Sleep Deprivation Is Hyperalgesic in Patients With Gastroesophageal Reflux Disease

Transcription

1 GASTROENTEROLOGY 2007;133: Sleep Deprivation Is Hyperalgesic in Patients With Gastroesophageal Reflux Disease RON SCHEY,* RAM DICKMAN,* SAIRAM PARTHASARATHY, STUART F. QUAN, CHRISTOPHER WENDEL,* JONAH MERCHANT,* JEANNETTE POWERS,* BING HAN,* DANIEL VAN HANDEL,* and RONNIE FASS* *The Neuroenteric Clinical Research Group, Section of Gastroenterology, Southern Arizona VA Health Care System, and Arizona Respiratory and Sleep Disorders Centers, College of Medicine, University of Arizona, Tucson, Arizona See LeeJetalon page 1392 in CGH. Background & Aims: Studies have demonstrated that gastroesophageal reflux disease (GERD) can cause sleep deprivation because of nighttime heartburn or short, amnestic arousals during sleep. Sleep deprivation has been associated with reports of increased GERD severity. Our aim was to determine whether sleep deprivation enhances perception of intraesophageal acid in patients with GERD vs healthy controls. Methods: Ten healthy controls and 10 patients with erosive esophagitis (grades B D) were included in the study. All subjects were randomized to either sleep deprivation (1 night with <3 hours of sleep) or sufficient sleep (3 days with >7 hours sleep/night). Patients crossed over to the other arm after a washout period of 1 week. To ensure proper sleep time, we objectively monitored subjects with an actigraph. The morning after sufficient sleep or sleep deprivation, patients underwent stimulus response functions to esophageal acid perfusion. Results: Ten healthy controls and 10 GERD patients completed all stages of the study. GERD patients demonstrated a significant decrease in lag time to symptom report ( vs sec, respectively, P.02), increase in intensity rating ( vs cm, respectively, P.02), and increase in acid perfusion sensitivity score ( vs sec cm/100, respectively, P.02) after sleep deprivation as compared with nights of good sleep. Normal subjects did not demonstrate any differences in stimulus response functions to acid between sufficient sleep and sleep deprivation ( vs sec, vs cm, and vs sec cm/100, respectively, all P NS). Conclusions: Sleep deprivation is hyperalgesic in patients with GERD and provides a potential mechanism for increase in GERD symptom severity in sleep-deprived patients. Sleep deprivation (less than 7 hours per night) is very common in the United States and affects nearly 70% of the adult population. 1 Gastroesophageal reflux disease (GERD) is also very common, affecting 44% of the adult population at least once a month and 20% at least once a week. 2,3 The relationship between GERD and sleep has been recently established. Studies have demonstrated that GERD may adversely affect sleep primarily through 2 important mechanisms. The first is by awakening patients from sleep during the night, leading to reports of sleep deprivation and sleep abnormalities. The second is through multiple, short arousals because of gastroesophageal reflux events during sleep, leading to sleep fragmentation. 4 Between 47% and 57% of the GERD patients report having heartburn that awakens them from sleep during the night. 3,5,6 Approximately 25% of the general population reports having heartburn that awakens them from sleep during the night. 5 Of the GERD patients, 63% reports that they are unable to sleep well, 42% is unable to sleep through a full night, 39% has to take naps during the day, and 34% has to sleep in a seated position because of nighttime heartburn. 6 Furthermore, frequency of sleep difficulties in subjects with GERD correlates with the frequency of GERD symptoms. 6 The aforementioned studies suggest that GERD can disturb sleep, leading to sleep deprivation and sleep abnormalities, but do not exclude a more complex relationship in which sleep deprivation per se can adversely affect GERD. It is possible that GERD and sleep deprivation might have a relationship in which GERD exacerbates sleep deprivation, and, in turn, sleep deprivation exacerbates GERD. Central factors such as stress and anxiety have been shown to modulate esophageal and gastric perception of intraluminal stimuli. 7 Recently, Chiu et al have demonstrated through a postal survey that subjects with disturbed sleep have lower pain thresholds as compared with subjects with normal sleep (odds ratio [OR], 2.2; 95% confidence interval [CI]: ). 8 Hakki Onen et al evaluated tolerance thresholds to mechanical and thermal pain during sleep deprivation using a pressure dolorimeter and a thermode. 9 The authors found that healthy adult volunteers demonstrated a hyperalgesic ef- Abbreviations used in this paper: APSS, acid perfusion sensory score; ESS, Epworth Sleepiness Scale; GERD, gastroesophageal reflux disease; NERD, nonerosive reflux disease; SSS, Stanford Sleepiness Scale by the AGA Institute /07/$32.00 doi: /j.gastro

