Digestive and Liver Disease 46 (2014) 682–687
Contents lists available at ScienceDirect
Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld
Alimentary Tract
Efficacy of a CO2 -releasing suppository in dyschezia: A double-blind, randomized, placebo-controlled clinical trial夽 Anne Laure Tarrerias a , Laurent Abramowitz b , Marc M.L. Marty c,∗ , Pierre Coulom d , Ghislain Staumont d , Christophe Merlette e , Véronique Berger c , Bernard Savarieau c,f , Philippe Ducrotté g a
General and Digestive Surgery Unit, Foch Hospital, Suresnes, France Medical Surgical Proctology Unit, Bichat-Claude Bernard Hospital, France c Nukléus, Clinical Trials Department, Paris, France d Proctology Unit, Clinique St Jean Languedoc, Toulouse, France e General Medicine Unit, Fresnes, France f Hepatogastroenterology Unit, Intercommunal Hospital, Créteil, France g INSERM Unit U1073 and Gastroenterology Department, Rouen University Hospital, Rouen, France b
a r t i c l e
i n f o
Article history: Received 19 December 2013 Accepted 15 April 2014 Available online 11 June 2014 Keywords: Clinical trial Dyschezia Functional G.I. disease
a b s t r a c t Background: Constipation has a significant impact on quality of life. Aim of this study was to evaluate the safety and the efficacy for relieving dyschezia symptoms of a CO2 -releasing suppository in a randomized, placebo-controlled, clinical trial. Methods: Fifty-three office-based primary care physicians and 24 gastroenterologists conducted the study in France, between November 2010 and January 2012. Patients (aged 18–75 years) with dyschezia were eligible. Patients were randomly allocated a once-a-day suppository (CO2 -releasing suppository or placebo) for 21 days. Primary endpoint was the change, from Day 0 to Day 21, in the intensity of discomfort related to dyschezia based on a self-assessed 0–100 visual analogue scale. Results: A total of 323 patients were randomized, i.e. 166 into the intervention group and 157 into the placebo group. Co-variance analysis showed a greater reduction in discomfort visual analogue scale score in the intervention group (−34.5 mm; standard error of the mean: 1.8 mm) than in the placebo group (−26.2 mm; standard error of the mean: 1.9 mm; p < 0.001). The greater efficacy of the CO2 -releasing suppository was confirmed for all secondary efficacy parameters. No significant side effects for either treatment were observed. Conclusion: A CO2 -releasing suppository is more effective than a placebo for the relief of symptoms of dyschezia. This efficacy is associated with a good safety profile. © 2014 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
1. Introduction Constipation can be classified either as normal-transit constipation (often coexisting with irritable bowel syndrome), slow transit constipation, or functional defecation disorder. Each of these conditions, which can occur in isolation or co-exist [1], has a significant impact on the patient’s quality of life [2,3]. In functional defecation disorders, pathophysiological features are a normal or
夽 ClinicalTrial.gov number: NCT01243723. ∗ Corresponding author at: Nukléus, Clinical Trials Department, 55 Rue Bobillot, Paris 75013, France. Tel.: +33 (0)1 45 88 66 88. E-mail address: [email protected] (M.M.L. Marty).
