Address for Correspondence Prof. Dr. Guy E. E. Boeckxstaens, Department of Gastroenterology, University Hospital Leuven, Catholic University of Leuven, Herestraat 45, 3000 Leuven, Belgium. Tel: +32 16 330237; fax: +32 16 330238; e-mail: email@example.com
Background Visceral hypersensitivity to distension is thought to play an important role in the pathophysiology of the irritable bowel syndrome (IBS). Cannabinoids are known to decrease somatic pain perception, but their effect on visceral sensitivity in IBS remains unclear. Therefore, we evaluated the effect of the mixed CB1/CB2 receptor agonist delta-9-tetrahydrocannabinol (Δ9-THC, dronabinol) on rectal sensitivity.
Methods Ten IBS patients and 12 healthy volunteers (HV) underwent a barostat study to assess rectal sensitivity using an intermittent pressure-controlled distension protocol before and after sigmoid stimulation. Repetitive sigmoid stimulation is a validated method to increase visceral perception in IBS patients, consisting of a 10-min period of 30 s stimuli (60 mmHg), separated by 30 s of rest (5 mmHg). The effect of placebo and Δ9-THC (5 and 10 mg in healthy volunteers and 10 mg in IBS patients) on rectal sensitivity was evaluated on respectively three and two separate days in a double blind, randomized, crossover fashion.
Key Results All participants (HV and IBS) reported central side effects during the highest dose of Δ9-THC, most frequently increased awareness of the surrounding, light-headedness and sleepiness, whereas no side effects where reported during placebo. Although blood pressure was not affected, heart rate increased in both HV and IBS, but was most pronounced in IBS patients. The cannabinoid agonist Δ9-THC did not alter baseline rectal perception to distension compared to placebo in HV or IBS patients. Similarly, after sigmoid stimulation there were no significant differences between placebo and Δ9-THC in sensory thresholds of discomfort.
Conclusions & Inferences These findings imply that Δ9-THC does not modify visceral perception to rectal distension and argue against (centrally acting) CB agonists as tool to decrease visceral hypersensitivity in IBS patients.
The irritable bowel syndrome (IBS) is the most common disorder seen in the out-patients practice of gastroenterologists and consists of chronic abdominal pain or discomfort in the absence of an organic explanation.1 Visceral hypersensitivity or enhanced perceptual responses to controlled distension of the colon and rectosigmoid is a consistent finding and recognized as an important mechanism in IBS, and is present in 33–65% of IBS patients.2–8 In the remainder of IBS patients, but not in healthy controls, visceral perception is increased by central sensitization (e.g. by noxious sigmoid stimulation), further supporting an important role for visceral hypersensitivity in IBS patients.9 The pathophysiology of this enhanced intestinal perception is, however, largely unknown.
Cannabinoid (CB-) receptors are members of the G-protein-coupled superfamily and are widely expressed in the gastrointestinal (GI) tract.10 To date, two CB-receptors have been identified and cloned of which CB1 is mainly located on neuronal cells and CB2 on immunocytes. The peripheral afferent nerves of the GI tract express CB1 and are involved in processing sensory information from the gut.11,12 Recently, an animal model of visceral pain provided evidence for the involvement of the endocannabinoid system in visceral perception. In a rat model of acid-induced colitis CB1 and CB2 antagonists resulted in an increased visceral hypersensitivity to rectal distension, whereas CB1 and CB2 agonists reduced basal sensitivity and colitis induced hypersensitivity.13,14 Similarly, in a mouse model of acute pancreatitis, CB-receptor activation was antinociceptive, providing further support for a potential role of cannabinoids in targeting visceral pain.15 Hence, modulation of the endocannabinoid system could be an option in targeting visceral hypersensitivity and subsequently in the treatment of IBS. So far, the effect of cannabinoids has only been evaluated in healthy subjects showing no effect on pain perception,16 The effect on visceral perception in IBS patients however has not been studied yet. Therefore, in the present study, we evaluated the effect of the mixed CB1/CB2 receptor agonist delta-9-tetrahydrocannabinol (Δ9-THC) on rectal perception in IBS patients and healthy volunteers, before and after sigmoid stimulation.
