Abstract
Background
Patients with a short bowel need to restrict oral fluid to reduce stoma/fistula output. Some patients require oral nutritional supplements (ONS) to meet nutritional requirements. There is limited research on ONS for patients with a short bowel. This study aimed to compare polymeric verses semi-elemental ONS on jejunal wet weight and sodium output.
Materials and methods
Patients were included if they had < 200 cm of small bowel to a jejunostomy or an enterocutaneous fistula. Patients fasted from midnight then were randomly allocated to consume 200 ml of either Vital 1.5® or Ensure Plus® over 30 min. Jejunal output and urine were collected for 6 h during which patients remained nil by mouth. Jejunal and urine outputs were weighed, and samples sent for analysis. After a day washout, patients were swapped to the other ONS. Differences between the two study periods were assessed using multilevel regression.
Results
Ten patients completed, 4 females: 6 males, mean age 59.2 years, mean small bowel length 95 cm. Jejunal wet weight (P < 0.001) and jejunal sodium (P = 0.003) were significantly higher on Vital 1.5® compared to Ensure Plus® with a mean difference of 0.1 kg and13mmol respectively. There was no difference in faecal sodium balance (P = 0.16), urine sodium (P = 0.38) or urine volume (P = 0.56).
Conclusion
A semi-elemental ONS results in a greater jejunal wet weight loss compared to a polymeric one. The amount is small but the daily cumulative loss will be greater.
Keywords: Short bowel, Intestinal failure, High output stoma, Enterocutaneous fistula, Oral nutritional supplements
Introduction
Patients with a short bowel have either a jejuno-colic anastomosis (JCA) or an end jejunostomy. Patients with a jejunostomy have immediate problems post-operatively due to large losses from the jejunostomy leading to fluid, sodium and magnesium depletion. Malnutrition may also occur and so nutritional support with oral nutritional supplements (ONS), enteral nutrition and/or parenteral nutrition may be required to maintain health (and or growth in children).
The focus of treatments is to reduce jejunal losses, so preventing dehydration, electrolyte disturbances and malnutrition. This involves restricting oral fluid intake and taking an oral rehydration solution containing 90–120 mmol/L of sodium. The sodium content of jejunal fluid averages 88 mmol/L (range 60–118) [1]. When fluids containing < 90 mmol/L of sodium are consumed, the jejunum secretes fluid and sodium from the plasma into the lumen of the intestine which is then lost from the body, resulting in dehydration and sodium depletion. The osmolality of plasma is 275–295 mOsm/kg and therefore the ideal osmolality of an ONS would be similar as recommended in the British Society of Gastroenterology guidelines on the management of short bowel [2]. Most ONS currently available have an osmolality higher than this recommendation. A higher osmolality will cause fluid to move by osmosis into the jejunal lumen. It is currently unknown which are the most appropriate ONSs for patients with a jejunostomy to minimise fluid and sodium losses and promote macronutrient absorption.
Due to malabsorption, malnourished patients with short bowel not receiving parenteral nutrition are recommended to consume a diet high in energy and protein [3], [4]. For patients who have difficulties achieving recommended intakes from food, additional nutrients may be provided by ONS. Clinical studies are few and compared a variety of ONS with different constituents including different protein sources (whole proteins, peptides and amino acids) different fat sources (long chain and medium chain triglycerides) and different carbohydrate sources (maltodextrin, sucrose and glucose). From balance studies in patients with a jejunostomy, a fibre-free, polymeric or semi-elemental ONS appeared most efficacious [5], [6], [7].
This study aimed to assess the effect of a polymeric (Ensure Plus®) verses a semi-elemental (Vital 1.5®) ONS on jejunal wet weight and sodium losses.
Materials and methods
All patients with a jejunostomy or established small bowel enterocutaneous fistula admitted to a UK tertiary referral hospital were considered for participation in the study. All patients were on a high output stoma protocol which includes a proton pump inhibitor, loperamide and codeine (variable doses depending on effect). They were also on a fluid restriction of 1 L/day and an oral rehydration solution 1 L/day.
