Advertisement

Gastroesophageal Reflux Symptoms in Patients With Celiac Disease and the Effects of a Gluten-Free Diet

      Background & Aims

      Celiac disease (CD) patients often complain of symptoms consistent with gastroesophageal reflux disease (GERD). We aimed to assess the prevalence of GERD symptoms at diagnosis and to determine the impact of the gluten-free diet (GFD).

      Methods

      We evaluated 133 adult CD patients at diagnosis and 70 healthy controls. Fifty-three patients completed questionnaires every 3 months during the first year and more than 4 years after diagnosis. GERD symptoms were evaluated using a subdimension of the Gastrointestinal Symptoms Rating Scale for heartburn and regurgitation domains.

      Results

      At diagnosis, celiac patients had a significantly higher reflux symptom mean score than healthy controls (P < .001). At baseline, 30.1% of CD patients had moderate to severe GERD (score >3) compared with 5.7% of controls (P < .01). Moderate to severe symptoms were significantly associated with the classical clinical presentation of CD (35.0%) compared with atypical/silent cases (15.2%; P < .03). A rapid improvement was evidenced at 3 months after initial treatment with a GFD (P < .0001) with reflux scores comparable to healthy controls from this time point onward.

      Conclusions

      GERD symptoms are common in classically symptomatic untreated CD patients. The GFD is associated with a rapid and persistent improvement in reflux symptoms that resembles the healthy population.

      Keywords

      Abbreviations used in this paper:

      CD (celiac disease), CI (confidence interval), GERD (gastroesophageal reflux disease), GFD (gluten-free diet), GSRS (Gastrointestinal Symptoms Rating Scale), IgA, a-tTG (immunoglobulin A antitissue transglutaminase antibodies), PPI (proton pump inhibitor)
      See editorial on page 192.
      Celiac disease (CD) and gastroesophageal reflux disease (GERD) are common and distinct clinical conditions with clearly different clinical and histological features and outcome. CD is an immune-mediated chronic inflammatory disorder due to permanent gluten intolerance that mainly affects the small bowel. It has an approximate prevalence of 1% in the western hemisphere.
      • Fasano A.
      • Catassi C.
      Current approaches to diagnosis and treatment of celiac disease: an evolving spectrum.
      The clinical presentations of CD vary tremendously among CD patients with a minority reporting gastrointestinal symptoms and a large proportion of patients with few gastrointestinal symptoms but numerous extraintestinal symptoms. Some CD patients have few or no symptoms at all.
      • Ciclitira P.J.
      Celiac disease: a technical review.
      • Green P.H.
      The many faces of celiac disease: clinical presentation of celiac disease in the adult population.
      GERD is a chronic condition that develops when the reflux of stomach contents causes esophageal exposure to gastric fluids.
      • Spechler S.J.
      Epidemiology and natural history of gastro-esophageal reflux disease.
      However, reflux of gastric acid is not the only pathophyisiology of GERD symptoms.
      • DeVault K.R.
      • Castell D.O.
      Updated guidelines for the diagnosis and treatment of gastroesophageal reflux disease.
      While 30% to 40% of patients with GERD have erosive reflux disease, up to 70% have no macroscopic changes in the gastroesophageal junction (ie, nonerosive reflux).
      • DeVault K.R.
      • Castell D.O.
      Updated guidelines for the diagnosis and treatment of gastroesophageal reflux disease.
      A characteristic of GERD is the relapse of symptoms in many cases successfully treated with proton pump inhibitors (PPI).
      • Reimer C.
      • Søndergaard B.
      • Hilsted L.
      • et al.
      Proton-pump inhibitor therapy induces acid-related symptoms in healthy volunteers after withdrawal of therapy.
      CD patients often complain of GERD symptoms.
      • Tursi A.
      The treatment of gastro-esophageal reflux disease in adult celiac disease.
      Moreover, endoscopic investigation of patients with GERD symptoms has sometimes resulted in the diagnosis of an unexpected CD enteropathy.
      • Tursi A.
      The treatment of gastro-esophageal reflux disease in adult celiac disease.
      Clinical experience has shown that this observation is more commonly associated with GERD symptoms refractory to PPI treatment.
      • Reyes H.
      • Niveloni S.
      • Moreno M.L.
      • et al.
      A prospective evaluation of endoscopic markers for identifying celiac disease in patients with high and low probability of having the disease.
      • Bai J.C.
      • Morán C.
      • Martinez C.
      • et al.
      Celiac sprue after surgery of the upper gastrointestinal tract.
      • Cuomo A.
      • Romano M.
      • Rocco A.
      Reflux oesophagitis in adult coeliac disease: beneficial effect of a gluten free diet.
      Cuomo et al
      • Cuomo A.
      • Romano M.
      • Rocco A.
      Reflux oesophagitis in adult coeliac disease: beneficial effect of a gluten free diet.
      have suggested that GERD symptoms are frequent in CD patients and that a gluten-free diet (GFD) reduces the relapse rate of these symptoms. Although the first statement is not shared by others,
      • Collin P.
      • Mustalahti K.
      • Kyrönpalo S.
      • et al.
      Should we screen reflux oesophagitis patients for coeliac disease?.
      most authors agree that treatment with a GFD may prevent the recurrence of upper gastrointestinal symptoms.
      • Tursi A.
      The treatment of gastro-esophageal reflux disease in adult celiac disease.
      • Collin P.
      • Mustalahti K.
      • Kyrönpalo S.
      • et al.
      Should we screen reflux oesophagitis patients for coeliac disease?.
      • Usai P.
      • Manca R.
      • Cuomo R.
      • et al.
      Effect of gluten-free diet on preventing recurrence of gastroesophageal reflux disease-related symptoms in adult celiac patients with nonerosive reflux disease.
      However, the prevalence of GERD symptoms and the effectiveness of GFD in CD patients still remain to be elucidated. Furthermore, the pathogenesis of GERD symptoms in CD remains unknown.
      Our present study had 2 aims: (1) to assess the prevalence of GERD symptoms in CD patients at the time of diagnosis; and (2) to explore the long-term impact of treatment with a GFD on GERD symptoms. The present study had a double design, a cross-sectional analysis of a series of consecutive patients evaluated at the time of diagnosis, and a prospective, longitudinal, long-term assessment of symptoms at different time points after initiation of a GFD.

