Randomized, Placebo‐Controlled, Double‐Blind 8‐Week Trial on the Efficacy of A Proprietary Kiwifruit Extract on Constipation‐Predominant Irritable Bowel Syndrome

ABSTRACT

Background

Green kiwifruit ( Actinidia deliciosa var Hayward) extract improves constipation. This study aimed to determine its efficacy in patients with constipation‐predominant irritable bowel syndrome (IBS‐C).

Methods

A randomized, multicenter, double‐blind, parallel‐group, placebo‐controlled trial was conducted in 186 IBS‐C patients (Rome III criteria). Patients received either placebo or kiwifruit extract (575 mg twice daily for 4 weeks, followed by 575 mg daily for 4 weeks). Outcomes included measures of bowel movement frequency, Bristol Stool Scores, and abdominal pain and related measures (100 mm visual analog scale). The primary efficacy end point was the combined improvement of the number of complete spontaneous bowel movements and reduction of weekly average abdominal pain symptom score by at least 30% for at least half of the weeks during treatment.

Results

On kiwifruit extract, the proportion of subjects with increased frequency of spontaneous bowel movements (54% vs. 36%, p = 0.012), improved Bristol Stool Score (87 vs. 73%, p = 0.014), and abdominal pain (74% vs. 59%, p = 0.023) was greater than in controls. However, no difference was observed in the combined two‐variable primary end point (24% vs. 26%; p = 0.798). In post hoc analyses of 49 subjects with severe pain (≥ 50 mm), kiwifruit extract improved the primary end point (33% vs. 8%, p = 0.028) and normalized or maintained normal bowel actions with kiwifruit extract (44% vs. 24%, p = 0.005).

Conclusions

In patients with IBS‐C, kiwifruit extract improves bowel habits and abdominal pain. The predefined end point for the whole study population was not met because the 30% or greater improvement of pain only occurred in patients with more pain.

Trial Registration

Australian Clinical Trial Research Network (ACTRN 12613001222730)

1. Introduction

Constipation‐predominant irritable bowel syndrome (IBS) is highly prevalent in the population globally [1] and a frequent driver for healthcare consultation. The pathophysiology is as yet incompletely understood. Most likely, several mechanisms, including neuro‐immune interactions [2] and psychological comorbidities [3], are key drivers of the syndrome's symptomatology. While treatment options are limited, prokinetics, laxatives, or tricyclic antidepressants are frequently used [4]. Although various pharmacological treatments are modestly efficacious, these treatments may have adverse effects such as diarrhea and bloating [5, 6].

In a systematic review and meta‐analysis comprising seven randomized controlled trials (n = 399 participants) on the effects of fresh green kiwifruit ( Actinidia deliciosa var Hayward) or its extract on symptoms in constipated populations, constipation and abdominal discomfort was, in general, improved with efficacy similar to that of psyllium [7]. These findings were confirmed in an international RCT that compared two green kiwifruit per day and psyllium in healthy controls and patients with chronic constipation and IBS‐C [8]. The efficacy of kiwifruit in constipation is supported by its known physiological effects of increasing the water content in the small intestine and total colonic volume without change in colonic transit time [9]. Which of its multiple components, spanning fiber, prebiotics, polyphenols, and a protease complex, is or are responsible for these effects is not known.

The aim of the present study was to explore the efficacy of the green kiwifruit extract in a placebo‐controlled clinical trial on symptoms in patients with constipation‐predominant IBS. It was hypothesized that, in IBS‐C patients, the consumption of the green kiwifruit extract would significantly improve the FDA‐recommended IBS‐C clinical trial measures of frequency of complete spontaneous bowel movement (CSBM) and reduce abdominal pain.

