Introduction

Humans have used foods medicinally for thousands of years. The Greek physician Hippocrates, considered to be the father of modern Western medicine, is often attributed the quote; “Let food be your medicine, and medicine be your food.” He and his followers pioneered the concept that diseases were caused by natural phenomena and often used food as treatments. Traditional Chinese medicine dating as far back as the West Zhou dynasty (circa 1000 BCE) used food and nutrition as treatments [1]. Medical texts from the Tang dynasty in China (circa 618 CE) provide examples of food-based treatments, including a type of rice porridge used to treat beriberi and pork liver for night blindness [1].

Even today, using medicinal plants is common practice among a large percentage of the world’s population [2]. In this context, it is little wonder that many pharmaceutical products have their origins in food substances. Researchers continue to find inspiration for new drugs in ingredients isolated from food. Curcumin from turmeric, resveratrol from grapes, and epigallocatechin-3-gallate from green tea have been studied extensively for their effects on diseases such as cancer and cardiac disease [3,4,5,6]. More recently, avocatin B, a lipid derived from avocados, has been shown to reduce human primary acute myeloid leukemia cell viability without affecting normal peripheral blood stem cells [7]. A modern day medical food, in the contemporary sense of the term, inhabits a vague therapeutic space between food and prescription drugs.

Functional gastrointestinal disorders (FGIDs) are a heterogeneous group of chronic illnesses that are defined by the Rome Committee based on their symptomatology [8••]. The recently updated Rome IV outlines both criteria and the classification of 33 recognized adult and 20 pediatric FGIDs [8••]. These conditions are common and have a major impact on society in terms of impact on quality of life (QOL), health care expenditure, and lost productivity. Overall, FGIDs account for the most common gastrointestinal complaints among ambulatory patients seen by gastroenterologists and primary care providers and affect roughly 1 in 4 individuals [9]. Research advances over the last three decades have assisted in accurately and confidently diagnosing patients with FGIDs, and there have been commensurate advances in the therapeutics, with numerous FDA-approved medications available for the treatment of these disorders. Most of these therapies are directed towards functional dyspepsia (FD) and irritable bowel syndrome (IBS), representing two of the most common FGIDs [10, 11], and are typically symptom oriented.

A significant proportion of patients with FGIDs turn to self-directed care and use of so-called alternative medicines. There may be several factors at play that make individuals seek these treatments in lieu of prescription medications, including the ease of obtaining over the counter therapies, the high costs of physician office visits or limitations associated with insurance status, the abundance of internet sites promoting self-diagnosis and treatment, or a general distrust of the medical system. The following review will examine the background surrounding the class of therapies known as medical foods and the published evidence for these therapies as treatments for common FGIDs.

Medical Foods Regulatory Landscape

The interested reader is directed to an excellent summary by Burnett and Levy describing the regulatory evolution of medical foods since the inception of the FDA in 1938 to the present [12••]. The first “medical food” was an infant formula approved by the FDA in 1957 to replace phenylalanine in patients with phenylketonuria. The development of the global regulation of foods for use as medicine began in the early 1960s by two United Nations organizations. In 1963, the Food and Agriculture Organization (FAO) and the World Health Organization (WHO) established the Codex Alimentarius, a set of standards and guidelines for food production. These guidelines paved the way for what would eventually become (among other things) “medical foods” in the USA, “foods for special medical purposes” in the European Union and Australia, and “foods for special dietary use” in Canada.

To encourage innovation, an amendment was added to the 1988 Orphan Drug Act update that established medical foods as a separate therapeutic category, distinct from FDA-approved prescription drugs as well as foods for special dietary use [13]. In this legislation, a medical food was defined as “a food which is formulated to be consumed or administered internally under the supervision of a physician and which is intended for specific dietary management of a disease or condition for which distinctive nutritional requirements based on scientific principles, are established by medical evaluation.” In the Federal Register of January 6, 1993, the FDA incorporated this statutory definition of medical foods in its regulations at “Section 101.9(j)” [14]. Further, a medical food was defined to encompass that “it is intended for the dietary management of a patient who, because of therapeutic or chronic medical needs, has limited or impaired capacity to ingest, digest, absorb, or metabolize ordinary foodstuffs or certain nutrients, or who has other special medically determined nutrient requirements, the dietary management of which cannot be achieved by the modification of the normal diet alone.” The FDA also included several additional mandatory definitional requirements around medical foods in this regulation that governed the labeling requirements for these therapies.