Inclusion criteria
Patients included were over 18 years old, with a residual length of small bowel < 200 cm (determined either at operation or radiologically) to an established stoma or enterocutaneous fistula.
Exclusion criteria
Patients were excluded if: urine sodium concentration was less than 20 mmol/L, a jejuno-colic anastomosis, current sepsis of any cause, a change to anti-diarrhoeal or antisecretory agents in the last 3 days, severe gastrointestinal obstruction or structural abnormality of the intestine, active Crohn’s disease as assessed using two of the three parameters (CRP>10, albumin <30 g/L, platelets >400 ×109/L), participation in another intervention trial which may affect intestinal absorption, nil by mouth or unable or unwilling to provide consent.
Study design
The study was a randomised controlled crossover trial (Fig. 1). The crossover design was chosen due to the heterogeneous nature of the patient population so that patients acted as their own control. Written consent was obtained and patients were randomised to receive Ensure Plus® or Vital 1.5®. The composition of the two ONS is shown in Table 1. Randomisation was done by the pharmacy department using a computer-generated randomisation schedule grouped into blocks of six (Episys RPAS 3 v1.52). Patients fasted overnight from 12.00 am on day one and at 08.00 am were asked to pass urine and empty their stoma bag. They were asked to drink 200 ml of Ensure Plus® or Vital 1.5® over a 30-minute period. Urine and jejunal output were collected into two separate containers over the next 6-hours. The 6-hour collection was based on previous work [7], [8]. During this time the patients remained nil by mouth. After the 6-hour collection patients could eat and drink as per usual intake. The urine and jejunal output were weighed and an aliquot stored in a − 20°C freezer before being sent for analysis. Jejunal output is referred to as jejunal wet weight. Day two was a washout day when the patient was able to eat and drink normally. On day 3, the patient crossed over onto the alternate ONS and the study was repeated. No changes were made to the patient’s usual medications including the anti-diarrhoeal and/or antisecretory agents or parenteral support during the study. All patients were on our high output stoma protocol which includes a proton pump inhibitor (omeprazole or lansoprazole), loperamide and codeine (variable doses depending on effect). They were also on a fluid restriction of 1 L/day and an oral rehydration solution 1 L/day. All patients required parenteral nutrition. Urine and jejunal sodium were measured by flame photometry (Sherwood 420 flame photometer) at the Department of Biochemistry, Great Ormond Street Hospital, London, England.
Fig. 1.
Study design.
Table 1.
Composition of the oral nutritional supplements.
| Per 100 ml | Vital 1.5® | Ensure Plus® |
|---|---|---|
| Energy (kcal) | 150 | 150 |
| Protein (g) | 6.75 (peptide) | 6.25 (polypeptide) |
| Carbohydrate (g) | 18.4 (Sucrose & maltodextrin) | 20.2 (Sucrose & maltodextrin) |
| Fat (g) | 5.50 (MCT) | 4.92 (LCT) |
| Sodium (mmol) | 7.35 | 4.00 |
| Osmolarity (mOsm/L) | 487 | 509 |
| Osmolality (mOsm/kg H2O) | 630 | 660 |
MCT = medium chain triglyceride, LCT = long chain triglyceride
The primary outcome measures were the difference in jejunal wet weight and sodium losses after consumption of the two ONS.
The secondary outcome measures were jejunal sodium balance, urine sodium and urine volume.
Jejunal sodium balance (amount of sodium absorbed by the gut) was calculated as follows:
Sodium intake from the ONS (mmol) – [Jejunal wet weight (L) x jejunal sodium concentration (mmol/L)].
Ethics
Ethical approval was granted (Ref 15/SC/0281) and informed consent obtained. Patients consented to publishing the data with the knowledge that their identity would not be disclosed.
