What Is Known
A significant subset of patients with cystic fibrosis experience abdominal symptoms despite sufficient pancreatic enzyme replacement therapy.
Fecal calprotectin has been used as an indicator for intestinal inflammation.
Smaller studies have reported intestinal inflammation in many patients with cystic fibrosis.
What Is New
Two-thirds of >2400 stool specimens showed increased fecal calprotectin values.
Fecal calprotectin values were higher in subjects with pancreatic insufficiency, whereas no relevant associations with abdominal symptoms, forced expiratory volume in 1 second, or body mass index were observed.
The applicability of increased fecal calprotectin concentrations as an indicator for intestinal inflammation in cystic fibrosis remains to be determined.
The majority of patients with cystic fibrosis (CF) have pancreatic insufficiency. Despite high-dose pancreatic enzyme replacement therapy to treat malabsorption, many children and adult patients do not achieve normal body weight (1)
Werlin et al (2) (3)
Calprotectin is an inflammatory marker derived from neutrophils, which can be measured in different body fluids and specimens (4) (5) (4) (6) (7,8)
FC has also been evaluated in small groups of patients with CF. Increased values were observed with mean FC concentrations between 70 (9) (10) (10,11) (9) (2) (4)
To date, no repeated measurements of FC in patients with CF have been published. This excludes studies evaluating the effects of certain treatments on FC (9,10)
METHODS
Cystic Fibrosis Center, Innsbruck
The Cystic Fibrosis Center at Innsbruck University is a certified center for children, adolescents and adults with CF. Highly structured outpatient clinic visits involving all CF team members are performed every 3 months (12)
Study Design
We performed a longitudinal retrospective analysis of FC and other values that were routinely collected as part of regular clinical care according to standardized diagnostic procedures used at our center. Data from all patient encounters, both outpatient visits and hospital treatments, are stored in a patient database, which was established in 1995. The present analysis includes the period between July 2003 and August 2015. Ethics approval was obtained from the ethics committee at Medical University Innsbruck (UN5050SN324/4.10).
Participants
Since routine measurements of FC concentrations were established in 2003, all patient encounters after June 2003 were eligible for analysis. We included data from patients who had at least 1 documented FC measurement in the patient database.
Fecal Calprotectin Measurements
Patients collected stool samples at home before visiting the CF clinic for a routine assessment. FC measurements were performed 3 days a week by the laboratory using a commercially available enzyme-linked immunosorbent assay kit (Calprest, Eurospital, Italy, sensitivity 95%, specificity 93%, http://www.calprotectintest.com/english/calprest.html (13)
Other Variables and Normal Values
Data evaluated for the present analysis included age, sex, cystic fibrosis transmembrane conductance regulator (CFTR) mutation (F508del homozygosity, F508del heterozygosity, or other mutations), pancreatic insufficiency (defined as fecal elastase value <100 μg/g stool), pancreatic enzyme replacement therapy, body weight (kilograms), height (centimeters), and body mass index (BMI) expressed as z score (14) (15) 1 ), expressed as z score (16) (17) Lactobacillus rhamnosus GG (LGG) therapy.
Inflammatory parameters in serum included C-reactive protein (CRP) and immunoglobulin G, measured by routine laboratory methods. Chronic infection with Pseudomonas aeruginosa was determined using the Leeds criteria (18)
Statistical Methods
We used linear mixed models to evaluate whether FC concentrations were associated with changes in age or disease parameters, for example, FEV1 z score, BMI z score or CFTR mutations. To maintain maximal generalizability and have a good tradeoff between flexibility and the number of parameters of our model, we chose the scaled identity matrix as the residual correlation structure. As we assume conditional independence between the subjects, we also used the scaled identity covariance matrix for the random effect. If the values showed a skewed rather than a symmetric distribution and ranged over several orders of magnitude, modeling was performed using log base 10 or log base 2 transformed values. This applied to FC as well as to CRP, immunoglobulin G, steatocrit, and fecal chymotrypsin.
We performed bivariate instead of multivariate analyses, as the presence of some variables (eg, steatocrit) with many missing values would lead to a removal of the respective observations from the multivariate model, and would thus result in a loss of information of the other available variables.