slightly slowed colonic transit overall, but a preferential storage of residues for prolonged periods in the rectum, which explains the patient’s symptoms, i.e. feeling of incomplete evacuation, sensation of anorectal blockage or straining. In this instance, the primary failure is an inability to adequately evacuate rectal contents [4]. Functional defecation disorder occurs in approximately 7% of adults [5], although isolated defecatory dysfunction seems to be present in 25–40% of patients with chronic constipation [6]. According to the Rome III criteria, the diagnosis of functional defecation disorders requires abnormal findings detected by a balloon expulsion test, recto-anal manometry, imaging, or electromyography in order to discuss a targeted treatment. For instance, when the functional defecation disorder is related to pelvic floor dyssynergia, biofeedback is the treatment of choice
http://dx.doi.org/10.1016/j.dld.2014.04.005 1590-8658/© 2014 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
A.L. Tarrerias et al. / Digestive and Liver Disease 46 (2014) 682–687
683
with an efficacy in approximately 70–80% of the patients, and a long-term success rate in 50% [7,8]. However, in clinical practice, functional tests are not available in many cases. Therefore, the first-line treatment is given when a defecation disorder is only suspected according to symptoms of dyschezia including sensation of blocked stools, feeling of incomplete evacuation, repeated attempts to defecate, straining during bowel movements and need to practice manual manoeuvres to facilitate defecation, even if we have shown that some symptoms, such as straining at stool, are highly predictive of anorectal manometric disturbances [9]. In such cases, diet advice and laxatives are rarely effective [4]. Therefore, other options of treatment for dyschezia symptoms, which can be easily used in primary care, are of interest. The daily use of a CO2 releasing suppository (EductylTM -TechniPharma, Monaco) could be an alternative for some patients with dyschezia symptoms. This CO2 -releasing suppository contains 1.15 g of potassium acid tartrate, 0.7 g of sodium bicarbonate, and excipients including solid semisynthetic glycerides, soy lecithin, and talc. Within a few minutes after introduction in the rectum, this suppository releases approximately 100 mL of gas and triggers defecation. In previous trials, a CO2 -releasing suppository has been shown to promote more frequent stools with normal consistency than placebo [10]. In addition, the studied CO2 -releasing suppository seemed to be more effective than psyllium in reducing the sensation of incomplete evacuation and straining during defecation [11]. Therefore, we proposed the hypothesis that the daily use of a CO2 -releasing suppository could be effective to alleviate symptoms of dyschezia. We thus designed this multicentre trial to assess the efficacy and safety of this CO2 -releasing suppository in patients suffering from dyschezia.
2010 and January 2012. All participating physicians operated at their private practice. Two visits were scheduled: one at baseline (Day 0) and one after 3 weeks (Day 21). At baseline, after verification that the subjects fulfilled the inclusion criteria, they received comprehensive information about the study and provided written informed consent prior to inclusion. Demographic data and history of dyschezia symptoms were recorded. Both at baseline and Day 21, presence of stools in the rectum, intensity of discomfort related to dyschezia on a VAS [13], and Bowel Functional Index (BFI) [14] were assessed, as well as the Patient Assessment of Constipation Quality of Life (PACQOL) questionnaire [15] was filled. Additionally, patients gave their opinion about the treatment using the patient global improvement of change (PGIC) [16] at Day 21. At the end of the inclusion visit, patients were given a selfevaluation diary in which, at home each evening, they recorded the intensity of discomfort related to dyschezia on an 11-point numeric rating scale (NRS-11), ranging from 0 (no discomfort) to 10 (unbearable discomfort) from Day 1 to Day 21.
2. Materials and methods
2.4. Randomization and masking
2.1. Patients
Investigators, patients, and study sponsor were masked to treatment assignment. An independent statistician established a computer-generated randomization scheme balanced by centre. Allocation of treatment was carried out using a sequentially numbered container.
Patients aged between 18 and 75 years were eligible if they described symptoms listed among the Rome III criteria to define functional defecation disorders [12], in the presence of 2 or more of the following symptoms: (a) straining during defecation; (b) feeling of incomplete defecation; (c) feeling of anorectal obstruction/blockage at defecation; (d) need of manual manoeuvres to facilitate defecation. These signs had to be present in at least 25% of defecations and for the previous 3 months with symptom onset at least 6 months prior to inclusion. Patients also had to fulfil the following criteria: presence of stool in the rectum at digital rectal examination and intensity of discomfort related to dyschezia of at least 40 mm as assessed by the patient on a visual analogue scale (VAS) from 0 mm (no discomfort) to 100 mm (maximal discomfort). Exclusion criteria were the presence of neurological disease (e.g. paraplegia, multiple sclerosis), anal pain, colonic or rectal organic lesion (including rectocele), and chronic inflammatory colon disease. Clinical examination prior to inclusion led also to exclude patients with anal stenosis, anal fissure, rectal prolapse, haemorrhoid flare-up, and descended perineum. The use of antidepressants, psychotropic drugs, or laxatives was allowed if the dose, which was not to be modified during the study, had been stable for the previous 15 days. Opioids and anorectal treatments (other types of suppositories, cleaning enemas, or biofeedback) were prohibited throughout the study. 2.2. Design Fifty-three office-based primary care physicians and 24 gastroenterologists conducted the study in France, between November
2.3. Interventions Patients were randomly allocated to a once-a-day treatment in the morning, consisting of either a CO2 -releasing suppository (intervention group) or a placebo suppository containing the same excipients (solid semisynthetic glycerides, soy lecithin, and talc) (control group) but no CO2 . Treatment was given for 21 days. Size and appearance of both types of suppositories were strictly identical. All patients received usual advice with regard to diet (diet rich in fibre, adequate fluid intake) and lifestyle (need for a physical activity), despite these counsels were not standardized.
