The treatment of functional dyspepsia with red pepper


  • The results in this paper appear as a Letter to the Editor in the New England Journal of Medicine. Bortolotti M, Coccia G, Grossi G. Red pepper and functional dyspepsia. N Engl J Med 2002; 346: 947–8.

Correspondence to: Professor M. Bortolotti, Via Massarenti 48, 40138 Bologna, Italy. E-mail: ;


Aim : To decrease the intensity of dyspeptic symptoms by impairing the visceral nociceptive C-type fibres with capsaicin, contained in red pepper powder.

Methods : The study was performed on 30 patients with functional dyspepsia and without gastro-oesophageal reflux disease and irritable bowel syndrome. After a 2-week washout period, 15 patients received, before meals randomly and in a double-blind manner, 2.5 g/day of red pepper powder for 5 weeks, and 15 patients received placebo. A diary sheet was given to each patient to record, each day, the scores of individual and overall symptom intensity, which subsequently were averaged weekly and over the entire treatment duration.

Results : The overall symptom score and the epigastric pain, fullness and nausea scores of the red pepper group were significantly lower than those of the placebo group, starting from the third week of treatment. The decrease reached about 60% at the end of treatment in the red pepper group, whilst placebo scores decreased by less than 30%.

Conclusions : Red pepper was more effective than placebo in decreasing the intensity of dyspeptic symptoms, probably through a desensitization of gastric nociceptive C-fibres induced by its content of capsaicin. It could represent a potential therapy for functional dyspepsia.


Functional dyspepsia is one of the most common chronic digestive disorders affecting humans, with a prevalence in the community ranging from 7% to 41%.1 Although it is not a life-threatening condition, it may lead to a poor quality of life,2 and represents a high cost for society in terms of prescriptions and sick leave.3 The cause of functional dyspepsia is unknown. Gastric secretory and motility disorders, as well as diet, alcohol, tobacco, non-steroidal anti-inflammatory drug (NSAID) abuse, psychosocial factors, Helicobacter pylori infection, failure of fundal relaxation on meal ingestion and hypersensitivity to gastric distension and acid have been examined, but none has been found to be a definitive, unique cause of the disease.4 The lack of a well-established and unitary pathophysiology explains why no curative therapy exists and why the treatment is mostly empirical.5 Because the only certainty in dyspepsia is the presence of symptoms, such as pain or discomfort centred in the upper abdomen in the absence of endoscopic and biochemical abnormalities, therapy aimed at the impairment of the genesis and transmission of nociceptive impulses in the visceral sensorial fibres before they reach the central nervous system, where they give rise to symptoms, could be a remedy to the failure to find and cure the causes responsible for pathological activation of the visceral nociceptive neurones. This manner of dealing with dyspepsia follows an emerging trend in the treatment of functional gastrointestinal diseases,6 and is justified by the fact that the analgesic effect may be effective if the cause of dyspepsia is a lowered pain threshold,7, 8 whereas, if the visceral sensitivity is considered to be normal,9 the treatment simply blocks the transfer of nociceptive signals to the central nervous system, whatever the functional cause of dyspeptic symptoms. Some attempts have already been made to decrease pharmacologically the visceral sensitivity of functional gut disorders. The use of fedotozine, a peripheral K opioid agonist able to increase the visceral nociceptive threshold of the stomach,10 was investigated in patients with functional dyspepsia,11, 12 but showed rather disappointing results6, 13 and, probably because of its marginal clinical benefit, is not yet commercially available. Other drugs, including serotonin receptor (5-HT3 and 5-HT4) antagonists and the somatostatin analogue octreotide, are able to decrease visceral sensitivity,14 but have not been shown to be clinically efficacious in functional dyspepsia.6

Capsaicin (8-methyl-N-vanillyl-6-nonenamide), the pungent component of the various species of Capsicum (red pepper), can impair selectively the activity of nociceptive C-type fibres carrying pain sensations to the central nervous system.15–17 The analgesic properties of capsaicin have been known for more than a century, and a rich harvest of studies have demonstrated that the topical application of capsaicin is able to relieve post-herpetic neuralgia of the skin and oral cavity, trigeminal neuralgia and cluster headache,18–20 as well as painful diabetic neuropathy and vesical neuropathic pain.21, 22 In addition, recent experiments have demonstrated that repeated exposures of the oesophageal mucosa to capsaicin-containing red pepper sauce ingested by patients suffering from heartburn produces an analgesic effect after an initial phase of symptom worsening, suggesting the induction of ‘sensitization’ followed by ‘desensitization’ of the oesophageal mucosa by capsaicin.23 The results of these studies prompted us to investigate whether chronic ingestion of red pepper by patients with functional dyspepsia would lead to a decrease in their symptoms.