Repeated measurements analysis of variance were used to determine differences between means of certain patient groups, for example, values from female or male patients. Results with a significance level of P < 0.05 were considered statistically significant. R version 3.3.0 (R Core Team, 2016, R Foundation for Statistical Computing, Vienna, Austria) was used for statistical analysis, and the R-package ggplot2 for graphical analysis.
RESULTS
Participants
The study cohort comprised those 171 of 217 patients (79%) who had at least 1 FC value documented in the patient database between 2003 and 2015. At the most recent visit in 2015, half of the patients were adults (52.6%); the median age of the cohort was 18.9 years. Median BMI and FEV1 z scores were −0.61 and −1.29, respectively. Most patients (90.6%) were pancreatic insufficient. Further patient characteristics are displayed in Table 1 .
TABLE 1: Patient characteristics at the most recent visit in 2015
Fecal Calprotectin Concentrations
A total of 2434 FC measurements were documented from 171 patients during a total observation period of 1686 patient-years, with a median observation period of 7 years per patient. The median (interquartile range, Q1, Q3) FC concentration was 60.9 (27.5, 103.4) μg/g. Elevated concentrations >50 μg/g stool were observed in 61% of samples. Some patients displayed large fluctuations of FC values with time, particularly those with very high FC concentrations >500 μg/g. Calprotectin values did not differ between age groups (Fig. 1 ).
FIGURE 1: Fecal calprotectin (FC) concentrations (log10) during a 12-year period in N = 2434 specimens across age groups. Normal values are <50 μg/g, corresponding to a value of 1.7 log10Calprotectin on the y axis. The number of samples in each age group is depicted at the bottom of the x axis. FC concentrations did not change with age.
Male and female patients had comparable FC concentrations (median: 70 vs 66 μg/g, P = 0.16). Pancreatic insufficiency led to considerably higher FC values compared with normal pancreatic function (median 68 vs 29 μg/g, P < 0.0001, Fig. 2 ).
FIGURE 2: Fecal calprotectin (FC) concentrations (log10) in samples from patients with (N = 2061) and without (N = 176) pancreatic insufficiency. Pancreatic insufficiency was associated with significantly higher FC values than normal pancreatic function (median 68 vs 29 μg/g, P < 0.0001).
F508del homozygous subjects had higher median FC values than F508del heterozygous patients (median 71 vs 62 μg/g), but the difference was not statistically significant (P = 0.17). The dosage of pancreatic enzymes was not significantly associated with FC concentrations, when expressed as lipase units per day (P = 0.09) or as lipase units per kg body weight per day (P = 0.35). Furthermore, there was no association between FC concentrations and the severity of gastrointestinal symptoms, that is, bloating (P = 0.66), diarrhea (P = 0.25), and abdominal pain (P = 0.22).
Associations Between Fecal Calprotectin Concentrations and Disease Parameters
We performed a linear mixed model analysis with log10(FC) as the outcome variable and several disease parameters as independent variables (Table 2 ). Statistically significant associations among FC concentrations and age, BMI z score, or fecal elastase concentrations were observed. Most effect sizes were however not clinically meaningful (Table 2 ). For example, an increase in FEV1 z score of 1 was associated with a decrease in calprotectin of only −3.88%, and a 2-fold increase in CRP (log2 CRP) led to an increase in FC of 11% (Table 2 ).
Table 2: Linear mixed model analysis of log10 FC values
Fecal Calprotectin by Calendar Year
A significant trend toward lower FC concentrations was observed during the 12 year-observation period (Supplemental Digital Content 1, Fig., https://links.lww.com/MPG/A913 Table 1 ). The mixed model analysis revealed a significant 6.3% decline in FC values per calendar year (P < 2 × 10−16 ).
Longitudinal Changes in Patients With Increased Fecal Calprotectin Concentrations
A subgroup of 111 of the 171 patients had at least 1 stool specimen with a markedly increased FC content of >200 μg/g. When the first occurrence of such a value was used as a starting point, individual follow-up for up to 11 years showed a significant decrease (P < 0.0001) in FC of 10% per year (Supplemental Digital Content 2, Fig., https://links.lww.com/MPG/A914
About one-third of these patients were treated with LGG (2
capsules per day). Stool specimens from 28 patients on LGG therapy had 24.7% lower FC concentrations than those from untreated subjects (P = 0.0253) according to a linear mixed model analysis.