Over the past two decades, the FDA has attempted to further clarify its position regarding medical foods by issuing a series of “Frequently Asked Questions (FAQs) About Medical Foods,” most recently in May 2016. These documents emphasized that medical foods do not require regulatory pre-approval but remain subject to FDA oversight. To be considered a medical food, a product must comply with FDA regulations and labeling requirements, be comprised of ingredients that are generally recognized as safe (GRAS), and be specially formulated and processed (as opposed to naturally occurring). Importantly, although medical foods do not require a prescription, they should be used under medical supervision.

The recent FAQs have also created confusion and controversy in this field. Specific controversies include the lack of FDA approval for medical foods, leading to (1) minimal approval or awareness by regulators or medical professionals regarding the use of medical foods as chronic therapies, (2) lack of standards for clinical trials of medical foods, (3) lack of a formal definition of what constitutes a “nutritional requirement,” and (4) lack of consensus/complete understanding of the requirement for physician supervision. To further complicate matters, some payers have begun to interpret the express requirement of the “non-prescription” status of medical foods to imply that they are over the counter (OTC) and thus not subject to coverage. This prevalent uncertainty with regard to medical foods has created significant challenges in the development and commercial viability of some agents in this class.

The need to help clinicians expand their armamentarium beyond conventional prescription drugs is becoming ever more apparent. Due primarily to problems with restricted access and affordability of prescription drugs, only about half of the people who suffer from chronic conditions adhere to treatment recommendations [15]. In the USA, over half the enrollees in commercial plans are now exposed to high deductibles and co-insurance compared to about 25% as recently as 5 years ago. This further affects patient adherence which, in the USA, has now sunk to about 25% only 6 months after the first filling of higher-priced prescription drugs [16]. By providing options that are relatively inexpensive, safe, and effective, medical foods can be seen as important complements to these therapies.

Medical Foods for FGIDs

Irritable Bowel Syndrome

Peppermint Oil

The first known published work regarding the medicinal use of the peppermint plant was in 1753 by Carl Linnaeus [17]. Since then, the use of peppermint as a digestive aid has been widely studied for a variety of conditions, including IBS and dyspepsia. Peppermint oil (PO) and its active ingredient, L-menthol, are known to provide smooth muscle calcium channel antagonism [18], normalization of orocecal transit time [19], carminative effects [20], kappa opioid agonist activity [21], anti-infective [22] and anti-inflammatory [23] effects, and serotonergic (5-HT3) antagonism [24].

Enteric-coated formulations of PO for medicinal use were first developed in the 1970s; however, adverse events were frequently encountered. Heartburn and nausea were commonly reported, presumably due to premature rupture of PO capsules in the stomach [25]. Recently, a new formulation of PO has been made available OTC as a medical food. IBgard® utilizes site-specific technology (SST) to predictably deliver a sustained release of PO over a 4-h period to the small bowel [26••]. It is triple coated and has an enteric polymer middle coat, allowing IBgard® to remain insoluble at the gastric pH and dissolve in the more basic intestinal environment. This novel delivery system is thought to decrease the likelihood of treatment-related side effects and improve tolerability with this preparation.

The Irritable Bowel Syndrome Reduction Evaluation and Safety Trial (IBSREST) was the first clinical trial conducted that compared the efficacy of the PO SST formulation with placebo (Table 1). Peppermint oil was significantly more effective than placebo at multiple time points in this trial. Twenty-four hours after the first dose, patients in the PO SST arm reported an 18.8% reduction in the total IBS symptom score (TISS) from baseline, compared with a 9.8% reduction reported by patients in the placebo arm—a statistically significant difference of nearly twice the rate of global symptom improvement (P = 0.0092). For individual symptoms at 24 h, there was a trend towards a greater improvement with PO SST vs placebo in all eight of the primary IBS symptoms measured by the TISS. This difference reached statistical significance for abdominal pain or discomfort, with patients reporting a 21% reduction from baseline vs a 10% reduction with placebo (P < .05). Patients in the PO SST arm also had a significantly greater reduction in the intensity of bowel movement urgency at 24 h compared with patients in the placebo arm (25 vs 6%; P = .0374). At 4 weeks, there was a greater separation in the reduction of individual IBS symptoms between PO SST and placebo, and several of these differences were statistically significant. The primary endpoint of the study, reduction in the TISS at 4 weeks, was significantly greater with PO SST vs placebo (40 vs 25%; P = .0246).