      Materials and Methods

       Subjects

       Patients and controls

      From December 2004 to December 2005, a series of 151 consecutive adult patients who were newly diagnosed with CD (117 female and 34 male) at the Small Bowel Diseases Clinic of the “Dr. Carlos Bonorino Udaondo” Gastroenterology Hospital were prospectively enrolled in the cross-sectional part of the study. At the time of initial diagnosis, these patients were consuming a regular unrestricted diet. Patients without biopsy-proven CD or already on a GFD at the time of the first visit were not included in the assessment. We also excluded patients who declined to or were unable to participate in the initial and follow-up evaluations.
      The diagnosis of CD was based on clinical, serological, and histological criteria. The official diagnosis of CD required the presence of a characteristic CD enteropathy in the duodenal biopsy and was supported by the concomitant presence of positive antitissue transglutaminase antibodies, antideamidated gliadin peptide antibodies, or endomysial antibodies as well as histological response to a GFD in cases with enteropathy but a negative serology. Histological damage in the intestinal biopsy was graded according to the Marsh′s modified classification where the most severe damage assessed was reported.
      • Rostami K.
      • Kerckhaert J.
      • Tiemessen R.
      • et al.
      Sensitivity of antiendomysium and antigliadin antibodies in untreated celiac disease: disappointing in clinical practice.
      Based on the clinical status of patients at the time of diagnosis, they were categorized as presenting classically symptomatic CD (mainly gastrointestinal symptoms), atypical forms (subclinical) of CD (chronic anemia, hypertransaminasemia, autoimmune diseases, etc), or a silent clinical course (asymptomatic or silent CD).
      • Rostami K.
      • Kerckhaert J.
      • Tiemessen R.
      • et al.
      Sensitivity of antiendomysium and antigliadin antibodies in untreated celiac disease: disappointing in clinical practice.
      Patients with silent CD were detected during evaluation of first degree relatives of index cases.
      • Nachman F.
      • Mauriño E.
      • Vázquez H.
      • et al.
      Quality of life in celiac disease patients Prospective analysis on the importance of clinical severity at diagnosis and the impact of treatment.
      After applying the inclusion and exclusion criteria, 133 patients were enrolled in the study at the time of diagnosis. Nineteen other patients were excluded due to different reasons: 10 patients lived too far from the site for follow-up visits, 4 declined participation, 3 were reluctant to start treatment with a GFD, and 2 had already started treatment at the time of the first visit. Among the enrolled patients, 98 patients were categorized as having a classical clinical presentation; 35 were considered as presenting atypical (n = 25) or silent CD (n = 10).
      A cohort of 70 healthy adults with matched age and gender and without any known former or current disease was included and assessed as controls.

       Longitudinal long-term study

      Fifty-three patients (47 women and 6 men, mean age 38 years) completed the long-term assessment visits at 3 months, 6 months, 1 year, and beyond 4 years and were included in the prospective longitudinal analysis if data for all time points were available.

