What Is Known
- Most patients with CF have an exocrine pancreatic insufficiency leading to fat malabsorption and malnutrition.
- A good nutritional status contributes to an improved pulmonary function and survival.
- Pancreatic enzyme replacement therapy (PERT) is the proven therapy to substantially reduce fat malabsorption.
What Is New
- A coefficient of fat absorption (CFA) outcome below 85% was found in more than 20% of the measurements.
- There is an enormous variability in the response to PERT among patients, with no clear correlation between the CFA and PERT.
- Patients with a CFA consistently below 85% had a body weight below the group average.
In most patients with cystic fibrosis (CF), the absence or dysfunction of the CF transmembrane regulator in pancreatic duct cells results in exocrine pancreatic insufficiency (1), causing an inadequate digestion and leading to fat malabsorption and malnutrition. As a good nutritional status contributes to an improved pulmonary function and survival (2,3), the major aim in CF care is to prevent malabsorption. To prevent malabsorption, pancreatic enzyme replacement therapy (PERT) is the proven and, therefore, essential therapy, as it will substantially reduce fat malabsorption (4,5). The appropriate PERT generally results in a fat absorption of more than 85% in patients with CF and more than 90% in many patients (6). However, very few details of the daily practice on PERT and the coefficient of fat absorption (CFA) in large groups of patients with CF are known, as previous studies were rather small and were mainly conducted in the context of a trial (6). We therefore set out to record the daily practice of PERT and CFA during regular follow-ups of a large cohort of pancreatic insufficient pediatric patients with CF. Subsequently, we analyzed PERT and CFA in a subgroup of patients receiving enteral tube feeding. Finally, we addressed the effect of the interventions done in patients with CFA outcomes <85%.
This retrospective study includes Dutch children (born between 1988 and 2012) with proven CF and who received medical care in the CF Centre of the University Medical Centre Utrecht. Each child was confirmed as having CF by a positive sweat test and the presence of 2 CF mutations, as well as the clinical signs of CF and a positive family history.
Pancreatic insufficiency was confirmed in children up to 4 years of age, as having a fecal elastase concentration <15 μg g−1 stool (normal values: >200 μg g−1 stool) or chymotrypsin activity <3 U g−1 stool (normal values: >6.0 U g−1 stool), and in children 4 years and older as having a documented history of fat malabsorption with a CFA outcome <85%, and a fecal elastase concentration <15 μg g−1 stool or chymotrypsin activity <3 U g−1 stool. The body weight and body height were measured during each routine clinical care visit. Yearly, the dietary data were collected through 3-day dietary food records, which also included a registration of pancreatic enzyme ingestion. Simultaneously, a stool collection was done to measure the fecal fat excretion.
Generally, in newly diagnosed patients a starting dosage of 5000 LU kg−1 day−1 was advised. If a CFA below 85% was found, several options were available, that is, increasing the lipase dose and starting or increasing proton pump inhibitor (PPI), which were chosen by the physician in charge. The exact steps or its order were not defined in a protocol, however.
In all of the patients, the PERT dosage was primarily tailored to their individual dietary fat intake and subsequently adjusted when CFA was found to be below 85%. Thus, the study's database contained longitudinal data regarding the clinical parameters and the demographics of all of the patients who received medical care for CF at the University Medical Centre Utrecht from 1996 until December 2013. All of the parents provided a written informed consent for the storage and analysis of the data of their children. The study was performed in accordance with the guidelines of the medical ethics board of the University Medical Centre Utrecht.
The body weight, expressed in kilograms, was measured with a digital weight balance to the nearest 0.1 kg, and the body height, expressed in centimeters, was measured with a stadiometer (Holtain, Crymich, UK) to the nearest 0.5 cm. Yearly, all of the patients with CF received written instructions on completing a 3-day dietary food record, including a registration of pancreatic enzyme ingestion. The children and parents were encouraged to maintain the child's usual dietary intake. They have been advised to administer an individually worked out 1 single dosage of the pancreatic enzymes before every fat-containing meal, including snacks, fat-containing drinks, and the enteral tube feeding. On each yearly evaluation, the PERT was evaluated and advices were given to match the PERT dosage to the quantity of fat consumed and to administer PPIs.
