J, Murray IP, Wayne EJ. Statistical methods applied to the clinical diagnosis of thyrotoxicosis. Q J Med 1959;28:211–34.
6. Tobin MV, Fisken RA, Diggory RT, et al. Orocaecal transit time in health and in thyroid disease. Gut 1989;30:26–9.
7. Deen KI, Seneviratne SL, de Silva HJ. Anorectal physiology and transit in patients with disorders of thyroid metabolism. J Gastroenterol Hepatol 1999;14:384–7.
8. Clinical Laboratory Fee Schedule. Health and Human Services, Center for Medicare and Medicaid Services. https.http://www.cms.gov/clinicallabfeesched/02_clinlab.asp
. Accessed June 1, 2011.
9. American Academy of Pediatrics, Committee on Genetics. American Academy of Pediatrics: health supervision for children with Down syndrome. Pediatrics 2001;107:442–9.
10. Vezali E, Elefsiniotis I, Mihas C, et al. Thyroid dysfunction in patients with chronic hepatitis C: virus- or therapy-related? J Gastroenterol Hepatol 2009; 24:1024–1029.
11. Verrotti A, Laus M, Scardapane A, et al. Thyroid hormones in children with epilepsy during long-term administration of carbamazepine and valproate. Eur J Endocrinol 2009; 160:81–86.
12. Hollowell JG, Staehling NW, Flanders WD, et al. Serum TSH, T(4), and thyroid antibodies in the United States population (1988 to 1994): National Health and Nutrition Examination Survey (NHANES III). J Clin Endocrinol Metab 2002;87:489–99.