Bedtime H2 blockers improve nocturnal gastric acid control in GERD patients on proton pump inhibitors


Correspondence to: Dr Katz Graduate Hospital, Suite 501, Pepper Pavilion, 1800 Lombard Street, Philadelphia, PA 19146, USA. E-mail:



Proton pump inhibitors taken twice daily before meals (proton pump inhibitor b.d. AC) effectively controls daytime gastric pH; however, nocturnal gastric acid breakthrough (NAB) occurs in more than 75% of patients. Adding an H2-blocker at bedtime decreases NAB in normal subjects. The efficacy of this regimen has not been evaluated in GERD patients. The aim of this study was to assess the effects of proton pump inhibitor b.d., both with and without bedtime H2-blocker on intragastric pH and the occurrence of NAB in GERD patients.


Prolonged ambulatory pH studies in GERD patients were reviewed. Group A: 60 patients (mean age 53 years, male 30) taking either omeprazole 20 mg or lansoprazole 30 mg b.d. Group B: 45 patients (mean age 49 years, male 23) on proton pump inhibitor b.d. (omeprazole 20 mg or lansoprazole 30 mg) plus an H2-blocker at bedtime (ranitidine 300 mg, famotidine 40 mg or nizatidine 300 mg). Eleven patients were evaluated during treatment with both regimens (group C). The percentage time of nocturnal and daytime intragastric pH > 4 and per cent of patients with gastric NAB were analysed. In the patients with NAB, its duration and associated oesophageal acid exposure also were analysed.


Median percentage time intragastric pH > 4 overnight was 51% in group A, compared to 96% in group B (P < 0.0001). Median percentage daytime pH > 4 was 73% in group A and 79.8% in group B (P=0.14). Median percentage time intragastric pH >p 4 overnight increased from 54.6% without H2RA to 96.5% after adding bedtime H2RA (P=0.0013) in group C patients. NAB occurred in 82% patients in group A and 40% in group B (P < 0.0001). The mean duration of oesophageal acid exposure during NAB was significantly shorter in group B (18 ± 6 min) than in group A (42 ± 9 min, P=0.04).


Adding a bedtime H2-blocker to the treatment enhanced nocturnal gastric pH control and decreased NAB compared to the proton pump inhibitor b.d. regimen. A bedtime H2-blocker also decreased oesophageal acid exposure during NAB.


Adding a bedtime H2-blocker to a proton pump inhibitor b.d. regimen should be considered in patients who require continued nocturnal gastric acid control whilst taking proton pump inhibitor b.d.


Gastro-oesophageal reflux disease (GERD) is a common condition that usually requires regular treatment.1 The current pharmacological treatment for GERD is based on suppression of gastric acid secretion to produce symptom relief and mucosal healing. Proton pump inhibitors are the most effective medical therapy to realise these goals, achieving relief of symptoms in a mean of 78% of patients and healing oesophagitis in mean of 83% over a 4–8-week period.2 Some patients require greater degrees of acid suppression. These include patients with severe oesophagitis, scleroderma, Barrett’s oesophagus, and some patients with atypical manifestation of GERD,3 a group in whom recumbent reflux is common.4 We have previously shown that proton pump inhibitors given twice daily do not suppress intragastric pH sufficiently overnight to prevent recovery of acidity to potentially harmful pH levels. Nocturnal acid breakthrough (NAB), which we have defined as an intragastric pH < 4 for more than one continuous hour overnight, occurs in 75% of GERD patients and normal subjects,5 and may be accompanied by oesophageal reflux. Adding bedtime ranitidine 150–300 mg compared to additional proton pump inhibitor significantly decreased nocturnal gastric acidity and the frequency of NAB6 in normal volunteers. The efficacy of this regimen over time in the control of intragastric pH in GERD patients has not been evaluated. The aims of this study were to assess1 the effects of adding bedtime H2-blocker to proton pump inhibitor b.d. on intragastric pH control and the frequency of NAB and2 its effects on nocturnal oesophageal reflux in GERD patients. We separately compared these issues in an age and gender matched group of patients on proton pump inhibitor b.d. A small group of patients treated with both proton pump inhibitor b.d. and subsequently with proton pump inhibitor b.d. plus H2-blocker at bedtime was also assessed.