2.5. Outcomes and follow-up 2.5.1. Primary endpoint The primary endpoint was the change from Day 0 to Day 21 in the intensity of the global discomfort related to symptoms of dyschezia. This was assessed by the patient during the last 3 days using a VAS rated from 0 mm (no discomfort) to 100 mm (maximal discomfort). While no specific and validated clinical criterion is currently available to assess dyschezia, a VAS is a common and validated tool used to assess pain and/or discomfort related to disease, including bowel disorders [13]. 2.5.2. Secondary efficacy endpoints Secondary efficacy endpoints were changes in the intensity of discomfort related to dyschezia during the study, assessed each evening by the patients in their diary with the NRS-11. Other secondary endpoints included changes in BFI [14] and PAC-QOL questionnaire [15] between baseline and Day 21. PGIC [16] at Day 21 was also considered. The BFI is a simple, clinician-administered, patient-reported, 3item questionnaire used to evaluate opioid-induced constipation in cancer and non-cancer chronic pain patients. Each question is rated from 0 to 100, and the mean of 3 values is then considered. The questions assess ease of defecation, feeling of incomplete evacuation, and personal judgement of constipation [14].
684
A.L. Tarrerias et al. / Digestive and Liver Disease 46 (2014) 682–687
The PAC-QOL questionnaire [15] was developed to address the need for a standardized measure of patient-reported outcomes to evaluate constipation-related QOL over time. The PAC-QOL comprises 28 items forming 4 subscales (worries and concerns, physical discomfort, psychosocial discomfort, and satisfaction) and an overall scale. Multinational studies have demonstrated that the PAC-QOL is internally consistent, reproducible, valid, and responsive to improvements over time. The PGIC is a tool for patient assessment of change using a 7-point scale (1 = very much improved, 2 = much improved, 3 = minimally improved, 4 = no change, 5 = minimally worse, 6 = much worse, 7 = very much worse) [16]. 2.5.3. Overall safety Overall safety during the treatment period was evaluated using reports of adverse events.
Table 1 Baseline demographic and clinical characteristics of the full analysis set population. Variable
Intervention N = 162
Control N = 150
p-Value
Female gender N (%) Mean age (years) Mean disease duration (years) Mean discomfort VAS score (0–100 mm) Patients who had received treatment for dyschezia in the previous 3 months N (%)
127 (78%)
120 (80%)
0.72
54.2 ± 14.6 5.4 ± 8.8
53.9 ± 14.8 5.4 ± 10.1
0.85 0.99
69.1 ± 10.9
68.8 ± 12.7
0.70
82 (51%)
93 (62%)
0.043
VAS, visual analogue scale.
