Occurrence and Health Risk Assessment of Aflatoxins through Intake of Eastern Herbal Medicines Collected from Four Districts of Southern Punjab—Pakistan

Eastern herbal medicines (HMs) are plant-derived naturally occurring substances with minimum or no industrial processing that have long been used in traditional medicine. Aflatoxins are frequent contaminants of plants. Therefore, these mycotoxins are likely to contaminate HMs and pose a health risk to individuals using them on a regular basis as preventive or curative treatments of various diseases. The present study aimed to determine aflatoxin levels in the most popular Pakistani HM formulations and to assess the health risk associated with the intake of aflatoxins. A total of 400 samples of HM formulations collected from four districts of Punjab were analyzed for the quantification of aflatoxins, out of which 52.5% were found to be contaminated. The average daily dose (ADD) of AFB1 and AFs through the intake of HM formulations ranged between 0.00483 and 0.118 ng/kg bw/day and between 0.00579 and 1.714 ng/kg bw/day, respectively. The margin of exposure (MOE) and population cancer risk ranged from 99.49 to 29378.8 and from 0.00011 to 0.0325 liver cancer cases/105 individuals/year (0.0075–2.455 liver cancer cases/105 individuals/75 years), respectively. Despite the low exposure to aflatoxins from HM formulations in the four studied Punjab (Pakistan) districts, the frequent contamination of the analyzed samples suggests that official measures should be considered to manage the associated risk.


Introduction
Complementary and alternative medicines (CAMs), including traditional medicines (TMs), have long been practiced worldwide. The Eastern system of medicines (Unani medicine system) is an ancient TM system, which is widely used in South Asian countries, including India, Pakistan, Bangladesh, and Iran, and is also gaining acceptance in other parts of the world to prevent and cure various ailments. Around 65-80% of the population in developing countries principally rely on HMs for their day-to-day healthcare essentials [1,2]. The primary reasons behind the usage of TMs lie in that they are easily accessible at affordable prices, are patient-oriented, and are strongly related to patients' beliefs. Additionally, being natural, this practice is believed by the general population to be safer and non-toxic compared with chemically synthesized alternatives [3,4].
Eastern herbal medicines (HMs), also known as Tibb-i-Unani, are plant-derived naturally occurring substances with minimum or no industrial processing that have been used to cure health issues within regional or local healing practices [5]. The plants and their parts, such as leaves, roots, bark, rhizomes, and flowers, contain several important The aim of the present study was to evaluate the occurrence of aflatoxins in the most commonly used HM formulations in four districts of Punjab, Pakistan. The potential exposure and health risk of aflatoxins resulting from consumption of aflatoxin-contaminated HM formulations was also evaluated.

Sampling
A total of 400 samples of 20 HM formulation types produced in Pakistan (20 samples from each type) were purchased from local herbal medical stores located in four districts (Multan, Bahawalpur, Rahim Yar Khan, and Dera Ghazi Khan) of the province of Punjab (Pakistan). From each district, 100 samples were used in this study during the period of June to September 2020. The samples were packed in air-tight plastic bags and stored in a dark and dry place until analysis. Table 1 shows the types of HM formulations analyzed in this study, along with their composition, intake rate, and claimed therapeutic effects. The criterion used for selection was the extent of consumption in the region; only high consumption formulations were considered.

Extraction and Immunoaffinity Clean-Up
The samples of HM formulations were extracted with 100 mL methanol/water (60/40 v/v) in an orbital shaker (Thermo Scientific, Waltham, MA, USA), MaxQ 4000) for 4-5 h at 200 rpm. The extracts were filtered using Whatman filter paper No. 42, and the resulting filtrates (4 mL) were diluted with 16 mL of 50 mM phosphate-buffered saline (PBS) having pH 7.4. The diluted extracts were passed through immunoaffinity columns (Eurofins, Siegen, Germany) according to the manufacturer's recommendations.