All of the food and beverages consumed in portion sizes or weights, and the quantity of the pancreatic enzymes actually taken with each meal, were recorded during 2 weekdays and 1 weekend day whenever possible. Where weights were not specified, the portion size weights were obtained from the reference data. The enteral tube feeding was recorded as type and volume within the diaries. Registered dieticians coded and analyzed the food records according to a standardized approach using the Dutch Food Composition Table (2010) of the Dutch Nutrition Centre (7). The mean daily dietary fat intake was calculated for each assessment and expressed as both grams and energy percentage (En%) of the total energy intake. The recorded pancreatic enzyme ingestion varied in the amount of lipase unit (LU) per dosage and reported as the total LU intake per day, LU per gram of fat per day, and LU per kilogram per day.
A fat balance study was performed to measure the fat excretion in feces, analyzed by near infrared spectroscopy, and to calculate the CFA. To this effect, in conjunction with the 3-day dietary food record, a home-based 72-hour stool collection was obtained, starting on day 2 and ending on day 4, to determine the mean fecal fat content of this 3-day collection. The CFA was then calculated from the mean dietary fat intake of the 3-day dietary record and the mean pooled fecal fat output, and expressed as percentage.
The descriptive statistics of the categorical variables of the study sample were examined. Individual children had repeated CFA measurements at different years of age. So that different numbers of measurements did not bias the results, the children were stratified according to age year (eg, 0 year = birth to < 1 year, 1 year = 1 to < 2 years) for the cross-sectional analyses. We described the mean (± standard deviation) of body weight, dietary fat intake, and the median (25th–75th percentile) of LU per day, LU per gram of fat per day, LU per kilogram per day, and CFA for the total group and apart for those who have enteral tube feeding. We then determined the cross-sectional relation between LU per gram of fat per day, LU per kilogram per day, and CFA for the total group and apart for those with a CFA outcome <85% by using the Spearman's correlation coefficient. We also compared the CFA between those with and without enteral tube feeding by using the Mann-Whitney test. Subsequently, we compared the intake of LU per gram of fat per day and LU per kilogram per day among the categorized groups of CFA by using the Kruskal-Wallis test. For this purpose, the children were subdivided based on their CFA as having an outcome <85%, between 85% and 90%, between 90% and 95%, and >95%. Additionally, the effect of the interventions done in patients with CFA outcomes <85% was addressed. For this purpose, the patients with at least 2 consecutive yearly measurements, with initial CFA outcomes <85%, were included. We described the intervention given and compared the age, PERT, nutritional status expressed as z scores weight for age, height for age, and weight for height, the use of PPIs and ursodeoxycholic acid (UDCA) among patients with CFA outcomes <85% and ≥85% after an intervention done by using the Student t test, Mann-Whitney test, or chi-square test. The statistical analyses were performed using the Statistical Package for the Social Sciences program (SPSS Inc, version 20, IBM SPSS Statistics, Armonk, NY). All of the values were considered significant at P < 0.05.
The data of 224 patients (98% white, 48% women) with CF were analyzed. All were confirmed to be pancreatic insufficient and had PERT. A total of 1719 completed 3-day dietary food records, including the pancreatic enzyme intake registrations, and 1373 completed CFA assessments were obtained. The study enrolled 32 patients (accounted for 234 food records and 194 CFA measurements) diagnosed by a meconium ileus, 44 patients (148 records and 104 CFA) who were provided enteral tube feeding, 107 patients (665 records and 554 CFA) who had been prescribed PPIs, 67 patients (375 records/67 CFA) who had been prescribed UDCA, of which 3 patients (11 dietary food records/9 CFA) were known to have liver cirrhosis.
Dietary Fat Intake and PERT
The included patients were stratified according to age. Patients obtained between 34% and 36 En% of the total energy intake from fat. With the exception of infants, PERT, expressed as median LU per gram of fat per day, seems relatively stable. When PERT was expressed as LU per kilogram per day, the children up to 9 years of age had overall intakes >5000 LU per kilogram per day, and thereafter the supplementation fell markedly with age (Fig. 1 displays PERT for each year of age, see also Supplemental Table 1 for detailed information [https://links.lww.com/MPG/A445]). A PERT intake >10,000 LU kg−1 day−1 was found in 26 measurements (16 children).
The CFA varied between 86% and 91%, with a large range in the outcomes in the different age groups (Fig. 2 displays the CFA for each year of age, see also Supplemental Table 1 [https://links.lww.com/MPG/A445] for detailed information). For patients who had both a PERT and a CFA measurement, we found no relation between either LU per gram of fat per day and CFA or LU per kilogram per day and CFA, apart from the age group 17 (rs = 0.526, P = 0.04). Also in 325 records (24%), in which the CFA was <85%, we found no correlation between the PERT and CFA. Subsequently, the PERT was analyzed in 4 categories: CFA <85%, between 85% and 90%, between 90% and 95%, and >95%, in which we found the same distribution of PERT (P ≥ 0.055).