2.6. Statistical analysis A sample size of 153 patients randomly assigned per group was required to provide 80% statistical power and to detect an 8-mm difference on the VAS between the groups in the primary endpoint, assuming a standard deviation of 25 mm, using a two-sided test and an alpha level of 5%. The intent-to-treat (ITT) population included all randomized patients. The safety population included all randomized patients who took at least 1 dose of the study treatment. The full analysis set (FAS) included all patients from the safety population who had 1 primary endpoint measurement after the baseline visit. A per-protocol (PP) population was also defined on the basis of the absence of major deviation to the protocol, an exposure to study treatment during at least 16 days, and availability of primary endpoint. The main analysis of the primary endpoint was conducted on the FAS population using a model of analysis of covariance that was adjusted for treatment and baseline VAS score. The last observation carried forward (LOCF) imputation method was used for missing data. Sensitivity analyses of the primary endpoint were conducted on the ITT population (with LOCF and baseline observation carried forward (BOCF) imputation methods) and the PP population. Also, the 50% responder’s rate in terms of discomfort was analysed. Changes in the intensity of discomfort related to dyschezia during the study, as assessed each evening by the patient, were analysed by a repeated measure analysis of variance. Changes of PAC-QOL and BFI were analysed in the same way as the primary endpoint. PGIC was analysed with a Wilcoxon–Mann–Whitney test. All inferential analyses were performed with a two-sided level of significance of 5%. All statistical analyses were conducted using SAS software (version 9.1.4; SAS Institute, Cary, NC). 2.7. Regulatory aspects The protocol and its amendments were approved by the Ile de France IV independent Ethics Committee and authorized by the French Regulatory Agency. The study was conducted in accordance with the International Conference on Harmonization Guideline for Good Clinical Practice and the Declaration of Helsinki. This trial was registered with ClinicalTrial.gov, number NCT01243723. 3. Results 3.1. Enrolment and baseline characteristics The flow chart of the study is shown in Fig. 1. A total of 323 patients were randomized (166 in the intervention group and 157 in the control group). Among them, 256 patients (78.5%) were
recruited by general practitioners and 67 (20.7%) by gastroenterologists. Twenty-four patients (7.4%) withdrew from the study. The most common reasons for discontinuation in both groups were lack of efficacy (18 patients) and adverse events (6 patients). In the FAS (312 patients), 162 patients were in the intervention group and 150 in the control group. Demographic and clinical characteristic at baseline are shown in Table 1. Mean age was 54.2 ± 14.6 years, and 79.2% of patients were women. All patients fulfilled the inclusion criteria for dyschezia, with a mean duration of 5.4 ± 9.4 years. The mean baseline discomfort related to dyschezia VAS score was 69.1 ± 10.9 mm in the intervention group and 68.8 ± 12.7 mm in the control group. No significant clinical difference was observed between the 2 groups, or between patients recruited either by general practitioners or gastroenterologists. 3.2. Efficacy 3.2.1. Primary efficacy endpoint A greater reduction in dyschezia-related discomfort VAS score from baseline to Day 21 was observed in the intervention group (−34.5 mm; SEM: 1.8 mm) than in the control group (−26.2 mm; SEM: 1.9 mm). In the FAS population, covariance analysis of the global VAS score adjusted for treatment and baseline VAS scores showed a statistically significant difference of −8.2 [95% CI −12.9; −3.6] (p < 0.001), in favour of the intervention group (Table 2). Similar significant differences were observed for the analyses on the ITT and PP populations (Table 3). 3.2.2. Secondary efficacy endpoints In the FAS population, the number of 50% responders (with a decrease of at least 50% in dyschezia-related discomfort VAS score) at Day 21 was 82 (50.6%) for the intervention group and 50 (33.3%) for the control group (p < 0.01). This corresponds to an OR of 2.1 [95% CI 1.3–3.2]. Similar results were obtained for the PP population. The analysis of the daily intensity of discomfort related to dyschezia also showed a statistically significant difference in favour of the intervention group (p < 0.001) (Fig. 2). At Day 21, the 4 subscales (worries and concerns, physical discomfort, psychosocial discomfort, and satisfaction) and the overall scale of the PAC-QOL as well as the BFI scores showed clinical and statistical differences between the 2 groups in favour of the intervention group (Table 4). PGIC data confirmed the greater improvement in the intervention group versus the placebo group. The percentage of patients self-assessed as “very much improved”, “much improved”, “minimally improved”, “with no change”, “minimally worse”, “much worse”, and “very much
A.L. Tarrerias et al. / Digestive and Liver Disease 46 (2014) 682–687
685
Fig. 1. Patient flowchart. ITT, intent-to-treat; FAS, full analysis set; PP, per-protocol.
Table 2 Primary efficacy end point – main analysis (full analysis set population – last observation carried forward) – changes in discomfort intensity assessed using a VAS between Day 0 and Day 21.
