Randomised Double-Blind Placebo-Controlled Trial Of Aloe Vera For Irritable Bowel Syndrome
K. Davis; S. Philpott; D. Kumar; M. Mendall
Published: 11/02/2006
Aloe vera (AV) is suggested to be beneficial in treating irritable bowel syndrome (IBS) symptoms, but no scientific trials exist to confirm this. We aim to assess the efficacy of AV on IBS in refractory secondary care patients. Patients with IBS were randomised to receive AV or matching placebo for a month. Symptoms were assessed at baseline, 1 and 3 months. Fifty-eight patients randomised, 49 completed the protocol to 1 month and 41 to 3 months. Eleven of thirty-one (35%) AV patients, and 6 of 27 (22%) placebo patients responded at 1 month (p = 0.763). Diarrhoea predominant patients showed a trend towards a response to treatment at 1 month (10/23 V 2/14, p = 0.07). There was no evidence that AV benefits patients with IBS. However, we could not rule out the possibility that improvement occurred in patients with diarrhoea or alternating IBS whilst taking AV. Further investigations are warranted in patients with diarrhoea predominant IBS, in a less complex group of patients.
Introduction
Irritable bowel syndrome (IBS) is a common disorder affecting up to 50% of patients attending gastroenterology outpatients and 10-20% of the population.[1-5] It is a multi-symptom condition characterised by abdominal pain/discomfort, change in bowel habit (either towards constipation, diarrhoea or combination of both), feeling of incomplete evacuation, distension or bloating and passage of mucus. Patients tend to follow a chronic course of symptoms with intermittent exacerbations. There are a number of specific treatments available for this disorder. Conventional medical management options include treatment with antispasmodics or constipating agents/laxatives and bulking agents. Patients towards the more severe end of the spectrum of the disorder may require anxiolytics or even antidepressants. These conventional treatments are of modest efficacy only, and only muscle relaxants[6] and low-dose antidepressant therapy have shown to be of any benefit in double-blind placebo-controlled trials.[7] It is not surprising therefore that two-thirds of refractory patients fail to respond to such treatments. Patients in secondary and tertiary care are even less responsive to standard therapies. Complementary therapies include dietary changes, hypnotherapy, cognitive behavioural therapy and Aloe vera (AV), and although there is plenty of anecdotal evidence, these have not been scientifically evaluated. Aloe vera is a plant that can produce latex and gel. The gel is extracted from the leaf, and it is this substance that is most used as a treatment. AV has been evaluated in a number of different clinical contexts and some promising results have been found for its use in controlling cardiovascular risk factors[8] and diabetes,[9] besides being beneficial in areas of dermatology.[10,11] One explanatory factor for this is the anti-inflammatory properties of the plant.[12] More recently, AV has been evaluated in patients with active ulcerative colitis (UC),[13] and this study found that patients who had taken AV over 4 weeks had a higher frequency of clinical remission and improvement compared to those taking the placebo. It is possible that these anti-inflammatory effects of AV could have beneficial effects on symptoms of IBS by reducing gut hypersensitivity. The aim of this study was to assess the efficacy of AV using the Natural Living Products (NLP) formulation on symptoms of IBS in refractory secondary care patients. In this preparation alloin, which can cause diarrhoea, is nearly eliminated with a remaining concentration of less than 1 p.p.m. It is the polysaccharides in the preparation which NLP claim exert an anti-inflammatory action. It was decided to evaluate a 1-month course of treatment with two further months of follow-up, as the primary aim was to determine whether there was any beneficial effect with short-term treatment, and if so whether this was maintained after cessation of therapy
This was a placebo-controlled double-blind trial, with the patient randomised to either receive the drug or the placebo for 1 month and followed up for a further 2 months. The study protocol was approved by the ethics committees of the two study centres.
