Digestive Diseases and Sciences, Vol. 37, No. 10 (October 1992), pp. 1576-1582
Effect of Acupuncture on Gastric Acid Secretion in Healthy Male Volunteers GERVAIS TOUGAS, MDCM, LI YU YUAN, MD, JOHAN W. RADAMAKER, MD, STEVEN G. CHIVERTON, MD, and RICHARD H. HUNT, MD
Six randomized, placebo controlled studies were performed to investigate the effect o f electroacupuncture on gastric acid output in 38 healthy males. Electroacupuncture decreased basal acid output when compared to placebo acupuncture [from 3.50 +- 0.59 mmol/hr to 2.54 +- 0.56 mmol/hr (P < 0.05)] as well as sham feeding-stimulated acid output [from 18.52 +- 2.25 mmol/hr to 5.38 +- 2.11 mmol/hr (P < 0.005)], but had no effect on the pentagastrin stimulated acid output. The inhibitory effect of acupuncture on sham feeding-stimulated acid output was not affected by local anesthesia o f the acupoint, but was prevented by a prior intravenous naloxone injection. Acupuncture did not alter plasma gastrin levels (20.7 +- 7.6 txg/liter, vs control 21.2 +-- 7.2 tzg/liter) but naloxone increased it (26.1 + 14.5 txg/liter) (P < 0.05). We conclude that the antisecretory effects o f electroacupuncture do not result from decreased gastrin release or decreased parietal cell sensitivity to gastrin, but are mediated through naloxone-sensitive opioid neural pathways and vagal efferent pathways. KEY WORDS: acupuncture; electroacupuncture; gastric acid secretion; pentagastrin; gastrin; naloxone.
Acupuncture has been used empirically in China for centuries and a considerable literature has confirmed its efficacy in anesthesia and analgesia. Other studies have claimed that acupuncture can relieve epigastric pain and even heal peptic ulcers (1, 2). However, until recently, no systematic studies have investigated the effects of acupuncture on gastrointestinal function. In conscious dogs, acupuncture inhibits basal acid output (3), sham feeding-stimulated acid output, histamine- (4) and pentagastrin-stimulated Manuscript received September 25, 1991; revised manuscript received January 27, 1992; accepted February 27, 1992. From the Division of Gastroenterology, McMaster University Medical Centre, Hamilton, Ontario, Canada. Dr. Tougas was supported by a Fellowship of the Medical Research Council of Canada. This work was presented in part at the Annual Meeting of The American Gastroenterological Association, New Orleans, May 1988; Washington, May 1989; and San Antonio, May 1990. Address for reprint requests: Dr. Gervals Tougas, Division of Gastroenterology, Room 4W8, McMaster University Medical Centre, 1200 Main Street West, Hamilton, Ontario, Canada L8N 3Z5.
1576
maximal acid output (5), and also increases gastric secretion of bicarbonate and sodium (3). In two uncontrolled clinical trials in patients with duodenal ulcer, basal acid output, pentagastrin-stimulated (6), and histamine-stimulated maximal acid output (7) were decreased by acupuncture. Few studies have examined the mechanism of action of acupuncture on gastric secretion. In dogs the somatic afferent nerves might convey the sensory input signals of acupuncture, as the effect on gastric secretion of bicarbonate and sodium could be blocked by local anesthesia at the acupuncture point (acupoint) (3). Two conflicting studies examined the vagal mediation of acupuncture on gastric secretion in dogs; one showed that atropine abolished the effect of acupuncture on gastric bicarbonate and sodium secretion (3), while the other showed that vagotomy did not abolish the effect of acupuncture on gastric acid secretion (4). The analgesic effect of acupuncture is mediated by endogenous opioid pathways (8-11). The inhibDigestive Diseases and Sciences, Vol. 37, No. 10 (October 1992)
0163-2116/92/1000-1576506.50/0 9 1992PlenumPublishingCorporation
