Diseases of of the the Esophagus Esophagus (2015) (2016) ••, 29,••–•• 367–376 Diseases DOI: 10.1111/dote.12329 10.1111/dote.12329 DOI:

Original article

Value of the Gastroesophageal Reflux Disease Questionnaire (GerdQ) in predicting the proton pump inhibitor response in coronary artery disease patients with gastroesophageal reflux-related chest pain S. He,1 Y. Liu,2 Y. Chen,1 Y. Tang,1 J. Xu,1 C. Tang2 Departments of 1Gastroenterology and 2Cardiology, Suining Central Hospital, Suining, Sichuan, China

SUMMARY. Chest pain experienced by patients with coronary artery disease can be partly due to gastroesophageal reflux-induced chest pain (GERP). Empirical proton pump inhibitor (PPI) therapy has been recommended as an initial clinical approach for treating GERP. However, PPI use may lead to some health problems. The Gastroesophageal Reflux Disease Questionnaire (GerdQ) may represent a noninvasive and costeffective approach for avoiding PPI misuse and for identifying the appropriate patients for the PPI trial test. The aim of this pilot study was to prospectively evaluate the association between GerdQ scores and PPI response in patients with coronary artery disease (CAD) and GERP to determine whether the GerdQ predicts the PPI response in patients with CAD and GERP and to further validate the clinical application value of the GerdQ. A total of 154 consecutive patients with potential GERP were recruited to complete a GerdQ with subsequent PPI therapy. Based on the PPI trial result, patients were divided into a PPI-positive response group and a PPI-negative response group. The difference in the GerdQ scores between the two groups was assessed. The receiver operating characteristic (ROC) curve of GerdQ score was drawn according to the PPI response as the gold standard. The ability of GerdQ to predict the PPI response was assessed. A total of 96 patients completed the entire study; 62 patients (64.6%) were assigned to the PPI-positive response group, and 34 patients (35.4%) to the PPI-negative response group. The GerdQ score of the PPI-positive response group (8.11 ± 3.315) was significantly higher than that of the PPInegative response group (4.41 ± 2.743), and the difference was statistically significant (t = 5.863, P = 0.000). The ROC curve was drawn according to a PPI response assessment result with a score above 2 as the gold standard. The area under curve was 0.806. When the critical value of GerdQ score was 7.5, Youden index was up to 0.514, the diagnostic sensitivity was 0.661, and the diagnostic specificity was 0.853. A GerdQ score greater than 7.5 better predicts the response to the PPI trial therapy. There is a strong association between the GerdQ score and the response to PPI therapy. Higher GerdQ scores were predictive of a positive PPI response in CAD patients with GERP. The GerdQ may be a reasonable screening tool for GERP in patients with CAD who are prepared to accept PPI therapy. KEY WORDS: coronary artery disease, gastroesophageal reflux, Gastroesophageal Reflux Disease Questionnaire, noncardiac chest pain, proton pump inhibitor.

INTRODUCTION The distal esophagus and heart share a common afferent vagal supply.1–4 Chest pain experienced by patients with coronary artery disease (CAD) can be Address correspondence to: Ms Suyu He, Master, Department of Gastroenterology, Suining Central Hospital, Suining, Sichuan 629000, China. Email: [email protected] Specific author contributions: Suyu He and Yongjun Chen designed the study and performed the study; Yijun Liu, Yi Tang, Chuansu Tang and Jianyu Xu performed the study and analyzed the data; Suyu He and Yijun Liu wrote the manuscript. Conflicts of interest: The authors declare that they have no conflict of interest. C 2015 International Society for Diseases of the Esophagus V ©

partially of noncardiac origin, and the symptoms are frequently caused by gastroesophageal refluxinduced chest pain (GERP). A number of previous studies have provided evidence that the reflux prevalence is more than 40% in patients with CAD, with 40–70% of cases of reflux being directly associated with chest pain symptoms.5–7 Patients with known CAD are a challenging group to manage if concomitant GERP is present. Diagnostic tests are commonly employed to diagnose GERP in patients with CAD, and these may include upper endoscopy and pH monitoring. However, these diagnostic studies can be expensive, and many are associated with some risks in 367 1