       Methods

       Assessment of GERD symptoms

      To evaluate GERD symptoms, we used a subdimension in the Gastrointestinal Symptoms Rating Scale (GSRS) that focuses on heartburn and acid regurgitation symptoms and is scored on a 7-graded Likert scale. The GSRS is a disease-specific instrument developed based on reviews of gastrointestinal symptoms to evaluate common symptoms of gastrointestinal disorders.
      • Nachman F.
      • Mauriño E.
      • Vázquez H.
      • et al.
      Quality of life in celiac disease patients Prospective analysis on the importance of clinical severity at diagnosis and the impact of treatment.
      The instrument includes 5 subdimensions, 1 of which is related to gastroesophageal reflux symptoms. In the case of GERD, the questionnaire has shown good reliability, discriminates symptom severity, and is useful for evaluating treatment outcomes.
      • Collin P.
      • Mustalahti K.
      • Kyrönpalo S.
      • et al.
      Should we screen reflux oesophagitis patients for coeliac disease?.
      • Revicki D.A.
      • Wood M.
      • Wiklund I.
      • et al.
      Reliability and validity of the Gastrointestinal Symptom Rating Scale in patients with gastroesophageal reflux disease.
      The GSRS scores range from 0 to 6 for both heartburn and regurgitation, where 0 signifies no symptoms at all and 6 represents the highest severity. An average score of ≥3 for both items is considered as indicative of moderate to severe GERD.
      • Collin P.
      • Mustalahti K.
      • Kyrönpalo S.
      • et al.
      Should we screen reflux oesophagitis patients for coeliac disease?.
      Patients underwent periodical clinical assessments, routine laboratory tests (hemoglobin and serum albumin concentrations are reported here), and measurement of serum concentration of immunoglobulin A antitissue transglutaminase antibodies (IgA a-tTG) at each time point. The cutoff value for IgA a-tTG antibodies used for positive results was that provided by the manufacturer (>20 U/mL).

       Assessment of compliance with the gluten-free diet

      Compliance with the GFD was assessed at 1-year and 4-year visits after diagnosis. Based on previous studies,
      • Nachman F.
      • Mauriño E.
      • Vázquez H.
      • et al.
      Quality of life in celiac disease patients Prospective analysis on the importance of clinical severity at diagnosis and the impact of treatment.
      • Sugai E.
      • Nachman F.
      • Váquez H.
      • et al.
      Dynamics of celiac disease-specific serology after initiation of a gluten-free diet and use in the assessment of compliance with treatment.
      compliance was determined by a combination of: (1) the assessment by the treating physician; (2) a meticulous enquiry by an experienced dietitian; and (3) self-report by the patient. A 4-day food diary of complete intakes by the patient was also considered. The CD-related serology was used as comparators with other parameters. Based on all these aspects, the physician in charge categorized the compliance as: (1) strict adherence (absolute observance of the GFD); (2) partial adherence (2–3 lapses or more frequent lapses per month or once or more per week); and (3) no gluten restriction at all.

       Study design, ethics, and statistical analysis

      The study was prospectively performed on consecutive patients diagnosed at 1 institution. This analysis is a part of a larger research project to assess the quality of life in CD patients, which was reported separately.
      • Nachman F.
      • Mauriño E.
      • Vázquez H.
      • et al.
      Quality of life in celiac disease patients Prospective analysis on the importance of clinical severity at diagnosis and the impact of treatment.
      Upon diagnosis, all patients underwent a standardized upper endoscopic procedure where duodenal biopsies were obtained. Endoscopic characteristics of the upper gastrointestinal mucosa were reported. After diagnosis of CD enteropathy, a written consent was obtained. Patients who agreed to participate in the prospective protocol underwent a complete clinical evaluation. Serum samples were obtained for routine hematological tests and specific serology: at the time of diagnosis while patients were consuming a regular unrestricted diet and at 3 months, 6 months, 9 months, 1 year, and 4 years after they were started on dietary restriction. The GSRS questionnaire was administered at the same time points. Expert dieticians evaluated patients' adherence to the GFD and reinforced the importance of the treatment. The study and the consent form were approved by the Research and Ethical Committees of the Gastroenterology Hospital.
      Data were analyzed using the MedCalc version 9.3.8.0 (MedCalc Software; Broekstraat, Mariakerke, Belgium). Descriptive statistics are reported as mean and 95% confidence intervals (95% CI) or median and range according to the data distribution pattern. Two statistical analyses were performed in the study. First, we made a cross-sectional assessment of GERD symptoms in CD patients by clinical characterization at diagnosis and compared with healthy controls. Comparisons between clinical subgroups and between CD patients and controls were performed using the Student t test, Mann–Whitney U test, χ2, or Fisher exact test as appropriate. Second, we assessed the effect of GFD on GERD symptom outcomes only in patients who completed all assessments. Comparisons between baseline and posttreatment outcomes were performed using the Student pair t test and Wilcoxon′s pair test. In addition, we compared GERD symptoms in CD patients by their degree of compliance with a GFD using the analysis of variance or Kruskal–Wallis test as appropriate. Univariate and multivariate linear regression analyses were performed.