ACUPUNCTURE AND GASTRIC SECRETION itory effect of acupuncture on gastric motility can also be p r e v e n t e d by prior intravenous infusion of the opioid antagonist naloxone (12). Finally naloxone can abolish the inhibitory effect of acupuncture on gastric acid secretion in the rat (13). So far no controlled study has examined the effect o f acupuncture on gastric acid secretion, and its mechanism of action in humans. We therefore studied the effect of acupuncture on basal and sham feeding- and pentagastrin-stimulated gastric acid secretion in healthy male volunteers. We examined the pathways and mechanisms mediating the acupuncture effect by infiltration of the acupuncture point (acupoint) by local anesthetic and b y intravenous injection of the opioid antagonist naloxone prior to acupuncture. Finally we studied the effect of acupuncture on serum gastrin levels. We used electroacupuncture because it is easily standardized and considered a more effective form o f acupuncture than manual twirling of the needles (3). MATERIALS AND M E T H O D S Subjects. All studies were performed in healthy males from whom written consent was obtained after explanation of the procedure. The study was approved by the Ethical Committee and the Research Project Advisory Committee of McMaster University Medical Centre. Thirty-eight nonsmoking, healthy, males (age 19-43 years, mean 27 years) were recruited. None had a history of gastrointestinal disease or reported taking any medication, drugs, or alcohol for 10 days prior to the study. Determination of Gastric Acid Output. After an overnight fast, a nasogastric tube (Salem Sump, 14 French, Sherwood Medical Industries) was passed into the stomach and, with the subject supine, correct placement of the tube was determined by the water-recovery method (14). Gastric juice was collected in 10-min aliquots by continuous aspiration, the volume of each sample was measured, and acid concentration determined by automatic titration to pH 7 (Radiometer, Copenhagen, Denmark). The gastric acid output was derived from measurements in the two consecutive periods of 30 rain immediately before and after acupuncture, with the exception of the pentagastrin dose-response study (see below), in which it was obtained only from measurements of the last 30-min period of each incremental step dose. Acid output is expressed as millimoles per hour. Acupuncture. New, sterilized stainless-steel acupuncture needles (length 5 cm, diameter 0.46 mm) were used. The electrical stimulus was generated as a continuous wave, biphasic pulse of 8 Hz and administered by an electrostimulator (Yenshan WQ-10, Beijing, China). The stimulus intensity was gradually increased to the maximum tolerated without discomfort or fasciculation of the surrounding muscles. For electroacupuncture, needles were inserted perpendicularly 2.5-3.5 cm into the acuDigestive Diseases and Sciences, Vot. 37, No. 10 (October 1992)
J.
Fig 1, Acupuncture points commonly used in gastrointestinal disorders. The acupoint we used for placebo acupuncture (lateral aspect of forearm) is not known to be a therapeutic acupoint.
point known as "zusanli" (stomach channel). Zusanli is located on the left leg, four finger widths below the depression inferior to the patella, lateral to the patellar ligament and one finger breadth from the anterior crest of the tibia. In traditional Chinese Medicine, zusanli is frequently used for the treatment of gastric symptoms and is shown in Figure 1. The needle was inserted until the subject experienced "qi" (needling reaction, a sensation of soreness, numbness or perceived distention around the point). For placebo acupuncture, needles were placed under the skin at a point one third of the way between the elbow and wrist on the lateral aspect of the nondominant forearm, where no known acupoint is located (Figure 1). The depth of insertion was similar to that used with EA (2.5-3.5 cm) and electrical stimulation was performed in an identical manner to electroacupuncture. The subjects reported the sensation elicited at the placebo site by the needle insertion to be similar to what they experienced at zusanli. Sham Feeding. The "chew-and-spit" technique was used (15), with subjects served a beef hamburger and instructed to chew but not swallow the food for a 15-min period. Accidental swallowing of food was monitored by examining the gastric aspirate for the presence of food particles. If accidental swallowing occurred, the study was repeated. Experiment 1: Effect of Electroaeupuncture on Basal Acid Output. This was a randomized, placebo controlled, crossover study observing the effect of electroacupuncture on basal acid output. On two different days, in random order, subjects (N = 8) received 15 min of either electroacupuncture or placebo acupuncture. Acid output was measured for 60 min, immediately prior to and following the stimulus.