Discomfort intensity (Day 0) (0–100 mm) Discomfort intensity (Day 21) (0–100 mm) Change (Day 0–Day 21) Change adjusted for baseline value Difference in change adjusted for baseline value
Intervention N = 162
Control N = 150
p-Value (2-sided)*
69.1 (10.9) 34.5 (20.4) −34.5 (22.6) −34.5 (1.8)**
68.8 (12.7) 42.7 (21.4) −26.1 (22.7) −26.2 (1.9)**
0.82
Contents lists available at ScienceDirect
Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld
Alimentary Tract
Efficacy of a CO2 -releasing suppository in dyschezia: A double-blind, randomized, placebo-controlled clinical trial夽 Anne Laure Tarrerias a , Laurent Abramowitz b , Marc M.L. Marty c,∗ , Pierre Coulom d , Ghislain Staumont d , Christophe Merlette e , Véronique Berger c , Bernard Savarieau c,f , Philippe Ducrotté g a
General and Digestive Surgery Unit, Foch Hospital, Suresnes, France Medical Surgical Proctology Unit, Bichat-Claude Bernard Hospital, France c Nukléus, Clinical Trials Department, Paris, France d Proctology Unit, Clinique St Jean Languedoc, Toulouse, France e General Medicine Unit, Fresnes, France f Hepatogastroenterology Unit, Intercommunal Hospital, Créteil, France g INSERM Unit U1073 and Gastroenterology Department, Rouen University Hospital, Rouen, France b
a r t i c l e
i n f o
Article history: Received 19 December 2013 Accepted 15 April 2014 Available online 11 June 2014 Keywords: Clinical trial Dyschezia Functional G.I. disease
a b s t r a c t Background: Constipation has a significant impact on quality of life. Aim of this study was to evaluate the safety and the efficacy for relieving dyschezia symptoms of a CO2 -releasing suppository in a randomized, placebo-controlled, clinical trial. Methods: Fifty-three office-based primary care physicians and 24 gastroenterologists conducted the study in France, between November 2010 and January 2012. Patients (aged 18–75 years) with dyschezia were eligible. Patients were randomly allocated a once-a-day suppository (CO2 -releasing suppository or placebo) for 21 days. Primary endpoint was the change, from Day 0 to Day 21, in the intensity of discomfort related to dyschezia based on a self-assessed 0–100 visual analogue scale. Results: A total of 323 patients were randomized, i.e. 166 into the intervention group and 157 into the placebo group. Co-variance analysis showed a greater reduction in discomfort visual analogue scale score in the intervention group (−34.5 mm; standard error of the mean: 1.8 mm) than in the placebo group (−26.2 mm; standard error of the mean: 1.9 mm; p < 0.001). The greater efficacy of the CO2 -releasing suppository was confirmed for all secondary efficacy parameters. No significant side effects for either treatment were observed. Conclusion: A CO2 -releasing suppository is more effective than a placebo for the relief of symptoms of dyschezia. This efficacy is associated with a good safety profile. © 2014 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
1. Introduction Constipation can be classified either as normal-transit constipation (often coexisting with irritable bowel syndrome), slow transit constipation, or functional defecation disorder. Each of these conditions, which can occur in isolation or co-exist [1], has a significant impact on the patient’s quality of life [2,3]. In functional defecation disorders, pathophysiological features are a normal or
夽 ClinicalTrial.gov number: NCT01243723. ∗ Corresponding author at: Nukléus, Clinical Trials Department, 55 Rue Bobillot, Paris 75013, France. Tel.: +33 (0)1 45 88 66 88. E-mail address: [email protected] (M.M.L. Marty).