2. Material and Methods

2.1. Study Design

A randomized, 8‐week, multicenter, double‐blind, parallel‐group, placebo‐controlled trial involving 186 patients with Rome III IBS symptoms was conducted in 13 centers in Australia and New Zealand from December 2013 to July 2015. Patients received either the encapsulated green kiwifruit extract 1150 mg twice daily for 4 weeks, followed by 4 weeks of 575 mg twice daily, or equivalent doses of placebo. Randomization was performed centrally 1:1 without blocking by means of a computer‐generated schedule, and concealed allocation was assured. Patients, trial‐center personnel, and sponsor staff were not aware of group assignments. Compliance with medication was assessed by returned pill count. Patients were considered compliant if they took ≥ 75% of doses and completed at least 4 weeks of treatment. The trial was designed, conducted, and reported in accordance with the principles of Good Clinical Practice guidelines. Before participating in the trial, patients at each center reviewed and signed an informed‐consent document that had been approved by the relevant institutional review board. The trial was conducted in accordance with the protocol. The study was registered with the Australian Clinical Trial Research Network (ACTRN 12613001222730).

2.2. Patients

Patients were recruited from site patient groups and via internet media and print advertisements. Men and women aged 18–75 years were required to fulfill Rome III criteria for IBS‐C (recurrent abdominal pain or discomfort at least 3 days/month in the last 3 months associated with two or more of the following: improvement of pain or discomfort with defecation; onset associated with a change in stool frequency; and onset associated with a change in stool form or appearance) prior to inclusion in the study. In addition to the Rome III criteria, subjects had to report hard or lumpy stools during at least 25% of bowel movements and loose watery stools at less than 25% of bowel movements. Patients were included if they had no more than 3 CSBM/week during the 2‐week run‐in period, and they also had at least a worst weekly abdominal pain score of 30 mm out of 100 mm as assessed on a visual analog scale (VAS). Patients with known relevant gastrointestinal or extra‐intestinal comorbidities were excluded. All patients with a history of kiwifruit allergy, a history of latex allergy, or recent gastrointestinal illness, nausea, vomiting, or diarrhea were excluded.

The use of drugs (prescription, over‐the‐counter, or herbal) with specific effects on bowel function was not permitted during the study. These include (but were not limited to) magnesium oxide, magnesium citrate, sodium biphosphate/sodium phosphate, methylcellulose, psyllium, docusate, magnesium hydroxide, garlic supplements, and oral aloe vera extract. Analgesic medication containing codeine or other opiate/opioid compounds was not permitted during the study. Rescue medication for pain of any origin was restricted to paracetamol and non‐steroidal anti‐inflammatory drugs. Rescue medication for constipation (bisacodyl 5–10 mg daily) was permitted but had to be recorded. Patients requiring rescue medication during the treatment phase were considered non‐responders for the week during which the rescue medication was used. The use of antidepressants for any reason (including low‐dose treatment of IBS) was permitted during the study provided that subjects were on stable doses for at least 3 months prior to screening and that no changes were initiated for the duration of the study.

2.3. Study Medication

The participants received capsules containing 575 mg of the green kiwifruit extract (Phloe, Douglas Pharmaceuticals, Auckland, New Zealand) or identically appearing placebo capsules that contained isomalt (670.8 mg per capsule), silica colloidal anhydrous (26.2 mg per capsule), vegetable‐sourced magnesium stearate (6.5 mg per capsule) and copper chlorophyll for coloring (3.8 mg per capsule). The maximum daily intake of isomalt for the placebo group was 2.683 g during the first 4 weeks of treatment, after which the dose was halved for the remaining four treatment weeks.

2.4. Assessments

Assessments of efficacy were done via a daily eDiary. Patients completed, via their smartphone or via the internet, a standardized symptom assessment. In addition to a complete description of bowel movements via the Bristol Stool Score, data regarding abdominal pain, abdominal discomfort, bloating, flatulence, distension, and urgency were collected using a 100‐point VAS. Furthermore, the IBS‐specific quality of life questionnaire (IBS‐QOL) [10], the SF‐36 Short Form general quality of life questionnaire [11, 12], and the IBS Symptom Severity Scale (IBS‐SSS) [12] were used. Safety and tolerability were assessed by adverse event reporting.

2.5. Study End Points

Based on pilot data, beneficial effects of treatment on symptoms (abdominal pain) and the number of CSBMs per week were hypothesized. FDA and EMA guidance documents (US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER), Guidance for Industry Irritable Bowel Syndrome‐Clinical Evaluation of Drugs for Treatment, May 2012; European Medicines Agency, Guidelines on the evaluation of medicinal products for the treatment of IBS, September 25, 2014) require for clinical studies in IBS an a priori defined overall responder definition. Considering the anticipated effects, the definition of a responder was as follows: completion of at least half of the weeks of study treatment and meeting both of the following criteria for at least 50% of the treatment time: (a) increased weekly stool frequency by at least one CSBM from baseline (determined by frequency during the 2 weeks between screening and randomization) and (b) decreased weekly average of abdominal pain score of at least 30% compared with baseline as self‐assessed using a 100‐mm VAS.