1577
TOUGAS ET AL Experiment 2: Effect of Electroacupuncture on Sham Feeding-Stimulated Acid Output. This was a randomized, placebo controlled study observing the effect of electroacupuncture on sham fed acid output. Ten subjects were randomly assigned to either receive electroacupuncture at zusanli or placebo acupuncture. Following the first hour of basal collection, subjects were simultaneously sham fed during 15 rain while receiving either electroacupuncture or placebo acupuncture. Acid output was then measured for an hour. Experiment 3: Effect of Eiectroacupuncture on DoseResponse of Acid Output to Pentagastrin. This was a randomized, placebo controlled, crossover study. On different days, following the first hour of basal study, subjects (N = 5) received a pentagastrin intravenous infusion with four-step dose increments: 0.03, 0.15, 1.7, and 7.5 ~g/kg/hr. Each step lasted 1 hr. Either electroacupuncture or placebo acupuncture with alternating cycles of 15 min on and off was given simultaneously from the time of the first dose throughout the study. The acid output was measured hourly for each pentagastrin dose increment. Experiment 4: Effect of Local Xylocaine Anesthesia on Action of Electroacupuncture on Sham Feeding-Stimulated Acid Output. This was a randomized, placebo controlled, crossover study to determine whether local anesthesia at zusanli abolished the inhibitory effect of electroacupuncture on sham fed acid output. On different days, 5 min before the end of the first hour of basal study, the acupoint zusanli was locally infiltrated by either 1% xylocaine (1.5 ml) or normal saline. This xylocaine dose was previously found to abolish the perception of pain from a needle entering the acupoint; however, the subjects were still aware of a pressure sensation in the area upon insertion of the needle. The subjects (N = 5) were then sham fed and simultaneously received either electroacupuncture or placebo acupuncture for 15 min. Acid output was measured for an hour. Experiment 5: Abolition by Naloxone of Effect of Eiectroacupuncture on Sham Feeding-Stimulated Acid Output. This was a randomized, placebo controlled, crossover study to examine whether the opioid antagonist naloxone abolished the inhibitory effect of electroacupuncture on sham-fed acid output. On different days, following the first hour of basal study, subjects (N = 5) simultaneously received 15 rain of sham feeding together with one of the following interventions: naloxone 4 mg intravenously (in a l-ml saline bolus) + 15 min electroacupuncture, saline intravenously + 15 min electroacupuncture, or saline intravenously + 15 rain placebo acupuncture. Acid output was then measured for a subsequent hour. Experiment 6: Effect of Eiectroacupuncture and Naloxone on Sham Feeding-Stimulated Acid Output and Circulating Serum Gastrin Concentration. This was a randomized, placebo controlled, crossover study to determine the effect of electroacupuncture on circulating gastrin levels, measured by radioimmunoassay. On different days, following the first hour of basal study, subjects (N = 5) received either placebo acupuncture or electroacupuncture with or without naloxone 40 txg/kg intravenously. Acid output was then measured for an hour and
1578
TABLE 1. EFFECTS OF ACUPUNCTURE AND PLACEBO ACUPUNCTURE ON BAO*
N Mean --- SEM P vs control P vs EA
Control
EA
PA
16 3,33 -- 0.35
8 2.54 --+ 0.56
Effect of Acupuncture on Gastric Acid Secretion in Healthy Male Volunteers GERVAIS TOUGAS, MDCM, LI YU YUAN, MD, JOHAN W. RADAMAKER, MD, STEVEN G. CHIVERTON, MD, and RICHARD H. HUNT, MD
Six randomized, placebo controlled studies were performed to investigate the effect o f electroacupuncture on gastric acid output in 38 healthy males. Electroacupuncture decreased basal acid output when compared to placebo acupuncture [from 3.50 +- 0.59 mmol/hr to 2.54 +- 0.56 mmol/hr (P < 0.05)] as well as sham feeding-stimulated acid output [from 18.52 +- 2.25 mmol/hr to 5.38 +- 2.11 mmol/hr (P < 0.005)], but had no effect on the pentagastrin stimulated acid output. The inhibitory effect of acupuncture on sham feeding-stimulated acid output was not affected by local anesthesia o f the acupoint, but was prevented by a prior intravenous naloxone injection. Acupuncture did not alter plasma gastrin levels (20.7 +- 7.6 txg/liter, vs control 21.2 +-- 7.2 tzg/liter) but naloxone increased it (26.1 + 14.5 txg/liter) (P < 0.05). We conclude that the antisecretory effects o f electroacupuncture do not result from decreased gastrin release or decreased parietal cell sensitivity to gastrin, but are mediated through naloxone-sensitive opioid neural pathways and vagal efferent pathways. KEY WORDS: acupuncture; electroacupuncture; gastric acid secretion; pentagastrin; gastrin; naloxone.
Acupuncture has been used empirically in China for centuries and a considerable literature has confirmed its efficacy in anesthesia and analgesia. Other studies have claimed that acupuncture can relieve epigastric pain and even heal peptic ulcers (1, 2). However, until recently, no systematic studies have investigated the effects of acupuncture on gastrointestinal function. In conscious dogs, acupuncture inhibits basal acid output (3), sham feeding-stimulated acid output, histamine- (4) and pentagastrin-stimulated Manuscript received September 25, 1991; revised manuscript received January 27, 1992; accepted February 27, 1992. From the Division of Gastroenterology, McMaster University Medical Centre, Hamilton, Ontario, Canada. Dr. Tougas was supported by a Fellowship of the Medical Research Council of Canada. This work was presented in part at the Annual Meeting of The American Gastroenterological Association, New Orleans, May 1988; Washington, May 1989; and San Antonio, May 1990. Address for reprint requests: Dr. Gervals Tougas, Division of Gastroenterology, Room 4W8, McMaster University Medical Centre, 1200 Main Street West, Hamilton, Ontario, Canada L8N 3Z5.
