Secondary Logo

Journal Logo

Original Articles: Hepatology and Nutrition

Vitamin D Deficiency and Low Bone Mineral Density in Pediatric and Young Adult Intestinal Failure

Ubesie, Agozie C.*; Heubi, James E.*; Kocoshis, Samuel A.*; Henderson, Carol J.*; Mezoff, Adam G.*; Rao, Marepalli B.; Cole, Conrad R.*

Author Information
Journal of Pediatric Gastroenterology and Nutrition: September 2013 - Volume 57 - Issue 3 - p 372-376
doi: 10.1097/MPG.0b013e31829c10eb
  • Free


Intestinal failure (IF) occurs when severe intestinal malabsorption caused by short bowel syndrome (SBS), congenital transport defects, or motility disturbances mandates artificial nutrition to be administered parenterally (1). This is because of a critical reduction of the functional gut mass below the level necessary to maintain adequate digestion and absorption of nutrients and fluid for normal growth (2–5). In children, the predominant underlying cause of IF is SBS usually caused by resection for congenital (intestinal atresia, malrotation with volvulus) or acquired (necrotizing enterocolitis, vascular thrombosis, or trauma) disorders. Regardless of the cause for IF, nutrient malabsorption is common (6).

Vitamin D as a hormone interacts with its receptor in the small intestine to increase the efficiency of intestinal calcium and phosphate absorption. In addition, the regulation of renal calcium and phosphorous handling is mediated by vitamin D. Vitamin D3 is synthesized either in the skin under the influence of ultraviolet light of the sun, or it is obtained from food, especially fortified milk (7,8). Once it is produced in the skin or ingested from the diet, it is converted sequentially in the liver to 25-hydroxyvitamin D (25-(OH) D) and by the kidneys to its biologically active form, 1, 25-dihydroxyvitamin D (9). Ultimately, vitamin D sufficiency is necessary for muscle and bone health (7,9–12). Malabsorption of vitamin D occurs in patients with small bowel resection (13,14), cholestasis (15,16), and fat malabsorption (17). Yang et al (18) reported a high prevalence of vitamin D deficiency (68%) among children with IF even after transitioning to 100% enteral nutrition. Prolonged home parenteral nutrition (PN) has been identified as a risk factor for both vitamin D deficiency and insufficiency (19). There is also a seasonal variation on cutaneous synthesis of vitamin D associated with significant increase in cutaneous vitamin D synthesis during the summer compared with winter (8,20,21). Deficiency of vitamin D leads to abnormal bone health, muscle weakness, and fractures (7). Bone health can be assessed using dual-energy x-ray absorptiometry (DXA), which is relatively inexpensive, fast, and delivers a low radiation dose (22). The goal of this study was to evaluate serum 25-(OH) D and bone mineral density (BMD) z score in this group of patients during a 5-year period. Our hypothesis was that patients with IF are at increased risk for vitamin D deficiency and reduced BMD.


Study Design

This was a retrospective study of patients seen between August 1st 2007 and July 31st 2012 at the Intestinal Care Center of Cincinnati Children's Hospital Medical Center, Cincinnati, OH.


One hundred twenty-three children (ages 3 years and older) seen during the period under review were included in the study because they met criteria for IF defined by the need for PN support for >30 days. The study was approved by the institutional review board of Cincinnati Children's Hospital Medical Center.


Relevant data of eligible patients were retrieved from the electronic medical record and an existing database. Information obtained included date of birth, age, sex, ethnicity, anthropometric centiles (weight, height, body mass index, weight for length), serum 25 (OH) D levels, lumbar spine (LS) BMD, and LS BMD z scores. Data were obtained longitudinally for every visit when available throughout the 5-year follow-up period. The months of March through August were defined as spring/summer, whereas September to February were defined as fall/winter. Customarily, patients underwent measurement of serum 25-(OH) D measurements every 6 to 12 months and a DXA scan every 12 months.