2 1788 SCHEY ET AL GASTROENTEROLOGY Vol. 133, No. 6 fect related to sleep deprivation and an analgesic effect related to slow wave sleep. The analgesic effect of slow wave sleep was apparently greater than the analgesia induced by level I (World Health Organization) 9 analgesic compounds in mechanical pain experiments. Roehrs et al found that sleep deprivation (4 hours) shortened by 25% finger withdrawal latency from a heat source as compared with a night of good sleep (8 hours). 10 The authors concluded that poor sleep may be hyperalgesic. Thus, we hypothesized that sleep deprivation may worsen GERD by modulating esophageal perception thresholds for pain. Consequently, the aim of the study was to determine whether sleep deprivation by itself can exacerbate GERD by evaluating stimulus response functions to acid in GERD patients vs healthy controls, who experience sleep deprivation and sufficient sleep conditions. Materials and Methods Participants This was a prospective, randomized, controlled trial with a crossover design. Ten healthy controls and 10 GERD patients with erosive esophagitis (Los Angeles classification grades B through D) were enrolled into this study. All GERD patients had a history of heartburn at least twice a week for a minimum of 3 months before study inclusion. GERD patients were recruited from the Southern Arizona VA Health Care System, the University of Arizona Health Sciences Center, and the general population of Tucson. Only patients who were found to have erosive esophagitis on upper endoscopy were invited to participate in this study. Patients who were consuming antireflux medications, such as proton pump inhibitors and H 2 blockers, or sleep-altering medications, such as psychotropics, narcotics, or benzodiazepines, were excluded from the study. Additionally, patients with known psychologic abnormalities (depression, anxiety, and others), severe underlying comorbid illnesses that may interfere with sleep (cardiopulmonary, renal, endocrine, and others) were also excluded. Patients with history of prior upper gastrointestinal (GI) surgery, diabetes mellitus, any type of neuropathy, seizure disorder, or sleep apnea were excluded from the study as well. Moreover, patients who were unable or unwilling to complete fully all stages of the study and were unable or unwilling to provide informed consent were also excluded. The Human Subjects Committee of the University of Arizona approved this study. Prior to the onset of the study, all subjects had to complete 2 screening questionnaires: the Berlin questionnaire, which assesses the level of risk for obstructive sleep apnea, and the Epworth Sleepiness Scale (ESS), which assesses excessive daytime sleepiness. 1 Patients with medium or high risk for obstructive sleep apnea or those with evidence of excessive daytime sleepiness were excluded from the study. Subsequently, all subjects completed 2 validated questionnaires that included a demographic survey and the GERD Symptom Questionnaire. Thereafter, all subjects were randomized to undergo stimulus response functions to acid using a modified acid perfusion test after having either a minimum of 7 hours of sleep during 3 consecutive days (sufficient sleep) or no more than 3 hours of sleep after falling asleep (sleep deprivation) the night prior to the test. Thus, 5 subjects from each group were randomized to one of the arms and after 1 week of washout period crossed over to the other arm. All subjects wore an actigraph on their nondominant arm to measure objectively sleep time for a period of 1 week prior to each of the acid perfusion tests to ensure compliance with the sleep time requirements of the protocol. Subjects received the actigraph 7 days before the planned acid perfusion test and were instructed to sleep for at least 7 hours during the last 3 days prior to the acid perfusion test. The morning after the last night of good sleep, the subjects actigraphs were removed, and sleep time during the last 3 days was analyzed. If subjects did not meet the criteria of 3 nights of good sleep, they were sent home for another 7 days (with 3 nights of at least 7 hours of sleep). Similarly, the actigraph was given to the subjects 7 days prior to the acid perfusion test with the instruction that subjects have to sleep only 3 hours during the last night before the acid perfusion test. The morning after the night of bad sleep, the subjects actigraphs were removed, and sleep time during the prior night was determined by analyzing the actigraph s recording. If the subjects did not meet the criteria for a night of bad sleep, they were sent home for another 7 days. In the morning and prior to each acid perfusion test, patients completed the Stanford Sleepiness Scale (SSS), which was used to assess the effect of sleep condition. Upper Endoscopy After an overnight fast, patients were placed in the left lateral position. Sedation was achieved with a combination of midazolam (Roche, Nutley, NJ) and meperidine (Sanofi Winthrop, New York, NY). The endoscope (GIF 100; Olympus, Center Valley, PA) was inserted via the mouth and into the esophagus. Careful examination of the distal esophagus was performed to determine the presence of mucosal injury. The stomach and duodenum were evaluated for possible mucosal lesions. The extent of esophageal mucosal damage was assessed by using the Los Angeles classification: Grade A: 1 (or more) mucosal break no longer than 5 mm that does not extend between the tops of the 2 mucosal folds; Grade B: 1 (or more) mucosal break more than 5 mm long that does not extend between the tops of 2 mucosal folds; Grade C: 1 (or more) mucosal break that is continuous between the tops of 2 or more mucosal folds