The trial was registered on Clinical trails.govtrails.gov. NCT02686606
Statistical analysis
The sample size was based on detecting a difference between treatment groups for jejunal sodium. Previous research using the same research design [9] suggested that the standard deviation of the differences between treatments was 17 mmol. A difference in outcomes between treatments of 20 mmol would be regarded as being clinically important. Using a 5 % significance level and 90 % power, it was calculated that 8 subjects were required for the study. Based on a likely drop-out rate of 10–20 % 10 subjects were recruited. Due to the crossover nature of the trial and to allow for repeated measurements, the analysis was performed using multilevel regression. Two level models were used with results from individual ONS nested within patients. Two terms were included in the models, the period of the study (first or second period) and the ONS received. Data were examined to assess carryover effect. A P value of < 0.05 was considered statistically significant. The data was normally distributed and so are presented as mean ± standard deviation.
Results
Ten patients completed the study (Table 2 and Fig. 2). The full volume of ONS was consumed by each participant. All received parenteral support.
Table 2.
Patient demographics.
| Age (years) | 59.2 ± 13 (range 43-83) |
|---|---|
| Gender | 4 Female:6 Male |
| Body Mass Index (kg/m2) | 23.9 ± 3.7 (range 20.0-30.0) |
| Aetiology | Post operative complication n = 6 Crohn’s disease n = 2 Mesenteric infarction n = 2 |
| Length of jejunum (cm) | 95 ± 48 (range 20-180) |
| Volume of parenteral support (L/d) | 1.8 ± 0.7 (range 1-3.5) |
| Sodium in parenteral support (mmol/d) | 166 ± 68 (range 60-300) |
| Baseline urine sodium (mmol/L) | 55 ± 28 (range 21-95) |
Fig. 2.
Consort diagram.
Results of the balance studies are shown in Table 3, and Fig. 3, Fig. 4. Jejunal wet weight (P < 0.001) and sodium (P = 0.003) were significantly higher on Vital 1.5® compared to Ensure Plus® with a mean difference of 0.1 kg and 13 mmol respectively. There was no difference in jejunal sodium balance (P = 0.16). There was no difference in urine sodium (P = 0.38) or urine volume (P = 0.56).
Table 3.
Results (mean ± SD) of the balance studies.
| Ensure Plus® | Vital 1.5® | Mean difference (95 %CI) | P value | |
|---|---|---|---|---|
| Jejunal wet weight (kg) | 0.33 ± 0.15 | 0.41 ± 0.20 | 0.10 (0.05, 0.16) | < 0.001 |
| Jejunal sodium (mmol) | 34 ± 13 | 44 ± 21 | 13 (4, 21) | 0.003 |
| Jejunal sodium balance | 26 ± 13 | 29 ± 21 | 6 (−2, 14) | 0.16 |
| Urine sodium (mmol) | 70 ± 48 | 82 ± 59 | 8 (−10, 26) | 0.38 |
| Urine volume (L) | 0.27 ± 0.08 | 0.27 ± 0.10 | -0.02 (−0.08, 0.04) | 0.56 |
Fig. 3.
Differences in jejunal wet weight between the two oral nutritional supplements.
Fig. 4.
Differences in jejunal sodium between the two oral nutritional supplements.
Discussion
This is the first study in patients with a jejunostomy to show that a semi-elemental ONS containing peptides and MCT, resulted in higher jejunal wet weight and sodium loss than a polymeric ONS containing polypeptides and LCT. Whilst not large, the increased losses with a semi-elemental ONS will cause significant fluid and sodium depletion with time. This provides evidence to encourage the administration of a polymeric ONS over a semi-elemental in patients with a jejunostomy.
Patients with short bowel are at risk of malnutrition due to malabsorption and can become malnourished quickly. Some patients with short bowel can maintain their nutritional status for years but then slowly deteriorate and become malnourished. If this occurs, patients may benefit from an ONS, enteral or parenteral nutrition. Enteral tubes and intravenous catheters are associated with complications (e.g., misplacement, infection and thrombosis). Therefore, all attempts to maintain patients on oral nutrition should be attempted.
