Reflux esophagitis in the pathogenesis of paroxysmal atrial fibrillation: results of a pilot study.

Background: We sought to assess whether proton pump inhibitor (PPI) therapy of gastroesophageal reflux disease (GERD) in patients with lone paroxysmal atrial fibrillation (PAF) leads to a reduction of PAF-related symptoms.

Methods: The records of patients with reflux esophagitis were screened for the diagnosis of lone PAF. All patients with reflux esophagitis and lone PAF were invited for a follow-up visit, at which PAF- and GERD-related symptoms, medication, and electrocardiogram were recorded.

Results: Among 89 patients, 18 (6 women, aged 39-69 years) had lone PAF. Decrease or disappearance of at least one PAF-related symptom occurred in 14 of 18 patients (78%) after PPI therapy. In two of the remaining four patients, GERD-related symptoms persisted. Antiarrhythmic drugs were discontinued in five patients, and none had to be increased in dosage or newly prescribed. The electrocardiogram showed sinus rhythm in all patients.

Conclusion: In lone PAF, GERD should be investigated as a potential pathogenetic mechanism. PPI therapy reduces not only GERD-related but also PAF-related symptoms.

Key Words: atrial fibrillation, gastroesophageal reflux disease, proton pump inhibitor, reflux esophagitis

**********

Clinical observations and a case report in the literature (1) suggest that gastroesophageal reflux disease (GERD) may induce atrial fibrillation (AF) and that therapy of GERD decreases the frequency of attacks of paroxysmal atrial fibrillation (PAF). Since proton pump inhibitors (PPIs) have fewer side effects and are less expensive than antiarrhythmic drugs or electrotherapy, therapy of GERD in patients with PAF might be an attractive alternative to the conventional and often not completely successful therapy of PAF. Therefore, the following pilot study was carried out to see whether therapy of GERD in PAF patients without underlying heart disorders ("lone" PAF) leads to a reduction of PAF attacks or even to a complete disappearance of PAF.

Patients and Methods

Patient Selection

For this pilot study, the endoscopic reports of all patients from the cardiologic department of the Krankenanstalt Rudolfstiftung who underwent gastroscopy between January 1999 and October 2001 were screened for the diagnosis of reflux esophagitis. After that, the hospital and outpatient clinic records of all patients with reflux esophagitis were screened for the diagnosis of lone PAF. All patients with a gastroscopic diagnosis of reflux esophagitis in the endoscopic report who had received PPI treatment for at least 2 months after endoscopy and who had a cardiologic diagnosis of lone PAF (as defined below) were included in this pilot study and invited for a follow-up visit in February and March 2002.

Diagnosis and Therapy of Esophagitis 

Gastroscopy was carried out using Fujinon-410HR Type S videoendoscopes (Fuji Photo Optical Co., Uetake, Japan). The severity of reflux esophagitis was classified according to Savary and Miller. (2) Additional endoscopic diagnoses were noted. Immediately after having established the diagnosis of reflux esophagitis, treatment with PPIs in standard dose for at least 2 months was prescribed.

Diagnosis of Lone PAF

PAF was diagnosed if there were at least two episodes of AF alternating with sinus rhythm documented by electrocardiography within 3 months before gastroscopy. Patients were excluded if they had chronic AF, mitral or aortic valve disease, dilative cardiomyopathy, hypertrophic cardiomyopathy, acute myocardial infarction or acute coronary syndromes, angiographically proven coronary artery stenoses, hyperthyroidism, chronic obstructive pulmonary disease, acute infections, or AF that occurred only within 2 weeks postoperatively.

Follow-up

At the follow-up visit, the patient was asked by one of the authors (MW) for the following parameters according to a preset questionnaire: history and onset of PAF, presence or absence of the PAF-related symptoms (palpitations, dizziness, exertional dyspnea, and weakness), and presence or absence of the GERD-related symptoms (heartburn, regurgitation, dysphagia, and vomitus). In addition, it was asked whether since the initiation of PPI therapy, frequency and duration of palpitations had changed and whether PAF- or GERD- related symptoms had increased, decreased, remained unchanged, disappeared, or newly appeared. The response to PPI therapy on PAF-related symptoms was quantified by a scoring system. If a PAF-related symptom increased or was newly reported, it was scored as +1. If a PAF-related symptom remained unchanged or was never reported, it was scored as 0. If a PAF-related symptom decreased or disappeared, it was scored as -1. The presence of arterial hypertension, diabetes mellitus, chronic respiratory infections, or asthma was assessed. Patients were asked about the present medication and changes in their medication and intercurrent disorders since the initiation of PPI therapy, and an electrocardiogram was recorded. The institutional review board approved the study. Informed consent was obtained from the patients.