The records of all patients who had undergone prolonged dual channel ambulatory pH monitoring for GERD between May 1996 and October 1999 were retrospectively reviewed. One hundred and five consecutive patients who were previously diagnosed with GERD and treated with proton pump inhibitor b.d., or proton pump inhibitor b.d. plus bedtime H2 blockers (proton pump inhibitor + H2RA) were selected for this review. These patients were divided into the following groups. Group A included 60 patients (mean age 53 years, range 24–81, male 30) on proton pump inhibitor twice daily and served as control. Group B included 45 patients (mean age 49 years, range 14–82, male 23) on proton pump inhibitors twice daily plus H2RA at bedtime. Group C consisted of 11 patients who were studied on proton pump inhibitor b.d. and subsequently re-tested after adding a bedtime H2RA because of continued symptoms. The interval between the two studies was at least 4 weeks. The tracings of these 11 patients were re-analysed (from 22.00 to 08.00 hours) separately to examine the effects of adding bedtime H2 blocker on nocturnal gastric acid control and oesophageal acid reflux. These patients are thus included in group A, B and also reanalysed separately as group C.

Medications included a combination of omeprazole or lansoprazole with ranitidine, famotidine or nizatidine. Table 1 shows details of the distribution of drug regimen among all patients. Patients in all groups were on their respective regimens for at least 28 days and were all uniformly instructed to take their proton pump inhibitor half an hour before breakfast and dinner, and H2RA at bedtime. The indication for pH study was either to confirm the efficacy of treatment or to evaluate continued symptoms. Endoscopy was not performed systematically and H. pylori status was not investigated. The majority of patients studied in our laboratory have either extra-oesophageal (atypical) symptoms, refractory typical GERD symptoms, oesophagitis or Barrett’s oesophagus. Specific disease severity was not assessed, nor were groups compared for type of symptoms.

Table 1.   Distribution of drugs taken Thumbnail image of

Prolonged ambulatory pH monitoring

Prolonged ambulatory pH studies were performed with either a Medtronics (Medtronic Functional Test Inc Minneapolis, MN) or Sandhill (Sandhill Scientific Inc, Littleton, CO) recording system. The dual electrodes were initially calibrated in buffer solutions of pH 7, pH 4 and pH 1. The probe was introduced transnasally. The distal oesophageal pH electrode was placed 5 cm above the proximal border of the LES, which was located manometrically prior to placement, with the intragastric electrode located 15 cm distally about 5–7 cm below the LES in the gastric fundus. Recordings lasted for 24 h. The patients were instructed to follow their usual daily routine.

Data analysis

Digital recordings were transferred to an IBM compatible computer and were analysed using standard commercial software (Esophagram, Synectics Medical, or GERD check, Sandhill Scientific Inc.). The percentage times of gastric pH < 4 were calculated for the upright and recumbent periods. The percentage time of gastric pH < 4 was converted to pH > 4 for the purpose of illustration of gastric acid control. The nocturnal (recumbent) periods (from 22.00 to 08.00 hours) of the pH recordings were further analysed for the presence of nocturnal acid breakthrough (gastric pH < 4 lasting for more than a continuous hour), and accompanying oesophageal acid reflux, defined as any time of oesophageal pH < 4. In the patients with NAB, the duration and median pH of the NAB were also measured and the mean of the medians was reported. All meals were excluded from analysis. All data were subjected to a normality test. A paired t-test or unpaired t-test was used when the data passed the normality test, otherwise a nonparametric test was used. A χ2 test was used for percentage data. A P-value of less than 0.05 was considered as significant. Statistical software, GRAPHPAD PRISM (GraphPad Software, Inc. San Diego, CA) was used in this study for all analyses.


Intragastric pH control

Overall the median percentage time of gastric pH > 4 during the recumbent period was 51% (IQR 38–88%) in group A and 96% (IQR 66–100%) in group B (P < 0.0001, Figure 1). The median percentage time pH > 4 in the upright period was 73.4% (IQR 53–94%) in group A and 79.8% (IQR 65–96%, P=0.14) in group B. Thus, the addition of bedtime H2 blockers significantly enhanced gastric acid control in recumbent period without decreasing the effects of the morning dose of proton pump inhibitor.