368 Diseasesofof Esophagus 2 Diseases thethe Esophagus

patients with CAD, especially in patients with Canadian Cardiovascular Society (CCS) functional classes III–IV. Mortality rates as high as 1% have been reported in patients with recent AMI (acute myocardial infarction) undergoing EGD (esophagogastroduodenoscopy).8 This is higher than the 0.1% mortality rate reported for patients undergoing EGD for the evaluation of gastrointestinal hemorrhage and several log-fold higher than the baseline 0.0004% mortality rate of EGD.9,10 Previous studies have confirmed the value of the proton pump inhibitor (PPI) trial test in GERP.11 The diagnostic value of a PPI trial test has been confirmed, and a ≥50% symptom reduction under PPI treatment showed both a sensitivity and specificity of almost 90%. A high PPI dose (double reference dose, twice daily) can quickly provide important diagnostic information in patients with unexplained chest pain.12,13 However, PPI use has been shown to lead to some health problems. Firstly, PPI use may lead to nutrition deficiency. Such as B12 deficiency, it has shown that among patients with incident diagnoses of Vitamin B12 deficiency, 12% were dispensed a 2 or more years’ supply of PPIs.14 Secondly, PPI use has been shown to lead to some infections, including clostridium difficile-associated diarrhea (CDAD) and pneumonia. For the former, a recent meta-analysis has found a 65% increase in the incidence of CDAD among hospitalized patients who were PPI users.15 For the latter, the incidence of both community and hospital-acquired pneumonia was found increase in PPI users.16 Thirdly, PPI use may lead to the presence of polyps. Previous studies also found that PPI intake was the strongest risk factor associated with the presence of fundic gland polyps.17 The elderly might be at increased risk from PPI therapy, which indicates that an alternative noninvasive and cost-effective procedure is needed to avoid PPI misuse and to identify the appropriate patients for the PPI trial test. The Gastroesophageal Reflux Disease Questionnaire (GerdQ) is a patient-centered and self-reported diagnostic instrument for gastroesophageal reflux disease (GERD). Previous studies have proved that a GerdQ score over 8 can achieve an overall diagnostic accuracy similar to that of a gastroenterologist, with a sensitivity of 64.6% and a specificity of 71.4%.11,12 However, the utility of the GerdQ and the possible relationship with PPI response in GERP has not yet been addressed. We suppose that GerdQ may be useful for screening right patients with CAD and GERP to accept PPI trial test. The aim of this study was to prospectively evaluate the association between GerdQ scores and PPI response in patients with CAD and GERP. To evaluate the possibility of the GerdQ in predicting PPI response in patients with CAD and GERP, and further validate the clinical application value of GerdQ.

MATERIALS AND METHODS Patients Consecutive patients referred to our cardiovascular clinic for potential GERP between January 2011 and January 2014 were recruited for this study. All patients in the study were not remunerated. Requirements for participation included any one of the following: (i) the presence of chest pain lasting for at least 2 months and at least three episodes of chest pain per week; (ii) at least 50% narrowing of the coronary vessels on angiography that was not amenable to revascularization, as declared by at least two interventional cardiologists; (iii) chest pain that was not relieved with aggressive medical and/or surgical therapy; (iv) 40–70 years of age; and (v) CCS functional classes II–III. The exclusion criteria were as follows: (i) contraindications to any test used in the study; (ii) have a history of upper gastrointestinal surgery and/or connective tissue disease; (iii) have abnormal upper gastrointestinal endoscopy findings for reasons other than reflux disease, such as peptic ulcer (gastric and/or duodenal ulcer) and malignancy; (iv) have esophageal motility abnormality other than reflux disease, including nutcracker esophagus and nonspecific esophageal motor disorder, that might be associated with non-cardiac chest pain;18 (v) anti-reflux therapy or anti-reflux medication in the previous month; (vi) refusal to participate; and (vii) pregnancy. The study protocol was approved by the Ethics Committee of Suining Central Hospital, and written informed consent was obtained from all subjects prior to enrollment. Study design This study was a pilot study. It is a part of a Chinese project on acid-related chest pain in patients with CAD. All patients were informed of the purpose and principles of the study. Initial patient assessments included a medical history, physical examination, and cardiac function tests. All patients had undergone coronary vessel angiography, gastrointestinal endoscopy, and esophageal manometry followed by GerdQ and PPI therapy. All included patients continued their usual activities and cardiac medications. Gastroduodenoscopy and esophageal manometry Gastroduodenoscopy was conducted by experienced endoscopists with standard endoscopes (XQ-260, Olympus Optical Co. Ltd., Tokyo, Japan) after an overnight fast. The distal portion of the esophagus was carefully examined to determine the presence of mucosal injury. Esophageal manometry was performed using a multi-channel water-perfused C 2015 International Society for Diseases of the Esophagus V ©

Prediction ofofPPI Prediction PPIresponse responsebybyGerdQ GerdQ 369 3

manometry catheter (Medical Measurement Systems, B.V., Enschede, the Netherlands). The catheter was passed transnasally and positioned with the most distal channel in the stomach. The esophageal motility and lower esophagus sphincter (LES) function were measured by station pull-through techniques. With the patient supine, ‘wet’ swallows, each consisting of 5 mL of water, were given at intervals of 30 seconds. Peristaltic activity and LES function were observed and if any abnormalities were detected a further five 5 mL ‘wet’ swallows were given (‘full’ manometric study). On completion of the swallows, the catheter was slowly withdrawn, then, the upper esophageal sphincter position and function observed, and the catheter removed, and a series of 10 wet swallows was performed. The diagnosis of each esophageal motility abnormality was verified, according to the accepted published criteria.19

GerdQ GerdQ were applied in all qualified patients. The GerdQ is a self-administered diagnostic questionnaire.20 It consists of six symptom-related items comprising four reflux-related items that are positively related to GERD and two items that are negatively related to GERD. The first two questions (1 and 2) are positive predictors of GERD, and a higher score suggests a higher symptom frequency. Questions 3 and 4 address dyspeptic symptoms that lower the probability of GERD, i.e., they are negative predictors of GERD. The two final questions (5 and 6) assess the impact of GERD symptoms on patient lives and are also positive predictors of GERD. Briefly, the frequencies of the six symptoms, including heartburn, regurgitation, and upper abdominal pain, are evaluated using a 4-point Likert scale (Table 1).21 The patients are asked to recall, with the guidance of the investigator, how often they experienced the events described in the various questions during the preceding week and to score their answers using a 4-point scale ranging 0 to 3 for positive predictors and from 3 to 0 for negative predictors. The

total scores determine the GerdQ score, which ranges from 0 to 18. A higher score indicates a greater possibility of GERD.