Table 1 Clinical trials with medical foods for FGIDs (US trials)
Full size table

The safety analysis from the IBSREST demonstrated that PO SST was well tolerated. The incidence of treatment-emergent adverse events (TEAEs) as 5.7% in the PO SST arm compared with 10.8% in the placebo arm. Specific adverse events (AEs) reported in the PO SST arm included one episode of dyspepsia (2.9%) and one upper respiratory tract infection (2.9%) that was unlikely associated with therapy. In the placebo arm, AEs included flatulence, gastroesophageal reflux disease symptoms, viral gastroenteritis, upper respiratory tract infection, and back pain, reported in one patient each (2.7%). There were no discontinuations due to TEAEs in either arm, and no serious TEAEs or deaths occurred during the study.

Serum-Derived Bovine Immunoglobulin/Protein Isolate (EnteraGam®)

Serum-derived bovine immunoglobulin/protein isolate (SBI) is an oral prescription medical food which has been studied in HIV-associated enteropathy, inflammatory bowel disease (IBD), and IBS-D. It is comprised of a beef protein powder containing > 50% IgG and other proteins similar to colostrum/milk and is free of milk, soy, and gluten products. A pilot study evaluating the impact of SBI on GI symptoms and QOL in patients with IBS-D was published in 2013 by Wilson et al. (Table 1). This was a randomized double-blind placebo-controlled trial (RCT) of 66 patients diagnosed with IBS-D (Rome II) who were given 5 g/day SBI, 10 g/day SBI, or placebo for 6 weeks. It was observed that subjects treated with 10 g/day SBI had statistically significant within-group reductions in abdominal pain, stool form, bloating, flatulence, urgency, and any symptom at the end of the trial vs baseline [27]. Due to lack of statistical power, comparison between groups could not be made. While SBI was well tolerated, there were equivocal effects on QOL in the patients randomized to this therapy and the high drop-out rate of 35% observed in this trial further limits conclusions.

Recently, Valentin et al. published an open-label study regarding the effects of 5 g SBI BID on 15 patients with IBS-D (Table 1). Their aim was to evaluate the potential mechanisms of action of SBI as well as to evaluate efficacy. Although patients did report improved bowel function (frequency, ease of passage, evacuation) and abdominal pain after 8 weeks of therapy, there were no demonstrable changes in stool microbiome, bile acid metabolism, intestinal permeability, or gene expression (90 genes related to inflammation, immune function, and tight junction integrity) [28]. Thus, validation of the possible mechanisms of action of SBI remains elusive.

VSL #3

VSL#3 is the only probiotic classified as a medical food. It consists of eight separate strains of live organisms: four Lactobacillus, three Bifidobacterium, and one Streptococcus. Since the bacteria in VSL#3 are live, it does require refrigeration. VSL#3 has been most studied in the dietary management of patients with ulcerative colitis, ileal pouch, and IBS. We will focus our discussion of VSL#3 on the published literature pertaining to this therapy in patients with IBS.

Some of the data regarding VSL#3 in IBS dates back to the early 2000s. A study by Brigidi et al. in 2001 describes its effects on the composition and enzymatic activity of the human fecal microbiota in patients with IBS (Table 1). This clinical trial, while small, concluded that consumption of VSL#3 provided clinical improvement in 9/10 study patients [29]. It also postulated that there was a change in the fecal microflora of these patients as analyzed by strain-specific PCR detection. This was the first study to suggest a possible transient change in the composition of the fecal microbiota following ingestion of live bacteria. A subsequent RCT in 2003 by Kim et al. included 25 patients with IBS-D who were randomly assigned to either VSL#3 or placebo BID for 8 weeks [30] (Table 1). Although there was a trend towards improved bloating in the treatment arm, however, there were no changes in colonic transit times or symptom scores with VSL#3 compared to placebo. These investigators later published the results of a 48 patient double-blinded placebo-controlled RCT in 2005 that compared VSL#3 (in yogurt) vs placebo (Table 1). This study found that patients treated with VSL#3 had reduced flatulence scores as well as slower colonic transit; however, there were no statistically significant changes in bloating, stool-related symptoms, or abdominal pain [31]. Although the bloating severity scores were not different in the two groups, the proportion of responders was 14% higher with VSL#3.