Results

Baseline Characteristics

Between January 1999 and October 2001, 640 patients from the cardiologic department underwent gastroscopy at the gastroenterologic department of the Krankenanstalt Rudolfstiftung. The endoscopic reports mentioned reflux esophagitis in 89 patients. In 18 of these 89 patients, the diagnosis of lone PAF was established in the absence of any exclusion criterion. These 18 patients were invited and all came to the follow-up visit. All patients had received treatment with PPI in standard or double standard doses after endoscopic diagnosis of reflux esophagitis. The clinical characteristics of the patients and their endoscopic findings are listed in Table 1. In Patient 4, surgery had been performed 2 months before because of a highly differentiated intraductal breast cancer, and she was undergoing radiotherapy at the time of the follow-up visit. The duration of PAF varied considerably, from 2 to 32 years.

Clinical Effects of PPI

The symptoms reported by the patients are listed in Table 2. After PPI therapy, a decrease or disappearance of at least one PAF-related symptom occurred in 14 of 18 patients (78%). Interestingly, in two of the remaining four patients, GERD-related symptoms also persisted or appeared newly (Patients 4 and 12). The PAF score increased in only one patient (Patient 18), remained unchanged in three (Patients 4, 9, and 12), and decreased from -1 to -5 in the remaining patients (Table 2). Because of the decrease in PAF-related symptoms, [beta]-blockers or propafenone were discontinued in five patients (Patients 5, 7, 11, 15, and 16) 4 to 32 weeks after initiation of PPI therapy. In none of the patients did antiarrhythmic drugs have to be increased in dosage or newly prescribed. The electrocardiogram, recorded at the follow-up visit, showed sinus rhythm in all patients.

Discussion

In this pilot study, PPI therapy led to a decrease in PAF-related symptoms in 78% of cases with lone PAF and reflux esophagitis. In 28% of the patients, antiarrhythmic drugs could even be discontinued. Thus, therapy of GERD seems to be a promising strategy in patients with lone PAF.

PAF is the most common arrhythmia, with a mean annual incidence of 6.2 per 10,000 population per year. (3) PAF accounts for 25 to 62% of cases of AF. PAF has underlying causes and embolic risks similar to chronic AF. (4) PAF significantly impairs the quality of life due to palpitations, dizziness, exertional dyspnea, and weakness. (5) In patients with lone PAF, autonomic imbalances have been postulated as causative. (5), (6) Vagally mediated PAF occurs predominantly at night, during rest, and postprandially. Adrenergically induced PAF occurs predominantly in daytime, while under stress, or during exercise. Both forms of PAF may coexist in one patient, due to interactions between the two limbs of the autonomic nervous system. (6) The management of PAF is directed toward prevention of paroxysms and long-term maintenance of sinus rhythm. For this purpose, Class 1C antiarrhythmic drugs and amiodarone are highly effective, but they all have potentially dangerous side effects. Nonpharmacologic therapeutic options including pacemakers, electrophysiologic techniques, and implantable atrial defibrillators seem promising, but imply the risk and costs of an interventional procedure. (4)

GERD describes the symptoms and changes of the esophageal mucosa that result from reflux of stomach contents into the esophagus. GERD is the most frequent disease of the esophagus and is increasing in incidence and prevalence, with an annual incidence of 12 per 10,000 population. (7) GERD patients present with heartburn, regurgitation, dysphagia, and vomitus. (7) Complications of GERD include bleeding, esophageal stenosis, Barrett esophagus, and esophageal adenocarcinoma. (7) Several extraesophageal manifestations of GERD have been described. GERD has been reported to be associated with respiratory diseases such as pharyngitis, tonsillitis, bronchitis, pneumonia, emphysema, asthma, bronchiectasis, and empyema. (8) In addition, hoarseness (9) and otitis media (10) have been reported to be associated with GERD. Mechanisms for these extraesophageal manifestations of GERD may be aspiration, (11) mechanical factors, or autonomic dysregulation. (12) GERD can be treated by acid suppression therapy, among which the PPIs are the most successful. (13), (14)