PERT and CFA in Patients Receiving Enteral Tube Feeding
All patients receiving (nocturnal) enteral tube feeding used a nonelemental, which provided up to 9%–99% of the total energy intake. The patients receiving enteral tube feeding had more or less similar intakes, varying over the age groups between 1359 and 2043 LU g−1 fat day−1, and 3959 and 7627 LU−1 kg−1 day−1. The CFA varied between 91% and 96% and was higher in the children with the enteral tube feeding, and significantly so in the age groups 3, 5, 6, 7, 9, and 12 (P ≤ 0.024) (Fig. 3).
Interventions in Patients With Consecutive Yearly CFA Measurements and a CFA <85% in the First Measurement
In 170 measurements, at least 2 consecutive yearly measurements, with a CFA <85% in the first measurement, were available, enabling us to study the effect of interventions done (Table 1). After the first intervention, in which PERT and PPI dosage was adjusted in almost all patients (Table 1), the CFA remained below <85% in 57 measurements. Subsequently, follow-up data of a second intervention was available for 47 out of the 57 measurements. In these patients, the CFA remained <85% in 19 measurements (Table 2). Follow-up data of a third intervention was available for 18 out of the 19 measurements (Table 1). Despite an increase in PERT and PPIs, the CFA remained <85% in 11 measurements. We found no differences in age, or the use of PERT, PPIs, and UDCA between patients with a CFA ≥85% and those with a CFA <85%. However, patients with a CFA <85% during 3 consecutive yearly measurements had significantly lower z scores for weight for age and weight for height (P = 0.01) than those with a CFA ≥85% after the third intervention (Table 2).
The present study, including 224 patients who completed 1719 3-day dietary food records and 1373 CFA measurements, is the largest study at present available describing the daily use of PERT and the resulting CFA as measured in a 72-hour stool collection in a cohort of patients with CF. With the CFA outcomes between 86% and 91% throughout the age groups, although with large ranges in outcomes, our results are in concordance with the previous results (6,8,9). An enormous variability in the response to PERT among patients was found (10–12), with no clear correlation between the CFA and PERT, as previously reported (11–13). The PERT dosage, expressed as LU per gram of fat per day, was fairly constant, but gradually became lower when expressed as LU per kilogram per day, as a consequence of the lower caloric and fat intake per kilogram body weight with increasing age. The present international guidelines aim to keep PERT below 10,000 LU kg−1 day−1(4,5) to avoid fibrosing colonopathy (14). In the present study, in less than 2% of the measurements an intake in excess of this limit was found. It is unlikely that increasing PERT beyond the threshold of 10,000 LU kg−1 day−1(4,5) will produce any benefit with respect to either fat malabsorption (15,16) or growth (17). This is supported by a study in which children who initially received more than 11,000 LU kg−1 day−1 reduced their PERT dosage to less than 10,000 LU kg−1 day−1, but still had CFA outcomes of 92%, while growth parameters even improved (18).
High-strength preparations were used in 7% (121 out of 1719 measurements), with only 1 patient exceeding the 10,000 LU kg−1 day−1 threshold. Therefore, the suggestion that the use of high-strength enzyme preparations carries a greater risk of exceeding the upper limit (19) seems at variance with our results. However, the low frequency we found may be because of the clinical controlled dosage, as opposed to the self-determined intake in the earlier study (19), although this warrants further clinical research.
Our patients with (nocturnal) nonelemental enteral tube feeding were advised to ingest PERT in a single dosage at the beginning of the tube feeding. Nevertheless, despite the often long infusion period, these patients had better CFA outcomes than the group without the tube feeding, that could be related to the liquid emulsion form of enteral tube feeding, which may provide a more optimal substrate for pancreatic enzyme activity. The present study supports the common advice of a single dosage of enzymes before the beginning of the (nocturnal) nonelemental enteral tube feeding.