Patients were recruited over a 9-month period from the gastroenterology clinics of two large south London Hospitals. Patients were eligible for inclusion if they satisfied the Rome 2 criteria for IBS,[14] were aged 18-65 and had previously tried but failed conventional management with antispasmodics, bulking agents and dietary intervention. Patients were excluded if they were on medications other than those related to IBS, had other medical conditions, including those related to the bowel, were pregnant or at risk of pregnancy. Detailed signed informed consent was sought from all selected patients. A previously validated questionnaire[15] was administered to the subjects by a trained research nurse at baseline and repeated at 1 and 3 months follow-up. All subjects were treated with either AV using the NLP formulation or matching placebo at a dose of 50 ml taken four times a day for 1 month. The AV formulation came as a pink syrup flavoured with mango and the placebo was matched to this in colour and flavour. All patients received the same batch from the same manufacturer. The patients, doctors and researchers remained blind to the treatment they received throughout the trial.
Patients were randomly assigned to the treatment group using a computerised random numbers table. Once recruited, patients received their allocated medication from the central pharmacies of the two hospitals. All medication was kept in numbered bottles which were held in the pharmacy department. The patients, dispensing pharmacists and the researchers were all blind to the treatment group allocated to each patient. The code was held in pharmacy and only broken at the end of the study for the purpose of analysis.
Sample Size and Study End Points
The primary end point that was assessed was the change in global summated symptom score for abdominal pain, distension, satisfaction with bowel habit, and global impact of symptoms on well-being, with a change of 50 points indicating improvement.[15] The symptom score with a maximum of 500 was derived by adding the scores for the individual symptoms, and the proportion of days with symptoms. Each individual symptom was scored from 0% to 100% on a visual analogue scale. Responses of individual symptom scores were also assessed as secondary end points. Most studies of IBS find a placebo response rate of 35%, with a 65% response rate being commonly found in studies of smooth muscle relaxants.[16] For AV to be a potentially useful treatment in view of the inconvenience of taking the medication, and given the safety and palability of other medications, we felt that a substantial therapeutic benefit needed to be established for AV to be widely used was of a 40% higher response rate. If it was assumed that the placebo response rate was 35%, then to detect a 75% response rate with 80% power required 27 subjects per treatment group or a total study size of 54 subjects. Analysis
Chi-square tests were used to test proportions, the Wilcoxon rank sum test to compare non-normally distributed data, and the Student's t-test to compare normally distributed continuous data. All analysis was carried out in STATA (version 8). The Kurtosis test for normality was performed on the data to establish whether to use parametric or nonparametric analysis. An intention-to-treat analysis was performed with subjects who dropped out being assumed not to have responded. Analyses on the follow-up data were adjusted for baseline scores using logistic regression.
The numbers of subjects completing each stage of the study are shown on the attached flow chart (Figure 1). Fifty-eight patients consented and were randomised to the treatment groups (31 to the active treatment group and 27 to the placebo group). Between randomisation and the start of the trial, two patients from each group withdrew, leaving 54 patients. At 1 month follow-up, two patients from the placebo group and three from the active group withdrew, and at 3 months follow-up a further eight patients withdrew (five in the placebo group and three in the active group). The main reason for failure to complete the study was nausea and vomiting with the study medication (four in placebo and two in the active groups) and non-responders. At entry, there were no significant differences between the two groups of patients ( Table 1 ).
| | Figure 1. Study flow chart. |
Figure 1.
Study flow chart.
Table 2 gives the change in scores at 1 and 3 months. By intention to treat, at 1 month, 11 of 31 (35%) patients in the active group and 6 of 27 (22%) in the placebo group responded to treatment. Per protocol, 11 of 26 (42%) of patients in the active group and 6 of 23 (26%) in the placebo group had responded to treatment with an improvement in their IBS score of 50 and over (p = 0.234). There were trends towards improvement in pain score, proportion of days with pain, proportion with distension in the past week, bowel habit satisfaction score and interference with life score, but these results were not significantly better in the active group. The distension score worsened in the active group, but the mean change was not statistically different between the active and placebo groups. At 3 months, although there was a continued improvement in pain score in the active group, this was not significant (p = 0.08), and no other improvements were observed. The distension score continued to get worse for the active group, but again this was not significant.