Patients and Methods
Between May 2006 and February 2008 12 healthy volunteers and 10 IBS patients were included. Female subjects of childbearing potential and not using appropriate methods of contraception were excluded. Healthy volunteers were recruited by public advertisement. None of the healthy volunteers had a history of GI complaints or was taking any mediation other than a birth control pill. For IBS patients, selection criteria included a positive diagnosis by Rome II criteria. A minimum work-up to exclude organic disease included careful history taken, a normal physical examination, a normal sigmoidoscopy or colonoscopy, normal thyroid stimulating hormone levels and blood counts and negative stool examinations. Patients had to be free of concomitant diseases, including psychiatric disorders. Medication likely to interfere with GI function and analgesics were discontinued at least 7 days before entering the study. Frequent cannabis or any other drug abuse was an exclusion criterion for study participation. Participants had to be free of cannabis use for at least 2 months prior to the study. The study protocol was approved by the Medical Ethics Committee of the Academic Medical Center. All participants gave their written and informed consent to participate in the study.
Dronabinol (Marinol®) or the mixed CB1/CB2 receptor agonist Δ9-THC, is used in clinical practice as antiemetic in chemotherapy-associated nausea and as appetite stimulant in AIDS-related anorexia. The absorption after oral administration is 90–95%. Delta-9-tetrahydrocannabinol is highly lipophilic and distributes to adipose tissue, whereas approximately 10–20% of the administered dose reaches the systemic circulation. The dose of Δ9-THC selected for this study (5 and 10 mg) was based on a previous study evaluating the effect of Δ9-THC 10 and 20 mg on esophageal sphincter relaxation. After intake of 20 mg, half of the subjects experienced nausea and vomiting leading to premature termination of the study, whereas 10 mg was well-tolerated.17 Therefore, in the current study 10 mg was the highest dosage given. Serum levels of THC increase at 30 min to peak at 1 h after oral administration.17 Therefore, we assessed the effect on rectal sensation between 30 and 100 min after oral administration of Δ9-THC.
The study was designed in a randomized, double blind, placebo controlled, cross-over fashion, performed on three separate days in healthy volunteers (placebo, 5 mg and 10 mg Δ9-THC) and on two separate days in IBS patients (placebo and 10 mg Δ9-THC), at least 6 days apart. All barostat studies were performed before noon after an overnight fast. Caffeine, nicotine and alcohol were not allowed during at least 24 h prior to each experiment.
A double balloon catheter was used consisting of two polyethylene bags, attached to a double lumen polyvinyl tube (Salem Sump tube 14 Ch.; Sherwood Medical, St Louis, MO, USA). The diameter of both balloons was 9 cm. The rectal balloon had a length of 14 cm, the sigmoid balloon of 9 cm and the distance between the balloons was 5 cm. Subjects received a tap water enema 45 min before balloon placement. The catheter was introduced during sigmoidoscopy and was attached to the mucosa at 35 cm from the anal verge by a vascular clip. The catheter was then connected to the barostat device and the subject was placed in the left lateral decubitus position. Subjects were allowed to recover from the endoscopic procedure during 30 min. Thirty minutes before the assessment of rectal sensitivity, subjects received either placebo or Δ9-THC (HV: 5 and 10 mg, IBS: 10 mg). Heart rate, systolic and diastolic blood pressure were measured every 15 min throughout the protocol, using an automatic sphygmomanometer (Boso, Jungingen, Germany).
Visceral sensitivity To assess sensitivity to rectal distension, an electronic barostat that automatically corrected for the compressibility of air (Distender series IITM; G&J Electronics Inc., Toronto, Ontario, Canada) was used. Minimal distending pressure (MDP) was defined as the minimum pressure at which the intrabag volume was >30 mL. Subjects scored the perception of sensation evoked by rectal distension using a 6-point scale with verbal descriptors (0 = no sensation; 1 = first sensation; 2 = first urge to defecate; 3 = normal urge to defecate; 4 = severe urge to defecate; 5 = discomfort/pain). If a subject reported discomfort or pain, the bag was instantaneously deflated. The rectal distension series was performed according to a phasic, semi-randomly ascending isobaric distension protocol consisting of pressure increments of 3 mmHg above MDP (3, 6, 12, 9, 18, 15, 24, 21, 30 mmHg, etc.), each step lasting 1 min and separated by 30-s intervals at baseline (MDP). The inflation rate was set at 38 mL s−1. Sensations were scored after 30 s of each distension step.