Post-Column Derivatization
The derivatization of aflatoxins was accomplished as stated by the AOAC official method 2005.08 [40]. Briefly, the dried extracts of samples and standards were dissolved in 200 µL of hexane and added with 50 µL trifluoro-acetic acid (TFA). The vials were then closed tightly and placed in the dark for 5-6 min before the addition of 1.95 mL mixture of double distilled water and acetonitrile (9:1) to each of them, followed by vortex mixing for 1-2 min. The lower aqueous layer containing aflatoxins was removed and filtered via syringe filter (0.45 µm) prior to chromatographic analysis.

Chromatographic Analysis
A high-performance liquid chromatographic (HPLC) system, S 500 routine series using S1125 isocratic pump (Sykam, Eresing, Germany), coupled with a fluorescence detector (Sykam, RF-20A) was used. An isocratic mobile phase of water/methanol/acetonitrile (55/22.5/22.5 v/v/v) was used at a flow rate of 1.0 mL/min. A Welchrom (Welch Material, Inc., Austin, TX, USA) silica gel reverse phase C-18 column (4.6 × 250 mm) was used as a stationary phase. The run time for each standard and sample was 20 min, and the injection volume was 20 µL. The excitation and emission wavelengths of the fluorescence detector were 365 nm and 440 nm, respectively, and the column oven temperature was set at 37 • C. The retention times for AFB 1 , AFG 1 , AFB 2 , and AFG 2 were 5.09, 4.28, 8.78, and 6.66 min, respectively.

Method Validity
A mixed solution of aflatoxins in acetonitrile from Sigma-Aldrich (St. Louis, MO, USA) was used as a standard. The solution contained the four major aflatoxins: AFB 1 (0.5 µg/mL), AFG 1 (0.5 µg/mL), AFB 2 (0.25 µg/mL), and AFG 2 (0.25 µg/mL) of HPLC grade purity (≥98%) for each aflatoxin. Working standard solutions at four different concentrations (0.005, 0.010, 0.025, and 0.050 µg/mL) were prepared in acetonitrile and were used in recovery experiment and for preparation of calibration standards (solventmatched). The recovery percentages were computed by spiking the aflatoxin-free samples of HMs at three different concentrations, 12, 24, and 48 µg/kg, with the ratios of AFB 1 , AFB 2 , AFG 1 , and AFG 2 being 1.0:0.5:1.0:0.5, respectively. The standards of aflatoxins were quantified independently in nine replicates. The spiked samples were allowed to stand for 12 h to ensure the adsorption of aflatoxins within the samples before they were prepared for chromatographic analysis according to the procedure mentioned above. The recovery percentages were calculated by using Equation (1).

Exposure Assessment
The average daily dose (ADD) (expressed as ng/kg bw/day) of aflatoxins was computed based on the concentration of toxin detected and the intake rate of studied HM formulations. The HMs analyzed in the present study were a mixture of herbs, finely ground and suspended in water or milk at given ratios to be taken orally as recommended by the medical herbalist. The ADD, expressed in ng/kg bw/day, was calculated by using Equation (2) [42].
where C is the concentration of aflatoxin (ng/kg). IR is the intake rate (kg of HM/day) calculated for children or adults according to the practitioner recommendations for the studied HM formulations (Table 1). EF is the exposure frequency; a figure of 90 days/year was used as recommended earlier [8,[42][43][44][45]. ED is the exposure duration; 70 years was taken as the current average human lifespan. AT is the average time (ED × 365 days/year). W AB is the average body weight; the respective values of 32.7 kg and 72 kg were used for Pakistani children and adults [46,47]. The left-censored data (data below LOD and LOQ) were processed by applying the substitution method of EFSA [48]. Two exposure scenarios were considered: a lower bound (LB) scenario, in which zero was assigned to samples showing aflatoxins concentration below LOD/LOQ, and an upper bound (UB) scenario, in which the value of LOD was assigned to the samples in which the aflatoxins concentration was below the detection limit, and the LOQ value was assigned to the samples where aflatoxins were present at levels below the LOQ [48].

