Adverse Effects of Plant Food Supplements Self-Reported by Consumers in the PlantLIBRA Survey Involving Six European Countries

Background The use of food supplements containing botanicals is increasing in European markets. Although intended to maintain the health status, several cases of adverse effects to Plant Food Supplements (PFS) have been described. Objectives To describe the self-reported adverse effects collected during the European PlantLIBRA PFS Consumer Survey 2011–2012, with a critical evaluation of the plausibility of the symptomatology reported using data from the literature and from the PlantLIBRA Poisons Centers' survey. Subjects/Setting From the total sample of 2359 consumers involved in the consumers' survey, 82 subjects reported adverse effects due to a total of 87 PFS. Results Cases were self-reported, therefore causality was not classified on the basis of clinical evidence, but by using the frequency/strength of adverse effects described in scientific papers: 52 out of 87 cases were defined as possible (59.8%) and 4 as probable (4.6%). Considering the most frequently cited botanicals, eight cases were due to Valeriana officinalis (garden valerian); seven to Camellia sinensis (tea); six to Ginkgo biloba (Maidenhair tree) and Paullinia cupana (guarana). Most adverse events related to the gastrointestinal tract, nervous and cardiovascular systems. Conclusions Comparing the data from this study with those published in scientific papers and obtained by the PlantLIBRA Poisons Centers' survey, some important conclusions can be drawn: severe adverse effects to PFS are quite rare, although mild or moderate adverse symptoms can be present. Data reported in this paper can help health professionals (and in particular family doctors) to become aware of possible new problems associated with the increasing use of food supplements containing botanicals.


Introduction
The European Union (EU) Directive on Food Supplements (2002/46/EC) defines food supplements (which include PFS): ''. . .foodstuffs the purpose of which is to supplement the normal diet and which are concentrated sources of nutrients or other substances with a nutritional or physiological effect, alone or in combination, marketed in dose form, namely forms such as capsules, pastilles, tablets, pills and other similar forms, sachets of powder, ampoules of liquids, drop dispensing bottles and other similar forms of liquids and powders designed to be taken in measured small quantities". Their market is growing significantly both in Europe and the USA [1]. Although there exists some overlap/confusion with traditional herbal medicinal products [2], plant food supplements cannot be sold as having any diagnostic, preventative or therapeutic properties; their role is only complementary to the diet.
The consumption of Plant Food Supplements (PFS) is usually estimated on the basis of market data, and mainly from import/export of raw ingredients, but since botanicals are used in both food and medicinal areas, the extrapolation to PFS is quite difficult [3]. Data on the use of dietary supplements reported by consumers are very limited and normally include only those products containing vitamins and minerals [4]; other available data come from studies relating to complementary/traditional medicine [5].
To provide new data on PFS usage patterns, a survey was performed with consumers of PFS in the framework of the European Project PlantLIBRA (n. 249159); it involved 2359 adults from Finland, Germany, Italy, Romania, Spain and UK. The main results of the PlantLIBRA PFS Consumer Survey (2011-2012) were published in 2014 [6]; the present paper deals with the adverse effects self-reported by the consumers participating in the survey.
Adverse effects to PFS have been reported by several authors; most of the studies were: a) case reports describing a specific acute event, or b) reviews of cases in a specific clinical area (cardiovascular, gastrointestinal, etc.) [7][8]. A critical limitation of the information reported to date in the scientific literature is a lack of assessment of causality; in other words, the strict association between the intake of a specific plant and the clinical event is rarely demonstrated by measuring biomarkers or by the de-challenge/re-challenge approach. On this basis, a systematic review of the data on adverse effects due to PFS/botanical ingredients, including misidentification and interactions of PFS/botanicals with pharmaceutical drugs or nutrients was undertaken [9]. Data were collected for 66 botanicals, which are common ingredients of PFS; all papers were classified for causality according to the WHO guidelines [10] and grouped as "certain, probable, possible and uncertain/unclassifiable events". Among the 492 papers selected, 402 (81.7%) dealt with adverse effects due to the botanical as such or in a PFS, and 89 (18.1%) described interactions with conventional drugs. Misidentification was confirmed in one case [9].
The aims of this paper are: 1) to identify the adverse effects reported by the European participants in the PlantLIBRA PFS Consumer Survey, and 2) to critically evaluate the plausibility of the symptomatology reported as being related to PFS.