Secondary outcomes included abdominal pain scores, change of Bristol Stool scale, stool habits, and assessments of related symptoms, including bloating, flatulence, urgency, and distension.

2.6. Statistical Analysis

Considering treatment effects in previous studies [10] and assuming an alpha of 0.05% and 80% power in a binary analysis for a 1:1 randomization with an expected experimental treatment response rate of 42% and a placebo response rate of 24%, frequencies estimated from the results of four previously conducted randomized, placebo‐controlled, double‐blinded green kiwifruit extract clinical trials, a sample size of 176 patients was required to demonstrate an effect of therapy.

All end points and safety analyses were analyzed using the modified intention‐to‐treat (ITT) population (all subjects who were randomized and received at least one dose of the study drug). The primary end point was also analyzed using the per‐protocol population (defined as compliance with medication > 75% of doses and at least 4 weeks of treatment) as a sensitivity analysis. For all data, when appropriate, frequencies and mean values were calculated. Drop‐outs and missing data were recorded as missing data without replacement and analyzed accordingly. For the comparisons of treatment groups, chi‐square and a logistic regression model with the treatment being the classification factor and baseline CSBM average and the mean abdominal pain score as covariates were used. Analyses for secondary end points used the same logistic regression modeling technique for binary end points such as weekly CSBM response rates, response to abdominal pain, discomfort, and composite end points. For continuous end points, mixed models repeated measures analysis of covariance on the change from the baseline was performed. In these cases, the baseline value for the end point was used as the covariate, and the treatment group was the classification factor. SPSS version 23 was used for all analyses. Statistical analyses were completed by the Data Analysis Group of Datapharm Australia Pty Ltd., the Contract Research Organization engaged to coordinate this clinical trial.

3. Results

3.1. Characteristics of Study Population and Treatment Groups

Of 289 patients recruited, 12 failed screening and 89 failed to meet the selection criteria during the 2‐week run‐in period. Thus, 188 were randomized. Two patients withdrew before taking any study drug and were, therefore, excluded from the ITT population. One hundred eighty‐six patients received medication (mITT population). The per‐protocol population included 146 patients. The patient flow is shown in Figure 1. Patients treated with active medication or patients on placebo were not different with regard to gender, age, or severity of symptoms, or any other baseline indices (Table 1).

FIGURE 1.

Subject disposition from enrollment to completion, by treatment arm. “Modified intention‐to‐treat” included any participant receiving at least one dose of study treatment. “Per‐protocol population” included participants who complied with medication > 75% of doses and at least 4 weeks of treatment. The safety population included all participants receiving at least one dose of study treatment.

TABLE 1.

Demographic data from all subjects who took at least one dose of study medication (safety population).

Index		Kiwifruit extract (n = 94)	Placebo (n = 92)
Number of female participants (%)		84 (89%)	81 (87%)
Ethnicity, n (%)	Asian	5 (5%)	4 (4%)
	Caucasian	80 (85%)	81 (87%)
	Other	9 (10%)	8 (9%)
Mean age (SD), years		43.7 (12.2)	43.8 (12.4)
Mean height (SD), cm		165 (8)	165 (8)
Mean weight (SD), kg		74 (17)	71 (16)

Prior to randomization, subjects who presented with historical symptoms consistent with Rome III IBS‐C symptoms were enrolled in a 2‐week run‐in period to confirm the inclusion criteria of a minimum worst abdominal pain score of 30 mm, a maximum of 3 CSBM/week, and the ability to complete the daily e‐diary. While subjects randomized to the trial met these run‐in phase inclusion criteria, they demonstrated a wide range of average weekly abdominal pain scores, weekly spontaneous bowel movements (SBMs) and predominant stool form as self‐assessed using the Bristol Stool Chart. The relationships among SBM frequency, average abdominal pain, and weekly predominant Bristol Stool Score scores during the run‐in period showed remarkably low correlations (Spearman's r for SBM frequency by average abdominal pain = −0.02; average abdominal pain by Bristol Stool Score = −0.18; and SBM frequency by Bristol Stool Score = 0.18).