       Cross-sectional assessments upon diagnosis

      Table 1 summarizes the demographic and clinical characteristics of the CD patients and healthy controls in the study. Patients and controls were well matched in age and the predominance of females. CD patients had a low mean body mass index and a mean hemoglobin concentration below the normal range. Baseline endoscopies showed hiatal hernia in 4 CD patients and Barrett′s disease in 1 case (confirmed by histopathology). At enrollment, 6 control subjects were using PPI.
      Table 1Demography, Baseline Clinical Characteristics, and Baseline GERD Symptom Scores of Patients With CD and Healthy Controls and Vs Classical CD
      CharacteristicsCD patientsHealthy controls
      Number of subjects (females/males)133 (114/19)70 (55/15)
      Age mean (range)38.1 (16–72)39.5 (19–71)
      Baseline clinical categorization, n (%)
       Classical CD98
       Atypical/silent CD35
      Serology: IgA a-tTG
       Number of cases with positive test (%)130 (97.7)
      Body mass index, median ± SEM, kg/cm220.2 ± 3.1
      Serum albumin, mean ± SEM, g %4.0 ± 1.5
      Hemoglobin, mean ± SEM, g %11.7 ± 2.2
      GERD symptom scores, mean (95% CI)
       Overall CD patients1.8 (1.5–2.0)0.7 (0.5–1.0)
      P < .01.
       Classic CD1.9 (1.6–2.3)
       Atypical/silent CD1.2 (0.7–1.7)
      P < .03.
      Number of cases with scores ≥3 (%)
       Overall CD patients40 (30.1)4 (5.7)
      P < .0001.
       Classic CD35 (35.0)
       Atypical/silent CD5 (15.2)
      P < .03.
      Patients with erosive esophagitis (%)5 (3.8)
      Patients receiving PPIs (%)12 (9.0)6 (8.6)
      SEM, standard error of the mean.
      a P < .01.
      b P < .03.
      c P < .0001.

       GERD symptoms

      Compared with healthy controls, patients had significantly higher GERD symptom scores (P < .001). The proportion of individuals with moderate to severe symptoms (≥3 points for the average of heartburn and regurgitation) was 6-fold higher among CD patients than healthy controls (P < .01) (Table 1). When patients were categorized according to CD presentation at the time of diagnosis, those with classical clinical manifestations had higher GERD symptom scores (P < .01) and a higher proportion of cases with moderate or severe reflux symptoms (P < .03) compared with patients with atypical/silent CD. Only a quarter of patients with GERD symptoms were using PPIs at the time of CD diagnosis and a minority of cases had erosive esophagitis based on the baseline endoscopy. Figure 1 depicts the number of CD patients (Figure 1A) and healthy controls (Figure 1B) with different scores at diagnosis and among patients by clinical presentation at baseline (Figures 1C and D). Patients with GERD symptoms clustered among the classically symptomatic CD patients, while atypical/silent cases resemble distribution in healthy controls.
      Figure thumbnail gr1
      Figure 1Numbers of individuals with different GERD symptom scores at the time of CD diagnosis. (A) Overall CD population; (B) healthy controls; (C) classic CD; (D) atypical/silent CD. Gray bars represent number of subjects with scores ≥3.