Health Risk Characterization
The risk characterization originating from the oral exposure to aflatoxins was computed using two approaches; the qualitative margin of exposure (MOE) approach established by EFSA for substances that are both genotoxic and carcinogenic [49] and the quantitative approach to liver cancer risk estimation proposed by the Joint FAO/WHO Expert Committee on Food Additives (JECFA) [50]. The MOE is the ratio between the point of departure (POD) for carcinogenesis, a toxicological reference point corresponding to a dose at which a low but measurable adverse response is first observed and the human exposure to the substance. The MOE value was calculated using Equation (3): where BMDL 10 is the benchmark dose lower confidence limit (BMDL 10 ) for 10% increased cancer risk. The value of 170 ng/kg bw/day suggested by the EFSA on the basis of animal data with the application of uncertainty factors [51] was used in this study. ADD is the average daily dose used to estimate the exposure levels, as calculated in Equation (1). Because substances that are both genotoxic and carcinogenic can pose health risks at any dose level, to explain the significance of our results, we followed the recommendation of JECFA and applied an MOE of 10,000.A calculated MOE value lower than 10,000 implies that the exposure to a carcinogenic and genotoxic substance is of concern to public health and should be given high-priority for risk management [52]. It should be mentioned, however, that the MOE value is not a measure of cancer risk per se, but it rather provides an estimation of the level of concern [49,53]. Therefore, the risk for liver cancer from AFB 1 exposure via HM intake was calculated by a deterministic approach on the basis of AFB 1 carcinogenic potency (P cancer ) resulting from synergistic carcinogenic effects of hepatitis B virus (HBV) infection and AFB 1 , expressed by using Equations (4) and (5): Cancer risk = P cancer × ADD (5) where, P cancer is the average carcinogenic potency of AFB 1 expressed as the number of cancer cases per 100,000 individuals per ng of AFB 1 per kg bw per day. PHBsAg + and PHBsAg − are P cancer of aflatoxins for hepatitis B surface antigen-positive (HBsAg + ) and hepatitis B surface antigen-negative (HBsAg -) individuals, respectively. The %Pop. HBsAg + and %Pop. HBsAg − are the prevalence of HBV and non-HBV carriers, respectively. For HBsAg + individuals, the P cancer of AFs was 0.3 cancers/year/10 5 individuals/ng AFB 1 /kg bw/day, while the P cancer of AFs for HBsAg − individuals was estimated to be 0.01 cancers/year/10 5 individuals/ng AFB 1 /kg bw/day [54]. Considering the prevalence rate of 3.3% for HBsAg + individuals reported by WHO for the Eastern Mediterranean Region [55], the P cancer related to HMs intake in the four studied districts of Punjab (Pakistan) was estimated to be 0.019 cancers/year/10 5 individuals/ng AFB 1 /kg bw/day. To calculate the cancer risk resulting from lifetime exposure (75 years), the resulting P cancer value was then multiplied by 75 [54].

Statistical Analysis
The data were analyzed using Statistix 8.1 (Informer Tech. Inc., Los Angeles, CA, USA). All the measurements were performed in triplicates. A probability value (p-value) less than 0.05 was considered statistically significant. The data are expressed as mean ± standard deviation, computed using Microsoft Excel 2013 version. The analysis of variance (ANOVA) followed by Least Significance Difference (LSD) test was used for statistical comparison of the data. A Shapiro-Wilk test of normality was run to check the normality of data and after recording the data as normal, the analysis of variance (ANOVA) followed by a Least Significance Difference (LSD) test was used for statistical comparison of the data.