Concerning the pathomechanism of PAF due to GERD, the following five hypotheses are proposed: first, the local inflammatory process penetrates also the esophageal wall and thus affects the adjacent vagal nerves. Inflammation of peripheral nerves--motor, sensory, and vegetative--have been described to affect myelination, discharge rate, and the propagation of stimuli along these fibers. (15) It has been recently shown that the innervation of the esophageal mucosa is altered in response to inflammation. (16) Second, local inflammation of the esophageal mucosa may induce afferent-efferent reflex mechanisms with involvement of the cerebral representation of the cardiac rhythm and thus lead to secondary stimulation of the vagal nerves. Possibly, local inflammation affects receptors that overstimulate the parasympathetic system via reflex loops involving the brain, similar to the bolus mechanism. Vagal overstimulation may lead to bradycardia and thus AF. (17) Third, propagation of the local inflammatory process through the esophageal wall may cause local pericarditis or atrial myocarditis, although esophagitis tends to be a mucosal and submucosal disease. Atrial myocardial biopsies in patients with lone AF have shown myocarditis in 66% of patients. (18) Fourth, GERD may lead to a release of inflammatory mediators, which may affect the atrial myocardium or the cardiac conduction system. Elevated C-reactive protein levels in patients with PAF have been reported recently. (19) Fifth, GERD may induce an autoimmune response that contributes to PAF. Autoantibodies against myosin heavy chain have been detected in the sera of patients with lone PAF. (20)

In four patients of the present study, the PAF-related symptoms remained unchanged or worsened. One explanation for the persistence of PAF may be that adrenergic mechanisms in these patients were more important than vagal

Key Points

* Paroxysmal atrial fibrillation is a common arrhythmia among adults that impairs quality of life.

* For the management of paroxysmal atrial fibrillation, antiarrhythmic drugs are effective, but they have potential dangerous side effects.

* Our observations suggest that in a subgroup of patients with paroxysmal atrial fibrillation, therapy of disease-causing esophagitis may be more effective, less dangerous, and less expensive than the established therapies. influences. This is substantiated by the occurrence of PAF during day or exertion in three of them (Patients 4, 9, and 18). However, Patient 12 had vagally induced PAF but an observational period of only 3 months. Because it is known that clinical remission of GERD may last more than 3 months, (14) it can be speculated that longer therapy would also have reduced the symptoms in this patient.

Limitations of this pilot study include the low number of patients, the retrospective design, and the lack of a control group. As all patients were informed about the diagnosis and treatment of reflux esophagitis immediately after gastroscopy, we cannot exclude that a placebo effect of PPIs contributed to the improvement of PAF in a subgroup of patients. Therefore, it is necessary to confirm the findings of our study in a prospective, randomized, double-blind, controlled trial. Since in our study the diagnosis of GERD was only based on endoscopy but not on 24-hour pH-metry, only patients with reflux esophagitis but not with endoscopy-negative GERD were included. However, it seems reasonable to speculate that patients with endoscopy-negative GERD and PAF might also benefit from PPI therapy.

Conclusion

If the results of this pilot study can be confirmed by a randomized trial, the clinical consequence would be to look for GERD in patients with lone PAF and to adequately treat GERD, if detected. Our observations suggest that, in a subgroup of patients with lone PAF, therapy of disease-causing esophagitis may be more effective, less dangerous, and less expensive than the established therapeutic options.

Key Points

* Paroxysmal atrail fibrillation is a common arrhythmia among adults that impairs quality of life.

* For the management of paroxysmal atrail fibrillation, antiarrhythmic drugs are effective, but they have potential dangerous side effects.

* Our observations suggest that in a subgroup of patients with paroxysmal atrail fibrillation, therapy of disease-causing esophagitis may be more effective, less dangerous, and less expensive than the established therapies.

Let us not look back in anger, Nor forward in fear, But around in awareness.

--James Thurber

Table 1. Clinical characteristics of the patients with reflux
esophagitis and paroxysmal atrial fibrillation (a)

Patient  Age/sex  Onset of PAF  Diagnosis  Months since
                                           endoscopy