These heterogeneous bowel habit symptoms during the run‐in period delineated subjects into four distinct segments: 53 subjects (29%) presented with fewer than an average of 3 SBM/week and were classified as constipated; 25 subjects (13%) presented with at least 2 days/week with a Bristol Stool Score of 6 or 7 and were classified as diarrheal, and the remaining 108 subjects were classified as “neither” constipated nor diarrheal, with 41 subjects (22%) presenting with a low weekly average abdominal pain (≤ 30 mm on a 100‐mm VAS) and 67 subjects (36%) presenting with moderate to high weekly average abdominal pain (> 30 mm on a 100‐mm VAS).

Furthermore, the average frequencies of CSBM and SBM were unrelated, with the average weekly SBM frequencies for the run‐in period CSBM rates of 0, 1, and 2 being 5.4, 5.8, and 5.9, respectively (Figure 2).

FIGURE 2.

Number of spontaneous bowel movements (SBMs) in patients without any complete spontaneous bowel movements (CSBMs), 1 CSBM/week, and 2 or more CSBM/week during the run‐in period.

3.2. Primary Study End Point

As shown in Figure 3, similar proportions of patients met the primary end point (combination of a 30% improvement of the pain score and an improvement of bowel movements by at least 1 CSBM/week) in the active (24%) and placebo treatment groups (26%; Χ ² = 0.066, p = 0.80). In the per‐protocol analysis, the primary end point was met in 41% for the active treatment and 33% for the placebo (p = 0.39).

FIGURE 3.

Primary end point—proportion of patients with improvement in both CSBM frequency and abdominal pain reduction for at least half of the treatment period—for the total sample, the non‐diarrheal with moderate‐to‐high pain group, and the balance of the trial's subjects (*p < 0.05; chi‐square test).

3.3. Post Hoc Analysis of Primary End Point

Due to the difficulty of subjects with diarrhea meeting a constipation‐relief end point and of subjects with low abdominal pain meeting a pain‐reduction end point, a subgroup analysis of patients with an average pain score > 50 mm and non‐diarrhea during the run‐in period was performed. It revealed that the primary end point was met in the 49 such patients, with response rates of 33% in subjects consuming the green kiwifruit extract compared with 8% placebo (Χ ² = 4.838, p = 0.028, Figure 3). In contrast, for the remaining 137 patients, there was no difference with 22% for those on the kiwifruit extract compared with 32% for those on placebo responding (Χ ² = 1.677, p = 0.20).

Separately, since the bowel habit measures were heterogeneous during the run‐in period, subjects were categorized into distinct subgroups determined based on these run‐in period bowel habit characteristics. Of the 54 subjects who had an average run‐in SBM frequency of fewer than 3 SBM/week, 42% receiving kiwifruit extract and 14% placebo improved their average treatment period SBM frequency by at least 2 SBM/week and had an average treatment period SBM rate of 3 or more per week (Χ ² =4.554, p = 0.033). Of the 25 subjects in the diarrhea group, 67% on the kiwifruit extract and 46% on placebo reduced abdominal pain and resolved their diarrhea without developing constipation during treatment (Χ ² = 1.066, p = 0.302). For the 108 subjects with neither constipation nor diarrhea, 39% on the kiwifruit extract and 23% on placebo reduced pain and maintained relatively normal bowel habits (Χ ² = 3.415, p = 0.65). Combining these results for the three main run‐in symptom segments, 44% of those on the kiwifruit extract and 24% of those on the placebo met the run‐in segment‐specific treatment objectives of normalizing or maintaining normality of bowel actions (Χ ² = 8.059, p = 0.005).

3.4. Secondary End Points

The results for secondary end points are shown in Figure 4. In the group of patients on active medication, the number of SBMs improved more frequently compared with controls (54% vs. 36%, Χ ² = 6.3462, p = 0.012). Significantly more patients on active medication had improvement of the Bristol Stool Score (87% vs. 73%, Χ ² = 6.0568, p = 0.014). The proportion of patients experiencing an improvement of abdominal pain was significantly greater in the active medication group (74% vs. 59%, Χ ² = 5.2049, p = 0.023).