In the 1970′s it was presumed that patients with short bowel were unable to digest a polymeric feed and that elemental feed would be beneficial due to the lack of digestion required. Small observational case reports and cohort studies assessed the effect of polymeric, semi-elemental and elemental feeds on nutrient absorption although none were randomised controlled trials [10], [11]. Levy et al. (1988) demonstrated improvements in nitrogen absorption with semi-elemental enteral nutrition in post-operative patients with short bowel. Mean jejunal wet weight did not increase above baseline (2.4 L/day) before enteral nutrition was instigated and gradually decreased to 1.4 L/day after 45 days of treatment. Despite not being a comparative study, they concluded that an elemental formula was not required and undesirable due to the high osmolarity [12].
As fluid and sodium losses dominate the issues, less attention has been given to the way macronutrients are administered. Previously the focus has been on the amount and type of fat given. Whatever the type of fat, the proportion of fat absorbed remains constant so a high fat diet is encouraged [13]. Jeppesen & Mortensen (1998) conducted a randomised controlled trial in which patients received a diet containing either 100 % LCT or a 50 % substitution of LCT with MCT [14]. In 9 patients with a stoma (small bowel length 203 ± 69 cm, range 125–300 cm) the addition of MCT resulted in an increase in stoma volume of 0.5 L/24 h. Therefore, the MCT in Vital 1.5® compared to the LCT in Ensure Plus® may contribute to the increased losses.
McIntyre et al. (1986) compared elemental versus polymeric ONS in 7 patients and found no significant differences in macronutrient absorption, jejunal wet weight or sodium losses between the two [5]. Cosnes et al. (1992) compared a polymeric and semi-elemental enteral feed and demonstrated improved nitrogen absorption on the semi-elemental compared to the polymeric feed although no other significant differences were observed [6]. These contrasting results in nitrogen absorption are likely to be due to the composition of the formulas investigated as the patients and the volumes infused were similar.
A prospective double blind randomised controlled crossover trial in infants with short bowel comparing a hydrolysed verses a non-hydrolysed enteral formula demonstrated no difference in weight gain, intestinal permeability or nitrogen balance [15]. The study did not measure jejunal wet weight and sodium losses but it does provide evidence that non-hydrolysed (polymeric) feeds are a suitable choice for infants with short bowel.
Concern that enteral nutrition could increase faecal losses in short bowel has been refuted. Joly et al. (2009) conducted a randomised controlled trial assessing the impact of the addition of 1000 kcal of a polymeric fibre free formula via a nasogastric tube over a 3-day period [16]. This study included 11 patients with a jejunocolic anastomosis and 4 with a jejunostomy (small bowel length 25–130 cm). The study demonstrated significantly increased absorption of energy, nitrogen and fat without increasing faecal losses. The continuous method of nasogastric feeding was credited with improving intestinal absorption and should be a treatment option for patients before proceeding to parenteral nutrition.
Much of the previous research in short bowel is confounded by including both patients with a jejunostomy and a jejunocolic anastomosis [3], [7], [12], [16]. It is known that patients with a jejunostomy benefit from a high fat diet compared to those with a jejunocolic anastomosis who benefit from a low-fat diet [17].
There is a lack of randomised controlled trials and over reliance on cohort studies or observational studies to inform practice. Other issues include a lack of standardised conditions during observations which render many results unreliable.
In this study the main difference between the two ONS was that Vital 1.5® consists of peptide and MCT whereas Ensure Plus® consists of polypeptides and LCT. We think that the reason for the increased wet weight and sodium loss via the stoma is most likely due to the differences in the protein source (i.e., polypeptide vs peptide) or the fat source (i.e., LCT vs MCT); though the small differences in osmolarity and sodium content could contribute.
All previous studies conducted have been short-term and it is not known if patients were able to improve or maintain their nutritional, hydrational and electrolyte status in the long term on oral nutritional supplements or enteral nutrition.