PPI therapy Regardless of the GerdQ findings, all qualified patients received anti-reflux therapy consisting of high-dose PPI twice daily for at least 4 weeks. Three kinds of the region’s most commonly used PPIs, including omeprazole, esomeprazole, and lansoprazole, were included in the study (specifically, double reference dose, twice daily: omeprazole 40 mg twice daily, esomeprazole 40 mg twice daily, or lansoprazole 30 mg twice daily). Patients were randomly divided into these three PPI groups (omeprazole group, esomeprazole group, and lansoprazole group). After the start of treatment, the patients were scheduled for a visit after 4 weeks of therapy, including assessments of adverse events (two questions would be asked about: (i) Does your chest pain get worse? (ii) Is there any adverse reaction, such as rashes, after your prescription of PPI?) and an response to treatment. During the treatment, patients should visit the investigator immediately if they experience some adverse events. At the end of 4 weeks, for patients with inadequate symptom control, they can switch to a different PPI or not for an additional 4 weeks therapy. If patients choose to switch to a different PPI, patients taken omeprazole or esomeprazole could switch to lansoprazole; patients taken lansoprazole could switch to esomeprazole. After 8 weeks, the study was terminated. Symptom improvement was assessed in a face-to-face interview at the end of the anti-reflux therapy using a 4-point Likert scale (0 = no improvement, 1 = mild improvement, 2 = moderate improvement and 3 = marked improvement). According to the PPI therapy result, patients were divided into a PPI-positive response group and a PPI-negative response group. GERP or positive PPI response was determined as the presence of moderate or marked improvement in NCCP symptoms. Otherwise, patients were classified into the PPI-negative

Table 1 GerdQ self-assessment questionnaire

Questions 1. How often did you have a burning feeling behind your breastbone (heartburn)? 2. How often did you have stomach contents (liquids or food) moving upwards to your throat or mouth (regurgitation)? 3. How often did you have a pain in the center of the upper stomach? 4. How often did you have nausea? 5. How often did you have difficulty getting a good night’s sleep because of your heartburn and/or regurgitation? 6. How often did you take additional medication or your heartburn and/or regurgitation, other than what the physician told you to take? SP, symptom presence. C 2015 International Society for Diseases of the Esophagus V ©

SP 0 days

SP 1 day

SP 2–3 days

SP 4–7 days

0 0

1 1

2 2

3 3

3 3 0

2 2 1

1 1 2

0 0 3

0

1

2

3

370 Diseasesofof Esophagus 4 Diseases thethe Esophagus

response group. The difference between the GerdQ scores in the two groups was assessed. The receiver operating characteristic (ROC) curve of GerdQ score was drawn according to the PPI response as the gold standard. The predictive accuracy of GerdQ for PPI response was assessed.

off score was selected to maximize the trade-off between sensitivity and specificity. Statistical calculations were performed using SPSS 20.0 software (Statistical Product and Service Solutions, Chicago, IL, USA).

RESULTS

Statistical analysis Data with normal distributions are expressed as the mean ± standard deviation, whereas data with skewed distributions are expressed as median values (interquartile). The unpaired Student’s t-test was used for comparison of the data where applicable. The chi-square test was applied to test the PPI response rate between 4 weeks and 8 weeks. Student– Newman–Keuls test was used for comparison of different endoscopic finding groups’ PPI response and GerdQ data. ROC curve analysis was employed to evaluate the ability of the GerdQ to discriminate between patients with positive PPI response and those without. The optimal cut-off point was determined based on the highest Youden index. The cut-

Demographics A total of 154 patients met the inclusion criteria of the study. Of the 154 patients, 145 completed upper endoscopy. Among them, 38 had abnormal endoscopy findings (Fig. 1). Twenty patients had hiatal hernia (HH). Erosive esophagitis (EE) was found in 18 patients. Among the 18 patients, 11 had HH. Except for 10 patients with ulcer and 1 with gastric malignant, 134 patients accepted esophageal manometry. One hundred and twenty-eight completed the manometry testing. And 24 patients had abnormal manometry findings (Fig. 2). One hundred and four patients with normal esophageal manometry motility accepted the GerdQ test, 2 patients declined

Eligible patients (n = 154)

9 patients dropped out for they cannot tolerate it.

Patients completed endoscopy (n = 145)

Abnormal (n = 38) (1 malignant, 10 ulcer, 18 EE, 20 HH)

1 malignant, 10 ulcer were excluded.

Normal (n = 107)

Patients accepted manometry (n = 18 + 116 = 134)

6 patients dropped out for they cannot tolerate it.

Patients completed manometry (n = 128)

Abnormal (n = 24)

2 patients declined to go on the

study.

PPI-positive response group (n = 62)

Normal (n = 104)

Patients completed GerdQ test and PPI therapy (n = 96)

PPI-negative response group (n = 34)

The difference of GerdQ score of the two group was assessed.

Patients unfinished GerdQ test and PPI therapy (n = 6)

3 patients’ ischemic conditon was worsen. 2 patients quit because of drug side effect. 1 patient dropped without giving any reason.

Fig. 1 Endoscopic findings. C 2015 International Society for Diseases of the Esophagus V ©

Prediction ofofPPI Prediction PPIresponse responsebybyGerdQ GerdQ 371 5

Patients completed GerdQ test (n = 96) 4 weeks

PPI-positive response (n = 57)

PPI-negative response (n = 39)

25 patients didn’t switch PPI

Patients switched PPI (n = 14) (5O, 3E switched to L; 6L switched to E) 4 weeks

4 weeks

PPI-positive response (n = 1)

PPI-positive response (n = 4) (2 patients switched to L; 2 patients switched to E)

PPI-positive response (n = 57 + 5 = 62) Fig. 2 Esophageal motility results.

high-dose PPI therapy before the PPI therapy test, and 102 patients accepted the double dose PPI therapy. In three patients, the ischemic condition worsened, and these patients dropped out of the study; two patients discontinued the PPI because of the drug’s side effects, and one patient dropped out without providing any reason. A total of 96 patients (94.1%) finished the double-dose PPI therapy. Among them, 31 patients taken omeprazole, 33 taken esomeprazole, and 32 patients taken lansoprazole.