slightly slowed colonic transit overall, but a preferential storage of residues for prolonged periods in the rectum, which explains the patient’s symptoms, i.e. feeling of incomplete evacuation, sensation of anorectal blockage or straining. In this instance, the primary failure is an inability to adequately evacuate rectal contents [4]. Functional defecation disorder occurs in approximately 7% of adults [5], although isolated defecatory dysfunction seems to be present in 25–40% of patients with chronic constipation [6]. According to the Rome III criteria, the diagnosis of functional defecation disorders requires abnormal findings detected by a balloon expulsion test, recto-anal manometry, imaging, or electromyography in order to discuss a targeted treatment. For instance, when the functional defecation disorder is related to pelvic floor dyssynergia, biofeedback is the treatment of choice
http://dx.doi.org/10.1016/j.dld.2014.04.005 1590-8658/© 2014 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
A.L. Tarrerias et al. / Digestive and Liver Disease 46 (2014) 682–687
683
with an efficacy in approximately 70–80% of the patients, and a long-term success rate in 50% [7,8]. However, in clinical practice, functional tests are not available in many cases. Therefore, the first-line treatment is given when a defecation disorder is only suspected according to symptoms of dyschezia including sensation of blocked stools, feeling of incomplete evacuation, repeated attempts to defecate, straining during bowel movements and need to practice manual manoeuvres to facilitate defecation, even if we have shown that some symptoms, such as straining at stool, are highly predictive of anorectal manometric disturbances [9]. In such cases, diet advice and laxatives are rarely effective [4]. Therefore, other options of treatment for dyschezia symptoms, which can be easily used in primary care, are of interest. The daily use of a CO2 releasing suppository (EductylTM -TechniPharma, Monaco) could be an alternative for some patients with dyschezia symptoms. This CO2 -releasing suppository contains 1.15 g of potassium acid tartrate, 0.7 g of sodium bicarbonate, and excipients including solid semisynthetic glycerides, soy lecithin, and talc. Within a few minutes after introduction in the rectum, this suppository releases approximately 100 mL of gas and triggers defecation. In previous trials, a CO2 -releasing suppository has been shown to promote more frequent stools with normal consistency than placebo [10]. In addition, the studied CO2 -releasing suppository seemed to be more effective than psyllium in reducing the sensation of incomplete evacuation and straining during defecation [11]. Therefore, we proposed the hypothesis that the daily use of a CO2 -releasing suppository could be effective to alleviate symptoms of dyschezia. We thus designed this multicentre trial to assess the efficacy and safety of this CO2 -releasing suppository in patients suffering from dyschezia.
2010 and January 2012. All participating physicians operated at their private practice. Two visits were scheduled: one at baseline (Day 0) and one after 3 weeks (Day 21). At baseline, after verification that the subjects fulfilled the inclusion criteria, they received comprehensive information about the study and provided written informed consent prior to inclusion. Demographic data and history of dyschezia symptoms were recorded. Both at baseline and Day 21, presence of stools in the rectum, intensity of discomfort related to dyschezia on a VAS [13], and Bowel Functional Index (BFI) [14] were assessed, as well as the Patient Assessment of Constipation Quality of Life (PACQOL) questionnaire [15] was filled. Additionally, patients gave their opinion about the treatment using the patient global improvement of change (PGIC) [16] at Day 21. At the end of the inclusion visit, patients were given a selfevaluation diary in which, at home each evening, they recorded the intensity of discomfort related to dyschezia on an 11-point numeric rating scale (NRS-11), ranging from 0 (no discomfort) to 10 (unbearable discomfort) from Day 1 to Day 21.
2. Materials and methods
2.4. Randomization and masking
2.1. Patients
Investigators, patients, and study sponsor were masked to treatment assignment. An independent statistician established a computer-generated randomization scheme balanced by centre. Allocation of treatment was carried out using a sequentially numbered container.
Patients aged between 18 and 75 years were eligible if they described symptoms listed among the Rome III criteria to define functional defecation disorders [12], in the presence of 2 or more of the following symptoms: (a) straining during defecation; (b) feeling of incomplete defecation; (c) feeling of anorectal obstruction/blockage at defecation; (d) need of manual manoeuvres to facilitate defecation. These signs had to be present in at least 25% of defecations and for the previous 3 months with symptom onset at least 6 months prior to inclusion. Patients also had to fulfil the following criteria: presence of stool in the rectum at digital rectal examination and intensity of discomfort related to dyschezia of at least 40 mm as assessed by the patient on a visual analogue scale (VAS) from 0 mm (no discomfort) to 100 mm (maximal discomfort). Exclusion criteria were the presence of neurological disease (e.g. paraplegia, multiple sclerosis), anal pain, colonic or rectal organic lesion (including rectocele), and chronic inflammatory colon disease. Clinical examination prior to inclusion led also to exclude patients with anal stenosis, anal fissure, rectal prolapse, haemorrhoid flare-up, and descended perineum. The use of antidepressants, psychotropic drugs, or laxatives was allowed if the dose, which was not to be modified during the study, had been stable for the previous 15 days. Opioids and anorectal treatments (other types of suppositories, cleaning enemas, or biofeedback) were prohibited throughout the study. 2.2. Design Fifty-three office-based primary care physicians and 24 gastroenterologists conducted the study in France, between November
2.3. Interventions Patients were randomly allocated to a once-a-day treatment in the morning, consisting of either a CO2 -releasing suppository (intervention group) or a placebo suppository containing the same excipients (solid semisynthetic glycerides, soy lecithin, and talc) (control group) but no CO2 . Treatment was given for 21 days. Size and appearance of both types of suppositories were strictly identical. All patients received usual advice with regard to diet (diet rich in fibre, adequate fluid intake) and lifestyle (need for a physical activity), despite these counsels were not standardized.