DISCUSSION
We report the first longitudinal analysis of FC values in patients with CF, comprising >2400 stool specimens from a median observation period of 7 years per patient. Two of 3 specimens showed increased FC concentrations >50 μg/g. Moreover, two-thirds of patients had markedly increased FC values (>200 μg/g) at least once during the observation period. In subjects with chronic inflammatory bowel disease, abnormal FC concentrations are related to intestinal inflammation. If we assume that this also applies to CF, the results of the present study suggest that intestinal inflammation is present in a considerable proportion of subjects with CF
This is the largest longitudinal collection of stool samples from CF patients reported in CF literature. Previous authors evaluated only small groups of patients with CF, with either a cross-sectional design or a second FC determination after several weeks, for example, before and after therapeutic interventions (4,6,11) (9) (4,11,19)
FC values were twice as high in patients with pancreatic insufficiency as in those with normal pancreatic function; this supported some previous reports in literature (2,6) (4) 1 , or BMI had no clinically relevant association with FC. Additionally, FC concentrations in our patients were not related to gastrointestinal symptoms or the dosage of pancreatic enzymes. The observed statistical associations with fecal elastase, fecal chymotrypsin, and steatocrit were quantitatively small. These data suggest that demographic and clinical parameters cannot explain why calprotectin concentrations are elevated in fecal specimens from CF patients. Thus, one could argue that routine measurements of calprotectin in stool samples are not necessary.
To our knowledge, this is the first report on the relation between FC levels and the CFTR mutation. In the present study, F508del homozygous patients had higher median FC values than heterozygous patients, albeit not statistically significant. We observed for the first time that increased FC is associated with inflammatory parameters in serum. A 2-fold increase in serum CRP was associated with an 11% increase in FC. This suggests a moderate association between intestinal and systemic inflammation. Previous authors compared calprotectin concentrations in blood and stool from subjects with CF and found that faecal concentrations were about 6 times higher than those in serum (4) (20) (21)
An unexpected finding was a clinically relevant decline of median FC values in the whole patient cohort from 2003/2004 to recent years. For each calendar year, FC concentrations decreased by 6%, despite no changes in sampling procedures or calprotectin determinations. One reason for this trend could be that physicians were aware of FC results during each consultation and initiated therapeutic interventions in patients with increased values. For example, a subgroup of 28 patients received LGG therapy in a dose of 2 capsules twice a day. Stool specimens obtained during treatment with LGG had 25% lower FC concentrations than those from subjects without LGG therapy. This observation is in line with earlier clinical trials in small patient groups, which demonstrated a 30% to 40% difference in FC concentrations between patients treated with probiotics and placebo patients (10,13,22)
The major strengths of our study are the longitudinal observation of FC values during 12 years and the large number of analyzed specimens. More than 80% of our patients provided stool samples and were from all age groups and had different degrees of nutritional status and lung disease. The major limitation of this study is that it was not a prospective study assessing morphologic changes in the gut in relation to FC concentrations. Furthermore, some patients refused to bring fecal specimens to the clinic. Patients in shared care had only 1 appointment in the outpatient clinic per year and only a small subset of these provided stool specimens. Nonetheless, we regard the present results as representative for the whole patient cohort.
In conclusion, longitudinal assessment for up to 12 years revealed that FC concentrations were increased in the majority of specimens from patients with CF. Values were higher in subjects with pancreatic insufficiency, whereas nutritional status or severity of lung disease was not meaningfully associated with calprotectin levels. It remains to be shown whether FC is a good indicator for intestinal inflammation in patients with CF. Future studies should investigate in more detail what abnormal FC concentrations mean and whether these can be modified through appropriate therapeutic interventions.
Acknowledgments
The authors thank Johannes Eder, MD, for assistance with data collection. The authors acknowledge the generous support by CF-TEAM Forschung, Innsbruck, Austria.
REFERENCES
1. Turck D, Braegger CP, Colombo C, et al. ESPEN-ESPGHAN-ECFS guidelines on nutrition care for infants, children, and adults with cystic fibrosis.
Clin Nutr 2016; 35:557–577.
2. Werlin SL, Benuri-Silbiger I, Kerem E, et al. Evidence of intestinal inflammation in patients with cystic fibrosis.
J Pediatr Gastroenterol Nutr 2010; 51:304–308.
3. Smyth RL, Croft NM, O’Hea U, et al. Intestinal inflammation in cystic fibrosis.
Arch Dis Child 2000; 82:394–399.