Aflatoxins Occurrence in Herbal Medicine Formulations
The results of aflatoxins occurrence in 400 samples of 20 HM formulations collected from four districts of the Punjab province (Pakistan) are summarized in Tables 2-4. Total AFs were detected in 52.5% of the analyzed samples, with AFB 1 being the most frequently occurring (46.3%), followed by AFG 1 (35.6%), AFB 2 (34.5%), and AFG 2 (27%) ( Table 2). The results also demonstrate that there was a significant difference (p < 0.05) in the concentration of the five studied groups of aflatoxins (AFB1, AFB2, AFG1, AFG2, and AFs) among the HM formulations (Table 3), while no significant difference (p > 0.05) was found between aflatoxin levels in samples collected from the four districts of Punjab (Table 4). Table 3 summarizes the concentrations of the different types of aflatoxins in the analyzed samples of the 20 HM formulations. This table shows that AFs ranged from <LOD to 17.5 ng/g, with an average of 1.95 ng/g, while individual aflatoxins were detected in the range of <LOD-8.4 ng/g, <LOD-3.23 ng/g, <LOD-12.83 ng/g, and <LOD-8.93 ng/g with average values of 0.58 ng/g, 0.20 ng/g, 0.78 ng/g, and 0.39 ng/g for AFB 1 , AFB 2 , AFG 1 , and AFG 2 , respectively. As for the contamination levels per type of HM, the highest average level of AFs (12.40 ng/g) was recorded in the Safoof-e-Lal, followed by Safoof-e-Mughaliz (4.21 ng/g) and Allergex (3.26 ng/g), while none of the analyzed aflatoxin types was detected in Akhseer-e-Pachish and Johar Hazim samples by the technique used (Table 3). It is worth mentioning that AFB 1 and AFs concentrations exceeded the EU MTLs of 2 ng/g and 4 ng/g in 5.75% (n = 23) and 10.25% (n = 41) of HMs samples, respectively (Table 5) [56,57]. However, the average concentrations of AFB 1 and AFs in the studied samples of HM formulations altogether were below the abovementioned MTLs; yet, the average concentration of AFs in each of two formulations, Safoof-e-Lal and Safoof-e-Mughaliz, exceeded the MTL of the EU and the MTL of the European Commission of 10 ng/g for spices ([58] Section 2.1.9) in one formulation, Safoof-e-Lal (Table 3), but they remained below the Pakistani MTL of 20 ng/g for all articles of food [34]. Moreover, the prevalence of AFB 1 and AFs in the samples of HM formulations in the four districts varied between 22% and 60%, depending on the aflatoxin type, with average concentrations of 0.5 to 0.63 ng/g for AFB 1 and 1.8 to 2.2 ng/g for AFs (Table 4).    Results are expressed as mean of 20 determinations ± standard deviation; statistically significant difference was observed among means having different letters within the columns (p < 0.05); AFs is the sum of AFB 1 , AFB 2 , AFG 1 , and AFG 2 ; LOD = limits of detection; * = mean of 400 determinations of 20 HM formulations. Results are expressed as mean of X determinations ± standard deviation; no statistically significant difference was observed among means within the columns (p > 0.05). Abbreviations are as in Table 3 above.  Occurrence of aflatoxins in Eastern Medicines/HMs have been reported in various parts of the world. In Southeastern Nigeria, 84.21% of 57 studied Eastern Medicines were found to be contaminated with AFs at levels varying from below the LOD to 20 ng/g, with an average concentration of 7.35 ng/g [23]. A Korean study also demonstrated that out of 700 analyzed samples of HMs, 2.43% (n = 17) were positive for AFs, with concentrations ranging between 4.51 and 108.42 ng/g and that 35.29% of the positive samples exceeded the regulatory limit of 10 ng/g set by Korean Food and Drug Administration [59]. Zhao