3 December 2007 GERD AND SLEEP 1789 but that involves less than 75% of the circumference; and Grade D: 1 (or more) mucosal break that involves at least 75% of the esophageal circumference. To ensure that the GERD population was as homogenous as possible, we included only those with Los Angeles classification B through D in this study. Berlin Questionnaire The Berlin Questionnaire is used to assess risk factors for sleep apnea including the presence and frequency of snoring, wake-time sleepiness, or fatigue, and history of obesity or hypertension. 11,12 The questionnaire is composed of 15 questions divided among 3 different domains. Determining the risk for obstructive sleep apnea is based on the responses to 3 different categories. In category 1, high risk is defined as persistent symptoms (more than 3 to 4 times a week) in 2 or more questions about snoring. In category 2, high risk is defined as persistent (more than 3 to 4 times a week) wake-time sleepiness, drowsy driving, or both. In category 3, high risk is defined as history of high blood pressure or a body mass index (BMI) more than 30 kg/m 2. To be considered as high risk for sleep apnea, a patient had to fulfill at least 2 symptom categories. ESS Questionnaire The ESS differentiates persons with excessive daytime sleepiness from alert individuals by measuring their sleep propensity, which has been described as the net interaction of the waking drive and the sleep drive, with questions such as the following: How likely are you to doze off or fall asleep in the following situations, in contrast to feeling just tired? This refers to your usual way of life in recent times. Even if you have not done some of these things recently try to work out how they would have affected you. Any subject with medium or high risk for obstructive sleep apnea or a score above 10 on the ESS was excluded from the study. Demographic Questionnaire The demographic questionnaire provided information about age, gender, ethnicity, college education, income, and other demographic parameters. GERD Symptom Questionnaire This questionnaire evaluates heartburn severity and frequency, acid regurgitation, chest pain, dysphagia, the effect that GERD symptoms have on quality of life and health-care utilization, respiratory complaints, other upper GI symptoms, esophageal disorders, and history of surgery. Actigraphy During the study period, subjects underwent actigraphy to objectively measure total sleep time. The Figure 1. Example of a wrist actigraph. actigraph (Figure 1), a watch-like device worn on the nondominant wrist, records motions with accelerometers that are stored digitally in the device (Ambulatory Monitoring Inc, Ardsley, NY). The actigraph can collect data continuously over an extended period of time (1 week). The stored digital information can then be downloaded and analyzed by proprietary software to yield periods of quiescence that can be inferred as time asleep. 13 Only after verification of sleep hours required by the protocol 1 night of bad sleep ( 3 hours) or 3 nights of good sleep ( 7 hours) (Figures 2 and 3) were the patients allowed to undergo the modified acid perfusion test. SSS Prior to the acid perfusion test, patients filled out the SSS, a validated, subjective instrument used to measure level of sleepiness. 14,15 The scale is composed of 7 statements that describe current state of sleepiness, from which the patients choose one. The measure has been shown to reflect the effect of sleep loss and can be administered multiple times. However, the scale only evaluates sleepiness at one point in time. The list of statements includes feeling active and vital, alert, wide-awake; functioning at a high level but not at peak, able to concentrate; relaxed, awake, not at full alertness, responsive; a little foggy, not at peak, let down; fogginess, beginning to lose interest in remaining awake, slowed down; sleepiness, prefer to be lying down, fighting sleep, woozy; and almost in reverie, sleep onset soon, losing struggle to remain awake. Stimulus Response Functions to Acid A catheter with a central lumen was inserted via the nasal passage and placed in the midesophagus, 10 cm above the upper border of the lower esophageal sphincter. By using a Harvard apparatus, saline was

4 1790 SCHEY ET AL GASTROENTEROLOGY Vol. 133, No. 6 Figure 2. Actigraph reading with arrow indicating the beginning of a bad night sleep ( 3 hours). infused initially into the esophagus at a rate of 10 ml/min for 2 minutes. Subsequently, without the patient s knowledge, 0.1 N hydrochloric acid was infused for 10 minutes at a similar rate. Patients were instructed to report whenever their typical heartburn was reproduced. Normal subjects were instructed to report whenever they experience a burning sensation behind the breastbone that may travel up to the throat. Only prior to the acid perfusion test were patients and normal subjects instructed on how and what symptoms to report. During the test, patients were not distracted with further instructions. Stimulus-response functions to acid were quantified by 3 parameters: lag time to first report of heartburn, sensory intensity rating, and an acid perfusion sensory score (APSS). 16 Lag time was defined as the time (in seconds) to initial heartburn, and sensory intensity rating was defined as the intensity of heartburn symptoms at the end of acid perfusion, using a previously validated verbal descriptor that ranged from no sensation to extremely intense (0 20, respectively). 17 The APSS was calculated from both the duration of typical symptom perception (T) expressed in seconds and sensory intensity rating at the end of acid perfusion (I). For convenience, the APSS was divided by 100 (APSS (I) (T)/100). Figure 3. Actigraph reading with arrow indicating 3 days with good night sleep ( 7 hours).