There are several possible mechanisms by which VSL#3 may function. Mennigen et al. studied the effect of VSL#3 on the epithelial barrier in a murine model of colitis. Mice were treated with VSL#3 or placebo during colitis-induction with dextran sodium sulfate. They concluded that the VSL#3-treated mice had a better-protected epithelial barrier through preserved tight junction protein expression and decreasing apoptosis [32]. The first study to describe a culture-independent molecular approach of the effect of VSL#3 on fecal microbiota was done by Michail et al. in 2011. They conducted a 24 patient double-blind placebo RCT for 8 weeks and concluded that VSL#3 did not change the gut microbiota [33]. Another recent analysis, by Wong et al., concluded that patients treated with VSL#3 who had improved abdominal pain and distension also had significantly higher salivary melatonin levels, introducing another possible mechanism of action of this medical food [34].

Other Probiotics, Prebiotics, and Synbiotics

Although these supplements are not “medical foods” per se, we have dedicated a small section regarding their use. In 2001, the Food and Agriculture Organization of the United Nations and the World Health Organization defined probiotics as live microorganisms, which, if administered in an adequate amount, confer a health benefit to the host. The science behind using probiotics for the treatment of FGIDs comes from the wealth of research aimed at decoding the distinct differences in each person’s gut flora. They are an attractive option in establishing, or re-establishing, a healthy gastrointestinal environment through their purported ability to enhance mucosal barrier protection, improve immune and anti-inflammatory effects, and competitively inhibit overgrowth of less desirable bacterial species [35].

Evidence supporting the use of probiotics in IBS does exist and multiple meta-analyses and systematic reviews suggest benefit. One recent study reported a positive impact on global IBS symptoms such as pain, bloating, and flatulence resulting in a more consistent GI lifestyle [36]. These analyses do, however, have several limitations. There is significant heterogeneity among individual/aggregate probiotic studies, and there is also the question of the appropriateness of consolidating clinical trials using different strains of bacteria into one analysis. Overall, individual study quality for probiotic trials is poor, with less than 50% meeting quality criteria to be included in rigorous meta-analytics (small, short, unclear entry criteria/endpoints). In addition, the overwhelming majority of studies conducted with probiotics have only included the IBS-D patient population.

Prebiotics are non-digestible, fermentable dietary components that are thought to confer a benefit to the host by altering the composition or activity of the gut microbiota. Examples of prebiotics include the constituents of the low fermentable oligo-, di-, monosaccharide and polyol (FODMAP) diet. There are few studies in the IBS population regarding prebiotics, and these have conflicting results. Fructo-oligosaccharides (FOS) and trans-galacto-oligosaccharide improved individual symptoms in two RCTs; whereas oligofructose, FOS, and high fermentable carbohydrate diets all worsened symptoms in subjects in three other RCTs [37].

Synbiotics are dietary supplements containing both prebiotics and probiotics in an effort to increase levels and activity of beneficial microbes in the gut. Curro et al. reviewed three RCTs with different combinations of synbiotics (Lactobacillus spp. + Bifidobacter + phytoextracts, Bacillus coagulans + FOS, and Cellulose + L-leucine + 29 probiotic spp.) and concluded that the data for synbiotics is still too limited to draw strong conclusions regarding their effects in patients with IBS [38].