et al. [60] analyzed 22 samples of Chinese HMs for the presence of aflatoxins and found that 63.6% (n = 14) of them were contaminated; AFs and AFB 1 concentrations varied from 0.2 to 7.5 ng/g and 0.2 to 4.8 ng/g, respectively. Another study of 20 samples of medicinal plants used for HM preparations in China reported the occurrence of AFs at a rate of 40%, with 10% (n = 2) exceeding the regulatory Chinese limit [61]. In India, Afs content in crude medicinal plants used for herbal medicine formulations ranged between <LOD and 24 ng/g. Ali et al. [62] reported that the average concentrations of AFB 1 , AFB 2 , AFG 1 , and AFG 2 in 23 samples of traditional herbal medicine preparations from Malaysia and Indonesia were 0.26 ng/g, 0.07 ng/g, 0.10 ng/g, and 0.03 ng/g, with a prevalence of 70%, 61%, 30%, and 4%, respectively. An analysis of 16 samples of HMs from South Africa revealed that the concentration of AFs in all of the analyzed samples was below the detectable level of 0.5 ng/g [24]. Nonetheless, such results may not be conclusive due to the low number of samples analyzed. In comparison with previous studies, the levels of aflatoxins found in the Pakistani HM formulations in this study fell in the same range as those reported in Thai HMs, where 18% of the samples were contaminated with levels varying between 1.7 and 14.3 ng/g [25]. The level of contamination of HM preparations made with organic medicinal plants may be higher than that of those made from conventional medicinal plants due to the restricted use of synthetic fungicides in organic agriculture. This was demonstrated in Turkish samples of different herbs commonly used in traditional medicine, where 86% of the analyzed samples were contaminated with AFB 1 , at mean levels varying between 5.7 and 44.5 ng/g, depending on the herb, and the percentage of the positive samples varied from 60% to 100%, with 65.6% of the positive samples (21 out of 32 samples) exceeding the EU MTL [28]. Conversely, our result are lower than those reported in HMs from Southeastern Nigeria [23], South Korea [59], and India [61] and are higher than those from China [60], South Africa [24], and Indonesia and Malaysia [62]. Many factors may account for discrepancies regarding aflatoxin contamination of HMs from different countries; these include the sampling season and procedure, the constituents of HM formulations, the varieties and chemotypes of raw medicinal plants, the climate and soil of the region where these plants grew, and the post-harvest conditions of preparation, packaging, and storage. In fact, within the same region the extent of contamination may vary greatly from one formulation to another. For example, contrary to the rest of the HM formulations studied herein, none of the five types of aflatoxins were detected in Akhseer Pachish and Johar Hazim formulations. On the contrary, samples of Safoof-e-Thandak, Majoon Azaraqi, Safoof-e-Lal formulations were contaminated with the highest levels of the five aflatoxin types. In Safoof-e-Lal and Safoof-e-Mughaliz formulations, the total AFs content exceeded the most stringent MTL of 4 ng/g set by the European Pharmacopoeia [56] (Table 3). Apart from environmental parameters and agricultural practices, such variations between HM formulations may be ascribed to the conditions of preparation, packaging, and storage by herbalists, which may account for the high contamination of some medicinal plants, despite their well-documented resistance to mold growth and/or toxigenesis [63][64][65][66], owing to their ability to produce antifungal bioactive substances [67][68][69].