Sigmoid stimulation In order to increase visceral perception, noxious sigmoid conditioning was applied after the assessment of basal visceral perception. This technique has previously been demonstrated to increase visceral perception in normosensitive IBS patients.9 The sigmoid stimulation consisted of 10 high pressure distensions (60 mmHg) lasting 30 s and separated by 30-s intervals of rest (5 mmHg).
The primary endpoint was the sensory threshold for discomfort/pain during rectal distension before and after sigmoid stimulation. Secondary endpoints were the thresholds for first sensation and the urge to defecate (i.e. a sensation score of 1 and 3 respectively) and rectal compliance. The threshold for discomfort/pain was determined by the pressure at which an individual reported a sensation score of 5. If a subject did not report discomfort/pain before the fixed pressure limit of 60 mmHg was reached, we conservatively determined the threshold by the highest pressure reached plus one pressure step. Sensory thresholds for first sensation and urge to defecate were calculated for each individual by the lowest pressures at which a certain sensation score (i.e. 1 and 3, respectively) was reported. To obtain a pressure-volume curve for each individual, corresponding volumes for each distending pressure were averaged. Rectal compliance was calculated as the slope of the pressure-volume curve over the first four distension steps (i.e. the steeper part of the curve).
The primary endpoint was the sensory threshold for discomfort/pain during rectal distension. We calculated the sample size to detect differences for the primary endpoint of at least one pressure step (i.e. a treatment effect of 3 mmHg). Based on a previous study in hypersensitive IBS patients, we assumed that the common standard deviation would be 4 mmHg.5 Therefore, a minimum of seven subjects was needed in each group to obtain a power of 80% at the 5% significance level (paired comparisons). Statistical evaluations were performed using commercially available software (SPSS 16.0; SPSS Inc., Chicago IL, USA). Continuous data were compared using the Student t-test for independent samples and the paired t-test for related samples. Nominal data were compared by the Mann–Whitney U-test for independent samples or the Wilcoxon signed rank test for related samples. Differences were considered significant at the 5% level. Data are presented as mean ± SEM.
We included 12 healthy volunteers (HV, seven female) and 10 IBS patients (eight female). Mean age was comparable between healthy volunteers (32 ± 4 years, range 20–52 years) and IBS patients (33 ± range 20–52 years). Similarly, body mass index (BMI, calculated by dividing weight in kilograms by the square of height in meters) was not different between groups (HV: 24 ± 1, range 19–29 kg m−3 and IBS: 26 ± 1, range 20–33 kg m−3). Based on the Rome II criteria of IBS, five IBS patients were considered constipation predominant, four diarrhea predominant and one had the alternating subtype of IBS. The mean duration of symptoms at intake was 11 ± 4 years (range 1–35 years).
Blood pressure was not affected by administration of placebo or Δ9-THC whereas heart rate increased 45 min after administration of Δ9-THC, in both HV and IBS patients (Fig. 2). This increase in heart rate was most pronounced in IBS patients.
All participants reported central side effects with the highest dose of Δ9-THC (10 mg), starting approximately 60–75 min after administration. The most frequent observed side effects were an increased awareness of the surrounding (IBS 80%, HV 58%), light-headedness (IBS 60%, HV 42%) and sleepiness (IBS 50%, HV 67%). During 5 mg of Δ9-THC six healthy volunteers reported sleepiness (50%), three (25%) had a mildly increased awareness and one (7%) reported light-headedness. None of the participants reported a central side effect during placebo.
Effect of Δ9-THC
Healthy volunteers Rectal sensitivity. The mean MDP was comparable on the three different treatment days (9 ± 1 mmHg). The rectal threshold of first sensation (placebo: 2 ± 1; Δ9-THC 5 mg: 3 ± 1 and Δ9-THC 10 mg: 5 ± 1 mmHg above MDP), urge (placebo: 13 ± 2; Δ9-THC 5 mg: 18 ± 5 and Δ9-THC 10 mg: 16 ± 3 mmHg above MDP) and discomfort (placebo: 38 ± 4; Δ9-THC 5 mg: 38 ± 5 and Δ9-THC 10 mg: 36 ± 4 mmHg above MDP) before sigmoid stimulation were not different between placebo, 5 mg or 10 mg Δ9-THC (Fig. 3). During placebo, two healthy volunteers (17%) were unable to complete the complete 10 min of sigmoid stimulation, whereas during low dose Δ9-THC three healthy volunteers (25%) and during high dose Δ9-THC one healthy volunteer (8%) did not complete sigmoid stimulation (NS). No differences were observed in the mean duration of tolerated sigmoid distension during placebo (580 ± 14 s), 5 mg (564 ± 22 s) or 10 mg (570 ± 30 s) of Δ9-THC. Sigmoid stimulation did not significantly decrease the rectal sensory threshold of first sense, urge or discomfort during placebo. Rectal sensory thresholds after sigmoid stimulation were comparable between placebo, 5 mg Δ9-THC or 10 mg Δ9-THC.