PPI-positive group and PPI-negative group (Table 3) in terms of the mean age, gender, body mass index, the presence of diabetes mellitus, high blood pressure, and CCS functional classes. And there were also no differences between the patients switched PPI or not (Table 4). Regarding different anti anginal drug administration, no differences were found among patients chose different PPIs (Table 5) or patients switched PPI or not (Table 6). However, compared with the PPI-negative group, the combination of CCB (calcium channel blocker), antiplatelet drugs,

PPI response after 4 weeks and 8 weeks PPI therapy

General characteristics and cardiac drug prescription There were no differences among patients taken different PPIs (Table 2) and no differences between the C 2015 International Society for Diseases of the Esophagus V ©

ROC curve

GerdQ score Reference

1.0

0.8

Sensitivity

After 4 weeks high-dose PPI therapy, there were 57 (59.4%) patients demonstrated moderate or marked improvement in NCCP (non cardiac chest pain) symptom. Thirty-nine (40.6%) patients who exhibited a PPI-negative response received an additional 4 weeks therapy. Among them, 14 patients switched to a different PPI. Four patients who switched to a different PPI and one patient who taken the same PPI as before exhibited a PPI-positive response (Fig. 3). After 8 weeks therapy, a total of 62 patients (64.6%) exhibited a PPI-positive response, and 34 (35.4%) exhibited a PPI-negative response (Fig. 4). The PPIpositive rate was 59.4% at 4 weeks and 64.6% at 8 weeks. There was no significant difference in the PPI therapy results between 4 weeks and 8 weeks (P = 0.457).

0.6

0.4

0.2

AUC = 0.806 0.0 0.0

0.2

0.4

0.6

0.8

1.0

1 - Specificity

Fig. 3 Proton pump inhibitor (PPI) therapy results and the strategy for changing PPIs (O: omeprazole; E: esomeprazole; L: lansoprazole).

372 Diseasesofof Esophagus 6 Diseases thethe Esophagus

Ulcer, 10

Gastric malignant, 1

Hiatal hernia, 20

Erosive esophagis, 18 Normal , 107

Fig. 4 Consort flow diagram through the study.

β-blockers, and nitrates (P = 0.012), were more common in the PPI-positive group (Table 7). Relationship between GerdQ data and PPI response The GerdQ score of the PPI-positive response group (8.11 ± 3.315) was significantly higher than that of the PPI-negative response group (4.41 ± 2.743), and the difference was statistically significant (t = 5.863, P = 0.000). Regarding the symptoms, the most important GerdQ question is question 1 and 2. Forty patients (42%) had heartburn feeling and 36 patients (37.5%) had regurgitation. Regarding the impacts of the disease, 38 patients (39.6%) answered that they have difficulty in getting a good night’s sleep because of their symptoms. The least important question is question 6. Eight patients (8%) answered that they have taken additional medicine for their heartburn or regurgitation. The ROC curve was drawn according to the PPI response assessment result higher than a score of 2 as the gold standard. The area under curve

was 0.806. When the critical value of GerdQ score was 7.5, Youden index was up to 0.514, the diagnostic sensitivity was 0.661, and the diagnostic specificity was 0.853 (Fig. 5). Relationship between endoscopy findings and response to PPI and GerdQ For patients who finished the PPI therapy and GerdQ, 24 patients with abnormal endoscopy findings, including 7 patients with EE, 7 patients with HH, 10 patients with EE and HH. All four groups, including EE, HH, EE, and HH, and normal endoscopy findings were compared in pairs. For patients who exhibited a PPI positive response, six patients with EE (85.7%), three patients with HH (42.9%), seven patients with EE and HH. However, these differences were not statistically significant in any paired comparison. The GerdQ score of the EE group (9.43 ± 2.992) was significantly higher than that of the normal endoscopy group (8.10 ± 4.228), and the dif-

Table 2 General characteristics of the three PPI groups Study population

Omeprazole (n = 31)

Esomeprazole (n = 33)

Lansoprazole (n = 32)

Age Gender, male BMI CCS II DM HB

56.91 ± 6.38 26 (84) 23.44 ± 3.41 19 (61) 7 (23) 20 (65)

59.47 ± 6.38 24 (73) 23.59 ± 3.10 18 (55) 6 (18) 20 (61)

60.16 ± 7.37 27 (84) 23.58 ± 3.17 21 (66) 6 (19) 22 (69)

Results expressed as mean ± SD. Gender, male, CCS, DM, and HB expressed as n (%). P > 0.05, compared between the groups. BMI, body mass index; CCS, Canadian Cardiovascular Society; DM, diabetes mellitus; HB, high blood pressure. C 2015 International Society for Diseases of the Esophagus V ©

Prediction ofofPPI Prediction PPIresponse responsebybyGerdQ GerdQ 373 7

Table 3 General characteristics of the two PPI response groups Study population

PPI (+)

PPI (−)

P value

Age Gender, male BMI HB CCS II DM

58.4 ± 7.05 48 (77) 23.8 ± 3.32 42 (65) 39 (63) 12 (19)

59.7 ± 6.35 28 (82) 23.2 ± 2.87 20 (69) 19 (56) 7 (21)

0.592 0.249 0.310 0.567 0.252 0.776

Results expressed as mean ± SD. Gender, male, CCS, DM expressed as n (%). BMI, body mass index; DM, diabetes mellitus; CCS, Canadian Cardiovascular Society; HB, high blood pressure.

ference was statistically significant (P = 0.038). There were a slighter difference between the other groups, but these differences were not statistically significant (Table 8).