Diagnostic Methods

Serum 25(OH) D was measured by radioimmunoassay using a direct competitive chemiluminescence immunoassay (Diasorin Liaison, Stillwater, MN) with intraassay coefficients of variation of 6.3% and 8.6% for mean concentrations of 21 and 65 ng/mL, respectively. The interassay coefficients of variation were 9.4% ± 1.6% and 7.7% ± 4.2%, respectively. Serum 25(OH) D <20 ng/mL was defined as vitamin D deficiency (23). LS DXA scans were measured using Hologic Discovery A, serial no. 81400 (Hologic Inc, Waltham, MA) to determine the BMD (22,24). The posterior-anterior spines (lumbar vertebrae L1–L4) were obtained using standardized positioning techniques in the supine position to generate measures of areal BMD (g/cm2). The analysis of the LS DXA was with auto low-density software version (1% reproducibility; Hologic). A DXA scan z score of ≤−2 was defined as reduced (25). The reported final BMD z scores were adjusted for age and sex, and for height for patients who were short for age and had reduced bone mass. “Short for age” was defined as height for age <5th percentile based on the Centers for Disease Control and Prevention growth standards (26). Concurrent DXA BMD and z score were defined as measures done within 6 months of serum 25 (OH) D measurements. In our practice, patients started taking vitamin D at a dose of 8000 IU daily or 50,000 IU weekly if their serum 25 (OH) D levels were <20 ng/dL. Weight, height, and body mass index (BMI) percentiles were based on Centers for Disease Control and Prevention growth charts.

Data Analysis

The data were analyzed using SPSS version 19 for Windows (SPSS Inc, Chicago, IL) and statistical software R (R Core Team, Vienna, Austria). The BMD z scores in short-for-age patients (<5th percentile for age) were adjusted to the 50th percentile of height age (26). Serum 25 (OH) D deficiency and categorized BMD z score were the outcome variables. The χ2 test was used to test association between categorical predictors and each of the outcome variables. The binary logistic regression model was used to test for association between risk factors and each of the outcome variables. The relevant odds ratios (ORs) and confidence intervals (CIs) were calculated. A P value of <0.05 was regarded as significant. All reported P values are 2-sided and all CIs reported are 95%.


Participant Characteristics

One hundred twenty-three patients ages 3 years or older with at least 1 documented serum 25 (OH) D level were enrolled in this study. This represented 90.4% (123/136) of the entire population of patients with IF ages 3 years or older seen at the Cincinnati Children's Hospital Medical Center Intestinal Care Center during the period under evaluation. Among them, 80 (65%) had a concurrent DXA scan done. The median age of entry into the study was 4 years (range 3–22 years), and 57.7% were boys. Children between ages 3 and 5 years accounted for 58.5% of the study population. Most of the patients were white (86.2%) and necrotizing enterocolitis was the most common primary gastrointestinal diagnosis (23.6%). The characteristics of the patients are shown in Table 1. There was no difference in the BMI of African Americans (AAs) and non-AAs (P = 1). Eleven patients with reduced BMD z scores but were short for age (<5th percentile height) had z scores adjusted to their 50th percentile height age.

Characteristics of the subjects

Vitamin D and Bone Health

Our patients had a total of 449 serum 25 (OH) D and 157 BMD measurements during the study interval. One hundred forty-three concurrent measurements of 25 (OH) D and BMD were done (within 6 months of one another). The overall mean difference in time between vitamin D and DXA measures was 37.4 ± 3.7 days. Among patients with low BMD z score, the mean difference in time was 36.8 ± 4.3 compared with 42.1 ± 7.4 among those with normal BMD z score (P = 0.56). There was no significant association between vitamin D and either short stature–adjusted (P = 1) or nonadjusted (P = 1) concurrent BMD z score status. Two hundred thirty-two (64.3%) of 276 vitamin D measurements during the winter/fall seasons were in the deficient range compared with 29% (129/173) during the summer/spring seasons (P = 0.015). Twenty-two of 97 (22.7%) DXA measures done during the winter/fall seasons were abnormal compared with 9 of 60 (15%) measured in summer/spring (P = 0.3).