Compliance. Rectal compliance, defined as the slope of the pressure-volume curve was not altered by sigmoid stimulation during placebo (presigmoid 14 ± 2 mL mmHg−1, postsigmoid 14 ± 1 mL mmHg−1, NS), 5 mg Δ9-THC (presigmoid 15 ± 1 mL mmHg−1, postsigmoid 15 ± 2 mL mmHg−1, NS) or 10 mg Δ9-THC (presigmoid 15 ± 2 mL mmHg−1, postsigmoid 15 ± 1 mL mmHg−1, NS). Neither was there a difference in compliance between placebo, 5 mg or 10 mg Δ9-THC.
Irritable bowel syndrome patients Rectal sensitivity. Rectal sensory thresholds of first sense, urge and discomfort were not significantly different in IBS patients compared to healthy volunteers. In IBS patients, MDP was similar during placebo (8 ± 1 mmHg) and 10 mg Δ9-THC (9 ± 1 mmHg, NS). Baseline (i.e. before sigmoid distension) rectal thresholds of first sensation (placebo: 3 ± 1; Δ9-THC: 2 ± 1 mmHg above MDP), urge (placebo: 11 ± 3; Δ9-THC: 11 ± 4 mmHg above MDP) and discomfort (placebo: 31 ± 4; Δ9-THC: 29 ± 4 mmHg above MDP) were not affected by administration of 10 mg of Δ9-THC (Fig. 3). Five IBS patients (50%) were unable to complete 10 min of sigmoid stimulation during placebo compared to seven IBS patients (70%) during Δ9-THC (NS). The mean duration of tolerated sigmoid was 416 ± 69 and 308 ± 72 s, respectively during placebo and Δ9-THC (NS). Sigmoid stimulation during placebo did not affect any of the rectal sensory thresholds. During Δ9-THC, the sensory thresholds after sigmoid stimulation were not significantly different from placebo. No significant differences were observed in rectal sensory thresholds after sigmoid stimulation during placebo and during Δ9-THC in IBS patients compared to healthy volunteers.
Compliance. In IBS, presigmoid rectal compliance was similar to healthy volunteers and there were no differences between placebo (14 ± 1 mL mmHg−1) and Δ9-THC (16 ± 3 mL mmHg−1). Rectal compliance was not affected by sigmoid stimulation during either of the conditions (postsigmoid placebo: 15 ± 2 mL mmHg−1, Δ9-THC 15 ± 3 mL mmHg−1, NS).
Visceral hypersensitivity of the colon is widely accepted as an important mechanism in the generation of abdominal pain in IBS patients. For centuries physicians have prescribed cannabis as analgesic agent in the treatment of abdominal pain, until the 20th century when the medical use of cannabis was ended due to restrictive federal legislation. Moreover, dosing of inhaled cannabis was difficult to titrate. A few decades ago, new cannabinoids, such as the mixed CB1/CB2 receptor agonist delta-9-tetrahydrocannabinol (Δ9-THC, dronabinol) were synthesized and introduced as potential treatment of chronic pain conditions.12Activation of CB-receptors reduced allodynia and hyperalgesia in animal models of inflammatory pain,18,19 neuropathic pain20,21 and capsaicin induced pain.22–24 In addition, in rats, both CB1 and CB2 agonism effectively reduces visceral hypersensitivity.13,14 indicating that cannabinoid agonists such as Δ9-THC could be interesting agents to target visceral hypersensitivity (and thus abdominal pain) in IBS patients. In the current study, however, administration of Δ9-THC did not reduce visceral perception to rectal distension when compared to placebo in either HV or IBS patients.