1576
maximal acid output (5), and also increases gastric secretion of bicarbonate and sodium (3). In two uncontrolled clinical trials in patients with duodenal ulcer, basal acid output, pentagastrin-stimulated (6), and histamine-stimulated maximal acid output (7) were decreased by acupuncture. Few studies have examined the mechanism of action of acupuncture on gastric secretion. In dogs the somatic afferent nerves might convey the sensory input signals of acupuncture, as the effect on gastric secretion of bicarbonate and sodium could be blocked by local anesthesia at the acupuncture point (acupoint) (3). Two conflicting studies examined the vagal mediation of acupuncture on gastric secretion in dogs; one showed that atropine abolished the effect of acupuncture on gastric bicarbonate and sodium secretion (3), while the other showed that vagotomy did not abolish the effect of acupuncture on gastric acid secretion (4). The analgesic effect of acupuncture is mediated by endogenous opioid pathways (8-11). The inhibDigestive Diseases and Sciences, Vol. 37, No. 10 (October 1992)
0163-2116/92/1000-1576506.50/0 9 1992PlenumPublishingCorporation
ACUPUNCTURE AND GASTRIC SECRETION itory effect of acupuncture on gastric motility can also be p r e v e n t e d by prior intravenous infusion of the opioid antagonist naloxone (12). Finally naloxone can abolish the inhibitory effect of acupuncture on gastric acid secretion in the rat (13). So far no controlled study has examined the effect o f acupuncture on gastric acid secretion, and its mechanism of action in humans. We therefore studied the effect of acupuncture on basal and sham feeding- and pentagastrin-stimulated gastric acid secretion in healthy male volunteers. We examined the pathways and mechanisms mediating the acupuncture effect by infiltration of the acupuncture point (acupoint) by local anesthetic and b y intravenous injection of the opioid antagonist naloxone prior to acupuncture. Finally we studied the effect of acupuncture on serum gastrin levels. We used electroacupuncture because it is easily standardized and considered a more effective form o f acupuncture than manual twirling of the needles (3). MATERIALS AND M E T H O D S Subjects. All studies were performed in healthy males from whom written consent was obtained after explanation of the procedure. The study was approved by the Ethical Committee and the Research Project Advisory Committee of McMaster University Medical Centre. Thirty-eight nonsmoking, healthy, males (age 19-43 years, mean 27 years) were recruited. None had a history of gastrointestinal disease or reported taking any medication, drugs, or alcohol for 10 days prior to the study. Determination of Gastric Acid Output. After an overnight fast, a nasogastric tube (Salem Sump, 14 French, Sherwood Medical Industries) was passed into the stomach and, with the subject supine, correct placement of the tube was determined by the water-recovery method (14). Gastric juice was collected in 10-min aliquots by continuous aspiration, the volume of each sample was measured, and acid concentration determined by automatic titration to pH 7 (Radiometer, Copenhagen, Denmark). The gastric acid output was derived from measurements in the two consecutive periods of 30 rain immediately before and after acupuncture, with the exception of the pentagastrin dose-response study (see below), in which it was obtained only from measurements of the last 30-min period of each incremental step dose. Acid output is expressed as millimoles per hour. Acupuncture. New, sterilized stainless-steel acupuncture needles (length 5 cm, diameter 0.46 mm) were used. The electrical stimulus was generated as a continuous wave, biphasic pulse of 8 Hz and administered by an electrostimulator (Yenshan WQ-10, Beijing, China). The stimulus intensity was gradually increased to the maximum tolerated without discomfort or fasciculation of the surrounding muscles. For electroacupuncture, needles were inserted perpendicularly 2.5-3.5 cm into the acuDigestive Diseases and Sciences, Vot. 37, No. 10 (October 1992)
J.