Subgroup Analyses
It was clear that treatment was ineffective in constipation predominant patients. Therefore,subgroup analysis was restricted to patients who were diarrhoea predominant or mixed. The baseline scores are shown in Table 3 and the changes at 1 and 3 months in Table 4 . There were 18 patients randomised to the placebo group and 26 to the active treatment group. At baseline those patients randomised to the active group had higher IBS and pain scores than the placebo group.
At 1 month, 2 of 18 (11%) randomised to the placebo group and 10 of 26 (38%) randomised to the active group showed a response to treatment. Per protocol, 2 of 14 (14%) patients in the placebo group and 10 of 23 (43%) in the active group responded to treatment (p = 0.07). Amongst patients in the active group there was a significant improvement in IBS score, proportion with pain in the past week, proportion of days with pain in the past week and bowel habit satisfaction score compared with the placebo group.
At 3 months, the response by intention to treat was 5 of 18 (28%) and 5 of 26 (19%) in the placebo and active group respectively. Per protocol, this was 5 of 11 (45%) and 4 of 21 (19%) in the placebo and active groups respectively (p = 0.12). There was a continued improvement in the IBS score (mean change 12.19 vs. −5.82, p = 0.57) and pain score in the active group, although these were not significant.
At 1 month, there were no differences in the changes in the distension score between the two groups. At 3 months the distension score had worsened for the active group, but this was not significantly different to the placebo group
In this, the first placebo-controlled trial of AV using the NLP formulation in IBS, there was no overall benefit found among patients taking the active treatment compared to patients taking a matching placebo. Although a slight benefit was seen in the diarrhoea predominant group whilst the treatment was being taken, this subgroup was not included as part of the protocol and therefore generates a hypothesis that warrants further study.
It is possible that the failure to find a benefit in the whole group was due to the study being underpowered from overoptimistic assumptions of the size of effect of the NLP AV preparation. A study size of 203 subjects per group would have been required for an 80% power to detect the differences observed in the whole group at the 5% level of significance based on an intention-to-treat analysis. However, our findings suggest that future studies should study diarrhoea predominant subjects, and based on the point estimates from the current study 46 subjects per group will be required.
The magnitude of the benefit we observed in the whole group, although not statistically significant, was equal to that observed for Alosetron, the IBS drug recently withdrawn for its side effects by Glaxo (41% versus 29% response rates for active and placebo respectively).[17] Furthermore, AV using the NLP formulation was generally well tolerated, with distension as the only apparent side effect. Six patients did withdraw from the study with nausea and/or vomiting. However, a majority of these came from the placebo group and so was not considered to be a significant side effect. This was a double-blind randomised placebo-controlled trial, thus reducing the potential for bias. All patients took the drug as requested and few patients were lost to follow-up. However, these results have to be considered in the light of the fact that the subjects included in this trial were all recruited from a pool of refractory patients in secondary and tertiary care, and as such the research protocol may have constituted an overly rigorous test of efficacy. By chance scores in the active group tended to be higher at baseline than in the placebo group and hence any apparent response in the group could be explained by regression to the mean, but the fact that the active group score returned to baseline after treatment had been discontinued argues against this. It is also possible that placebo responses were more marked in the active group because of their more severe symptoms at baseline. We have performed further analysis adjusting the response for the baseline score and this does little to change the message.
Aloe vera has long been recognised as having pain killing and healing properties. It is typically used for topical treatments of wounds, minor burns and skin irritations, but has also been used to treat constipation, ulcers, diabetes and cardiovascular risk factors. Despite a lack of evidence of its therapeutic effects, it is widely taken to control abdominal discomfort including IBS. Recently a trial was conducted investigating the efficacy of AV in patients with UC.[13] The researchers found that after taking AV gel for 4 weeks the UC patients had greater reductions in their colitis and histological scores. No adverse events were recorded. The mechanism of action of AV is unknown. AV juice is a complex mixture of more than 75 biologically active chemicals. It has proved difficult to isolate a single active ingredient and it is believed that there may be synergy between the different components. Mannose-6-phosphate, a polysaccharide, is one of these components, which for example is capable of activating fibroblast receptor and possibly promoting healing. The NLP formulation used in this study goes through a purification process to remove aloin, and enrich the polysaccharide content. Aloin is known to possess laxative effects and may be responsible for side effects when AV is ingested orally.