Based on a large body of evidence, baseline perception of rectal distension is normal in approximately half of all IBS patients.2–8 In these patients visceral perception can be experimentally increased following 10 min of intermittent noxious sigmoid stimulation.9–25 In order to increase the number of patients with visceral hypersensitivity, we adopted this protocol and tested the effect of Δ9-THC on visceral sensitivity before and after sigmoid stimulation in both HV and IBS patients. We hypothesized that Δ9-THC would reduce visceral perception, especially in subjects with proven hypersensitivity. In the current study, however, sigmoid stimulation failed to significantly increase visceral perception in IBS patients. This may be explained by the fact that half of the patients were unable to finish the 10-min period of intermittent sigmoid stimulation. However, as reported by the investigators who introduced the technique, completion of the entire distension protocol is not required to increase visceral perception.9 Not the duration or completion of the sigmoid stimulation determined whether visceral hypersensitivity developed in response to sigmoid distension, but rather the subtype of IBS, i.e. patients with pain-predominant symptoms.26 The latter could explain our inability to trigger visceral hypersensitivity in IBS patients, and would suggest that ideally, we should have selected IBS patients with a pain-predominant symptom profile. Alternatively, only patients with visceral hypersensitivity should have been included and randomized to evaluate whether Δ9-THC normalizes visceral perception in hypersensitive patients. In the current study, only three patients were hypersensitive after sigmoid distension. Therefore, our negative finding may be a type II error implying that we cannot exclude that Δ9-THC may be effective in IBS patients with visceral hypersensitivity. Nevertheless, our study shows that Δ9-THC does not affect visceral perception in unselected IBS patients.
Our data indicating that Δ9-THC does not decrease visceral perception in human, contrasts with animal data showing a reduction in visceral pain by both CB1 and CB2 receptors.13,14,27 Obviously, it is difficult to extrapolate findings from animal models to the human situation, especially given the complexity of the pathophysiology of IBS. Moreover, in animal models pain perception is measured by the contractile response of the abdominal muscles as part of the spinally mediated visceromotor response. The muscle contractions may be directly altered by CB agonists, potentially interfering with the response of the animal to painful stimuli.28,29 On the other hand, it could be argued that, interaction of cannabinoids with higher cortical functions may override a potential peripheral antinociceptive effect of CB-receptor stimulation. Indeed, central activation of the CB1 receptor in human is known to induce anxiety and an increased awareness of physical stimuli,30 both capable to enhance perceptual responses to rectal distension.31 Anxiety induced by mental stress (color word conflict and mental arithmetic test) for example lowered the thresholds of visceral perception in both healthy volunteers and IBS patients.32 Although we did not assess the state of anxiety in the present study, increased awareness of the surrounding, light-headedness and sleepiness were reported by both the healthy volunteers and IBS patients 1 hour to 90 min after administration of Δ9-THC. In addition, heart rate increased at 30 min to peak at 90 min after administration, like the central side effects coinciding with the peak in plasma concentration of Δ9-THC.17 To what extent the psychotropic side effects of Δ9-THC may have modulated the perception of rectal distension is impossible to determine, but this would indicate that if cannabinoid agonists will be developed to reduce visceral perception, they should be preferentially peripherally active.
In conclusion, we showed that the CB-receptor agonist Δ9-THC failed to reduce visceral perception to rectal distension in healthy volunteers and IBS patients, arguing against (centrally acting) cannabinoid agonists as therapeutic agents to treat IBS.
Authors Contribution and Conflict of Interest Statement
Tamira K. Klooker
I declare that I participated in the design and set up of the study, the data collection, analysis, interpretation and writing of the article. I have seen and approved the final version. I have no conflicts of interest.
Karolien E. Leliefeld
I declare that I participated in the data collection and approved the final version. I have no conflicts of interest.
Rene M. van den Wijngaard
I declare that I participated in the interpretation of the data and I have seen and approved the final version. I have no conflicts of interest.
Guy E. Boeckxstaens
I declare that I designed and set up the study, I participated in the interpretation of the data and writing of the article and I have seen and approved the final version. My work is funded by a grant of the Flemish Government (Odysseus program, Fonds Wetenschappelijk Onderzoek (FWO), grant G.0905.08) which had no influence on study design; or on the collection, analysis, interpretation of data; on the writing of the report; or on the decision to submit the paper for publication.