Fig 1, Acupuncture points commonly used in gastrointestinal disorders. The acupoint we used for placebo acupuncture (lateral aspect of forearm) is not known to be a therapeutic acupoint.
point known as "zusanli" (stomach channel). Zusanli is located on the left leg, four finger widths below the depression inferior to the patella, lateral to the patellar ligament and one finger breadth from the anterior crest of the tibia. In traditional Chinese Medicine, zusanli is frequently used for the treatment of gastric symptoms and is shown in Figure 1. The needle was inserted until the subject experienced "qi" (needling reaction, a sensation of soreness, numbness or perceived distention around the point). For placebo acupuncture, needles were placed under the skin at a point one third of the way between the elbow and wrist on the lateral aspect of the nondominant forearm, where no known acupoint is located (Figure 1). The depth of insertion was similar to that used with EA (2.5-3.5 cm) and electrical stimulation was performed in an identical manner to electroacupuncture. The subjects reported the sensation elicited at the placebo site by the needle insertion to be similar to what they experienced at zusanli. Sham Feeding. The "chew-and-spit" technique was used (15), with subjects served a beef hamburger and instructed to chew but not swallow the food for a 15-min period. Accidental swallowing of food was monitored by examining the gastric aspirate for the presence of food particles. If accidental swallowing occurred, the study was repeated. Experiment 1: Effect of Electroaeupuncture on Basal Acid Output. This was a randomized, placebo controlled, crossover study observing the effect of electroacupuncture on basal acid output. On two different days, in random order, subjects (N = 8) received 15 min of either electroacupuncture or placebo acupuncture. Acid output was measured for 60 min, immediately prior to and following the stimulus.
1577
TOUGAS ET AL Experiment 2: Effect of Electroacupuncture on Sham Feeding-Stimulated Acid Output. This was a randomized, placebo controlled study observing the effect of electroacupuncture on sham fed acid output. Ten subjects were randomly assigned to either receive electroacupuncture at zusanli or placebo acupuncture. Following the first hour of basal collection, subjects were simultaneously sham fed during 15 rain while receiving either electroacupuncture or placebo acupuncture. Acid output was then measured for an hour. Experiment 3: Effect of Eiectroacupuncture on DoseResponse of Acid Output to Pentagastrin. This was a randomized, placebo controlled, crossover study. On different days, following the first hour of basal study, subjects (N = 5) received a pentagastrin intravenous infusion with four-step dose increments: 0.03, 0.15, 1.7, and 7.5 ~g/kg/hr. Each step lasted 1 hr. Either electroacupuncture or placebo acupuncture with alternating cycles of 15 min on and off was given simultaneously from the time of the first dose throughout the study. The acid output was measured hourly for each pentagastrin dose increment. Experiment 4: Effect of Local Xylocaine Anesthesia on Action of Electroacupuncture on Sham Feeding-Stimulated Acid Output. This was a randomized, placebo controlled, crossover study to determine whether local anesthesia at zusanli abolished the inhibitory effect of electroacupuncture on sham fed acid output. On different days, 5 min before the end of the first hour of basal study, the acupoint zusanli was locally infiltrated by either 1% xylocaine (1.5 ml) or normal saline. This xylocaine dose was previously found to abolish the perception of pain from a needle entering the acupoint; however, the subjects were still aware of a pressure sensation in the area upon insertion of the needle. The subjects (N = 5) were then sham fed and simultaneously received either electroacupuncture or placebo acupuncture for 15 min. Acid output was measured for an hour. Experiment 5: Abolition by Naloxone of Effect of Eiectroacupuncture on Sham Feeding-Stimulated Acid Output. This was a randomized, placebo controlled, crossover study to examine whether the opioid antagonist naloxone abolished the inhibitory effect of electroacupuncture on sham-fed acid output. On different days, following the first hour of basal study, subjects (N = 5) simultaneously received 15 rain of sham feeding together with one of the following interventions: naloxone 4 mg intravenously (in a l-ml saline bolus) + 15 min electroacupuncture, saline intravenously + 15 min electroacupuncture, or saline intravenously + 15 rain placebo acupuncture. Acid output was then measured for a subsequent hour. Experiment 6: Effect of Eiectroacupuncture and Naloxone on Sham Feeding-Stimulated Acid Output and Circulating Serum Gastrin Concentration. This was a randomized, placebo controlled, crossover study to determine the effect of electroacupuncture on circulating gastrin levels, measured by radioimmunoassay. On different days, following the first hour of basal study, subjects (N = 5) received either placebo acupuncture or electroacupuncture with or without naloxone 40 txg/kg intravenously. Acid output was then measured for an hour and
1578
TABLE 1. EFFECTS OF ACUPUNCTURE AND PLACEBO ACUPUNCTURE ON BAO*
N Mean --- SEM P vs control P vs EA
Control
EA
PA
16 3,33 -- 0.35
8 2.54 --+ 0.56