Prevalence of Vitamin D Deficiency and Abnormal Bone Health

Vitamin D deficiency was identified in 40% (49/123) of the patients (95% CI 32%–49%). Twenty-nine of the 49 deficient patients (42.9%) had intermittent or persistent deficiency even after appropriate therapeutic doses of vitamin D had been started. Sixteen patients received vitamin D supplementation. Eight of the 49 deficient patients (16.3%) had been vitamin D supplemented before the discovery of deficiency. Twelve of the 16 vitamin D–supplemented patients received daily vitamin D dosing, whereas 4 patients were receiving weekly dosing. Six of 12 patients (50%) receiving daily supplementation developed deficiency compared with 2 of 4 (50%) among those receiving weekly supplementation. Ten (12.5%) of 80 patients that had concurrent DXA scans done had reduced LS z scores (95% CI 6.9%–21.5%) after adjusting for height age, and 8 of them (80%) were receiving vitamin D supplementation. The mean age of the children who were serum 25 (OH) D deficient was 7 years compared with 5 years among the nondeficient group (P = 0.04). Similarly, the mean age of the children that had abnormal BMD z score was 9.9 years compared with 5.2 years among those that had a normal BMD z score (P = 0.03). Univariate analysis using the χ2 test showed that age older than 10 years was significantly associated with serum 25 (OH) D deficiency (P = 0.01). Thirteen (68.4%) of 19 children older than 10 years compared with 36 (34.6%) of 104 children ages 10 years or younger had 25 (OH) D deficiency (P = 0.01). The risk for developing vitamin D deficiency when the subject is younger than 10 years is lower compared with the risk if older than 10 years (OR 0.23, 95% CI 0.08–0.66). Although race was not significantly associated with serum 25 (OH) D deficiency, a higher proportion of AA (52.9%) compared with non-AAs (37.3%) were vitamin D deficient (P = 0.29). There was also a higher proportion of the AAs (42.9%) compared with the other racial groups (20.2%) with intermittent or persistent vitamin D deficiency and this difference showed a trend (P = 0.09) that did not attain statistical significance.

Categorized age (10 years or younger or older than 10 years) was significantly associated with BMD z score (P = 0.02). Four of 10 children (40%) older than 10 years had reduced BMD z scores compared with 5 of 70 children (7.1%) ages 10 years or younger. The risk for having abnormal BMD versus normal BMD when the patient is 10 years or younger is lower than when the patient is older than 10 years (OR 0.12, 95% CI 0.03–0.51). All patients with persistent or intermittent vitamin D deficiency compared with 11 of 28 (39.3%) of those with only 1 documented vitamin deficiency had reduced BMD z score (P = 0.02). The mean PN duration among vitamin D–deficient patients was 28.3 months compared with 29.3 months among vitamin D–sufficient patients (P = 0.88). Three of 10 patients (30%) who were exclusively dependent on PN had vitamin D deficiency compared with 40 of 105 (38.1%) receiving enteral nutrition (P = 0.74). Conversely, 3 of 7 patients (42.9%) who were PN dependent compared with 6 of 70 (8.6%) of those receiving enteral nutrition (8.6%) (P = 0.03) had reduced BMD z scores. The mean PN duration among patients with reduced BMD z scores was 36.9 ± 26.6 months compared with 26.6 ± 3.8 months among those with normal BMD z scores (P = 0.52). A binary logistic regression analysis showed that only age was significantly associated with both serum 25 (OH) D deficiency (P = 0. 04) and reduced BMD z score (P = 0. 01) (shown in Tables 2 and 3). The OR of increasing age being associated with vitamin D deficiency was 0.91 (95% CI 0.85–0.1.0) and 0.8 (95% CI 0.67–0.94) for reduced bone mass (Fig. 1). When the effect of age (above and below 10 years) as a dependent variable was examined for both vitamin D and bone health status using a logistic regression model, only abnormal bone health remained significantly associated with age older than 10 years (P = 0.02). The OR was 0.16 (95% CI 0.03–0.77).

Logistic regression model showing risk factors associated with vitamin D deficiency
Logistic regression model for risk factors associated with reduced bone mineral density
Logistic regression showing the probability of developing vitamin D deficiency and reduced bone mineral density z score with increasing age.