Based upon this study and previous work a polymeric fibre free high fat (LCT and MCT) feed should be the first choice for an ONS in a patient with an end jejunostomy. This will minimise the wet weight and sodium losses from the jejunostomy. Further balance studies are required to assess the effect of different ONSs on macronutrient, electrolyte and fluid status over longer periods of time.
Limitations
The study design instructed patients to consume 200 ml of the ONS over 30 min. In clinical practice patients are advised to sip ONS slowly to aid absorption but 30 min is not uncommon. The patient conditions were not all standardised; the volume and sodium content of the parenteral fluid and nutrition patients received were not the same but were constant for each patient throughout the 3 days of study; and the food and fluid intake during the washout period was not standardised. There was no evidence of a carryover effect. A 6-hour collection is short but work by Rodrigues et al. (1989) showed that 98 % of a marker is collected by this time [7]. It is not clear which components of the ONS increased the output as there were differences between the two formulations (e.g., fat component and protein source). There was also a difference in the sodium content (7.35 mmol vs 4 mmol/100mls) which may have affected results but these were the closest comparative feeds we could find.
The macronutrient content of the jejunostomy output was not measured. We did not collect any patient reported outcomes such as adverse side effects or taste preferences.
Conclusion
A semi-elemental ONS results in greater jejunal and sodium loss compared to a polymeric ONS. Whilst the amount was small, the daily cumulative effect will be much greater.
Funding
The study was funded by the St Mark’s Foundation. They were not involved in the study design, data collection or analysis and did not have any input into the manuscript.
CRediT authorship contribution statement
Alison culkin: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Beth Rye: Data curation. Jeremy Mark Darcy Nightingale: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
We acknowledge Paul Bassett for statistical help. Thanks to the nurses on the unit for their help and to the patients who agreed to participate.
Consent to publish
Patients consented to publishing the data with the knowledge that their identity would not be disclosed.
Contributor Information
Alison Culkin, Email: alisonculkin@nhs.net.
Beth Rye, Email: bethrye@nhs.net.
Jeremy MD Nightingale, Email: jnight@globalnet.co.uks.
References
- 1.Nightingale J.M., Lennard-Jones J.E., Walker E.R., Farthing M.J. Jejunal efflux in short bowel syndrome. Lancet. 1990;336:765–768. doi: 10.1016/0140-6736(90)93238-k. PMID: 1976145. [DOI] [PubMed] [Google Scholar]
- 2.Nightingale J., Woodward J.M. Guidelines for management of patients with a short bowel. Gut. 2006;55(Suppl 4):iv1–iv12. doi: 10.1136/gut.2006.091108. (Small Bowel and Nutrition Committee of the British Society of Gastroenterology) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Woolf G.M., Miller C., Kurian R., Jeejeebhoy K.N. Nutritional absorption in short bowel syndrome. Evaluation of fluid, calorie, and divalent cation requirements. Dig Dis Sci. 1987;32:8–15. doi: 10.1007/BF01296681. PMID: 3792183. [DOI] [PubMed] [Google Scholar]
- 4.Messing B., Pigot F., Rongier M., Morin M.C., Ndeïndoum U., Rambaud J.C. Intestinal absorption of free oral hyperalimentation in the very short bowel syndrome. Gastroenterology. 1991;100:1502–1508. doi: 10.1016/0016-5085(91)90645-2. PMID: 1850371. [DOI] [PubMed] [Google Scholar]