2.5. Outcomes and follow-up 2.5.1. Primary endpoint The primary endpoint was the change from Day 0 to Day 21 in the intensity of the global discomfort related to symptoms of dyschezia. This was assessed by the patient during the last 3 days using a VAS rated from 0 mm (no discomfort) to 100 mm (maximal discomfort). While no specific and validated clinical criterion is currently available to assess dyschezia, a VAS is a common and validated tool used to assess pain and/or discomfort related to disease, including bowel disorders [13]. 2.5.2. Secondary efficacy endpoints Secondary efficacy endpoints were changes in the intensity of discomfort related to dyschezia during the study, assessed each evening by the patients in their diary with the NRS-11. Other secondary endpoints included changes in BFI [14] and PAC-QOL questionnaire [15] between baseline and Day 21. PGIC [16] at Day 21 was also considered. The BFI is a simple, clinician-administered, patient-reported, 3item questionnaire used to evaluate opioid-induced constipation in cancer and non-cancer chronic pain patients. Each question is rated from 0 to 100, and the mean of 3 values is then considered. The questions assess ease of defecation, feeling of incomplete evacuation, and personal judgement of constipation [14].
684
A.L. Tarrerias et al. / Digestive and Liver Disease 46 (2014) 682–687
The PAC-QOL questionnaire [15] was developed to address the need for a standardized measure of patient-reported outcomes to evaluate constipation-related QOL over time. The PAC-QOL comprises 28 items forming 4 subscales (worries and concerns, physical discomfort, psychosocial discomfort, and satisfaction) and an overall scale. Multinational studies have demonstrated that the PAC-QOL is internally consistent, reproducible, valid, and responsive to improvements over time. The PGIC is a tool for patient assessment of change using a 7-point scale (1 = very much improved, 2 = much improved, 3 = minimally improved, 4 = no change, 5 = minimally worse, 6 = much worse, 7 = very much worse) [16]. 2.5.3. Overall safety Overall safety during the treatment period was evaluated using reports of adverse events.
Table 1 Baseline demographic and clinical characteristics of the full analysis set population. Variable
Intervention N = 162
Control N = 150
p-Value
Female gender N (%) Mean age (years) Mean disease duration (years) Mean discomfort VAS score (0–100 mm) Patients who had received treatment for dyschezia in the previous 3 months N (%)
127 (78%)
120 (80%)
0.72
54.2 ± 14.6 5.4 ± 8.8
53.9 ± 14.8 5.4 ± 10.1
0.85 0.99
69.1 ± 10.9
68.8 ± 12.7
0.70
82 (51%)
93 (62%)
0.043
VAS, visual analogue scale.