Figure 1.

 Median percentage time gastric pH > 4 in patients on proton pump inhibitor b.d. and proton pump inhibitor b.d. + H2RA. In the recumbent period, the percentage time gastric pH > 4 was significantly greater in proton pump inhibitor b.d. + H2RA. *< 0.0001 Mann–Whitney test, There was no significant difference between the two groups in the upright period.

Patients were further divided into subgroups by the degree of their intragastric acid control during the recumbent period (Figure 2a). In group A, 4 of 60 (7%) patients had pH > 4 100% of the time, compared to 12 out of 45 (27%) patients in group B (P < 0.001). Similarly, 9 out of 60 (15%) patients achieved 90–99% time intragastric pH > 4 in group A, vs. 16 out of 45 (36%) patients in group B (P < 0.001). In contrast, patients having percentage time of gastric pH > 4 less than 50% was 48% in group A vs. 11% in group B (P < 0.0001). A similar analysis of the upright period did not show any significant difference (Figure 2b).

Figure 2.

 Percentage of patients who reached different level of intragastric pH control. (a) There was a significantly higher percentage of patients in group B who reached 100% and 90–99% of gastric pH > 4 in the recumbent period than in group A. On the other hand, there was a significantly lower percentage of patients who did not achieve more than 50% time intragastric pH > 4 in group B than in group A. *P < 0.0001, χ2 -test. (b) There was no significant difference between the two groups in gastric pH control in the upright period. PPI=proton pump inhibitor.

Nocturnal gastric acid breakthrough (NAB) and associated oesophageal acid reflux

NAB occurred in 49 out of 60 patients (82%) in group A, and in 18 out of 45 (40%) patients in group B (P < 0.0001, χ2 test). The duration of NAB was not significantly different between groups A and B (257 ± 19 min in group A vs. 198 ± 27 min in group B, P > 0.05). However, patients on H2RA at bedtime had a higher mean gastric pH during NAB (2.4 ± 0.08 in group A vs. 2.8 ± 0.15 in group B, P=0.02). Moreover, the duration of oesophageal acid exposure (pH < 4) was significantly shorter (42 ± 8.8 min vs. 18 ± 6.2 minute, P=0.04) in group B. There was a good correlation between percentage time intragastric pH < 4 and oesophageal acid exposure in patients in both groups (Figure 3a,b).

Figure 3.

 Correlation between nocturnal gastric pH control and abnormal oesophageal acid exposure. The pH tracing of nocturnal period (22.00–08.00 hours) was analysed for percentage time pH < 4. There was a good correlation between nocturnal gastric pH < 4 and abnormal oesophageal acid exposure in both groups A and B.

Patients on proton pump inhibitor b.d. with and without H2RA HS

Eleven out of 105 patients (group C) tested on the proton pump inhibitor b.d. only regimen were subsequently re-tested on proton pump inhibitor b.d. + H2RA because of continued symptoms. Median percentage time intragastric pH > 4 was 54.6% (IQR 46.5–66.7%) on proton pump inhibitor b.d. regimen and 96.5% (IQR 79.4–100%) on proton pump inhibitor b.d. plus H2RA (P=0.0013, Mann–Whitney test). Median percentage time oesophageal pH < 4 was 1.6% (IQR 0–18.5%) on proton pump inhibitor b.d. and 0% (IQR 0–0.55%) on proton pump inhibitor b.d. + H2RA (P=0.06, Mann–Whitney test). Individual data is shown in Figure 4.

Figure 4.

 Percentage time gastric pH > 4 and oesophageal pH < 4 in 11 patients on proton pump inhibitor b.d. regimen, with and without H2RA in the nocturnal period (22.00–06.00 hours).


The goals of treatment for patients with GERD are to eliminate symptoms, heal mucosal lesions, maintain symptomatic and endoscopic remission and, whenever possible, prevent complications.7 The healing rate of erosive oesophagitis at 8 weeks is directly related to the duration of suppression of intragastric acidity to pH > 4 over a 24-h period.8 Ladas et al. recently showed that the time intragastric pH < 4 has predictive value for recurrence of oesophagitis for patients on omeprazole (20 mg/day) maintenance therapy after oesophagitis has been healed,9 further suggesting the importance of pH control in maintenance of GERD healing.