No large studies have evaluated serum 25 (OH) D levels and BMD in children with IF. In this population of patients with IF, the prevalence of vitamin D deficiency (abnormal serum 25 (OH) D) and reduced bone mass by DXA (after height correction) were 39.8% (95% CI 31.6%–48.7%) and 12.5% (95% CI 6.9%–21.5%), respectively. Patients with vitamin D deficiency and low BMD z scores were more likely to be older than 10 years and exclusively on PN at study entry.

The high prevalence of vitamin D deficiency in this study of patients with IF compares with previous smaller reports (27–30). Vitamin D–deficient rickets has been reported in 4 children after weaning them off prolonged PN following small bowel resection (27). Forty percent to 87% of adult patients with intestinal resection have vitamin D deficiency (28–30). Vitamin D deficiency in patients with IF is likely caused, in part, by malaborption resulting from intestinal resection or dysmotility (2,29). Intestinal resection may lead to reduced surface area or impaired vitamin D solubilization because of interruption of enterohepatic circulation of bile acids with ileal resection (31). The high prevalence of vitamin D deficiency in our patients with IF may relate to poor compliance with vitamin D supplementation and inadequate intake of vitamin D. Our patients were significantly more likely to be vitamin D deficient during the winter/fall seasons than in summer/spring seasons (P = 0.02). The vast majority of our patients lived in latitudes between 35 degrees north (Knoxville) and 42 degrees north (Toledo) (Cincinnati is located at 39 degrees north). At those latitudes, significant variability exists between vitamin D levels measured in the summer and winter (32).

In a previous study of 24 children, Diamanti et al (33) reported 83% with reduced BMD z score. Their reported high prevalence contrasts with the 12.5% prevalence in our series. There may be a number of explanations: Diamanti et al defined reduced bone mass as a z score of <–1, whereas our study used the criteria of ≤–2 (our definition was based on recent recommendations of the International Society for Clinical Densitometry (25)); only patients receiving long-term PN (defined by at least 3 months of PN) were included in their study; the BMD was not corrected for height age. The association between long-term PN and metabolic bone disease has been described (34,35). Our study, however, did find that those receiving exclusive PN at study entry had a significantly higher rate of reduced bone mass but not vitamin D deficiency.

A major strength of this study was the analysis of age with vitamin D and bone mass status. We found that older age (older than 10 years) was significantly associated with both serum 25 (OH) D deficiency and reduced bone mass. These findings are similar to those described in children with chronic medical illness (36).

There was no significant association between serum 25 (OH) D and BMD; however, our patients with intermittent or persistent vitamin D deficiency had significantly reduced bone mass. There are conflicting reports regarding the relation between serum 25 (OH) D and BMD (28). Adults reported on by Haderslev et al (28) had low serum 25 (OH) D that correlated significantly with lower BMD z scores; however, Stein et al (37) found no significant association between serum 25 (OH) D and BMD after adjusting for season, race, age, and BMI among healthy young girls. Others have failed to show a correlation between serum 25 (OH) D levels and BMD (38,39).

The retrospective design of the study is a weakness, which limits the ability to generalize these findings; however, the data are longitudinal and hypothesis generating. A total of 449 serum 25 (OH) D and 160 BMD longitudinal measurements were performed during the study period and included in the analysis, ranking this among the largest series evaluating vitamin D and bone health in this population of children.


The risk of vitamin D deficiency and low BMD increases with age among patients with IF. Vitamin D deficiency was also significantly more common in the spring/fall seasons, as was reduced bone mass in patients exclusively receiving PN. Patients with repeatedly deficient serum 25 (OH) D had significantly reduced bone mass. We recommend that older patients with IF and those receiving chronic PN be monitored closely for risk of abnormal bone health and be considered for routine vitamin D supplementation.


The authors are grateful to Misty Troutt (Project Manager, Intestinal Rehabilitation Program) and Justina Dunigan (Administrative Assistant, Intestinal Care Center) for their assistance and support during this study. We are also thankful to the families of our patients, whose data were used in the study.

Of Weaknes of the Stomacke and Vomytynge

Many tymes the stomacke of the chyld is so feble that it cannot retayne eyther meate or drynke, in which case, and for all debilitye thereof, it is very good to wasshe the stomake with warme water of roses wherin a lytle muske hath bene dissolved, for that by the odour and natural heate gyveth a comforte to al the spirituall members.