Table 4 General characteristics of patients switched PPI or not Study population

C (n = 14)

NC (n = 25)

P value

Age Gender, male BMI CCS II HB DM

58.8 ± 6.82 9 (64) 23.1 ± 3.10 9 (64) 9 (65) 3 (21)

59.9 ± 6.31 19 (76) 23.1 ± 2.77 10 (40) 18 (69) 5 (20)

0.588 0.168 0.758 0.596 0.465 0.422

Results expressed as mean ± SD. Gender, male, CCS, DM expressed as n (number) (%). BMI: body mass index; C, change; CCS: Canadian Cardiovascular Society; DM: diabetes mellitus; NC, no change.

Table 5 Cardiac drug prescriptions of the three PPI groups

Drugs

Omeprazole Esomeprazole Lansoprazole (n = 31) (n = 33) (n = 32)

β-blockers N Antiplatelets CCB β-blockers +CCB + Antiplatelets + N

30 31 30 26 22

29 30 30 26 24

30 32 31 25 23

All results expressed as n (number). P > 0.05, compared between the groups. C, change; CCB, calcium channel blocker, N, nitrates; NC, no change.

Table 6 Cardiac drug prescriptions of patients switched PPI or not Drugs

C (n = 14)

NC (n = 25)

P value

β-blockers N Antiplatelets CCB β-blockers +CCB + Antiplatelets + N

12 14 12 9 8

21 31 22 18 10

0.781 0.152 0.694 0.365 0.548

All results expressed as n (number).C, change; CCB, calcium channel blocker, N, nitrates; NC, no change. C 2015 International Society for Diseases of the Esophagus V ©

Table 7 groups

Cardiac drug prescriptions of the two PPI response

Drugs

PPI (+) (n = 64)

PPI (−) (n = 32)

P value

β-blockers N Antiplatelets CCB β-blockers +CCB + Antiplatelets + N

58 61 59 50 47

31 32 32 27 22

0.130 0.136 0.665 0.776 0.012

All results expressed as n (number). CCB, calcium channel blocker; N, nitrates.

DISCUSSION Not all chest pain experienced by patients with CAD is definitely of cardiac origin. Chest pain in patients with CAD can be partly from GERP. Patients with known CAD are a challenging group to manage if concomitant NCCP is present. To date, there is no gold standard for diagnosing GERD-related NCCP, especially for diagnosing it in patients with CAD. Ambulatory esophageal pH monitoring allows objective quantitative assessment for excessive esophageal acid exposure and chest pain/reflux event association. While a strong symptom–reflux correlation is highly suggestive of GERD,1 most NCCP patients do not experience chest pain during conventional 24-hour pH study, making it difficult to determine the relationship between chest pain and acid reflux.4,5 Furthermore, pH monitoring or other invasive diagnosing procedures are costly and invasive, and many patients with CAD cannot tolerate. Thus, the diagnosis of GERP in patients with CAD remains challenging. In this study, we attempted to determine whether a symptom-based questionnaire, the GerdQ, can predict PPI response in patients with chest pain that might be related to CAD and GERP. The major finding of this study was that a GerdQ score greater than 7.5 can better predict the response to PPI therapy. Higher GerdQ scores were predictive of a positive PPI response in CAD patients with GERP. High-dose PPI treatment (double reference dose, twice daily) was prescribed in this study. Previous studies have shown that in patients with NCCP, the treatment response to high-dose PPI treatment provides important information regarding GERD, and should be considered first.12 High-dose PPI treatment can quickly provide important diagnostic information in patients with unexplained chest pain. And it should be administered before other invasive tests including ambulatory pH testing.22 However, the optimal duration of PPI trial for these patients remains uncertain. Most studies using PPIs have used just 2 weeks of PPI therapy to determine PPI response in patients with chest pain.11 This is the first study that 4 weeks time was used to assess patients’ response, and another 4

374 Diseasesofof Esophagus 8 Diseases thethe Esophagus

Non-specific esophageal molity disorder, 18

Nutcracker esophagus, 6

Normal esophageal molity, 104

Fig. 5 Receiver operating curve (ROC) of the Gastroesophageal Reflux Disease Questionnaire (GerdQ) in the prediction of PPI response. The area under curve (AUC), sensitivity, and 1-specificity by cut-off point of GerdQ score.

weeks can be added for patients with inadequate symptom control. It means that 8 weeks was given to test patients’ PPI response. Although mechanisms of chest pain in GERD patients are poorly understand, we have known that GERD-related NCCP can be attributed to esophageal pain hypersensitivity. Despite prompt relief of typical reflux symptoms after acid suppression, the process of ‘desensitization’ of acid-induced esophageal hypersensitivity may take much longer after PPI treatment, which leads to delayed symptom response of chest pain to PPI. It means a longer treatment time is needed in NCCPrelated patients. Xia et al. had compared the treatment response after PPI therapy for 4 weeks and 2 weeks. A treatment duration of 4 weeks (lansoprazole) resulted in a better LR- (likelihood ratio) (LR + 2.75; LR − 0.13)23 when compared with 2 weeks (omeprazole,24 LR + 2.7; LR− 0.15). Flook et al. also