This study has demonstrated that AV is well tolerated among patients with IBS, and that whilst the patients were taking it there was some improvement in symptoms, specifically in the IBS score, pain score and proportion of days with pain in the past week, although this was only significant among patients with diarrhoea predominant or mixed symptoms. These patients had all failed previous therapeutic regimes and so were deemed as complex patients. The fact that there was some improvement is therefore encouraging.
We conclude that AV is safe to take and could possibly benefit patients with diarrhoea predominant or alternating diarrhoea and constipated IBS, although we were formally unable to demonstrate this. For these patients there was an improvement in pain (proportion with pain in the past week, pain score and proportion of days with pain in the past week) and bowel habit satisfaction score. These effects were not sustained off treatment. Further studies are warranted to assess the efficacy of taking AV for a longer period of time, with less complex patients and with a more convenient preparation. It remains to be determined whether the treatment would be effective in patients seen in primary care who were less complex and less likely to have failed other treatments.
1. Jones R, Lydeard S. Irritable bowel syndrome in the general population. BMJ 1992; 304: 87-90.
2. Heaton KW, O'Donnel LJ, Braddon FE et al. Symptoms of irritable bowel syndrome in a British urban community: consulters and nonconsulters. Gastroenterology 1992; 102: 1962-7.
3. Osterberg E, Blomquist L, Krakau I et al. A population study on irritable bowel syndrome and mental health. Scand J Gastroenterol 2000; 35: 264-8.
4. Ehlin AGC, Montgomery AM, Ekbom A et al. Prevalence of gastrointestinal diseases in two British national birth cohorts. Gut 2003; 52: 1117-21.
5. Wilson S, Roberts L, Roalfe A et al. Prevalence of irritable bowel syndrome: a community survey. Br J Gen Pract 2004; 54: 495-502.
6. Poynard T, Regimbeau C, Benhamou Y. Meta-analysis of smooth muscle relaxants in the treatment of irritable bowel syndrome. Aliment Pharmacol Ther 2001; 15: 355-61.
7. Jackson JL, O'Malley PG, Tomkins G et al. Treatment of functional gastrointestinal disorders with antidepressant medications: a meta analysis. Am J Med 2000; 108: 65-72.
8. Agarwal OP. Prevention of artheromatous heart disease. Angiology 1985; 36: 485-92.
9. Yeh GY, Eisenberg DM, Kaptchuk TJ et al. Systematic review of herbs and dietary supplements for glycemic control in diabetes. Diabetes Care 2003; 26: 1977-94.
10. Klein AD, Penneys NS. Aloe vera. J Am Acad Dermatol 1988; 18: 714-20.
11. Choi SW, Son BW, Park YI et al. The wound healing effect of a glycoprotein fraction isolated from aloe vera. Br J Dermatol 2001; 145: 535-45.
12. Vazquez B, Avila G, Segura D et al. Anti-inflammatory activity of extracts from Aloe vera gel. J Ethnopharmacol 1996; 55: 69-75.
13. Langmead L, Feakins RM, Goldthorpe S et al. Randomized, double-blind, placebo-controlled trial of oral aloe vera gel for active ulcerative colitis. Aliment Pharmacol Ther 2004; 19: 739-74.
14. Thompson WG, Longstreth GF, Drossman DA et al. Functional bowel disorders and functional abdominal pain. Gut 1999; 45: 1143-7.
15. Francis CY, Morris J, Whorwell PJ. The irritable bowel severity scoring system: a simple method of monitoring irritable bowel syndrome and its progress. Aliment Pharmacol Ther 1997; 11: 395-402.
16. Pynard T, Naveau S, Mory B, Chaput J-C. Meta-analysis of smooth muscle relaxants in the treatment of irritable bowel syndrome. Aliment Pharmacol Ther 1994; 8: 499-510.
17. Camilleri M, Northcutt AR, Kong S et al. Efficacy and safety of alosetron in women with irritable bowel syndrome: a randomised, placebo-controlled trial. Lancet 2000; 355: 1035-40