Materials and methods

Patient selection

The study was performed on 30 patients who met the following criteria.4

  • (a) the presence of persistent or recurrent pain and/or discomfort centred in the upper abdomen associated with food intake, which compelled the patients to require a medical consultation. Discomfort was defined, according to Talley et al.,4, 5 as a subjective negative feeling that did not reach the level of pain and that included symptoms such as a feeling of post-prandial epigastric fullness, upper abdominal bloating without visible distension, epigastric burning, burping–belching, early satiety, nausea and/or vomiting.
  • (b) the persistence of symptoms for at least 3 months before the study and their occurrence at least three times a week.
  • (c) the absence of other functional gut disorders, such as gastro-oesophageal reflux disease (excluded on the basis of symptoms and, in some dubious cases, by 24-h pH-metry), irritable bowel syndrome (diagnosed on the basis of Manning's criteria) and biliary motility disorders (selected on the basis of symptoms and ultrasonography).

Only patients aged between 18 and 65 years were eligible.

The following conditions represented a contraindication for inclusion in the study: (i) organic alterations of the gut, such as hiatal hernia, oesophagitis, duodenitis, ulcers, erosions, stenosis, neoplasms, gallstones, etc.; (ii) previous surgery of the digestive tract (excluding appendectomy); (iii) history of gastric or duodenal ulcer; (iv) hepatic, cardiac, renal, neurological and systemic diseases, in which an involvement of the digestive tract was possible (diabetes, scleroderma, etc.); (v) pregnancy and lactation; (vi) alcohol intake > 20 g/day; (vii) coffee intake > 3 cups/day; (viii) smoking > 10 cigarettes/ day; (ix) drug addiction; (x) chronic use of NSAIDs and other ulcerogenic drugs, as well as drugs which may interfere with gut motility.

The patients were selected on the basis of history, physical examination, recent upper gut endoscopy–biopsy and abdominal ultrasonography, and, when necessary, biochemical tests were also performed to exclude general disease that might be responsible for dyspeptic symptoms. Informed consent was obtained from all subjects.

Drug composition and administration

Red pepper powder was obtained by grinding the entire dried fruit of Capsicum annuum, including the seeds, produced in the south of Italy. The content of capsaicin in red pepper powder was determined as follows. Capsaicin was extracted with organic solvents, filtered and analysed by means of high-pressure liquid chromatography with selective electrochemical detection.8-Methyl-N-vanillyl-6-nonenamide (capsaicin 98%), obtained from Sigma (Munich, Germany), and capsaicin 65% + dihydrocapsaicin 35%, obtained from Fluka (Buchs, Switzerland), were used as standards. The content of pure capsaicin and hydrocapsaicin was 0.7 mg/g of red pepper powder.

The red pepper powder was administered in opaque gelatine capsules, each containing 0.5 g, whilst the placebo capsules contained 0.5 g of tomato powder. After a 2-week washout period, the patients were randomized into two parallel groups to receive identical capsules of red pepper or placebo in a double-blind manner for 5 weeks. The patients were asked to take the capsules 15 min before meals for 5 weeks: one capsule before breakfast (continental), two before lunch and two before dinner. No concurrent medications that may influence dyspepsia symptoms were allowed during the study, and a 2-week washout period of prokinetic and antisecretory drugs was performed to assess the baseline symptoms in the week preceding treatment.

Assessment of symptoms

A diary sheet was given to each patient for symptom self-assessment during the baseline and treatment periods. On each diary sheet, there was a list of the days of treatment crossed with a list of the following symptoms: epigastric pain, epigastric fullness, epigastric burning, epigastric bloating, burping–belching, early satiety, nausea and/or vomiting. Symptom intensity was scored by the patients and marked on the diary each day during the baseline and treatment periods on the basis of a five-point Likert scale (1, absent; 2, slight, can be ignored when you do not think about it; 3, moderate, cannot be ignored, but does not influence daily activities; 4, strong, influences concentration on daily activities; 5, unbearable, markedly influences daily activities and/or requires rest). This scale was validated for dyspeptic symptoms by previous studies.24 The self-assessment by patients focused on individual symptoms as well as on their overall intensity. Side-effect assessment was based on spontaneous complaints reported in the diary or on response to specific questioning at the final visit. In addition, in order to ensure as high a compliance with the treatment as possible, the patients were allowed to contact the doctor by telephone for any problem related to the treatment.