Table 8 The relationship between endoscopic findings and response to PPI and endoscopic findings and GerdQ Endoscopic findings

GerdQ

PPI (+)

EE (n = 7)* HH (n = 7) EE + HH (n = 10) Normal (n = 72)*

9.43 ± 2.992 6.29 ± 4.071 8.10 ± 4.228 6.42 ± 3.528

6 (85.7%) 3 (42.9%) 7 (70%) 26 (63.9%)

*P < 0.05 (comparison between the two groups). EE + HH, patients with EE and HH. EE, erosive esophagitis; GerdQ, Gastroesophageal Reflux Disease Questionnaire; HH, hiatal hernia.

found that a 4-week course of high-dose esomeprazole provided statistically significant relief of unexplained chest pain in patients with NCCP.25 Recent investigations have suggested that patients with NCCP who are treated empirically or those with diagnosis of GERDrelated NCCP should receive double-dose PPI for at least 2 months.13 Considered some patients may benefit from switching PPI or prolonging the treatment, we chose 8 weeks of PPI therapy to determine response in patients with NCCP and CAD in our study. Actually, according to the two possibly effective change measures (omeprazole or esomeprazole to lansoprazole; lansoprazole to esomeprazole),26–28 four of the 14 patients (28.5%) received positive PPI response. We have compared the PPI-positive rate at 4 weeks (59.4%) and 8 weeks (64.6%). We found that there was no significant difference in the PPI therapy results between 4 weeks and 8 weeks (P = 0.457). Thus, we consider that 4 weeks may be a suitable duration of PPI trial for patients with NCCP. Although PPI therapy provides important information regarding GERD, in patients with unspecified GERD symptoms, the specificity of PPI therapy for a GERD diagnosis has been reported to be as low as 54%.29 It is also known that some patients with NCCP fail to respond to PPI treatment.30 In addition, there is growing concern regarding potential adverse effects of PPI therapy, and elderly, chronically ill patients are at especially increased risk of PPI therapy.31 Thus, a tool to help identify the appropriate patients for PPI therapy is needed. C 2015 International Society for Diseases of the Esophagus V ©

Prediction ofofPPI Prediction PPIresponse responsebybyGerdQ GerdQ 375 9

Current guidelines recommend a clinical diagnosis of GERD based on the presence of the most troublesome symptoms of heartburn and regurgitation.32 Generally, patients with obvious regurgitation and heartburn have a higher likelihood that their chest pain is due to reflux. Thus, the presence of GERP in patients with CAD may be predicted by the GerdQ because it primarily and quantitatively evaluates the frequency and intensity of these symptoms. Previous studies have found that the implementation of the GerdQ reduces the need for upper endoscopy and 24-hour pH monitoring and improves resource utilization.33,34 This is the first study to investigate the ability of the GerdQ to predict patients’ PPI response. We propose that the GerdQ, without indications for endoscopy and pH monitoring, accurately predicts the PPI treatment response. The use of the GerdQ would also ensure that physicians do not underestimate the symptom burden or overestimate the effect of medical treatment, as previously described.35 We found that the optimal GerdQ cut-off score for positive PPI response in patients with CAD and NCCP of suspected GERD was 7.5, which was consistent with that previously reported in patients with GERD,20 corresponding to a diagnostic sensitivity of 66.1% and a diagnostic specificity of 85.3%. In addition, we found that patients with EE had a higher GerdQ scores compared with patients with normal endoscopic findings (P = 0.038). This is consistent with previous study found that patients with reflux esophagitis had a higher GerdQ scores.36 However, when compared patients with different endoscopy findings. There were no differences found in any paired comparison. We think it might be explained by the use of high-dose PPI therapy. Previous studies have investigated the possible influence of cardiac drugs on GERD morbidity.37,38 In our study, we also analyzed the rate of anti-anginal drug administration in the two groups. Because we found that the combinations of CCB, antiplatelet drugs, and nitrates, and the combination of CCB, antiplatelet drugs, β-blockers, and nitrates were more frequent in the patients with GERD and CAD in our previous study,39 we compared the prescription of combined CCB, antiplatelet drugs, and nitrates, and the combination of CCB, antiplatelet drugs, β-blockers, and nitrates in this study, and obtained the same result. All research studies have limitations. First, our patient population was primarily Chinese, and it is possible that our findings do not accurately reflect the findings of the questionnaire if applied in other populations. In addition, the strict inclusion criteria possibly resulted in selection bias; therefore, the characteristics of the enrollees may not resemble those of a typical CAD chest pain population. Nevertheless, these patients are usually potential candidates for empiric anti-reflux therapy and should be C 2015 International Society for Diseases of the Esophagus V ©