       Longitudinal outcomes

      Table 2 shows clinical data at different time points for the study cohort with long-term follow-up. With GFD treatment, CD patients improved their body mass index and hemoglobin concentration at the 3-month assessment and remained within normal ranges thereafter. The IgA a-tTG antibody also significantly decreased after initiation of the GFD (P < .04). GERD symptom scores improved at the 3-month assessment and continued the trend (P < .0001). The proportion of cases with moderate to severe GERD symptom scores (>3) decreased over the same period (P < .001). Both parameters became similar to those of the healthy subjects at every time point assessed. At the assessment beyond 4 years, only 2 patients still had scores >3 (3.7%) and were using PPIs.
      Table 2Clinical and Biochemical Data and GERD Symptom Scores of CD Patients (n = 53) Assessed at Diagnosis, 3 Months, 6 Months, 1 Year, and Beyond 4 Years
      Baseline3 months6 months1 year4 years
      Serology, IgA a-tTG
       Cases with positive tests, n (%)51 (96.2)37 (69.8)
      P < .001 vs baseline values.
      30 (56.6)
      P < .001 vs baseline values.
      32 (60.4)
      P < .001 vs baseline values.
      21 (39.6)
      P < .0001 vs baseline values.
      Body mass index
       Median ± SEM20.6 ± 5.021.5 ± 4.822.0 ± 5.224.0 ± 3.5
      P < .001 vs baseline values.
      24.2 ± 3.5
      P < .001 vs baseline values.
      Serum albumin
       Mean ± SEM, g %4.1 ± 2.04.1 ± 0.54.2 ± 0.54.1 ± 0.44.3 ± 0.5
      P < .01 vs baseline values.
      Hemoglobin
       Mean ± SEM, g %11.5 ± 2.511.9 ± 2.512.6 ± 2.112.3 ± 1.813.4 ± 1.7
      P < .01 vs baseline values.
      GERD symptoms scores, mean (95% CI)
       Overall1.9 (1.4–2.4)0.8 (0.5–1.1)
      P < .001 vs baseline values.
      0.7 (0.6–1.3)
      P < .001 vs baseline values.
      0.8 (0.5–1.1)
      P < .001 vs baseline values.
      0.6 (0.4–0.8)
      P < .0001 vs baseline values.
      Number of patients with GERD scores ≥3 (%)
       Overall16 (30.1)4 (7.5)
      P < .01 vs baseline values.
      0
      P < .0001 vs baseline values.
      2 (3.7)
      P < .001 vs baseline values.
      2 (3.7)
      P < .001 vs baseline values.
      NOTE. Body mass index calculated as weight in kilograms divided by the square of height in meters.
      SEM, standard error of the mean.
      a P < .001 vs baseline values.
      b P < .0001 vs baseline values.
      c P < .01 vs baseline values.
      We also examined GERD symptoms by the degree of compliance with the GFD at the 4-year time point. At this point, 28 study cases were considered strictly compliant with the diet and the remaining patients were considered as partially compliant. Clinical parameters were comparable (P = not significant) between patients who were strictly and partially compliant with the dietary treatment (Table 3). The strictly compliant patients had significantly lower serum samples positive for the IgA a-tTG test (P < .04). Although the amount of gluten consumption varied widely among partially compliant cases, no patient was considered by the expert team as not adherent at all. The analysis did not show a significant difference based on compliance in terms of GERD symptom scores and proportion of cases with moderate to severe GERD symptoms (Table 3).
      Table 3Clinical Parameters and GERD Scores for CD Patients Assessed Beyond 4 Years After Diagnosis and Categorized as Strictly Complaint or Partially Compliant With the GFD
      CharacteristicStrict compliancePartial compliance
      Number (F/M)28 (23/5)25 (24/1)
      Serology, IgA a-tTG, cases with positive tests, n (%)25.0 (7.9–42.1)51.9 (31.7–72.0)
      P < .04.
      Body mass index, median ± SEM24.1 (22.7–25.5)24.4 (22.7–26.0)
      Serum albumin, mean ± SEM, g %4.3 (4.1–4.5)4.4 (4.2–4.6)
      Hemoglobin, mean ± SEM, g %13.4 (12.9–13.8)13.3 (12.6–14.0)
      GERD symptom scores, mean (95% CI)0.5 (0.2–0.9)0.6 (0.3–0.9)
      Number of patients with scores ≥3 (%)2 (7.0)0
      NOTE. Body mass index calculated as weight in kilograms divided by the square of height in meters.
      F, female; M, male; SEM, standard error of the mean.
      a P < .04.