5 December 2007 GERD AND SLEEP 1791 Table 1. Demographic Characteristics of the Study Participants Demographic parameters GERD patients Healthy controls Statistical Analysis t test P values Age, mean ( SD) 49.5 ( 14.76) 36.9 ( 17.08).1 BMI, mean ( SD) ( 6.36) ( 8.13).9 Sex (M/F) 8/2 5/5.2 Education High school: 1 Some college: 4 College: 5 Employment Full-time: 6 Retired: 2 Unable to work: 2 Income $25,000: 3 $40,000: 4 $60,000: 1 $60,000: 2 High school: 1 Some college: 2 College: 7 Full-time: 8 Retired: 1 Student: 1 $5,000: 2 $25,000: 1 $40,000: 4 $60,000: 3 Descriptive statistics were used to compare cases and controls with respect to demographics and underlying risk factors. All baseline aspects of this study were compared to determine whether there were any statistically significant differences between the groups. The evaluative tools were also each individually compared. Paired t tests were conducted to statistically identify any differences between the cases and controls. An level of the t tests was set at A P value less than.05 indicated a statistically significant difference between the cases and controls. The main analysis was a comparison of change in stimulus response function to acid. Normally distributed data were analyzed by using the Student t test, and nonnormally distributed data were analyzed by using the Wilcoxon signed-ranks test. This nonparametric test is ideally suited for analyzing small samples. All analyses were 2-sided, using an a priori level of We had 80% power to detect a minimum mean score change of 14%. The main outcome of evaluation is the time to onset of heartburn symptom (in seconds). We had 80% power to detect a difference as small as 79 seconds between groups when using a standard deviation of 60 seconds. This standard deviation is similar to that found in previous studies. With a population standard deviation of 3 units, we had 80% power to detect a difference in mean score intensity rating as small as 4 units. By expanding the standard deviation to 4.5, we had 80% power to detect a difference in intensity rating score as small as 6. All calculations were based on a sample size of 20 (10 matched pairs), an a priori of 0.05, and 2-sided hypothesis testing. Results Ten healthy controls and 10 GERD patients completed all stages of the study (Table 1). The control group consisted of 5 males and 5 females, whereas the patient group consisted of 8 males and 2 females. In general, the age, BMI, gender, education, employment, and income were not statistically significantly different between the cases and the controls at the P.05 significance level. Thus, there was overall homogeneity between the 2 subject groups. GERD Symptom Checklist Analysis Two (20%) of the GERD patients reported having heartburn duration of 7 months to 1 year; 2 (20%), 1 2 years; 3 (30%), 5 10 years; 2 (20%), years; and 1 (10%), more than 20 years. Five patients (50%) had daily symptoms, and 4 (40%) patients had symptoms several times a week. Seven (70%) patients had moderate severity of heartburn ( cannot be ignored but does not affect my lifestyle ). Eight (80%) of the patients reported having heartburn that awakens them from sleep during the night. Acid regurgitation was reported by 9 (90%) GERD patients. Four (40%) patients reported 5 10 years duration of acid regurgitation, and 2 (20%) patients reported duration of acid regurgitation 7 months to 1 year. Five (50%) patients reported moderate severity of acid regurgitation, and 8 (80%) patients stated that acid regurgitation awoke them from sleep during the night. Analysis of SSS The sleepiness ratings (1 7) were averaged to give mean ratings within each group (patients or healthy controls) for each of the 2 conditions (night of good or night of bad sleep). According to the SSS, the average sleepiness reported for the healthy controls after a night of good sleep was between 1 (functioning at high levels but not at peak; able to concentrate) and 2 (feeling active, vital, alert, or wide awake). The average sleepiness reported for the healthy controls after a night of bad sleep was between 3 (awake but relaxed; responsive but not fully alert) and 4 (somewhat foggy, let down). The difference in the relationship between healthy controls after a night of bad sleep vs healthy controls after a night of good sleep was statistically significant (P.002). According to the SSS, the average sleepiness reported by GERD patients after a night of good sleep was between 1 (functioning at high levels but not at peak; able to concentrate) and 2 (feeling active, vital, alert, or wide awake). The average sleepiness reported by the GERD patients after a night of bad sleep was between 3 (awake but relaxed; responsive but not fully alert) and 4 (somewhat foggy, let down). This relationship was also statistically significant (P.0016) (Table 2). The relationship between the patients and healthy controls was also analyzed and is shown in Table 3. There was no statistically significant difference in sleepiness ratings between the healthy controls and the patients who had a night of good sleep. Similarly, there was no statistically significant difference in sleepiness rating be-