screened using the GerdQ prior to therapy. Second, the study was not placebo controlled. Because we propose that the patients had both cardiac chest pain and GERP, it is expected that acid-reducing therapy would be helpful. Therefore follow-up without acidsuppression therapy was not available, particularly after realizing the positive impact of this therapy on patient management. Furthermore, the small number of enrolled patients did not permit any randomized use of a placebo. However, all of the patients presented here were well defined and homogenous. Third, our study results were mainly based on the patients’ subjective assessment. And the GerdQ demonstrated a suboptimal sensitivity and specificity in predicting GERP. Given that there is no single test with a high-accuracy rate for diagnosing GERD and that endoscopy or pH monitoring alone cannot identify all patients with GERD, a clinical diagnosis of GERD based on the presence of the most troublesome symptoms has been recommended.40,41 Although Lacy et al. have found when compared with 48-hour wireless pH testing, the GerdQ cannot accurately diagnose GERD.42 However, this study was done in patients with typical reflux symptoms. It also found that elevated GerdQ scores were associated with increased likelihood of an abnormal pH study. And GerdQ subscale scores for regurgitation were associated with an abnormal pH study. As a matter of fact, many patients with CAD cannot tolerate the invasive diagnostic procedures. Noninvasive clinical assessment might be more effective option. Thus, considering its invasive and expensive nature, pH monitoring is currently less cost-effective than the GerdQ of diagnosis of GERP. Fourth, the GerdQ was initially designed for GERD and lacks the items related to chest pain. However, the addition of chest pain evaluation items to the GerdQ might not likely improve its predictive accuracy because the character and timing of the pain do not provide clues regarding the cause of the pain. These issues would be investigated in our future study. In conclusion, there is a strong association between the GerdQ score and the response to PPI therapy. Higher GerdQ scores were predictive of a positive PPI response in CAD patients with GERP. The symptombased approach using the GerdQ may provide physicians with a tool for more structured care of patients with CAD and GERP. The GerdQ is recommended as a screening tool for GERP in patients with CAD who are prepared to accept PPI therapy. Acknowledgments We would like to thank all members of the Scientific Education section of Suining Central Hospital and the physicians who were involved in planning and conducting the study. We also wish to thank the local health departments of the study area.

376 Diseasesof ofthe theEsophagus Esophagus 10 Diseases

References 1 Wright R A, Miller S A, Corsello B F. Acid induced esophagobronchial-cardiac reflexes in humans. Gastroenterology 1990; 99: 71–3. 2 Verberne A J M, Saita M, Sartor D M. Chemical stimulation of vagal afferent neurons and sympathetic vasomotor tone. Brain Res Rev 2003; 41: 288–305. 3 Makk L J, Leesar M, Joseph A, Prince C P, Wright R A. Cardioesophageal reflexes: an invasive human study. Dig Dis Sci 2000; 45: 2451–4. 4 Johansson S, Wallander M A, Ruigómez A, García Rodríguez L A. Is there any association between myocardial infarction, gastroesophagealreflux disease and acid suppressing drugs? Aliment Pharmacol Ther 2003; 18: 973–8. 5 Garcia-Pulido J, Patel P H, Hunter W C, Douglas J E, Thomas E. Esophageal contribution to chest pain in patients with coronary artery disease. Chest 1990; 98: 806–10. 6 Talwar V, Wurm P, Bankart M J, Gershlick A H, de Caestecker J S. Clinical trial: chest pain caused by presumed gastro-oesophageal reflux in coronary artery disease – controlled study of lansoprazole vs. placebo. Aliment Pharmacol Ther 2010; 32: 191–9. 7 Mehta A J, de Caestecker J S, Camm A J, Northfield T C. Gastro-oesophageal reflux in patients with coronary artery disease: how common is it and does it matter? Eur J Gastroenterol Hepatol 1996; 8: 973–8. 8 Lin S, Konstance R, Jollis J et al. The utility of upper endoscopy in patients with concomitant upper gastrointestinal bleeding and acute myocardial infarction. Dig Dis Sci 2006; 51: 2377–83. 9 Gilbert D A, Silverstein F E, Tedesco F J. National ASGE survey on upper gastrointestinal bleeding: complications of endoscopy. Dig Dis Sci 1981; 26: 55–9. 10 Silvis S E, Nebel O, Rogers G et al. Endoscopic complications. Results of the 1974 American Society for Gastrointestinal Endoscopy Survey. JAMA 1976; 235: 928–30. 11 Cremonini F, Wise J, Moayyedi P, Talley N J. Diagnostic and therapeutic use of proton pump inhibitors in non-cardiac chest pain: a meta-analysis. Am J Gastroenterol 2005; 100: 1226–32. 12 Wertli M M, Ruchti K B, Steurer J, Held U. Diagnostic indicators of non-cardiovascular chest pain: a systematic review and meta-analysis. BMC Med 2013; 11: 239. 13 Hershcovici T, Achem S R, Jha L K, Fass R. Systematic review: the treatment of noncardiac chest pain. Aliment Pharmacol Ther 2012; 35: 5–14. 14 Lam J R, Schneider J L, Zhao W, Corley D A. Proton pump inhibitor and histamine 2 receptor antagonist use and vitamin B12 deficiency. JAMA 2013; 310: 2435–42. 15 Janarthanan S, Ditah I, Adler D G, Ehrinpreis M N. Clostridium difficile-associated diarrhea and proton pump inhibitor therapy: a meta-analysis. Am J Gastroenterol 2012; 107: 1001–10. 16 Eom C S, Jeon C Y, Lim J W, Cho E G, Park S M, Lee K S. Use of acid-suppressive drugs and risk of pneumonia: a systematic review and meta-analysis. CMAJ 2011; 183: 310–9. 17 Zelter A, Fernández J L, Bilder C et al. Fundic gland polyps and association with proton pump inhibitor intake: a prospective study in 1,780 endoscopies. Dig Dis Sci 2011; 56: 1743–8. 18 Fass R, Achem S R. Noncardiac chest pain: epidemiology, natural course and pathogenesis. J Neurogastroenterol Motil 2011; 17: 110–23. 19 Spechler S J, Castell D O. Classification of oesophageal motility abnormalities. Gut 2001; 49: 145–51. 20 Jones R, Junghard O, Dent J et al. Development of the GerdQ, a tool for the diagnosis and management of gastro-oesophageal reflux disease in primary care. Aliment Pharmacol Ther 2009; 30: 1030–8. 21 Tielemans M M, van Oijen M G. Online follow-up of individuals with gastroesophageal reflux disease using a patientreported outcomes instrument: results of an observational study. BMC Gastroenterol 2013; 13: 144. 22 Kahrilas P J, Smout A J. Esophagel disorders. Am J Gastroenterol 2010; 105: 747–56. 23 Xia H H, Lai K C, Lam S K et al. Symptomatic response to lansoprazole predicts abnormal acid reflux in endoscopy-