Exposure Assessment
The exposure of the south Punjab (Pakistan) population (children and adults) to AFB 1 and to total AFs from the intake of HM formulations was assessed by the ADD determinations; the results are summarized in Table 6. Irrespective of the age and for both lower bound (LB) and upper bound (UP) scenarios, the ADD of AFB 1 and total AFs through the consumption of HM formulations ranged between 0.01 and 0.12 ng/kg bw/day and between 0.01 and 1.71 ng/kg bw/day, respectively. For children, the mean LB exposure to AFB 1 and total AFs ranged from 0.01 to 0.11 ng/kg bw/day and from 0.01 to 1.40 ng/kg bw/day, and the mean UB exposure varied from 0.02 to 0.12 ng/kg bw/day and from 0.01 to 1.41 ng/kg bw/day, respectively. For adults, the mean LB exposure to AFB 1 and total AFs ranged from 0.003 to 0.11 ng/kg bw/day and from 0.01 to 1.70 ng/kg bw/day, and the mean UB exposure from 0.01 to 0.11 ng/kg bw/day and from 0.01 to 1.71 ng/kg bw/day, respectively. The average ADD of total AFs through consumption of HM formulations was higher for children (0.21 and 0.22 ng/kg bw/day for the LB and the UB, respectively) than for adults (0.17 and 0.18 ng/kg bw/day for LB and UB, respectively). This can be explained by the lower body weight of children compared with adults, which outweighs the effect of the intake rate. As regards the type of medicine, Safoof-e-lal contributed the highest level of exposure to total AFs (1.55 ng/kg bw/day), followed by Alhazim (0.16 ng/kg bw/day) and Senna Maki (0.13 ng/kg bw/day) (data not shown). Overall, none of the samples analyzed exceeded the respective MTLs of 2 ng/g and 4 ng/g for AFB 1 and AFs set by European Pharmacopoeia [56,70] or those set by the United States Pharmacopeia (USP) of 5 ng/g for AFB 1 and 20 ng/g for total AFs. Such results suggest that the exposure of the Southern Punjab Pakistani population to aflatoxins from HM formulations is too low to raise a serious public health concern. The risk may be even lower if the HM formulations were administered as infusions, where the plant material is separated from the beverage to be taken after the infusion process. This treatment of medicinal plants was reported to reduce aflatoxin content by 70% to 100% [71]. Nonetheless, our results indicate that aflatoxins are rather common contaminants of HM formulations in the studied region of Punjab, with 46.3% and 52.5% for AFB 1 and total AFs, respectively ( Table 2). The growing use of medicinal plant preparations in popular medicine in different countries around the world, and particularly in developing countries where they are usually informally marketed beyond any official control, is an additional risk factor [72]. Moreover, due to the inconsistent harvest, preparation, distribution, and storage conditions of HMs, significantly higher levels of aflatoxins in HMs from remote areas can be reasonably anticipated. Therefore, the potential risk that the consumption of such medicines pose to public health cannot be ruled out, especially for consumers in the 95th percentile if these products continue to be marketed without official control. Meanwhile, it is recommended that the health risk associated with aflatoxins in HMs in Pakistan be systematically assessed to serve as a scientific basis for the development of adequate regulatory standards. Although, no tolerable daily intake (TDI) can be used to define safe levels as a reference for putative regulations due to the lack of a threshold response of aflatoxins as carcinogenic and genotoxic toxicants, the "as low as reasonably achievable" (ALARA) approach can be adopted to ensure safe use of HM formulations. More comprehensive surveys of the prevalence and extent of contamination of HMs with aflatoxins are needed to provide sufficient data for a meaningful risk characterization, and hence to provide a sound basis for regulatory provisions [52]. Few studies, to our knowledge, have conducted a formal risk assessment of aflatoxins in medicinal plants, their formulations, or their extracts/infusions in other countries. Pallarés, Berrada, Fernández-Franzón and Ferrer [71] surveyed 224 samples of 56 different species of medicinal plants commercialized in Spain for the occurrence and the level of contamination with different mycotoxins, including the four major aflatoxin types (AFB 1 , AFB 2 , AFG 1 , AFG 2 ). The study showed that aflatoxins were found in the raw materials at mean concentrations ranging between 64.76 and 838.58 ng/g for the major aflatoxins; however, their prevalence and concentrations were drastically reduced in the infusions with the notable elimination of AFB 1 . The authors concluded that the health risk associated with the consumption of medicinal plant infusions was low, and yet, it should be managed with the ALARA approach. However, no inference in the study was made to the risk associated with the ingestion of the raw medicinal plants. Similar results were reported on Moroccan aromatic and medicinal plants, whose infusions were shown not to pose an appreciable health risk owing to a low exposure, in spite of the fact that the concentrations of total AFs in some raw plant material exceed the regulatory standards of 4 ng/g or 10 ng/g [73]. It is worth mentioning that AFB 1 , the most toxic aflatoxin, whose carcinogenicity for humans is well established, was not detected in any of the samples analyzed in the latter study. Exposure of Nigerian infants and young children (IYC) to abnormally high health risks from the consumption of aflatoxin-contaminated complementary foods with too low MOE values (0-70 for AFB 1 and 0-7 for AFs) [37] or a too high exposure (641 ng/kg bw per day) [36] was reported. However, in the latter studies, the main ingredients, e.g., maize, rice, oat, wheat, millet, and peanut of the surveyed foods are notoriously known for their vulnerability to contamination with various mycotoxins. For example, AFB 1 concentration reached a value as high as 51,192 ng/g in Tom bran, a whole meal from mixed grains, including maize and peanut [36]. This may account for the difference in our results on HM preparations, generally consisting of a mixture of medicinal plants with varying degrees of antifungal activities [69]. Akhseer Pachish and Johar Hazim Medicinal Herbal formulations were not considered for exposure assessment, as their aflatoxins concentrations were below the LOD; LB = lower bound scenario (censored numbers < LOD were given zero); UB = upper bound scenario (censored numbers <LOD were given LOD values and censored numbers <LOQ were given LOQ values); AFs = sum of AFB 1 , AFB 2 , AFG 1 , AFG 2 ; NS = not studied for this age group (herbal medicine is not for consumption by that particular age group); * = average of all medicines.