And then it is good to roste a quince tender, and with a lyttle pouder of cloves and suger to gyve it to the chyld to eate; conserva quinces with a lyttle cynamome and cloves is singuler good for the same entent. Also ye may make a juyce of quinces, and gyve it to the child to drynke with a lytle suger.*

Phaer, Thomas (c. 1510–1560), The Boke of Chyldren, London, 1544

*Sixteenth-century English had no fixed philological spelling rules; thus, note the various spellings for “stomach,” “child,” “little,” etc. Phaer's pharmacopeia was Dioscoridian. Rose water with musk (Mimulus moschatus) was a known calmative; cloves (Syzygium aromaticum), quince (Cydonia oblonga), and cinnamon (Cinnamomum vera) have been used since antiquity to relieve intestinal cramps. Interestingly, the eugenol content of cinnamon is being studied for its antiviral effects.

—Contributed by Angel R. Colón, MD


1. Cole CR, Ziegler TR. Duggan CP, Gura KM, Jaksic T. Etiology and epidemiology of intestinal failure. Clinical Management of Intestinal Failure. Boca Raton, FL:CRC Press; 2012. 3–12.
2. Wessel JJ, Kocoshis SA. Nutritional management of infants with short bowel syndrome. Semin Perinatol 2007; 31:104–111.
3. Goulet O, Fusaro F, Lacaille F, et al. Permanent intestinal failure. Indian Pediatr 2008; 45:753–763.
4. Tavakkolizadeh A, Whang EE. Understanding and augmenting human intestinal adaptation: a call for more clinical research. J Parenter Enteral Nutr 2002; 26:251–255.
5. O’Keefe SJD, Buchman AL, Fishbein T, et al. Short bowel syndrome and intestinal failure: Consensus definitions and overview. Clin Gastroenterol Hepatol 2006; 4:6–10.
6. Peterson J, Kerner JA. New advances in the management of children with intestinal failure. J Parenter Enteral Nutr 2012; 36 (suppl 1):36S–42S.
7. Wimalawansa SJ. Vitamin D: an essential component for skeletal health. Ann N Y Acad Sci 2011; 1240:E1–E12.
8. Cole CR, Grant FK, Tangpricha V, et al. 25–Hydroxyvitamin D status of healthy, low-income, minority children in Atlanta, Georgia. Pediatrics 2010; 125:633–639.
9. Christakos S. Recent advances in our understanding of 1, 25-dihydroxyvitamin D (3) regulation of intestinal calcium absorption. Arch Biochem Biophys 2012; 523:73–76.
10. Lips P. Interaction between vitamin D and calcium. Scan J Clin Lab Invest Suppl 2012; 243:60–64.
11. Holick MF. Vitamin D: a d-lightful solution for health. J Investig Med 2011; 59:872–880.
12. Stroud ML, Stilgoe S, Stott VE, et al. Vitamin D-a review. Aust Fam Physician 2008; 37:1002–1005.
13. Markestad T, Aksnes L, Finne PH, et al. Decreased vitamin D absorption after limited jejunal resection in a premature infant. J Pediatr 1982; 101:1001–1008.
14. Compston JE, Ayers AB, Horton LWL, et al. Osteomalacia after small intestinal resection. Lancet 1978; 1:9–12.
15. Strople J, Lovell G, Heubi J. Prevalence of subclinical vitamin K deficiency in cholestatic liver disease. J Pediatr Gastroenterol Nutr 2009; 49:78–84.
16. Heubi JE, Hollis BW, Specker B, et al. Bone disease in chronic childhood cholestasis. I. Vitamin D absorption and metabolism. Hepatology 1989; 9:258–264.
17. Lo CW, Paris PW, Clemens TL, et al. Vitamin D absorption in healthy subjects and in patients with intestinal malabsorption syndromes. Am J Clin Nutr 1985; 42:644–649.
18. Yang CF, Duro D, Zurakowski D, et al. High prevalence of multiple micronutrient deficiencies in children with intestinal failure: a longitudinal study. J Pediatr 2011; 159:39–44.
19. Thomson P, Duerksen DR. Vitamin D deficiency in patients receiving home parenteral nutrition. J Parenter Enteral Nutr 2011; 20:1–6.