Materials and Methods
The survey was conducted in 6 European countries (Finland, Germany, Italy, Romania, Spain and the United Kingdom), and recruitment of participants occurred in 4 cities in each country. In this study, "Botanical" means raw material and derived preparations made from plants, algae, fungi or lichens (http://www.efsa.europa.eu/en/topics/topic/botanicals). The botanicals to be included in the survey were clearly defined at the outset; PFS were defined as the "foodstuffs the purpose of which is to supplement the normal diet and which are concentrated sources of botanical preparations that have nutritional or physiological effect, alone or in combination with vitamins, minerals and other substances which are not plant-based". Herbal remedies, other medicinal products based on botanicals, herbal teas or juices were excluded [6].
In order to obtain a sample of 400 consumers/each country, approximately 2000 individuals were screened per country (total number close to 2400) [6]. Eligible consumers completed a detailed questionnaire on PFS usage, providing product/plant names, dosage forms, frequency of use, reasons for use, adverse effects, places and patterns of purchase and information sources on products. Data on a maximum of five different PFS for each consumer was recorded; when PFS were more than 5, the inclusion was based on the frequency of use. Responders' sociodemographic data, including age, gender, level of education and employment status, as well as height, weight and health-related lifestyle information, were also collected. Further details on the survey have been reported previously [6]. The composition of each PFS was obtained from the label, if at disposal, or by searching the PFS ingredients in the website of producers.
Regarding the collection of data on adverse effects, the following two questions were included for each product: 1. Have you experienced any adverse effects while taking this product?
2. If yes, which one? (list of symptoms provided, with "other" as an option). In all countries, informed consent was obtained from survey participants verbally after reading the survey information sheet. The data were collected anonymously on paper questionnaires and then transferred to an electronic database; all responders were assigned an ID number prior to data analyses.

Statistical analysis
All data were entered into the statistical package SPSS for Windows v. 18 (IBM Corporation, Somers, NY, USA), which was used for analysis.