4. Rumman N, Sultan M, El-Chammas K, et al. Calprotectin in cystic fibrosis.
BMC Pediatr 2014; 14:133.
5. Vaos G, Kostakis ID, Zavras N, et al. The role of calprotectin in pediatric disease.
Biomed Res Int 2013; 2013:542363.
6. Dhaliwal J, Leach S, Katz T, et al. Intestinal inflammation and impact on growth in children with cystic fibrosis.
J Pediatr Gastroenterol Nutr 2015; 60:521–526.
7. Vos M, de Louis EJ, Jahnsen J, et al. Consecutive fecal calprotectin measurements to predict relapse in patients with ulcerative colitis receiving infliximab maintenance therapy.
Inflamm Bowel Dis 2013; 19:2111–2117.
8. Yamamoto T, Shiraki M, Bamba T, et al. Fecal calprotectin and lactoferrin as predictors of relapse in patients with quiescent ulcerative colitis during maintenance therapy.
Int J Colorectal Dis 2014; 29:485–491.
9. Fallahi G, Motamed F, Yousefi A, et al. The effect of probiotics on fecal calprotectin in patients with cystic fibrosis.
Turk J Pediatr 2013; 55:475–478.
10. Bruzzese E, Raia V, Gaudiello G, et al. Intestinal inflammation is a frequent feature of cystic fibrosis and is reduced by probiotic administration.
Aliment Pharmacol Ther 2004; 20:813–819.
11. Bruzzese E, Callegari ML, Raia V, et al. Disrupted intestinal microbiota and intestinal inflammation in children with cystic fibrosis and its restoration with Lactobacillus GG: a randomised clinical trial.
PLoS One 2014; 9:e87796.
12. Ellemunter H, Eder J, Steinkamp G. Structured care in an ISO certified centre for patients with cystic fibrosis and their families.
Pneumologie 2011; 65:615–623.
13. Di NG, Oliva S, Menichella A, et al. Randomised clinical trial:
Lactobacillus reuteri ATCC55730 in cystic fibrosis.
J Pediatr Gastroenterol Nutr 2014; 58:81–86.
14. Kromeyer-Hauschild K, Wabitsch M, Kunze D, et al. Perzentile für den Body-mass-Index für das Kindes- und Jugendalter unter Heranziehung verschiedener deutscher Stichproben [Percentiles of body mass index in children and adolescents evaluated from different regional German studies].
Monatsschrift Kinderheilkunde 2001; 149:807–818.
15. Miller MR, Hankinson J, Brusasco V, et al. Standardisation of spirometry.
Eur Respir J 2005; 26:319–338.
16. Quanjer PH, Stanojevic S, Cole TJ, et al. Multi-ethnic reference values for spirometry for the 3-95-yr age range: the global lung function 2012 equations.
Eur Respir J 2012; 40:1324–1343.
17. Wenninger K, Staab D, Henry B, et al. Der Fragebogen zur Lebensqualität bei Cystischer Fibrose - Reliabilität und Validität der deutschen Adaptation [The Cystic Fibrosis Questionnaire (CFQ): Reliability and validity of the German version CFQ-14+].
Zeitschrift für Medizinische Psychologie 1999; 8:121–128.
18. Lee TW, Brownlee KG, Conway SP, et al. Evaluation of a new definition for chronic
Pseudomonas aeruginosa infection in cystic fibrosis patients.
J Cyst Fibros 2003; 2:29–34.
19. Flass T, Tong S, Frank DN, et al. Intestinal lesions are associated with altered intestinal microbiome and are more frequent in children and young adults with cystic fibrosis and cirrhosis.
PLoS One 2015; 10:e0116967.
20. Sagel SD, Thompson V, Chmiel JF, et al. Effect of treatment of cystic fibrosis pulmonary exacerbations on systemic inflammation.
Ann Am Thorac Soc 2015; 12:708–717.
21. Gray RD, Imrie M, Boyd AC, et al. Sputum and serum calprotectin are useful biomarkers during CF exacerbation.
J Cyst Fibros 2010; 9:193–198.
22. Del CR, Garriga M, Perez-Aragon A, et al. Improvement of digestive health and reduction in proteobacterial populations in the gut microbiota of cystic fibrosis patients using a
Lactobacillus reuteri probiotic preparation: a double blind prospective study.
J Cyst Fibros 2014; 13:716–722.