Statistical study

The statistical evaluation aimed to examine the effect of both capsaicin and placebo on the overall and individual symptom intensity over the entire treatment period (primary end-point) and the trends in symptom intensity over time. The study was carried out by an investigator unaware of the meaning of the symptom variations after the administration of capsaicin and placebo. In each patient, the daily scores of the individual and overall symptom intensities were averaged over 1-week periods and the entire treatment period. The overall and individual symptom intensity scores, assessed for the entire 5-week treatment period, of the red pepper group were compared with those of the placebo group.

To assess the course of the effect of capsaicin and placebo, the mean weekly scores of the overall symptom intensity and of two ‘main symptoms’, that is symptoms which had initially been reported by the patient to be the most disabling, were plotted against time in each group, and compared between groups and with the corresponding baseline values obtained from the week preceding treatment.

Parametric (analysis of variance for repeated measurements and Student's t-test for paired data) and non-parametric (Mann–Whitney U-test) tests were used as appropriate. P values less than 0.05 were considered to be statistically significant.


Two of the 15 patients in the red pepper group discontinued treatment because of the appearance of severe epigastric pain and/or burning after the first administration of red pepper, and one patient in the placebo group did not complete the trial. Consequently, the red pepper group contained 13 patients and the placebo group 14 patients, whose characteristics are summarized in Table 1. The most frequent individual symptoms before treatment were post-prandial fullness (92%), epigastric pain (85%), nausea (63%), early satiety (52%), epigastric burning (33%), burping–belching (29%) and epigastric bloating (26%). These symptoms were equally distributed in the two groups. The symptoms most often considered as ‘main’ by the patients were epigastric pain and epigastric fullness. The mean overall symptom intensity score for the entire treatment period and the individual scores for epigastric pain, epigastric fullness and nausea were significantly (P < 0.05) lower in the red pepper group than in the placebo group (Table 2), whilst the difference for early satiety was at the borderline of significance (P < 0.062). Epigastric burning, bloating and burping–belching in the red pepper group were not significantly different from those in the placebo group. Because three patients were withdrawn before the end of the study, the intergroup differences were also tested by performing an ‘intention-to-treat’ analysis with the ‘last value available’ method. The results of this analysis confirmed that the mean overall symptom intensity, epigastric pain, fullness and nausea scores of the red pepper group were significantly lower than those of the placebo group (Table 2).

Table 1.  Characteristics of the patients at entry
CharacteristicRed pepperPlacebo
  1. Results are expressed as mean ± s.d. The number of patients with the symptom is indicated in parentheses.

No. of patients1515
Sex (males/females)7/88/7
Age (years)40.2 (range, 19–63)43.3 (range, 21–58)
Helicobacter pylori positive9/158/15
Baseline self-evaluation score
 Overall symptom intensity3.31 ± 0.643.44 ± 0.72
 Epigastric pain3.09 ± 0.74 (14)3.16 ± 0.72 (12)
 Epigastric fullness2.98 ± 0.69 (14)3.10 ± 0.61 (13)
 Nausea and vomiting2.68 ± 0.59 (10)2.79 ± 0.56 (8)
 Early satiety2.45 ± 0.64 (7)2.34 ± 0.57 (7)
 Epigastric burning2.96 ± 0.83 (5)2.61 ± 0.61 (5)
 Epigastric bloating2.67 ± 0.54 (5)2.44 ± 0.51 (3)
 Burping–belching2.28 ± 0.47 (5)2.75 ± 0.94 (4)
Table 2.  Score values (mean ± s.d.) for overall symptom intensity and single symptom intensity obtained during 5 weeks of treatment with red pepper or placebo. The results of the ‘intention-to-treat’ analysis are indicated in parentheses
  1. Asterisks indicate a statistically significant (P < 0.05) difference between red pepper group and placebo group.