 1       41/M     R, E, D, N    BE, HH           5

 2       62/F     R, N          BE, HH          22

 3       65/M     E, D, N       E1, HH          21

 4       52/F     E, D          E1, HH          16

 5       56/M     R, E, D, N    E2              11

 6       55/F     R, E, D, N    E1, CG           6

 7       57/M     R, E, D, N    E1, PU           9

 8       60/M     R, D, N       BE, HH          19

 9       67/M     R, E, D       E1               6

10       61/F     R, E, D       E1              14

11       69/M     R, N          BE              30

12       66/M     R, N          E1               3

13       54/F     R, N          E1, HH          17

14       43/F     R, D, N       E1               7

15       55/M     R, D          E1, HH          11

16       39/M     R, E, N       E2, HH          13

17       69/M     R, E, D, N    E1, HH           9

18       50/M     R, D, N       E1              26

Patient   Cardiac    Concomitant
medication   disorders

 1       P, BB       0

 2       BB, ACI     H

 3       P, Ami      RA

 4       0           DM, RA, C

 5       0           0

 6       BB, ACI, P  H, DM

 7       0           H, RA

 8       BB, OA      DM

 9       BB, OA      H

10       BB,         H
   Moxonidin

11       0           0

12       P, BB       0

13       0           0

14       BB          0

15       0           0

16       0           0

17       ACI, BB,    H
   Doxazos in

18       AMI, P      0

(a) PAF, paroxysmal atrial fibrillation; A, arterial hypertension; ACI,
angiotensin-converting enzyme inhibitor; Ami, amiodarone; BB,
[beta]-blocker; BE, Barrett esophagus; C, cancer; CG, chronic
gastritis; D, day; DM, diabetes mellitus; E, exercise; E1 and E2,
esophagitis Stage I and II according to Savary and Miller (2); HH,
hiatal hernia; N, night; OA, oral anticoagulation; P, propafenon; PU,
prepyloric ulcer; R, rest; RA, chronic respiratory infections or
asthma; M, male; F, female.

Table 2. Cardiac and gastrointestinal symptoms after proton-pump inhibitor therapy (a)

Patient      Palpitations,       Palpitations,
               frequency           duration

 1            [down arrow]             [up arrow]
 2            [down arrow]       [left and right arrow]
 3            [down arrow]            [down arrow]
 4       [left and right arrow]  [left and right arrow]
 5                 0                       0
 6             [up arrow]             [down arrow]
 7       [left and right arrow]       [down arrow]
 8            [down arrow]            [down arrow]
 9       [left and right arrow]  [left and right arrow]
10            [down arrow]       [left and right arrow]
11            [down arrow]       [left and right arrow]
12       [left and right arrow]  [left and right arrow]
13                 0                       0
14            [down arrow]       [left and right arrow]
15            [down arrow]            [down arrow]
16            [down arrow]            [down arrow]
17            [down arrow]            [down arrow]
18             [up arrow]              [up arrow]

Patient        Dizziness          Exertional
                                   dyspnoea

 1                 -                       0
 2                 -                       -
 3                 -                       -
 4       [left and right arrow]            -
 5                 0                       -
 6                 0                       0
 7                 -             [left and right arrow]
 8                 -                       -
 9                 -                       -
10                 -                       -
11                 -                       -
12                 -                       -
13                 -                       -
14       [left and right arrow]  [left and right arrow]
15                 -                       -
16       [left and right arrow]  [left and right arrow]
17                 0                       0
18                 -             [left and right arrow]

Patient         Weakness         Sum of          Heartburn
                                  PAF
                                 score (b)

 1                 0                -2                -
 2                 -                -1                -
 3                 0                -3                0
 4       [left and right arrow]      0      [left and right arrow]
 5                 0                -4                0
 6                 -                -2                -
 7       [left and right arrow]     -1                0
 8       [left and right arrow]     -2                -
 9       [left and right arrow]      0                -
10                 -                -1                0
11                 -                -1                0
12       [left and right arrow]      0                N
13                 0                -3                -
14       [left and right arrow]     -1                0
15                 -                -2                -
16                 -                -2                0
17                 0                -5                0
18                 -                +2                0

Patient      Regurgitation             Dysphagia

 1                 -                       -
 2                 -                       -
 3                 -                       -
 4       [left and right arrow]  [left and right arrow]
 5                 0                       -
 6       [left and right arrow]            -
 7                 -                       -
 8                 0                       -
 9                 0                       -
10                 0                       -
11                 -                       -
12       [left and right arrow]            -
13                 0                       -
14       [left and right arrow]  [left and right arrow]
15                 -                       -
16       [left and right arrow]            -
17       [left and right arrow]            -
18                 -                       -

Patient         Vomitus

 1                 -
 2                 0
 3                 -
 4       [left and right arrow]
 5                 -
 6                 -
 7                 -
 8                 -
 9                 -
10                 0
11                 -
12                 -
13                 0
14       [left and right arrow]
15                 -
16                 -
17                 -
18       [left and right arrow]

(a) PAF, paroxysmal atrial fibrillation; [up arrow], increased; [down
arrow], decreased; [left and right arrow], unchanged; 0,
disappearance;--, never reported; N, newly reported.
(b) PAF score: If a PAF-related symptom increased or was newly reported,
it was scored as +1. If a PAF-related symptom remained unchanged or was
never reported, it was scored as 0. If a PAF-related symptom decreased
or disappeared, it was scored as -1.