FIGURE 4.

Key secondary end points (all subjects; n = 186)—proportion of patients with improvement in the frequency of spontaneous bowel movements (increase by one or more per week over run‐in), proportion of patients with improvement in abdominal pain, and the proportion of patients with improvement in stool form as measured by Bristol Stool Score (BSS) (*p < 0.05; chi‐square test).

3.5. Adverse Events

Fewer subjects receiving the kiwifruit extract (67%) reported adverse events during the study compared to those receiving the placebo (81%). Events considered by the investigators as possibly or probably related to the treatment were similar in each treatment group (18% for both). The most common reported adverse events are shown in Table 2. Most adverse events in both groups were mild. Five (5.3%) subjects in the kiwifruit extract group and six (6.5%) in the placebo group had six and nine severe events, respectively (Table 3). There was one serious adverse event (megacolon) in the placebo group. Almost half of the events for subjects in both groups recovered without treatment; 4 events in the kiwifruit extract group and 2 in the placebo group resulted in temporary interruption of study treatment, while 7 and 12 events in the kiwifruit extract and placebo groups, respectively, resulted in study withdrawal. Most events in both groups were resolved, and none were fatal.

TABLE 2.

Most common treatment‐emergent adverse events with Medical Dictionary of Regulatory Activities (MedDRA) preferred terms with > 4% occurrence in either treatment group (safety population).

Preferred term	Kiwifruit extract (Phloe) (n = 94)	Placebo (n = 92)
Abdominal pain	13 (13.8) a	19 (20.4)
Headache	15 (16.0)	17 (18.3)
Abdominal distension	2 (2.1)	12 (12.9)
Migraine	5 (5.3)	11 (11.8)
Nausea	11 (11.7)	9 (9.7)
Nasopharyngitis	9 (9.6)	9 (9.7)
Back pain	8 (8.5)	5 (5.4)
Vomiting	8 (8.5)	4 (4.3)
Influenza	7 (7.4)	6 (6.5)
Diarrhea	5 (5.3)	6 (6.5)
Oropharyngeal pain	2 (2.1)	6 (6.5)
Abdominal discomfort	3 (3.2)	5 (5.4)
Abdominal pain upper	4 (4.3)	5 (5.4)
Upper respiratory tract infection	3 (3.2)	5 (5.4)
Arthralgia	1 (1.1)	4 (4.3)
Gastroenteritis	3 (3.2)	4 (4.3)
Muscle spasms	1 (1.1)	4 (4.3)
Sinusitis	3 (3.2)	4 (4.3)

^a n (%).

TABLE 3.

Severe treatment‐emergent adverse events by Medical Dictionary Regulatory Activities (MedDRA) system organ class (SOC) and preferred term (PT) in the safety population.

SOC	PT	Kiwifruit extract (Phloe) (n = 94)	Placebo (n = 92)
Any		5 (5.3) a	6 (6.5)
Gastrointestinal disorders		2 (2.1)	4 (4.3)
	Abdominal distension	0 (0.0)	2 (2.2)
	Abdominal pain	1 (1.1)	1 (1.1)
	Abdominal pain upper	1 (1.1)	0 (0.0)
	Megacolon	0 (0.0)	1 (1.1)
Infections and infestations		1 (1.1)	0 (0.0)
	Gastroenteritis	1 (1.1)	0 (0.0)
Musculoskeletal and connective tissue disorders		0 (0.0)	1 (1.1)
	Back pain	0 (0.0)	1 (1.1)
Nervous system disorders		1 (1.1)	2 (2.2)
	Headache	1 (1.1)	1 (1.1)
	Migraine	0 (0.0)	1 (1.1)
Reproductive system and breast disorders		1 (1.1)	0 (0.0)
	Ovarian cyst	1 (1.1)	0 (0.0)
Surgical and medical procedures		1 (1.1)	0 (0.0)
	Removal of internal fixation	1 (1.1)	0 (0.0)

^a n (%).