      Discussion

      Heartburn and regurgitation are clinical manifestations associated with GERD and both are common causes for clinical consultations. While a minor proportion of cases have erosive esophagitis (30% to 40%), a majority of patients have the nonerosive type of GERD.
      • Spechler S.J.
      Epidemiology and natural history of gastro-esophageal reflux disease.
      • DeVault K.R.
      • Castell D.O.
      Updated guidelines for the diagnosis and treatment of gastroesophageal reflux disease.
      PPI treatment can relieve GERD symptoms, but a high relapse rate has been reported when treatment is discontinued.
      • Reimer C.
      • Søndergaard B.
      • Hilsted L.
      • et al.
      Proton-pump inhibitor therapy induces acid-related symptoms in healthy volunteers after withdrawal of therapy.
      GERD symptoms are rarely studied in CD patients.
      • Reyes H.
      • Niveloni S.
      • Moreno M.L.
      • et al.
      A prospective evaluation of endoscopic markers for identifying celiac disease in patients with high and low probability of having the disease.
      • Bai J.C.
      • Morán C.
      • Martinez C.
      • et al.
      Celiac sprue after surgery of the upper gastrointestinal tract.
      • Cuomo A.
      • Romano M.
      • Rocco A.
      Reflux oesophagitis in adult coeliac disease: beneficial effect of a gluten free diet.
      While some studies have focused on endoscopic evidence of erosive esophagitis,
      • Reyes H.
      • Niveloni S.
      • Moreno M.L.
      • et al.
      A prospective evaluation of endoscopic markers for identifying celiac disease in patients with high and low probability of having the disease.
      • Cuomo A.
      • Romano M.
      • Rocco A.
      Reflux oesophagitis in adult coeliac disease: beneficial effect of a gluten free diet.
      others have only referred to the presence of GERD symptoms irrespective of mucosal damage.
      • Collin P.
      • Mustalahti K.
      • Kyrönpalo S.
      • et al.
      Should we screen reflux oesophagitis patients for coeliac disease?.
      • Usai P.
      • Manca R.
      • Cuomo R.
      • et al.
      Effect of gluten-free diet on preventing recurrence of gastroesophageal reflux disease-related symptoms in adult celiac patients with nonerosive reflux disease.
      In contrast, some investigators have considered the association between GERD and CD as null or weak.
      • Cuomo A.
      • Romano M.
      • Rocco A.
      Reflux oesophagitis in adult coeliac disease: beneficial effect of a gluten free diet.
      • Collin P.
      • Mustalahti K.
      • Kyrönpalo S.
      • et al.
      Should we screen reflux oesophagitis patients for coeliac disease?.
      Interestingly, the few studies investigating this association agree that GFD may alleviate GERD symptoms and reduce the risk of relapse after PPI treatment is discontinued.
      • Bai J.C.
      • Morán C.
      • Martinez C.
      • et al.
      Celiac sprue after surgery of the upper gastrointestinal tract.
      • Cuomo A.
      • Romano M.
      • Rocco A.
      Reflux oesophagitis in adult coeliac disease: beneficial effect of a gluten free diet.
      • Collin P.
      • Mustalahti K.
      • Kyrönpalo S.
      • et al.
      Should we screen reflux oesophagitis patients for coeliac disease?.
      • Usai P.
      • Manca R.
      • Cuomo R.
      • et al.
      Effect of gluten-free diet on preventing recurrence of gastroesophageal reflux disease-related symptoms in adult celiac patients with nonerosive reflux disease.
      The present study estimated the prevalence of GERD symptoms at the time of CD diagnosis and explored the impact of certain clinical factors. Furthermore, the systematic, prospective follow-up over a long period revealed the course of GERD symptoms and the effect of GFD treatment on these symptoms.
      In this CD patient sample, 30% of cases complained of moderate to severe GERD symptoms at the time of CD diagnosis, a rate 6-fold higher than the rate seen in healthy controls (4.7%). Only 1 of the 7 patients with moderate to severe symptoms had erosive esophagitis. Although this prevalence of GERD symptoms among untreated CD patients is the highest rate reported so far, it should not be generalized to the overall CD population because the data here reported were derived from patients attending a malabsorption clinic where classic CD presentation is dominant. As seen in this study, patients with a classic CD picture had significantly higher GERD scores and greater prevalence of moderate to severe symptoms compared with atypical/silent cases.
      The outcome of GERD symptoms after initiation of a GFD treatment showed a rapid resolution during the first 3 months after diagnosis, and the early benefits persisted thereafter with only a minimal proportion of patients remaining moderately symptomatic beyond 4 years. Despite the evidence supporting effectiveness of GFD in alleviating GERD symptoms, the symptom scores did not differ between patients who strictly complied with the diet and those with partial compliance over the long term. It is likely that patients partially compliant with the GFD had decreasing dietary indiscretions over time, as suggested by their clinical improvement of CD, which was comparable with patients who were strictly compliant to the GFD. However, the small number of patients with partial compliance did not allow for further subgroup analyses.