From the Departments of Medicine and Neurology, Krankenanstalt Rudolfstiftung, Vienna, Austria.

Reprint requests to Claudia Stollberger, MD, Steingasse 31/18, A-1030 Vienna, Austria. Email: [e-mail]

Accepted April 25, 2003.

Copyright [c] 2003 by The Southern Medical Association

0038-4348/03/9611-1128

References

1. Gordon J, Saleem SM, Ngaage DL, et al. Swallow syncope associated with paroxysmal atrial fibrillation. Eur J Cardiothorac Surg 2002;21:587-590.

2. Ottenjann R, Classen M (eds). Gastroenterologische Endoskopie: Lehrbuch und Atlas. Stuttgart, Ferdinand Enke Verlag, 1991, p 22.

3. Goudevenos JA, Vakalis JN, Giogiakas V, et al. An epidemiological study of symptomatic paroxysmal atrial fibrillation in northwest Greece. Europace 1999;1:226-233.

4. Lip GY, Hee FL. Paroxysmal atrial fibrillation. QJM 2001;94:665-678

5. van den Berg MP, Hassink RJ, Tuinenburg AE, et al. Quality of life in patients with paroxysmal atrial fibrillation and its predictors: Importance of the autonomic nervous system. Eur Heart J 2001;22:247-253.

6. Coumel P. Paroxysmal atrial fibrillation: A disorder of autonomic tone? Eur Heart J 1994;15(Suppl A):9-16.

7. Sonnenberg A, El-Serag HB. Clinical epidemiology and natural history of gastroesophageal reflux disease. Yale J Biol Med 1999;72:81-92.

8. Ruhl CE, Sonnenberg A, Everhart JE. Hospitalization with respiratory disease following hiatal hernia and reflux esophagitis in a prospective, population-based study. Ann Epidemiol 2001;11:477-483.

9. McNally PR, Maydonovitch CL, Prosek RA, et al. Evaluation of gastroesophageal reflux as a cause of idiopathic hoarseness. Dig Dis Sci 1989;34:1900-1904.

10. Poelmans J, Tack J, Feenstra L. Chronic middle ear disease and gastroesophageal reflux disease: A causal relation? Otol Neurotol 2001;22:447-450.

11. Tasker A, Dettmar PW, Panetti M, et al. Reflux of gastric juice and glue ear in children. Lancet 2002;359:493 (letter).

12. Harding SM. Gastroesophageal reflux, asthma, and mechanisms of interaction. Am J Med 2001;111(Suppl 8A):8S-12S.

13. Loffeld RJ. Reflux complaints, symptom score and the use of medication in patients with reflux esophagitis: Results of a long term follow-up study. Can J Gastroenterol 2001;15:505-508.

14. Chiba N, De Gara CJ, Wilkinson JM, et al. Speed of healing and symptom relief in grade II to IV gastroesophageal reflux disease: A metaanalysis. Gastroenterology 1997;112:1798-1810.

15. Smith KJ, Hall SM. Factors directly affecting impulse transmission in inflammatory demyelinating disease: Recent advances in our understanding. Curr Opin Neurol 2001;14:289-298.

16. Newton M, Kamm MA, Soediono PO, et al. Oesophageal epithelial innervation in health and reflux oesophagitis. Gut 1999;44:317-322.

17. Kanoupakis EM, Manios EG, Mavrakis HE, et al. Relation of autonomic modulation to recurrence of atrial fibrillation following cardioversion. Am J Cardiol 2000;86:954-958.

18. Frustaci A, Chimenti C, Bellocci F, et al. Histological substrate of atrial biopsies in patients with lone atrial fibrillation. Circulation 1997;96:1180-1184.

19. Dernellis J, Panaretou M. C-reactive protein and paroxysmal atrial fibrillation: Evidence of the implication of an inflammatory process in paroxysmal atrial fibrillation. Acta Cardiol 2001;56:375-380.

20. Maixent JM, Paganelli F, Scaglione J, et al. Antibodies against myosin in sera of patients with idiopathic paroxysmal atrial fibrillation. J Cardiovase Electrophysiol 1998;9:612-617.

Manuela Weigl, MD, Michael Gschwantler, MD, Edmund Gatterer, MD, Josef Finsterer, MD, and Claudia Stollberger, MD