4. Discussion

The treatment with green kiwifruit extract in patients with constipation‐predominant IBS resulted in a significant improvement in the number of bowel movements and a significant reduction of the abdominal pain score. However, based on the EMA and FDA recommendations for the conduct of clinical trials in patients with functional gastrointestinal disorders, the predefined combined end point of an improvement of CSBM by at least 1 CSBM/week and a 30% reduction of baseline abdominal pain score showed no difference between active and placebo. This was likely due to the fact that, at least in this study, CSBM frequency seemed less robust a measure of constipation than SBM frequency, and a substantial proportion of patients had only moderate or mild abdominal pain symptoms. In these patients, the 30% reduction of abdominal pain was not achieved. Consequently, a post hoc analysis of patients with severe pain (mITT, n = 49) revealed significant superiority for the active treatment over placebo.

The sample size of this study and the resulting type II error may have contributed to the failure to reach the predefined efficacy end point due to limited responses of patients with mild to moderate symptoms on the one hand and the heterogeneity of run‐in period bowel habits on the other hand. Indeed, many clinical trials in patients with IBS utilizing the FDA‐ and EMA‐recommended combined outcome measures used large or even very large sample sizes, sample sizes that are likely to counteract the effects of run‐in period symptom variability from general diagnoses. Some trials recruited in excess of 2000 patients [13] or, in constipation‐predominant IBS, sample sizes exceeded 800 patients [14]. In contrast, the present study included 186 patients. Nevertheless, the subsequent subgroup analysis revealed that the predefined combined EMA/FDA end point was met in only 49 patients. In patients with mild or moderate symptoms, a ceiling effect may have occurred that prevented patients from achieving the required 30% reduction in abdominal pain. In retrospect, this could have been addressed by tighter entry criteria for the study. The variability in and heterogeneity of the run‐in period bowel habit symptoms support the conclusion that IBS‐C symptoms are in many cases erratic and hence, a diary run‐in to help exclude variability may improve assessments of experimental treatments [15, 16]. The findings also suggest that the combined EMA/FDA end point that includes a 30% reduction in pain is not suitable for patients with only moderate symptoms. Nevertheless, symptoms other than “pain” are likely highly relevant for these patients, and the response of these symptoms is of value to the patients.

Despite all participants fulfilling IBS‐C clinical diagnostic criteria, not all patients had an increase in the number of bowel movements. Indeed, some patients experienced normal or even abnormally high bowel movement frequency during the run‐in phase, a fact obscured by the use of CSBM frequency as one of the primary inclusion criteria. In our study, patients with run‐in constipation symptoms, as measured by SBM frequency, significantly increased bowel movement frequency (as discussed above); those with diarrhea tended to have a normalization of bowel movements, and those without an abnormality of frequency of SBMs during the run‐in phase did not experience any significant changes in bowel frequency. However, overall there was a stronger improvement in abdominal pain among patients on the experimental treatment than the placebo. Therefore, the active medication normalized bowel habits and improved abdominal symptoms, an effect that is quite distinct from the relatively simple objective of increasing the number of SBMs.

The physiological effects of kiwifruit were demonstrated in a magnetic resonance study of 14 healthy volunteers, where kiwifruit increased water content in the small bowel and ascending colon, with an increase in total colonic volume without change in colonic transit time [9]. Gas challenge experiments in 11 volunteers showed looser and more frequent stools without a change in gas transit [17]. The mechanisms by which kiwifruit extract might influence gut function are, however, uncertain as it is a complex mixture of potentially active components. A range of nondigestible carbohydrates such as pectic, hemicellulosic, or cellulosic polysaccharides might act via bulking and other laxative effects of non‐starch polysaccharides [18], although the quantitative contribution to total fiber intake would be very small with the kiwifruit extract. Polyphenolics and the potential prebiotic effects of indigestible carbohydrates, including oligosaccharides, might alter the gut microbiome. Indeed, kiwifruit is bifidogenic and selectively promotes butyrate‐producing bacteria such as Faecalibacterium prausnitzii [19, 20, 21, 22, 23, 24, 25, 26, 27]. At the same time, the kiwifruit extract upregulated the expression of mucin (MUC)‐2, MUC3, toll‐like receptor (TLR)‐4 or trefoil factor‐3 genes [28]. On the other hand, kiwifruit also contains water‐soluble proteins that include the proteolytic enzyme, actinidin, and its inactive forms, a so‐called thaumatin‐like protein. Actinidin is considered to be the most important enzyme in kiwifruit and studies in vitro and in animals confirm green kiwifruit and its extract can play a role in aiding the digestive process by improving the digestion of proteins [29, 30, 31]. However, these effects on human gut function have not been fully explored and their physiological role in digestion in the human gut remains to be determined. In addition, there is a wide range of enzymes that are linked to the ripening of kiwifruit. Whether the activity of these enzymes has any effect during the consumption and digestion of kiwifruit is unknown [31].