It is believed that with close monitoring, PERT should be able to improve the CFA to more than 85% (5). However, in the present study, it failed in 24% of the CFA measurements, although PERT was similar in both the groups with a CFA <85% and a CFA ≥85%. A substandard individual response to PERT, resulting in a CFA below 85%, can be because of the lack of adherence to the treatment (20). Other possible causes are small bowel bacterial overgrowth because of the antibiotic treatment, which is seen in a substantial proportion of patients with CF (21) and mucosal disorders, including celiac disease (22). Constipation, which is often seen in patients with CF, is associated with lower CFA outcomes (23). Likely, in these situations, a laxative treatment may improve the CFA, as it normalizes the intestinal transit time; in addition, it may improve the bacterial overgrowth (24,25). The individual variation in the fecal bile salt loss, which is seen in CF (26), may lead to a reduced solubilization capacity of bile, resulting in fat malabsorption. However, this suggestion is not supported by the study results, describing that the fecal bile salt malabsorption does not contribute to fat malabsorption (26,27). Furthermore, it could be hypothesized that patients receiving UDCA, which was prescribed to patients known with CF-related liver disease (28), treatment have a decreased fat absorption, as UDCA is less hydrophobic and has an impaired capability in forming mixed micelles (29). Nevertheless, the results of the relatively small studies did not find an association between the UDCA treatment and CFA (30,31), nor did our study. PPIs were prescribed in patients with gastroesophageal reflux disease (32) and in most patients with suspected fat malabsorption. In the latter, the low intraluminal pH will compromise the pancreatic lipase activity because of a delay in the dissolution of the acid-resistant enteric coating (33). Increasing the gastric pH to a level that is favorable for the optimal release of enzymes by the concomitant use of PPIs may protect the inactivation of the enzymes and improve efficacy (4). However, a significant positive effect of the gastric acid suppressant therapy on CFA is not fully demonstrated (9,34,35) as was also found in the present study; patients receiving PPIs still had recurrent CFA <85%. The PPIs possibly only improves the efficacy of PERT in patients with a low duodenal pH and will not affect the CFA outcomes in those with an optimum pH for enzyme release already.
The calculation of the CFA is based on the assessment of the difference between the average 3-day dietary fat intake and the average daily fecal fat output of the 72-hour stool collection. Therefore, the accuracy of the CFA outcome may be influenced by overreporting and underreporting of the dietary (fat) intake, or by the failure to collect the stool for the entire period, whereby the excreted fat will not be completely measured. In addition, we could not objectively measure the adherence to PERT or the timing of PERT ingestion in relation to the meals, which is another limitation of the present study.
PERT in patients with CF is assumed to result in a CFA above 85%; nevertheless, in the present study, this threshold was not passed in more than 20% of the measurements. Furthermore, no correlation between the enzyme dosage and the degree of fat malabsorption was found, which may be the result of the factors residing in the gastrointestinal tract (eg, persistent low intraluminal pH, slow transit, and abnormal bile acid metabolism) or outside the gastrointestinal tract (eg, insufficient compliance).
The authors thank all of the participating children with CF who made the present study possible.
1. O'Sullivan BP, Freedman SD. Cystic fibrosis. Lancet
2. Corey M, McLaughlin FJ, Williams M, et al A comparison of survival, growth, and pulmonary function in patients with cystic fibrosis in Boston and Toronto. J Clin Epidemiol
3. Yen EH, Quinton H, Borowitz D. Better nutritional status in early childhood is associated with improved clinical outcomes and survival in patients with cystic fibrosis. J Pediatr
4. Borowitz D, Baker RD, Stallings V. Consensus report on nutrition for pediatric patients with cystic fibrosis. J Pediatr Gastroenterol Nutr
5. Sinaasappel M, Stern M, Littlewood J, et al Nutrition in patients with cystic fibrosis: a European Consensus. J Cyst Fibros
6. Littlewood JM, Wolfe SP, Conway SP. Diagnosis and treatment of intestinal malabsorption in cystic fibrosis. Pediatr Pulmonol
7. Nevo Foundation. NEVO Table, Dutch Food Composition Table. The Hague: The Netherlands Nutrition Centre Foundation; 2006.
8. Borowitz D, Konstan MW, O’Rourke A, et al Coefficients of fat and nitrogen absorption in healthy subjects and individuals with cystic fibrosis. J Pediatr Pharmacol Ther
9. Francisco MP, Wagner MH, Sherman JM, et al Ranitidine and omeprazole as adjuvant therapy to pancrelipase to improve fat absorption in patients with cystic fibrosis. J Pediatr Gastroenterol Nutr
10. Cohen JR, Schall JI, Ittenbach RF, et al Fecal elastase: pancreatic status verification and influence on nutritional status in children with cystic fibrosis. J Pediatr Gastroenterol Nutr
11. Kalnins D, Corey M, Ellis L, et al Combining unprotected pancreatic enzymes with pH-sensitive enteric-coated microspheres does not improve nutrient digestion in patients with cystic fibrosis. J Pediatr
12. Kalnins D, Ellis L, Corey M, et al Enteric-coated pancreatic enzyme with bicarbonate is equal to standard enteric-coated enzyme in treating malabsorption in cystic fibrosis. J Pediatr Gastroenterol Nutr
13. Durie P, Kalnins D, Ellis L. Uses and abuses of enzyme therapy in cystic fibrosis. J R Soc Med
1998; 91 (suppl 34):2–13.