Functional Dyspepsia

Caraway Seed Oil and Peppermint Oil

Caraway seed oil combined with PO has been widely studied for digestive symptoms, and there is compelling data published on its beneficial effects in patients with FD. An article published over 17 years ago by May et al. showed that fixed-dose enteric-coated caraway seed/PO in this patient population was both well tolerated and effective in improving symptoms (heaviness and fullness) with a 43% reduction in the treatment arm vs 22% for placebo [39]. This finding was expounded upon in systematic reviews by Coon et al. in 2002 and Shams et al. in 2015. Both of these reviews included several placebo-controlled RCTs which demonstrated a caraway seed/PO combination to be statistically significantly more effective than placebo for reducing FD symptoms [40, 41]. Most recently, Chey et al. presented data at the 2017 Digestive Disease Week from the FDREST study. This is an RCT comparing the efficacy and tolerability of a fixed combination of caraway seed/PO SST (FDgard®) and usual medications vs placebo and usual medications. At day 28, 61.2% of treated patients reported improvement of symptoms with the caraway seed/PO combination vs only 48.9% of controls, although the difference was not statistically significant [42].

STW 5 (Iberogast®)

STW-5 is a proprietary blend of nine botanical extracts. Aside from Iberis amara, from whence its name originates, it also contains caraway, chamomile, peppermint, lemon balm, liquorice, angelica, greater celandine, and milk thistle. Although it is not available in the USA, this phytomedicine has been commonly used throughout Europe for several decades. It is marketed towards the treatment of a variety of functional digestive symptoms, including FD and motility-related disorders.

The exact mechanisms of STW-5 remain unknown, and it is unclear whether each extract included in the preparation acts independently of or synergistically with the other extracts. STW-5 has been shown to affect acetylcholine, opioid, and adenosine receptors in the gastrointestinal tract and is thought to reduce visceral afferent nerve firing. These effects are largely inhibitory, although STW-5 has also been shown to have a pro-secretory effect on intestinal chloride channels and CFTR [43]. Another study demonstrated that STW-5 altered intestinal smooth muscle cell resting membrane potential resulting in a slow wave pattern [44].

The data surrounding the clinical use of STW-5 has been widely published and included in several meta-analyses. For FD, there are five studies of STW-5; four were RCT vs placebo and the fifth evaluated STW-5 vs cisapride. One double-blind, placebo-controlled RCT of 315 patients was performed by von Arnim et al. [45••] Participants received either STW-5 or placebo for 6 months with the primary endpoint measured as the change in a validated gastrointestinal symptom scale (GIS). The authors reported a statistically significant improvement in the STW-5 arm regarding the provision of greater symptom-free days compared to placebo. They also found STW-5 to be well tolerated with similar rates of AEs compared to placebo and no serious AEs overall. These findings were confirmed in other studies comparing STW-5 to placebo with statistically significant results [46]. Rosch et al. designed a double-dummy clinical trial to evaluate the non-inferiority of STW-5 vs cisapride [47]. This study included 186 patients with FD randomized to receive either STW-5 or cisapride over a 4-week period. Participants were followed for a total of 6 months following treatment. The primary endpoint was the change in GIS during treatment. It was found that scores improved from baseline in STW-5 and cisapride-treated patients to a similar degree, confirming non-inferiority of STW-5; however, there were numerically more AEs with cisapride than STW-5 (33 vs 21%).

Madisch et al. evaluated STW-5 for the treatment of IBS in a multicenter placebo-controlled RCT of 208 patients with IBS [48]. Participants received STW-5, two research compounds, or placebo over a 4-week period. The primary endpoint was the recorded change in participants’ abdominal symptom and pain profile, using a score developed by an expert GI panel. They reported the STW-5 treatment arm to have both clinical and statistical superiority to placebo. Participants were also found to have good tolerability of STW-5 without any serious AEs.

Conclusion

The regulatory environment surrounding medical foods is complex. Nevertheless, these therapies represent an attractive option for many patients suffering from FGIDs. While several of the available medical foods marketed in the FGID space have pharmacodynamic and physiologic data to support their effects, level 1 clinical trial data surrounding medical foods for IBS or FD is largely lacking. This may largely be due to the regulatory restrictions and definitions that are currently in place for this class of therapies, and should not lead to the conclusion that these therapies are ineffective for IBS or FD. Based on their favorable safety profiles and accessibility, the evidence supporting many of the medical food preparations available in the USA and Europe discussed above may be substantial enough to warrant consideration by clinicians and patients. Currently in the USA, proof of clinical efficacy is not strictly required for medical foods, and if the last decade is any prediction on what the future holds, we should expect to see a substantial increase in this market over the next 5 to 10 years.