6 1792 SCHEY ET AL GASTROENTEROLOGY Vol. 133, No. 6 Table 2. Average Sleepiness Ratings for Stanford Scale After the Different Sleeping Periods for Healthy Controls and GERD Patients Healthy controls Mean ( SD) t test P value GERD patients Mean ( SD) t test P value Good sleep 1.3 ( 0.675) ( 0.527).0016 Bad sleep 3.5 ( 1.58) 3.8 ( 1.93) tween the healthy controls and the patients who had a night of bad sleep. In general, participants were significantly sleepier after the night of bad sleep condition than they were after the night of good night condition. Overall, the groups did not differ in terms of their sleepiness rating, and the difference between a night of good sleep and a night of bad sleep was statistically similar for both groups. Acid Perfusion Test Results We analyzed each variable as a mixed design analysis of variance (ANOVA) (between subject variable: group [patient/control]; within subjects variable: condition [night of good sleep/night of bad sleep]). In each analysis, the effect of group was statistically significant, and the main effect of condition and the interaction of group and condition was significant in GERD patients, lag time to symptom report ( vs , respectively, P.02), increase in intensity rating ( vs , respectively, P.02), and increase in acid perfusion sensitivity score ( vs , respectively, P.02). However, these results were nonsignificant in healthy control subjects ( vs , vs , and vs , respectively, all P NS) (Figure 4). Of the GERD patients, 9 demonstrated worsening of all stimulus-response functions to acid after a night of bad sleep as compared with a night of good sleep (1 patient demonstrated no change). In contrast, only 2 healthy controls demonstrated a very mild alteration in stimulus response functions to acid after a night of bad sleep as compared with a night of good sleep. The healthy control group demonstrated a lack of statistical difference in all stimulus response functions to acid between nights of good sleep and nights of bad sleep. Discussion This is the first study to demonstrate that sleep deprivation by itself can modulate esophageal perception thresholds for pain in patients with gastroesophageal reflux diseases and not in healthy controls. The effect of sleep deprivation on esophageal sensation was so pronounced that we were able to demonstrate it even in a small sample of GERD patients. In contrast, no significant effect of sleep deprivation was observed in normal subjects. In this study, we included in the GERD group only symptomatic patients with erosive esophagitis. We excluded patients with nonerosive reflux disease (NERD), Los Angeles grade A erosive esophagitis, Barrett s esophagus, and GERD complications. This was done to ensure a homogenous group of GERD patients that demonstrated similar perceptual responses during sensory testing of the esophagus. 18 Patients with Barrett s esophagus, esophageal stricture, and NERD have all demonstrated different perceptual responses to intraesophageal stimuli. 19,20 We used the actigraph to ensure compliance with our sleep protocol. Although performing the study in a sleep laboratory would have provided us with better supervision of participants compliance with the sleep protocol, the need for 3 nights of good sleep in addition to 1 night of bad sleep would have made the cost of the study prohibitive. Moreover, recruitment into the study would have been markedly compromised. The actigraph has been shown in sleep studies to be relatively accurate in estimating a person s sleep time. 21 The technique has been validated and is widely used in sleep research. The simplicity of the actigraphy technique can make it a valuable tool in the future to better determine esophageal physiology during sleep in persons while pursuing their everyday activities, thus avoiding any artifact related to confinement in a sleep laboratory. The hyperalgesic effect of sleep deprivation has not been previously studied at the visceral level. However, somatic hyperalgesia has been clearly documented in various recent studies In this study, we compared the effect of sleep deprivation (1 night 3 hours sleep) vs sufficient sleep (3 nights 7 hours sleep) on esophageal pain perception. It appears at least from somatic sensitivity studies that a night of good sleep is analgesic. 9 Because people either sleep less or more than 7 hours per night, and the latter is considered as normal sleep, the previously documented analgesic effect of a night of Table 3. Comparison of Average Sleepiness Rating for Stanford Scale After the Different Sleeping Periods Between Healthy Controls and GERD Patients Controls, mean ( SD) GERD, mean ( SD) t test P value Mean difference CI of mean difference Good sleep 1.3 ( 0.675) 1.5 ( 0.53) to Bad sleep 3.5 ( 1.58) 3.8 ( 1.93) to 1.36