Proton pump inhibitor given in a single daily dose provides sufficient acid suppression to effectively treat most patients. Symptom relief can be expected in about 83% of cases (range 71–96%) and oesophagitis healing in 78% (range 62–94%).10 Omeprazole 20 mg and lansoprazole 30 mg substantially increase the median 24 h gastric pH.11–13 Intragastric pH is maintained above 4 for 11–18 h, however, intragastric pH control is poorer overnight than in daytime.14 Despite the efficacy of single dose proton pump inhibitor in gastric acidity control, symptom relief, and healing of erosive oesophagitis, some patients, particularly those with higher grades of oesophagitis, do not heal as well and may require increased proton pump inhibitor dosing. In addition, patients with extra-oesophageal presentations, such as asthma, cough or laryngitis appear to require higher doses of proton pump inhibitor to treat symptoms effectively.15–17 In these patients it has become a practice to give proton pump inhibitor b.d., since taking omeprazole 20 mg b.d. yields superior intragastric pH control compared to a once a day regimen of omeprazole 40 mg given either before breakfast or dinner in normal subject.18 However, despite proton pump inhibitor b.d., there is a recovery of gastric acid secretion overnight (gastric pH < 4 for more than a continuous hour) in 75% of subjects. While the clinical importance of NAB is not clear, we have shown that abnormal oesophageal acid exposure is associated with NAB in up to 50% of patients with Barrett’s oesophagus and scleroderma,19 and is more common in patients with decreased lower oesophageal sphincter pressure and ineffective oesophageal body motility.20 Gastric pH may decrease to a level that is potentially damaging to the oesophageal mucosa during NAB, therefore put these patients at risk due to the prolonged oesophageal acid clearance seen in patients with ineffective motility.20 This study confirms our previous observations that on proton pump inhibitor b.d., more than 75% of patients have overnight gastric acid breakthrough (pH < 4 for more than 1 h) and perhaps of greater importance, almost half of patients have over 50% of the sleeping period in which gastric pH < 4, increasing the potential for oesophageal reflux. GERD patients treated with proton pump inhibitor b.d. + H2RA had superior overnight intragastric pH control. In this group, 27% of patients had gastric pH > 4 for the entire overnight period and 63% patients for greater than 90% of the nocturnal period. In contrast to 48% of patients in group A, only 11% of patients in group B had gastric pH > 4 less than 50% of the nocturnal time. This should substantially reduce the potential for nocturnal oesophageal acid reflux. Furthermore, the duration of oesophageal acid exposure was shorter in the group on bedtime H2 blocker.

While strong conclusions are difficult to make because the entire group was not studied on both regimens, the findings from the small group (n=11) who were studied on both regimens clearly showed a significant improvement in nocturnal intragastric pH control and a strong trend (P=0.06) toward a significant reduction of oesophageal acid exposure with the addition of H2 blocker. This, coupled with the large number of subjects studied on both regimens, indicates that this is a superior regimen for intragastric control.

It is of particular importance that, despite the addition of H2 blockers at bedtime to proton pump inhibitor b.d., 40% of patients continue to have some overnight acid recovery. This observation reinforces the fact that total acid control is extremely difficult to achieve with modern pharmacology (or in fact surgery). Fortunately, pharmacological achlorhydria is rarely necessary to achieve maximal oesophageal acid control,21 when needed.

Two other observations are of note. Daytime pH control appears unaffected by night-time H2 blocker. In addition, the correlation between oesophageal acid exposure and intragastric pH is supported by this study.

In summary, the study suggests that adding H2RA at bedtime to high dose proton pump inhibitor can enhance nocturnal gastric pH control, decrease nocturnal gastric acid breakthrough, and decrease the duration of oesophageal acid reflux associated with NAB. This regimen might benefit certain subgroups of GERD patients in whom greater gastric pH control is desired, although further studies are required to document the clinical efficacy and relevance.