20. Levis S, Gomez A, Jimenez C, et al. Vitamin D deficiency and seasonal variation in an adult South Florida population. J Clin Endocrinol Metab 2005; 90:1557–1562.
21. Van der Mei IAF, Ponsonby A, Engelsen O. The high prevalence of vitamin D insufficiency across Australian populations is only partly explained by season and latitude. Environ Health Perspect 2007; 115:1132–1139.
22. Griffin LM, Kalkwarf HJ, Zemel BS, et al. Assessment of dual-energy x-ray absorptiometry measures of bone health in pediatric chronic kidney disease. Pediatr Nephrol 2012; 27:1139–1148.
23. Guthery SL, Pohl JF, Bucuvalas JC, et al. Bone mineral density in long-term survivors following pediatric liver transplantation. Liver Transplant 2003; 9:365–370.
24. Institute of Medicine Dietary reference intakes for calcium and vitamin D Published November 30, 2010. Accessed August 26, 2012.
25. Gordon CM, Bachrach LK, Carpenter TO, et al. Dual energy X-ray absorptiometry interpretation and reporting in children and adolescents: the 2007 ISCD Pediatric Official Positions. J Clin Densitom 2008; 11:43–58.
26. Centers for Disease Control and Prevention Overview of the CDC growth charts. Accessed March 1, 2013.
27. Touloukian RJ, Gertner JM. Vitamin D deficiency rickets as a late complication of the short gut syndrome during infancy. Pediatr Surg 1981; 16:230–235.
28. Haderslev KV, Jeppesen PB, Sorensen HA, et al. Vitamin D status and measurements of markers of bone metabolism in patients with small intestinal resection. Gut 2003; 52:653–658.
29. Culkin A, Rye B, Hanson C. OC-037 Vitamin D deficiency is common in intestinal failure patients. Gut 2012; 61:A16.
30. Braga CB, Vannucchi H, Freire CM, et al. Serum vitamins in adult patients with short bowel syndrome receiving intermittent parenteral nutrition. J Parenter Enteral Nutr 2011; 35:493–498.
31. Arnaud SB, Goldsmith RS, Lambert PW, et al. 25-hydroxy vitamin D3: evidence of an enterohepatic circulation in man. Proc Soc Exp Biol Med 1975; 149:570–572.
32. Tsiaras WG, Weistock MA. Factors influencing vitamin D status. Acta Derm Venereol 2011; 91:115–124.
33. Diamanti A, Bizzarri C, Basso MS, et al. How does long-term parenteral nutrition impact the bone mineral status of children with intestinal failure? J Bone Miner Metab 2010; 28:351–358.
34. Seligman JV, Basi SS, Deitel M, et al. Metabolic bone disease in a patient on long-term parenteral nutrition: a case report with review of literature. J Parenter Enteral Nutr 1984; 8:722–727.
35. Klein GL, Targoff CM, Ament ME, et al. Bone disease associated with total parenteral nutrition. Lancet 1980; 2:1041–1044.
36. Holmlund-Suila E, Koskivirta P, Metso T, et al. Vitamin D deficiency in children with a chronic illness-seasonal and age related variations in serum 25-hydroxy vitamin D concentrations. PLoS One 2013; 8:360856.
37. Stein EM, Laing EM, Hall DB, et al. Serum 25-hydroxyvitamin D concentrations in girls aged 4-8 years living in the southeastern United States. Am J Clin Nutr 2006; 83:75–81.
38. Kruavit A, Chailurkit LO, Thakkinstian A, et al. Prevalence of vitamin D insufficiency and low bone mineral density in elderly Thai nursing home residents. BMC Geriatr 2012; 12:49.
39. Kirbas A, Kirbas S, Anlar O, et al. Investigation of the relationship between vitamin D and bone mineral density in newly diagnosed multiple sclerosis. Acta Neurol Belg 2013; 113:43–47.

bone mineralization; intestinal failure; Vitamin D

© 2013 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,