14. Littlewood JM, Wolfe SP. Control of malabsorption in cystic fibrosis. Paediatr Drugs
15. Dodge JA. Pancreatic enzyme therapy in cystic fibrosis. Expert Rev Respir Med
16. Van de Vijver E, Desager K, Mulberg AE, et al Treatment of infants and toddlers with cystic fibrosis-related pancreatic insufficiency and fat malabsorption with pancrelipase MT. J Pediatr Gastroenterol Nutr
17. Baker SS, Borowitz D, Baker RD. Pancreatic exocrine function in patients with cystic fibrosis. Curr Gastroenterol Rep
18. Lowdon J, Goodchild MC, Ryley HC, et al Maintenance of growth in cystic fibrosis despite reduction in pancreatic enzyme supplementation. Arch Dis Child
19. Schibli S, Durie PR, Tullis ED. Proper usage of pancreatic enzymes. Curr Opin Pulm Med
20. Lisowska A, Kaminska B, Grzymislawski M, et al Unresponsive or non-compliant steatorrhea in cystic fibrosis? J Cyst Fibros
21. Fridge JL, Conrad C, Gerson L, et al Risk factors for small bowel bacterial overgrowth in cystic fibrosis. J Pediatr Gastroenterol Nutr
22. Borowitz D. Update on the evaluation of pancreatic exocrine status in cystic fibrosis. Curr Opin Pulm Med
23. Van der Doef HP, Kokke FT, Beek FJ, et al Constipation in pediatric cystic fibrosis patients: an underestimated medical condition. J Cyst Fibros
24. De Lisle RC, Roach E, Jansson K. Effects of laxative and N-acetylcysteine on mucus accumulation, bacterial load, transit, and inflammation in the cystic fibrosis mouse small intestine. Am J Physiol Gastrointest Liver Physiol
25. De Lisle RC, Sewell R, Meldi L. Enteric circular muscle dysfunction in the cystic fibrosis mouse small intestine. Neurogastroenterol Motil
26. O’Brien S, Mulcahy H, Fenlon H, et al Intestinal bile acid malabsorption in cystic fibrosis. Gut
27. Strandvik B, Einarsson K, Lindblad A, et al Bile acid kinetics and biliary lipid composition in cystic fibrosis. J Hepatol
28. Debray D, Kelly D, Houwen R, et al Best practice guidance for the diagnosis and management of cystic fibrosis-associated liver disease. J Cyst Fibros
2011; 10 (Suppl. 2):S29–S36.
29. Armstrong MJ, Carey MC. The hydrophobic–hydrophilic balance of bile salts. Inverse correlation between reverse-phase high performance liquid chromatographic mobilities and micellar cholesterol-solubilizing capacities. J Lipid Res
30. Colombo C, Battezzati PM, Podda M, et al Ursodeoxycholic acid for liver disease associated with cystic fibrosis: a double-blind multicenter trial. The Italian Group for the Study of Ursodeoxycholic Acid in Cystic Fibrosis. Hepatology
31. Merli M, Bertasi S, Servi R, et al Effect of a medium dose of ursodeoxycholic acid with or without taurine supplementation on the nutritional status of patients with cystic fibrosis: a randomized, placebo-controlled, crossover trial. J Pediatr Gastroenterol Nutr
32. Van der Doef HP, Arets HG, Froeling SP, et al Gastric acid inhibition for fat malabsorption or gastroesophageal reflux disease in cystic fibrosis: longitudinal effect on bacterial colonization and pulmonary function. J Pediatr
33. Hendriks JJ, Kester AD, Donckerwolcke R, et al Changes in pulmonary hyperinflation and bronchial hyperresponsiveness following treatment with lansoprazole in children with cystic fibrosis. Pediatr Pulmonol
34. Sander-Struckmeier S, Beckmann K, Janssen-van SG, et al Retrospective analysis to investigate the effect of concomitant use of gastric acid-suppressing drugs on the efficacy and safety of pancrelipase/pancreatin (CREON(R)) in patients with pancreatic exocrine insufficiency. Pancreas
35. Ng SM, Jones AP. Drug therapies for reducing gastric acidity in people with cystic fibrosis. Cochrane Database Syst Rev