7 December 2007 GERD AND SLEEP 1793 Figure 4. (A) Comparison of time to initial perception of heartburn during an acid perfusion test between GERD patients and healthy controls during good and bad night s sleep. (B) Comparison of reports of intensity rating at the end of the acid perfusion test between GERD patients and healthy controls during good and bad night s sleep. (C) Comparison of calculated acid perfusion sensitivity score between GERD patients and healthy controls during good and bad night s sleep. good sleep may represent the attenuation of the hyperalgesic effect of sleep deprivation. It still remains to be elucidated whether there is a correlation between the level of improvement in the hyperalgesic effect of sleep deprivation on the esophagus and increase in sleep time. Similarly, it is unclear whether the hyperalgesic effect will become more pronounced and accentuated the more sleep deprived the GERD patients are. Further studies are needed to explore our findings. The current GERD and sleep paradigm suggests a unilateral relationship in which GERD adversely affects sleep. However, our study proposes a new conceptual model for the relationship between GERD and sleep. GERD appears to adversely affect sleep quality through 2 mechanisms: (1) heartburn that awakens patients from sleep during the night and (2), more commonly, short, amnestic arousals in response to acid reflux events. 5,22 Our study demonstrated that sleep deprivation per se can adversely affect GERD by enhancing perception of intraesophageal acid (esophageal hyperalgesia), supporting a bidirectional relationship between GERD and sleep quality. Our new conceptual model also suggests a potential vicious cycle in which GERD leads to poor quality of sleep that in turn enhances perception of intraesophageal stimuli that further exacerbates GERD. Additionally, the bidirectional relationship between GERD and quality of sleep may also offer new therapeutic strategies for improving GERD. Several studies have clearly shown that antireflux treatment improves nocturnal heartburn and patients quality of sleep. 23,24 Our study suggests that improvement in sleep quality may potentially result in improvement of GERD symptoms. Hence, treatment with medications that improve sleep quality may have a positive effect on GERD symptoms. However, it is unlikely that all medications that are used to improve sleep quality will be effective in improving GERD symptoms. For example, recently, benzodiazepines have been shown to be a risk factor for heartburn that awakens patients from sleep during the night. 5 This class of drugs adversely affects upper gut motility by slowing gastric emptying and reducing lower esophageal sphincter basal pressure. Regardless, future studies are needed in GERD patients using medications that improve sleep to test our hypothesis. The results of this study may also explain our recent clinical observation that GERD patients who sleep less during the night are more likely to have symptoms that correlate with acid reflux events. 4 Anxiety and stress have also been demonstrated to enhance visceral perception of intraluminal stimuli. Dickhaus et al have demonstrated that psychologic-induced stress modulated visceral perception of rectal balloon distentions in patients with irritable bowel syndrome. 25 Irritable bowel syndrome patients, but not healthy controls, reported enhanced perceptual sensitivity to moderate rectal distention under conditions of auditory stress. Geeraerts at al have recently demonstrated that experimentally induced anxiety alters gastric sensor motor function. 26 Experimentally induced anxiety resulted in significantly lower intragastric balloon volume thresholds for discomfort as compared with neutral condition. Additionally, anxiety led to significantly higher symptoms score for satiety, fullness, and bloating after a standard nutrient challenge. Thus, our study suggests that, in addition to anxiety and stress, poor sleep may also modulate visceral pain, thus causing GERD patients to perceive low-intensity esophageal stimuli as being painful. 7 These central factors may alter esophageal perception through brain-gut mechanisms. Interestingly, a recent study by Farre et al demonstrated in an animal model that acute stress increased mucosal permeability to all molecules tested, and this was associated with the rapid appearance of dilated intercellular spaces in the esophageal mucosa. 27 Similar studies in humans and comparison of esophageal mucosal response to acute stress between GERD patients and normal subjects are still unavailable. Our study demonstrated a clear discrepancy in the perceptual responses to intraesophageal acid between healthy controls and GERD patients. Our sleep deprivation protocol had almost no effect on esophageal perception thresholds for pain of healthy subjects. These findings are in contradiction to somatic pain studies in sleep-