Red 1.75 ± 0.19* 1.61 ± 0.25*1.65 ± 0.22* 1.60 ± 0.19* 1.76 ± 0.30 1.37 ± 0.69 2.13 ± 0.20 1.97 ± 0.5
 pepper(1.98 ± 0.64*)(1.95 ± 0.89*)(1.75 ± 0.42*)(1.74 ± 0.47*)(1.76 ± 0.30)(1.89 ± 1.30)(2.49 ± 0.86)(1.97 ± 0.50
Placebo 2.49 ± 0.28 2.37 ± 0.352.35 ± 0.29 2.00 ± 0.22 2.10 ± 0.30 2.50 ± 0.26 2.70 ± 0.50 2.37 ± 0.67
(2.57 ± 0.41)(2.51 ± 0.59)(2.40 ± 0.34)(2.00 ± 0.22)(2.10 ± 0.30)(2.50 ± 0.26)(2.70 ± 0.50)(2.77 ± 1.10)

The course of the weekly mean symptom score for the overall symptom intensity and the ‘main’ symptoms (epigastric pain and fullness) is illustrated in Figures 1–3. The intergroup statistical comparisons showed that all of these parameters in the red pepper group were significantly lower than those of the placebo group from the third week until the end of treatment. The intragroup statistical comparisons showed that the parameters of the red pepper group were significantly lower (P < 0.05) than the baseline values from the second week until the end of treatment, reaching a maximum decrease at the fifth week, whilst those of the placebo group were significantly (P < 0.05) lower than the baseline values from the first week until the end of treatment, reaching a maximum decrease at the third to fourth week. The percentage decrease in overall symptom score at the end of treatment was 60.9% in the red pepper group and 29.8% in the placebo group, that of epigastric pain was 61.0% in the red pepper group and 25.8% in the placebo group, and that of epigastric fullness was 59.7% in the red pepper group and 29.2% in the placebo group.

Figure 1.

Course of the weekly mean overall symptom intensity scores during red pepper (▵) and placebo (○) administration. Filled triangles and circles indicate values statistically different from the corresponding baseline values (B), whilst asterisks indicate a statistically significant difference between the two groups.

Figure 2.

Course of the weekly mean epigastric pain intensity scores during red pepper (▵) and placebo (○) administration. Filled triangles and circles indicate values statistically different from the corresponding baseline values (B), whilst asterisks indicate a statistically significant difference between the two groups.

Figure 3.

Course of the weekly mean epigastric fullness intensity scores during red pepper (▵) and placebo (○) administration. Filled triangles and circles indicate values statistically different from the corresponding baseline values (B), whilst asterisks indicate a statistically significant difference between the two groups.

With regard to the side-effects observed in the patients in the red pepper group, seven patients complained of abdominal pain and/or discomfort after the ingestion of capsules during the first week of treatment, but this worsening was not sufficiently severe to compel the patients to stop treatment, and was followed by progressive improvement after a few days of treatment. The other side-effects in the red pepper group were headache (one case), faintness (one case) and diarrhoea at the beginning of treatment (three cases); the side-effects in the placebo group were headache (one case), fatigue (one case) and drowsiness (one case).


The results of this study demonstrate that red pepper powder taken before meals is significantly more effective than placebo in decreasing the symptom intensity of patients with functional dyspepsia. In particular, epigastric pain, epigastric fullness and nausea showed more significant improvement after red pepper than after placebo, whilst early satiety was at the borderline of significance. The number of patients was not as high as that of other similar studies, because selection was restricted to those with functional dyspepsia alone, and did not include those with associated gastro-oesophageal reflux and irritable bowel syndrome, to avoid confusion in the interpretation of results. Consequently, we cannot generalize our findings to those dyspeptic patients in whom other gut functional disorders coexist with dyspepsia. With regard to the course of the effect on symptoms, red pepper became significantly more effective than placebo from the third week of treatment and was slower than placebo in decreasing symptom intensity, starting from the second week of treatment and reaching a decrease of about 60% with respect to baseline values at the end of the study; the effect of placebo began from the first week, reaching a decrease of less than 30% at the end of treatment. Superficially, these results appear to be more encouraging than those obtained with fedotozine in dyspeptic patients.12 Indeed, the decrease in symptom intensity score due to fedotozine was about 30% and that due to placebo was 20%, an improvement which, although significant, was considered to be too small to represent a real clinical benefit.13