      A number of causes for GERD symptoms in CD patients can be hypothesized. Actual reflux of deleterious gastric content seems to be the most likely pathophysiological mechanism associated with these symptoms. In this context, the presence of erosive esophagitis supports the reflux etiology in at least 4% of all cases. The resolution of GERD symptoms after GFD and the higher prevalence of nonerosive GERD in this study population than previously reported in patients with well-established GERD
      • Spechler S.J.
      Epidemiology and natural history of gastro-esophageal reflux disease.
      • DeVault K.R.
      • Castell D.O.
      Updated guidelines for the diagnosis and treatment of gastroesophageal reflux disease.
      • Reimer C.
      • Søndergaard B.
      • Hilsted L.
      • et al.
      Proton-pump inhibitor therapy induces acid-related symptoms in healthy volunteers after withdrawal of therapy.
      suggests that some CD patients might have a different pathophysiological mechanism for GERD symptoms.
      Interestingly, it would be reasonable to assume that GERD symptoms would persist despite CD treatment if they were due to a true reflux disorder, but the symptom resolution in most study patients could mostly be attributable to the implementation of the GFD. Furthermore, we confirm former observations where GERD symptoms do not relapse despite PPIs are discontinued. Thus, a potential pathogenic role of unknown disturbances associated with CD could be hypothesised for the GERD symptoms. These mechanisms would explain both the very low proportion of cases with long-term persistence of moderate to severe GERD symptoms, the almost null report of patients with symptomatic relapses and a very low rate of PPI use in the long term. Considering the possibility of alternative or complementary mechanisms explaining symptoms we explored some interesting aspects of CD. Thus, there is some evidence of reduced upper gastrointestinal motility in patients with CD.
      • Usai P.
      • Bassotti G.
      • Usai Satta P.
      • et al.
      Oesophageal motility in adult coeliac disease.
      • Bassotti G.
      • Villanacci V.
      • Mazzocchi A.
      • et al.
      Antroduodenojejunal motor activity in untreated and treated celiac disease patients.
      • Usai P.
      • Usai Satta P.
      • Lai M.
      • et al.
      Autonomic dysfunction and upper digestive functional disorders in untreated adult coeliac disease.
      These few studies have shown abnormal esophageal motility in about 50% of untreated CD cases and discrete motor gastric dysfunction with delayed emptying in almost of 80% of patients.
      • Usai P.
      • Bassotti G.
      • Usai Satta P.
      • et al.
      Oesophageal motility in adult coeliac disease.
      Patients on a GFD have shown a trend to normalization of upper gut motility. Interestingly, a minority of patients with esophageal motility disturbances had acid reflux as assessed by pH-metry recording.
      • Bassotti G.
      • Villanacci V.
      • Mazzocchi A.
      • et al.
      Antroduodenojejunal motor activity in untreated and treated celiac disease patients.
      These findings suggest that upper gastrointestinal motility dysfunction could play a causal role in GERD symptoms; however, this hypothesis requires additional research.
      Recent data suggest that a permeability defect in the stratified esophageal epithelium, producing dilated intercellular space, is a frequent but nonspecific feature observed in patients with GERD.
      • Van Malenstein H.
      • Farré R.
      • Sifrim D.
      Esophageal dilated intercellular spaces (DIS) and nonerosive reflux disease.
      • Söderholm J.D.
      Stress-related changes in oesophageal permeability: filling the gaps of GORD?.
      • Farré R.
      • van Malenstein H.
      • De Vos R.
      • et al.
      Short exposure of oesophageal mucosa to bile acids, both in acidic and weakly acidic conditions, can impair mucosal integrity and provoke dilated intercellular spaces.
      Although the assertion remains controversial, it might potentially contribute to the presence of GERD symptoms. Notably, zonulin, a protein involved in the regulation of interepithelial permeability in the intestines of CD patients, was recently found to be expressed in the esophageal epithelial cells as well.
      • Wex T.
      • Mönkemüller K.
      • Kuester D.
      • et al.
      Zonulin is not increased in the cardiac and esophageal mucosa of patients with gastroesophageal reflux disease.
      Although the expression of zonulin was not shown to be associated with esophageal mucosal damage, a theoretical role of the protein might account for the occurrence of GERD symptoms in CD patients. Based on these findings, a true link between active CD and the presence of reflux symptoms cannot be ruled out.
      In conclusion, our study confirms that moderate to severe GERD symptoms are more frequent in patients with undiagnosed CD than in the healthy population. The symptoms rapidly resolved in 3 months after the initiation of a GFD and improvement persisted over the long term. Patients with strict and partial compliance with the GFD showed no significant difference in GERD symptoms in the long-term follow-up. Although the pathogenesis of GERD symptoms in CD remains elusive, it is likely multifactorial. The fact that CD treatment is effective against GERD symptoms and relapse suggests that components in the CD pathogenesis may be linked to GERD symptoms.