8 1794 SCHEY ET AL GASTROENTEROLOGY Vol. 133, No. 6 deprived, healthy controls, demonstrating hyperalgesic changes even in this group of subjects. 28 However, in this study, we assessed visceral sensation, used intraesophageal acid as the stimulus, and compared healthy controls with GERD patients, who demonstrated active esophageal inflammation (Los Angeles grades B D). Although our study results need to be confirmed by future studies, we propose several potential mechanisms that could explain the discrepancy in esophageal perceptual responses to acid stimulation between sleep-deprived healthy controls and GERD patients. Possibly, the sleep deprivation protocol had only a limited stressful effect that resulted in minimal modulation of esophageal perception in healthy controls. In contrast, patients with GERD demonstrated marked modulation in esophageal pain perception for the same level of stimulus because of differences in anxiety, habituation, or attentional and affective responses. This differential effect has also been observed by Dickhaus et al, 25 who exposed normal controls and patients with irritable bowel syndrome to auditory stress during rectal balloon distention. The authors demonstrated a significant accentuation in visceral perception during rectal balloon distention in irritable bowel syndrome patients but no effect in normal controls. The authors hypothesized that the auditory stimulus used in the study resulted in limited stress response that was insufficient to alter visceral perception in normal controls. Alternatively, intraesophageal acid stimulation has no effect on healthy controls esophageal perception, regardless if they are sleep deprived. This hypothesis is due to our lack of knowledge about acid sensitivity in healthy controls. Another explanation is the different effect of sleep deprivation on mood or attention in healthy controls vs GERD patients. A different explanation is the presence of esophageal inflammation in GERD patients, which predisposes them to accentuated nociceptive response after sleep deprivation as compared with healthy subjects. Last, the accentuated effect of sleep deprivation on GERD patients resulting in increase in pain perception may suggest a stronger sympathetic activity. The presence of esophageal sensitivity to acid at baseline in GERD patients as compared with healthy controls may also suggest higher inhibitory mechanisms at baseline. Sleep deprivation may lead to a reduction in the inhibitory mechanisms, which will have a greater releasing effect and a greater increase in pain in the context of a higher baseline pain (GERD patients), than in the lower baseline pain condition (healthy controls). 29,30 This study included only a small number of subjects in each participating group. However, the sleep deprivation effect on esophageal sensation was so pronounced that it was detected even in such a small number of participants. Thus, increasing the number of participants in each group would have been superfluous. Nevertheless, comparison of frequency and severity of symptoms in the GERD group after a night of bad sleep vs a night of good sleep may have further supported our conclusions. However, patients had to be non per os prior to their modified acid perfusion test, which took place during the morning hours. This would have confounded documentation of GERD symptoms. In summary, this is the first study to demonstrate that sleep deprivation per se is hyperalgesic in patients with GERD. It is likely, because more people in today s society suffer from chronic sleep deprivation, that this mechanism will become more important in producing symptom exacerbation of GERD patients. It is also the first study to demonstrate that there is a bidirectional relationship between GERD and poor sleep. This new paradigm opens new opportunities for GERD treatment. However, further studies are needed to confirm our findings, specifically the lack of response to esophageal acid stimulation that was demonstrated by sleep-deprived normal subjects as compared with GERD patients. References 1. Hiestand D, Britz P, Goldman M, et al. Prevalence of symptoms and risk of sleep apnea in the US population: results from the National Sleep Foundation in America 2005 poll. Chest 2006; 130: Gallup Organization. A Gallup survey on heartburn across America. The Gallup Organization, Washington, DC, Locke G III, Talley N, Fettl S, et al. Prevalence and clinical spectrum of gastroesophageal reflux: a