No severe side-effects were observed, except for two cases, in whom red pepper intake was followed by epigastric pain that compelled the patients to stop treatment. In other patients, red pepper ingestion induced less severe epigastric pain and/or discomfort, which decreased after a few days of treatment, followed by a progressive improvement of dyspeptic symptoms. This phenomenon, which suggests ‘sensitization’ followed by ‘desensitization’, has also been observed after repeated applications of capsaicin not only on the skin, but also on the mucous membranes of the oral and nasal cavities, bladder, oesophagus and stomach;18–23, 25 it is probably related to the mechanism of action. With regard to our study, it is reasonable to assume that capsaicin, contained in the ingested red pepper powder, acted on the gastric C-type fibres of visceral nociceptive neurones in the gastric mucosa. Capsaicin-sensitive neural elements have been shown in animal stomach,26 and are presumably also present in human stomach, as can be argued from the results of previous experiments which demonstrated an exacerbation of dyspeptic symptoms after ingestion of a meal containing capsaicin.23 Approximately 80% of C-fibre afferent neurones are polymodal nociceptors activated by noxious thermal, mechanical or chemical (low pH) stimuli,27, 28 that may be involved in the genesis of dyspeptic symptoms. Capsaicin induces an increase followed by a decrease of C-type fibre activity, corresponding to an accentuation followed by an attenuation of sensitivity to pain. The mechanism is as follows. Capsaicin first binds the VR1 vanilloid receptor,29 changing the membrane permeability to Ca2+ of the primary sensorial neurones, with consequent depolarization, which releases many neuropeptides, including substance P and calcitonin gene-related peptide. Afterwards, it locks the neuronal membrane in a depolarized state that prevents subsequent depolarization and the release of neuropeptides, so that the fibres become unresponsive to nociceptive stimuli.15, 16, 28 These effects are reversible, except when the treatment is applied during the neonatal period, leading in this case to complete ablation of theC-type fibre system.28

The course of symptom intensity after treatment with red pepper in our patients suggests that the improvement of dyspeptic symptoms could be related to a desensitization effect of capsaicin, which renders the gastric nociceptive C-type fibres less responsive to the stimuli that give rise to dyspeptic symptoms. However, as antisecretory and prokinetic drugs may improve dyspeptic symptoms in some patients,5 one could advance the hypothesis that the symptomatic improvement of red pepper treatment could be due to an inhibition of acid secretion or a stimulation of gastrointestinal motility, rather than to a desensitization of visceral sensorial neurones. The literature results on the acute effect of capsaicin on gastric acid secretory activity are contrasting. Some investigators found an inhibition of basal and stimulated acid secretion, but others observed no influence on basal gastric acid secretion or even stimulation.30–32 Similar contrasting results have been reported when the effect of capsaicin on gastrointestinal motility was considered. The human studies carried out with the intragastric administration of capsaicin or red pepper showed an acceleration of gastric emptying,33 a delay25, 34, 35 or no variation.23, 36 On the basis of these results, it is reasonable to assume that the beneficial effect of capsaicin on dyspeptic symptoms cannot be ascribed to modifications in gastric secretion and/or motility, although the long-term effects of capsaicin on human gut functions are not known and require further study.

Finally, there is concern that prolonged treatment with capsaicin might induce alterations of the gastrointestinal or other systems. In addition to the fact that the neurotoxic effects of capsaicin have been demonstrated only at very high dosages37 or during the neonatal period,28 the best reply to this concern is that millions of people in the world use red pepper daily during their entire lifetime at a dose markedly higher than that used in this study without apparent side-effects. The ingestion of highly spiced meals or large amounts of red pepper did not induce acute or chronic gastroduodenal mucosal damage in nearly all endoscopic studies carried out in both normal subjects38–41 and in patients with duodenal ulcer.42 Conversely, red pepper seems to display a protective effect on the gastric mucosa.43

In conclusion, the chronic administration of red pepper was more effective than placebo in decreasing the intensity of dyspeptic symptoms, and could represent an attractive possible treatment for functional dyspepsia under the label of ‘natural’ therapy. These positive results in functional dyspepsia suggest that visceral analgesic therapy with red pepper should be investigated in other functional gut disorders, especially those characterized by increased visceral sensitivity, such as ‘irritable oesophagus’ and irritable bowel syndrome. Capsaicin, which is responsible for the visceral analgesic effect of red pepper, seems to have all the necessary requirements to play a new clinical role in dyspepsia, with possible entry into the restricted group of visceral analgesics, whilst maintaining at the same time the role of a useful pharmacological tool for investigating the function of primary sensory neurones, as it selectively impairs the activity of nociceptive C-type fibres. The main flaw of this type of analgesic therapy is the initial worsening of pain in some patients. However, this problem could be solved by using resiniferatoxin, a capsaicin analogue,44 which does not release substance P and other neuropeptides responsible for the initial worsening of symptoms, but retains the ability to desensitize C-type fibres, so giving visceral analgesia without the initial side-effects. However, to date, no studies are available to support its candidature as a visceral analgesic drug, and whilst we wait for them, we should settle for red pepper.


We acknowledge Dr A. Morselli-Labate for his help with the statistical study and CAPSOR, Bologna, for supplying red pepper powder.