Health Risk Characterization
The findings of the characterization of the risk for hepatocellular carcinoma (HCC) development upon exposure to AFs based on the MOE approach using ADD and BMDL 10 as well as by P cancer and ADD are presented in the  (18,287.55) were higher than the safe margin of 10,000, indicating that the exposure to this particular HM formulation poses a low health risk. Conversely, MOE values calculated for of the rest of the HM formulations indicate that they expose Punjabi users to a high risk, and hence they require official management measures. Eastern medicines having aflatoxins concentration < LOD (Akhseer Pachish and Johar Hazim) were not considered for exposure assessment; LB = lower bound scenario (censored numbers < LOD were given zero); UB = upper bound scenario (censored numbers < LOD were given LOD values and censored numbers < LOQ were given LOQ values); NS = not studied for this age group (herbal medicine is not for consumption by that particular age group); * = average of all medicines.

Conclusions
Despite the growing use of traditional medicine with HM formulations in Pakistan, no studies have been conducted on the potential risk they may pose to public health regarding their contamination with aflatoxins. This was the first study to investigate the occurrence of aflatoxins in 20 of the most commonly used HM formulations in four districts of Punjab (Pakistan) in order to perform a preliminary risk assessment of which population would be exposed in the region. Our results show that more than 46% and 50% of the analyzed samples were positive for total aflatoxins (AFs) and the most potent aflatoxin, AFB 1 , respectively. Although, generally low, the average concentrations of AFB 1 and total AFs were higher than the EU regulations in one and two HM formulations, respectively. Additionally, at the individual level, 5.7% and 10.25% of samples exceeded the latter MTL of AFB 1 and total AFs, respectively. The exposure data suggest that children are more at risk than adults, mainly because of their lower bodyweight. Although the overall health risk of aflatoxins and the calculated annual rate of extra-liver cancer cases caused by consumption of HM formulations were low, the risk may still be of concern, particularly with continuous exposure of the heavy consumers (95th percentile), and cannot be ignored. Further investigations are thus required, including the implementation of an adequate surveillance system and long-term monitoring of aflatoxin contamination in as many HM formulations and medicinal plants as possible. Thorough surveys of the extent and frequency of consumption of these products are also necessary to provide a more realistic risk assessment of outcomes, taking into account different consumption patterns, groups, and the percentiles of consumers. Studies to evaluate the co-occurrence of potentially harmful mycotoxins in the herbal medicine formulations and their synergistic or antagonistic effects can help refine any risk assessment conducted on aflatoxins as standalone hazards. Meanwhile, regulatory measures based on previously established MTLs in different countries can be issued to ensure the safety and quality of herbal medicines.

Conflicts of Interest:
The authors declare no conflict of interest.