Results and Discussion
A total of 11783 consumers (5799 males and 6004 females) were screened during the PlantLI-BRA survey, of which 2359 were considered eligible and included in the study. The number of consumers per country and the percentage self-reporting adverse effects are listed in Table 1.
Considering the entire survey, the percentage of consumers, who reported adverse effects was approximately 3.5%. Differences were observed between countries; the number of consumers reporting adverse effects ranged between 5 and 6% of the total interviewed in three countries (Finland, Germany and Spain), while they were less numerous in Romania (2%), Italy (1%) and the United Kingdom (0.3%).
There were no significant differences based on sex or age groups (Table 1 and Table 2). The data collected on adverse effects are presented in Table 3. Details of each of the 82 cases are recorded including: • ID of the 82 consumers reporting adverse effects. ID 1-23 were from Finland, ID 24-45 from Germany, ID 46-69 from Spain, ID 70-74 from Italy, ID 75-81 from Romania, ID 82 from the United Kingdom. When a consumer reported adverse effects for two PFS, letters A and B follows the ID; • Age and gender; • The botanical/s contained in the PFS associated with the reported adverse effect. For a more precise identification, Latin names have been used but the corresponding common names are reported in Table 4 (see table note for scientific sources); • The daily dose and the period of intake; • The reason for use reported by the consumer, i.e. the physiological effect expected by the consumer; • An assessment of the suitability of the botanical ingredient(s) present in the consumed PFS for the condition used, based on what the literature says about these ingredients. The main literature sources were the list of physiological effects published by the Italian Ministry of Health [11] and the EMA website [12]. In specific cases, other scientific papers were cited; • The general health status of the consumer. These data allow a better assessment of adverse effect causality; • Any reported simultaneous intake of conventional medicines and other food supplements; these data allow the assessment of possible interactions; • The adverse effects reported by the consumers; • A judgement as to the likelihood of causality, according to previous scientific citations and taking into consideration all available data.
Considering the suitability of the botanical product used in relation to the physiological effect expected by the consumers, the choice was considered appropriate in 88% of cases. One case (ID 4) was considered non pertinent and nine (ID 8,11,12,18,23,33,44,57,74) judged as unlikely or with limited evidence. One consumer (ID 08) used a product containing 12 herbal ingredients, but only Citrus aurantium could claim to have tonic properties due to the presence of active amines. None of the herbal ingredients present have any reported immune activity [11][12].
Since all cases were self-reported, it was not possible to establish causality of adverse effects on the basis of clinical evidence. The scientific literature was used to assess the likelihood of the       adverse effects being associated with the botanical used and 56 out of 87 (64%) cases were defined as possible (52) or probable (4) according to 1) the daily dose and period of intake, and 2) the frequency and strength of scientific evidence. The most significant references are reported in Table 3. The association was not confirmed for 28 cases. The interaction with conventional drugs was considered possible in three cases (ID 01, 15 and 17B). It is important to underline that comparing the list of conventional drugs used with the column "present or past main health problems", there are several incongruences; this is due to the self-reported nature of information collected. The frequency of self-reported adverse effects in relation to each botanical is reported in Table 4; the total number of botanical ingredients contained in PFS with reported adverse effects was 72 and the total counts were 144. In most cases (46%), the PFS involved contained one ingredient. Forty botanicals (55.6% of the total) were associated with a single adverse event and 80% of them were included in PFS containing two or more ingredients. Considering the  most prevalent botanicals associated with adverse effects, 14 were associated with 68 reported adverse effects, representing the 47.2% of the total counts. In particular, eight were due to Valeriana officinalis (seven of them in Spain), seven to Camellia sinensis, six to Ginkgo biloba and six to Paullinia cupana. The association of adverse effects with different organ systems is listed in Table 5.  Approximately 60% of adverse events were related to the gastrointestinal tract, distributed between gastric problems (where nausea was the most reported symptom) and intestinal effects (mainly diarrhoea).
The nervous system was the second most affected area with nine cases of insomnia, four of dizziness and two of migraine; the cardiovascular system was reported in four cases of tachycardia. The stimulating effects of botanicals containing caffeine can explain the insomnia and tachycardia reported by consumers 13, 46, 55, and 58, but in other cases, the effects were unexpected. An example is the case of insomnia due to Valeriana officinalis (ID 49 and ID 64), which is the opposite of the usual physiological effect, associated with this botanical [11][12]. Ginkgo biloba was involved in three cases of insomnia and one of dizziness.
Hair and skin were affected in three cases. A case of hair loss was reported by consumer 10, taking Equisetum arvense 3 units/day for 12 months. This adverse effect could be associated with the reported effect of E. arvense in reducing the bioavailability of thiamine after chronic consumption [50]. On the other hand, the role of thiamine deficiency in hair loss has been hypothesised but insufficiently documented [79]. Hepatotoxicity, defined as an increased level of liver enzymes, was reported by two consumers using red rice (Oryza sativa fermented by the fungus Monascus purpureus). Red rice is widely used in mild hypercholesterolemia, as a "natural" alternative to statins. Several side effects have been described in consumers using this ingredient, such as headache, dizziness, heartburn, gas and digestive tract discomfort, and it should be used cautiously by people suffering from liver disease and those at risk of it [52][53]. Allergies to Allium sativum and to a PFS containing Melissa officinalis and Olea europaea were reported by two consumers (ID 62 and ID 21, respectively).
A difficulty in swallowing was reported by a consumer (ID 19), using red rice to reduce blood cholesterol. Even though this effect has not been previously associated with red rice, there are some reports concerning the statins (having similar biological activity) for which the impaired swallowing was considered among possible symptoms of muscle degeneration [55]. Table 6 compares the plants most prevalently involved in adverse effects as reported by the PlantLIBRA project, in relation to: 1) data from the literature [9], 2) reports from Poisons Centers [80]; and finally from this study.
It is important to underline that the review from the literature did not separate cases due to botanicals used as food supplements or traditional medicines as was the case in the other two data reviews. Moreover, due to the very high number of botanicals in PFS, the review on the scientific literature included "only" 66 among the most frequently consumed botanicals. The lists of plants most reported by Poisons Centers and by the consumers' survey are similar, sharing five out 13 botanicals; among them Valeriana officinalis (garden valerian) and Camellia sinensis (tea) were in the first two positions. Moreover, a similar position in the ranking was occupied by Paullinia cupana (guarana), Cynara scolymus (globe artichoke), and Panax ginseng (chinese ginseng).

Conclusions
The cases of adverse effects described here were self-reported and thus without any supporting clinical evidence; the agreement with data published in scientific papers and in particular with the survey performed by the PlantLIBRA project among Poisons Centers allows the following conclusions: 1. As reported previously, severe adverse effects related to PFS are quite rare [80]; 2. Mild or moderate adverse symptoms can be present but most of them do not require clinical support; 3. Data reported in this paper confirm that some plants are more frequently involved in adverse effects than others and can help family doctors, among other health professionals, to become aware about the possible consequences of the increasing use of food supplements containing botanicals; 4. This information could also be used to educate the public as to the possibility of adverse effects associated with the consumption of these food supplements.