2.6. Statistical analysis A sample size of 153 patients randomly assigned per group was required to provide 80% statistical power and to detect an 8-mm difference on the VAS between the groups in the primary endpoint, assuming a standard deviation of 25 mm, using a two-sided test and an alpha level of 5%. The intent-to-treat (ITT) population included all randomized patients. The safety population included all randomized patients who took at least 1 dose of the study treatment. The full analysis set (FAS) included all patients from the safety population who had 1 primary endpoint measurement after the baseline visit. A per-protocol (PP) population was also defined on the basis of the absence of major deviation to the protocol, an exposure to study treatment during at least 16 days, and availability of primary endpoint. The main analysis of the primary endpoint was conducted on the FAS population using a model of analysis of covariance that was adjusted for treatment and baseline VAS score. The last observation carried forward (LOCF) imputation method was used for missing data. Sensitivity analyses of the primary endpoint were conducted on the ITT population (with LOCF and baseline observation carried forward (BOCF) imputation methods) and the PP population. Also, the 50% responder’s rate in terms of discomfort was analysed. Changes in the intensity of discomfort related to dyschezia during the study, as assessed each evening by the patient, were analysed by a repeated measure analysis of variance. Changes of PAC-QOL and BFI were analysed in the same way as the primary endpoint. PGIC was analysed with a Wilcoxon–Mann–Whitney test. All inferential analyses were performed with a two-sided level of significance of 5%. All statistical analyses were conducted using SAS software (version 9.1.4; SAS Institute, Cary, NC). 2.7. Regulatory aspects The protocol and its amendments were approved by the Ile de France IV independent Ethics Committee and authorized by the French Regulatory Agency. The study was conducted in accordance with the International Conference on Harmonization Guideline for Good Clinical Practice and the Declaration of Helsinki. This trial was registered with ClinicalTrial.gov, number NCT01243723. 3. Results 3.1. Enrolment and baseline characteristics The flow chart of the study is shown in Fig. 1. A total of 323 patients were randomized (166 in the intervention group and 157 in the control group). Among them, 256 patients (78.5%) were
recruited by general practitioners and 67 (20.7%) by gastroenterologists. Twenty-four patients (7.4%) withdrew from the study. The most common reasons for discontinuation in both groups were lack of efficacy (18 patients) and adverse events (6 patients). In the FAS (312 patients), 162 patients were in the intervention group and 150 in the control group. Demographic and clinical characteristic at baseline are shown in Table 1. Mean age was 54.2 ± 14.6 years, and 79.2% of patients were women. All patients fulfilled the inclusion criteria for dyschezia, with a mean duration of 5.4 ± 9.4 years. The mean baseline discomfort related to dyschezia VAS score was 69.1 ± 10.9 mm in the intervention group and 68.8 ± 12.7 mm in the control group. No significant clinical difference was observed between the 2 groups, or between patients recruited either by general practitioners or gastroenterologists. 3.2. Efficacy 3.2.1. Primary efficacy endpoint A greater reduction in dyschezia-related discomfort VAS score from baseline to Day 21 was observed in the intervention group (−34.5 mm; SEM: 1.8 mm) than in the control group (−26.2 mm; SEM: 1.9 mm). In the FAS population, covariance analysis of the global VAS score adjusted for treatment and baseline VAS scores showed a statistically significant difference of −8.2 [95% CI −12.9; −3.6] (p < 0.001), in favour of the intervention group (Table 2). Similar significant differences were observed for the analyses on the ITT and PP populations (Table 3). 3.2.2. Secondary efficacy endpoints In the FAS population, the number of 50% responders (with a decrease of at least 50% in dyschezia-related discomfort VAS score) at Day 21 was 82 (50.6%) for the intervention group and 50 (33.3%) for the control group (p < 0.01). This corresponds to an OR of 2.1 [95% CI 1.3–3.2]. Similar results were obtained for the PP population. The analysis of the daily intensity of discomfort related to dyschezia also showed a statistically significant difference in favour of the intervention group (p < 0.001) (Fig. 2). At Day 21, the 4 subscales (worries and concerns, physical discomfort, psychosocial discomfort, and satisfaction) and the overall scale of the PAC-QOL as well as the BFI scores showed clinical and statistical differences between the 2 groups in favour of the intervention group (Table 4). PGIC data confirmed the greater improvement in the intervention group versus the placebo group. The percentage of patients self-assessed as “very much improved”, “much improved”, “minimally improved”, “with no change”, “minimally worse”, “much worse”, and “very much
A.L. Tarrerias et al. / Digestive and Liver Disease 46 (2014) 682–687
685
Fig. 1. Patient flowchart. ITT, intent-to-treat; FAS, full analysis set; PP, per-protocol.
Table 2 Primary efficacy end point – main analysis (full analysis set population – last observation carried forward) – changes in discomfort intensity assessed using a VAS between Day 0 and Day 21.
Discomfort intensity (Day 0) (0–100 mm) Discomfort intensity (Day 21) (0–100 mm) Change (Day 0–Day 21) Change adjusted for baseline value Difference in change adjusted for baseline value
Intervention N = 162
Control N = 150
p-Value (2-sided)*
69.1 (10.9) 34.5 (20.4) −34.5 (22.6) −34.5 (1.8)**
68.8 (12.7) 42.7 (21.4) −26.1 (22.7) −26.2 (1.9)**
0.82