Of importance, no safety issues were identified. Indeed, treatment‐emergent adverse events were more frequent in the placebo group, although events considered possibly or probably related to the treatment itself were similar. The only issue associated with kiwifruit and its extract is the potential for allergic responses, but known or suspected kiwifruit allergy was an exclusion criterion for the present study.

The main limitations of the present study relate to the inclusion of patients who did not experience enough constipation, together with insufficient pain levels during the run‐in period, even though all patients were clinically diagnosed as having IBS‐C. This hampered the evaluation of the primary end point and necessitated post hoc subgroup analysis. Indeed, one may question the suitability of a combined end point if it does not reflect the clinical needs of the patient. Hence, the results must be considered as signals of benefit that require additional evaluation in that subgroup and cannot be generalized to all patients with IBS‐C.

In summary, this placebo‐controlled parallel‐group study in patients with clinically diagnosed constipation‐predominant IBS demonstrated significant improvement in abdominal pain and the number of bowel movements. However, for the combined end point (improvement of CSBMs and a 30% reduction of the abdominal pain score), no significant effect was observed. Interestingly, a post hoc subgroup analysis of patients with severe symptoms demonstrated that active therapy was efficacious even with regard to the combined outcome parameter. Thus, the data suggest that this green kiwifruit extract may be efficacious with regard to the improvement of pain and bowel habits in patients with IBS‐C.

Ethics Statement

The studies were approved by the relevant institutional review boards.

Consent

Prior to entry in the studies, patients signed written informed consent.

Conflicts of Interest

G.H.: Advisory board member for Australian Biotherapeutics, Glutagen, Bayer; received research support from Bayer, Abbott, Pfizer, Janssen, Takeda, Allergan. He serves on the Boards of the West Moreton Hospital and Health Service (WMHHS), Queensland, UQ Healthcare, Brisbane, and the Gastro‐Liga, Germany, and is Chair of the WMHHS Board Quality and Safety Committee. He has a patent for the Brisbane aseptic biopsy device and serves as Editor of the Gastro‐Liga Newsletter. N.J.T.: GutSee Ltd. (consulting microbiome, 2025), Brown University & Agency for Health Care Research and Quality (fiber and laxation) (2024), Biocodex (FD diagnostic tool) (present), Microba (consulting microbiome, 2025), Comvita Manuka Honey (FD trial consulting) (2025), BluMaiden (2025) outside the submitted work. In addition, Dr. Talley has a patent Nepean Dyspepsia Index (NDI) 1998, a patent Licensing Questionnaires Talley Bowel Disease Questionnaire licensed to Mayo/Talley, “Diagnostic marker for functional gastrointestinal disorders” Australian Provisional Patent Application 2 021 901 692, “Methods and compositions for treating age‐related neurodegenerative disease associated with dysbiosis” US Patent Application No. 63/537725. S.N.: Speaker Honoraria with Falk Pharma. P.R.G.: consultant or advisory board member for Anatara, Atmo Biosciences, Topas, and Comvita; research grants for investigator‐driven studies from Atmo Biosciences and Mindset Health, and speaker honoraria from Dr. Falk Pharma and Mindset Health Pty Ltd. Shareholder in Atmo Biosciences.

Holtmann G., Talley N. J., Nandurkar S., and Gibson P. R., “Randomized, Placebo‐Controlled, Double‐Blind 8‐Week Trial on the Efficacy of A Proprietary Kiwifruit Extract on Constipation‐Predominant Irritable Bowel Syndrome,” JGH Open 9, no. 8 (2025): e70250, 10.1002/jgh3.70250.

Data Availability Statement

The data and full protocol underlying this article will be shared on reasonable request to the corresponding author.