population-based study in Olmsted County, Minnesota. Gastroenterology 1997;112: Green C, Dekel R, Quan S, et al. The effect of sleep duration on symptoms perception of patients with gastroesophageal reflux disease (GERD). (abstr No S1729). Gastroenterology 2003;124:A Fass R, Quan S, O Connor G, et al. Predictors of heartburn during sleep in a large prospective cohort study. Chest 2005;127: Shaker R, Castell D, Schoenfeld P, et al. Nighttime heartburn is an under-appreciated clinical problem that impacts sleep and daytime function: the results of a Gallup survey conducted on behalf of the American Gastroenterological Association. Am J Gastroenterol 2003;98: Fass R, Tougas G. Functional heartburn: the stimulus, the pain, and the brain. Gut 2002;51: Chiu Y, Silman A, Macfarlane G, et al. Poor sleep and depression are independently associated with a reduced pain threshold. Results of a population-based study. Pain 2005;115: Hakki Onen S, Alloui A, Gross A, et al. The effects of total sleep deprivation, selective sleep interruption and sleep recovery on pain tolerance thresholds in healthy subjects. J Sleep Res 2001; 10: Roehrs T, Hyde M, Blaisdell B, et al. Sleep loss and REM sleep loss are hyperalgesic. Sleep 2006;29: Vgontzas A, Papanicolaou D, Bixler E, et al. Elevation of plasma cytokines in disorders of excessive daytime sleepiness: role of sleep disturbance and obesity. J Clin Endocrinol Metab 1997; 82: Netzer N, Stoohs R, Netzer C, et al. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Ann Intern Med 1999;131: Goldsmith G, Levin J. Effect of sleep quality on symptoms of irritable bowel syndrome. Dig Dis Sci 1993;38: Mitler M, Miller J. Methods of testing for sleepiness (corrected). Behav Med 1996;21:

9 December 2007 GERD AND SLEEP Hoddes E, Zarcone V, Smythe H, et al. Quantification of sleepiness: a new approach. Psychophysiology 1973;10: Bautista J, Wong W, Pulliam G, et al. The value of ambulatory 24-hour esophageal ph monitoring in clinical practice in patients who were referred with persistent gastroesophageal reflux disease (GERD)-related symptoms while on standard dose antireflux medications. Dig Dis Sci 2005;50: Gracely R, McGrath P, Dubner R. Ratio scales of sensory and affective verbal pain descriptors. Pain 1978;5: Fass R. Sensory testing of the esophagus. J Clin Gastroenterol 2004;38: Nagahara A, Miwa H, Minoo T, et al. Increased esophageal sensitivity to acid and saline in patients with nonerosive gastroesophageal reflux disease. J Clin Gastroenterol 2006;40: Fass R, Yalam J, Camargo L, et al. Increased esophageal chemoreceptor sensitivity to acid in patients after successful reversal of Barrett s esophagus. Dig Dis Sci 1997;42: Morgenthaler T, Alessi C, Friedman L, et al. Practice parameters for the use of actigraphy in the assessment of sleep and sleep disorders: an update for Sleep 2007;30: Dickman R, Green C, Fass S, et al. Relationships between sleep quality and ph monitoring findings in persons with gastroesophageal reflux disease. J Clin Sleep Med 2007;3: Johnson D, Orr W, Crawley J, et al. Effect of esomeprazole on nighttime heartburn and sleep quality in patients with GERD: a randomized, placebo-controlled trial. Am J Gastroenterol 2005; 100: Orr W, Goodrich S, Robert J. The effect of acid suppression on sleep patterns and sleep-related gastro-oesophageal reflux. Aliment Pharmacol Ther 2005;21: Dickhaus B, Mayer E, Firooz N, et al. Irritable bowel syndrome patients show enhanced modulation of visceral perception by auditory stress. Am J Gastroenterol 2003;98: Geeraerts B, Vandenberghe J, Van Oudenhove L, et al. Influence of experimentally induced anxiety on gastric sensorimotor function in humans. Gastroenterol 2005;129: Farre R, De Vos R, Geboes K, et al. Critical role of stress in increased oesophageal mucosa permeability and dilated intercellular spaces. Gut 2007;56: Lautenbacher S, Kundermann B, Krieg J-C. Sleep deprivation and pain perception. Sleep Med Rev 2006;10: Uijtdehaage SH, Thayer JF. Accentuated antagonism in the control of human heart rate. Clin Auton Res 2000;10: Thayer JF. On the importance of inhibition: central and peripheral manifestations of nonlinear inhibitory processes in neural systems. Dose-Response 2006;4:2 21. Received June 1, Accepted September 6, Address requests for reprints to: Ronnie Fass, MD, Professor of Medicine, University of Arizona, Head, the Neuro-Enteric Clinical Research Group, Southern Arizona VA Health Care System, GI Section (1-111G-1), 3601 S. 6th Avenue, Tucson, Arizona ronnie.fass@va.gov; fax: (520) Supported in part by a grant from Wyeth Pharmaceuticals. The sponsor had no role in the study design, collection, analysis, or interpretation of data. Conflict of interest: R.F. received research support from Wyeth Pharmaceuticals. All participants were informed that Wyeth Pharmaceuticals helped fund this study.