- 5.McIntyre P.B., Fitchew M., Lennard-Jones J.E. Patients with a high jejunostomy do not need a special diet. Gastroenterology. 1986;91:25–33. doi: 10.1016/0016-5085(86)90434-8. PMID: 3011582. [DOI] [PubMed] [Google Scholar]
- 6.Cosnes J., Evard D., Beaugerie L., Gendre J.P., Le Quintrec Y. Improvement in protein absorption with a small-peptide-based diet in patients with high jejunostomy. Nutrition. 1992;8:406–411. doi: 10.1016/0016-5085(86)90434-8. PMID: 3011582. [DOI] [PubMed] [Google Scholar]
- 7.Rodrigues C.A., Lennard-Jones J.E., Thompson D.G., Farthing M.J. The effects of octreotide, soy polysaccharide, codeine and loperamide on nutrient, fluid and electrolyte absorption in the short-bowel syndrome. Aliment Pharm Ther. 1989;3:159–169. doi: 10.1111/j.1365-2036.tb00202.x. PMID: 2491467. [DOI] [PubMed] [Google Scholar]
- 8.Newton C.R., Gonvers J.J., McIntyre P.B., Preston D.M., Lennard-Jones J.E. Effect of different drinks on fluid and electrolyte losses from a jejunostomy. J R Soc Med. 1985;78:27–34. doi: 10.1177/014107688507800106. PMID: 3968667; PMCID: PMC1289541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Culkin A., Gabe S.M., Nightingale J.M.D. A new palatable oral rehydration solution: a randomised controlled cross-over study in patients with a high output stoma. J Hum Nutr Diet. 2022;35:94–101. doi: 10.1111/jhn.12935. PMID: 34323341. [DOI] [PubMed] [Google Scholar]
- 10.Simko V., Linscheer W.G. Absorption of different elemental diets in a short-bowel syndrome lasting 15 years. Am J Dig Dis. 1976;21:419–425. doi: 10.1007/BF01072668. PMID: 820189. [DOI] [PubMed] [Google Scholar]
- 11.Simko V., McCarroll A.M., Goodman S., Weesner R.E., Kelley R.E. High-fat diet in a short bowel syndrome. Intestinal absorption and gastroenteropancreatic hormone responses. Dig Dis Sci. 1980;25:333–339. doi: 10.1007/BF01308056. PMID: 6989573. [DOI] [PubMed] [Google Scholar]
- 12.Levy E., Frileux P., Sandrucci S., Ollivier J.M., Masini J.P., Cosnes J., Hannoun L., Parc R. Continuous enteral nutrition during the early adaptive stage of the short bowel syndrome. Br J Surg. 1988;75:549–553. doi: 10.1002/bjs.1800750615. PMID: 3134973. [DOI] [PubMed] [Google Scholar]
- 13.Crenn P., Morin M.C., Joly F., Penven S., Thuillier F., Messing B. Net digestive absorption and adaptive hyperphagia in adult short bowel patients. Gut. 2004;53:1279–1286. doi: 10.1136/gut.2003.030601. PMID: 15306586; PMCID: PMC1774175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jeppesen P.B., Mortensen P.B. The influence of a preserved colon on the absorption of medium chain fat in patients with small bowel resection. Gut. 1998;43:478–483. doi: 10.1136/gut.43.4.478. PMID: 9824573; PMCID: PMC1727269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ksiazyk J., Piena M., Kierkus J., Lyszkowska M. Hydrolyzed versus nonhydrolyzed protein diet in short bowel syndrome in children. J Pedia Gastroenterol Nutr. 2002;35:615–618. doi: 10.1097/00005176-200211000-00005. PMID: 12454574. [DOI] [PubMed] [Google Scholar]
- 16.Joly F., Dray X., Corcos O., Barbot L., Kapel N., Messing B. Tube feeding improves intestinal absorption in short bowel syndrome patients. Gastroenterology. 2008;136:824–831. doi: 10.1053/j.gastro.2008.10.084. PMID: 19046971. [DOI] [PubMed] [Google Scholar]
- 17.Nordgaard I., Hansen B.S., Mortensen P.B. Colon as a digestive organ in patients with short bowel. Lancet. 1998;343(8894):373–376. doi: 10.1016/s0140-6736(94)91220-3. PMID: 7905549. [DOI] [PubMed] [Google Scholar]