24

25

26

27 28

29

30 31 32 33

34

35

36

37

38

39 40 41

42

negative patients with non-cardiac chest pain. Aliment Pharmacol Ther 2003; 17: 369–77. Pandak W M, Arezo S, Everett S et al. Short course of omeprazole – A better first diagnostic approach to noncardiac chest pain than endoscopy, manometry, or 24-hour esophageal pH monitoring. J Clin Gastroenterol 2002; 35: 307–14. Flook N W, Moayyedi P, Dent J et al. Acid-suppressive therapy with esomeprazole for relief of unexplained chest pain in primary care: a randomized, double-blind, placebo-controlled trial. Am J Gastroenterol 2013; 108: 56–64. Fass R, Murthy U, Hayden C W et al. Omeprazole 40 mg once a day is equally effective as lansoprazole 30 mg twice a day in symptom control of patients with gastro-oesophageal reflux disease (GERD) who are resistant to conventional-dose lansoprazole therapy-a prospective, randomized, multi-centre study. Aliment Pharmacol Ther 2000; 14: 1595–603. Fass R, Sontag S J, Traxler B, Sostek M. Treatment of patients with persistent heartburn symptoms: a double-blind, randomized trial. Clin Gastroenterol Hepatol 2006; 4: 50–6. Johnson D A, Stacy T, Ryan M et al. A comparison of esomeprazole and lansoprazole for control of intragastric pH in patients with symptoms of gastro-oesophageal reflux disease. Aliment Pharmacol Ther 2005; 22: 129–34. Numans M E, Lau J, de Wit N J, Bonis P A. Short-term treatment with proton-pump inhibitors as a test for gastroesophageal reflux disease: a meta-analysis of diagnostic test characteristics. Ann Intern Med 2004; 140: 518–27. Gasiorowska A, Fass R. The proton pump inhibitor (PPI) test in GERD: does it still have a role? J Clin Gastroenterol 2008; 42: 867–74. Sheen E, Triadafilopoulos G. Adverse effects of long-term proton pump inhibitor therapy. Dig Dis Sci 2011; 56: 931–50. Katz P O, Gerson L B, Vela M F. Guidelines for the diagnosis and management of gastro-esophageal reflux disease. Am J Gastroenterol 2013; 108: 308–28. Jonasson C, Wernersson B, Hoff D A, Hatlebakk J G. Validation of the GerdQ questionnaire for the diagnosis of gastrooesophageal reflux disease. Aliment Pharmacol Ther 2013; 37: 564–72. Bai Y, Du Y, Zou D. Chinese GerdQ Research Group. Gastroesophageal Reflux Disease Questionnaire (GerdQ) in real-world practice: a national multicenter survey on 8065 patients. J Gastroenterol Hepatol 2013; 28: 626–31. McColl E, Junghard O, Wiklund I, Revicki D A. Assessing symptoms in gastro-esophageal reflux disease: how well do clinicians’ assessments agree with those of their patients? Am J Gastroenterol 2005; 100: 11–8. Bai Y, Du Y, Zou D et al. Gastroesophageal Reflux Disease Questionnaire (GerdQ) in real-world practice: a national multicenter survey on 8065 patients. J Gastroenterol Hepatol 2013; 28: 626–31. Nasseri-Moghaddam S, Nokhbeh-Zaeem H, Saniee P, Pedramnia S, Sotoudeh M, Malekzadeh R. Oral nitrate reductase activity and erosive gastro-esophageal reflux disease: a nitrate hypothesis for GERD pathogenesis. Dig Dis Sci 2012; 57: 413–8. Ruszniewski P, Soufflet C, Barthélémy P. Nonsteroidal antiinflammatory drug use as a risk factor for gastro-oesophageal reflux disease: an observational study. Aliment Pharmacol Ther 2008; 28: 1134–9. Liu Y, He S, Chen Y et al. Acid reflux in patients with coronary artery disease and refractory chest pain. Intern Med 2013; 50: 1165–71. Tytgat G N, McColl K, Tack J et al. New algorithm for the treatment of gastro-oesophageal reflux disease. Aliment Pharmacol Ther 2008; 27: 249–56. Kahrilas P J, Shaheen N J, Vaezi M F. American Gastroenterological Association Institute, Clinical Practice and Quality Management Committee. American Gastroenterological Association Institute technical review on the management of gastro-esophageal reflux disease. Gastroenterology 2008; 135: 1392–413. Lacy B E, Chehade R, Crowell M D. A prospective study to compare a symptom-based reflux disease questionnaire to 48-h wireless pH monitoring for the identification of gastroesophageal reflux (revised 2-26-11). Am J Gastroenterol 2011; 106: 1604–11. C 2015 International Society for Diseases of the Esophagus V ©