      References

        • Fasano A.
        • Catassi C.
        Current approaches to diagnosis and treatment of celiac disease: an evolving spectrum.
        Gastroenterology. 2001; 120: 636-651
        • Ciclitira P.J.
        Celiac disease: a technical review.
        Gastroenterology. 2001; 120: 1526-1540
        • Green P.H.
        The many faces of celiac disease: clinical presentation of celiac disease in the adult population.
        Gastroenterology. 2005; 128: S74-S78
        • Spechler S.J.
        Epidemiology and natural history of gastro-esophageal reflux disease.
        Digestion. 1992; 51: 24-29
        • DeVault K.R.
        • Castell D.O.
        Updated guidelines for the diagnosis and treatment of gastroesophageal reflux disease.
        Am J Gastroenterol. 2005; 100: 190-200
        • Reimer C.
        • Søndergaard B.
        • Hilsted L.
        • et al.
        Proton-pump inhibitor therapy induces acid-related symptoms in healthy volunteers after withdrawal of therapy.
        Gastroenterology. 2009; 137: 80-87
        • Tursi A.
        The treatment of gastro-esophageal reflux disease in adult celiac disease.
        J Clin Gastroenterol. 2004; 38: 724-746
        • Reyes H.
        • Niveloni S.
        • Moreno M.L.
        • et al.
        A prospective evaluation of endoscopic markers for identifying celiac disease in patients with high and low probability of having the disease.
        Acta Gastroenterol Latinoamer. 2008; 38: 178-186
        • Bai J.C.
        • Morán C.
        • Martinez C.
        • et al.
        Celiac sprue after surgery of the upper gastrointestinal tract.
        J Clin Gastroenterol. 1991; 13: 521-524
        • Cuomo A.
        • Romano M.
        • Rocco A.
        Reflux oesophagitis in adult coeliac disease: beneficial effect of a gluten free diet.
        Gut. 2003; 52: 514-517
        • Collin P.
        • Mustalahti K.
        • Kyrönpalo S.
        • et al.
        Should we screen reflux oesophagitis patients for coeliac disease?.
        Eur J Gastroenterol Hepatol. 2004; 16: 917-920
        • Usai P.
        • Manca R.
        • Cuomo R.
        • et al.
        Effect of gluten-free diet on preventing recurrence of gastroesophageal reflux disease-related symptoms in adult celiac patients with nonerosive reflux disease.
        J Gastroenterol Hepatol. 2008; 23: 1368-1372
        • Rostami K.
        • Kerckhaert J.
        • Tiemessen R.
        • et al.
        Sensitivity of antiendomysium and antigliadin antibodies in untreated celiac disease: disappointing in clinical practice.
        Am J Gastroenterol. 1999; 94: 888-894
        • Nachman F.
        • Mauriño E.
        • Vázquez H.
        • et al.
        Quality of life in celiac disease patients.
        Dig Liv Dis. 2009; 41: 15-25
        • Revicki D.A.
        • Wood M.
        • Wiklund I.
        • et al.
        Reliability and validity of the Gastrointestinal Symptom Rating Scale in patients with gastroesophageal reflux disease.
        Qual Life Res. 1998; 7: 75-83
        • Sugai E.
        • Nachman F.
        • Váquez H.
        • et al.
        Dynamics of celiac disease-specific serology after initiation of a gluten-free diet and use in the assessment of compliance with treatment.
        Dig Liver Dis. 2010; 42: 352-358
        • Usai P.
        • Bassotti G.
        • Usai Satta P.
        • et al.
        Oesophageal motility in adult coeliac disease.
        Neurogastroenterol Motil. 1995; 7: 239-244
        • Bassotti G.
        • Villanacci V.
        • Mazzocchi A.
        • et al.
        Antroduodenojejunal motor activity in untreated and treated celiac disease patients.
        J Gastroenterol Hepatol. 2008; 23: e23-e28
        • Usai P.
        • Usai Satta P.
        • Lai M.
        • et al.
        Autonomic dysfunction and upper digestive functional disorders in untreated adult coeliac disease.
        Eur J Clin Invest. 1997; 27: 1009-1015
        • Van Malenstein H.
        • Farré R.
        • Sifrim D.
        Esophageal dilated intercellular spaces (DIS) and nonerosive reflux disease.
        Am J Gastroenterol. 2008; 103: 1021-1028
        • Söderholm J.D.
        Stress-related changes in oesophageal permeability: filling the gaps of GORD?.
        Gut. 2007; 56: 1177-1180
        • Farré R.
        • van Malenstein H.
        • De Vos R.
        • et al.
        Short exposure of oesophageal mucosa to bile acids, both in acidic and weakly acidic conditions, can impair mucosal integrity and provoke dilated intercellular spaces.
        Gut. 2008; 57: 1366-1374
        • Wex T.
        • Mönkemüller K.
        • Kuester D.
        • et al.
        Zonulin is not increased in the cardiac and esophageal mucosa of patients with gastroesophageal reflux disease.
        Peptides. 2009; 30: 1082-1087