Introduction

Aflatoxins are secondary metabolites produced by two related species of fungi, Aspergillus flavus and Aspergillus parasiticus1. These toxins contaminate many staple foods and are a significant concern for public health2,3. Several types of aflatoxins occur in nature, and the four major aflatoxins are AFB1, AFB2, AFG1 and AFG2, which are unsafe for humans and animals4.

Staple dietary foods such as maize, groundnuts, rice and milk are highly susceptible to aflatoxin contamination, making them a real threat to food security, safety and population health5,6. As a result, exposure to aflatoxins via the diet is associated with various health consequences7,8. The ingested aflatoxins are converted to aflatoxin-8,9-epoxide metabolites in the liver, which are responsible for many of the toxic effects in the body9. Acute toxicity includes hemorrhage, acute liver damage, edema, digestion problems and can sometimes precede death10, whereas chronic exposure leads to the alteration of intestinal integrity, resulting in stunting in children and immune suppression11,12. The International Agency for Research in Cancer classified AFB1 and mixtures of total aflatoxins as group I carcinogens that induce liver and colo-rectal cancer13. Notably, these implications are frequently reported in regions where the prevalence of aflatoxin occurrence is high3.

Aflatoxin contamination poses a global threat, as an estimated 25% or more of the world’s food crops are destroyed annually, posing a significant economic burden14. Warm and humid storage conditions allow the growth of molds, and aflatoxins accumulate in commodities throughout any stage of the value chain from preharvest to post harvest15. The incidence of aflatoxin contamination is reportedly highest in developing countries and, more specifically, in Asia16. This is due to their geographical location, as it provides favorable conditions for the proliferation of mycotoxins and the implementation of poor regulatory measures, resulting in potential exposure of more than 5 billion human populations17. For instance, in a study conducted by Jog Raj in 2021, 93% of the tested corn samples collected from India, Thailand, Vietnam, Taiwan and the Philippines were contaminated with one or multiple aflatoxins18.

To date, no study has been conducted on the contamination levels of food commodities by aflatoxins in Bhutan. Considering the country’s warm and humid climate due to global warming, low income and poor stringent rules and regulations, it is of paramount importance that the prevalence of aflatoxin contamination in major staple foods in the country be investigated.

Furthermore, it is important to conduct risk assessments because the complete elimination of aflatoxins is not possible. Risk assessment is the process of characterizing the potential hazards and associated risks resulting from exposure to chemicals present in food over a specific period19. This approach is used to guide food regulators and public health officials in implementing risk mitigation processes, such as setting regulatory limits on food supplies. Although Bhutan has set the maximum tolerable concentration of total aflatoxins to 20 µg/kg, no risk assessment has been conducted to date to evaluate the potential health implications of aflatoxin exposure. This is essential due to differences in dietary habits, exposure and characterization of the risk it possesses. Therefore, considering the scarcity of information on aflatoxin contamination in the region, the present study attempted to determine the occurrence of aflatoxin contamination in different food items in Bhutan, and a deterministic risk assessment was performed based on dietary exposure to aflatoxins.

Results

Validation of the analytical method

The calibration curve plotted using standard log concentration versus Logit B/Bo by linear regression showed excellent linearity, with a correlation coefficient > 0.999 for both total aflatoxin (AFT) and aflatoxin B1 (AFB1). The relative standard deviation of all the spiked samples ranged from 1.66 to 6.95%, and the recoveries of both AFT and AFB1 ranged from 87.19 to 112.02%. Furthermore, the limit of detection was adopted according to the manufacturer’s values of < 3 µg/kg for AFT and < 2 µg/kg for AFB1. Table 1 illustrates the findings.

Table 1 Total aflatoxin and aflatoxin B1 recoveries.
Full size table

Occurrence of aflatoxins

A total of 315 samples consisting of 10 different categories of food commodities were analyzed for the presence of AFT (B1, B2, G1 and G2), and those samples detected with > LOD of AFT were further subjected to analysis for aflatoxin B1 (Table 2). Of these, 48.81% of the samples were positive for AFT, and 79.35% were positive for AFB1. The mean levels of AFT and B1 in all the samples were 11.49 ± 12.83 µg/kg and 17.62 ± 23.99 µg/kg, respectively. The concentrations of AFT in the 95th and 50th percentiles were 41.98 and 6.84 µg/kg, respectively, while those of AFB1 were 85.89 and 10.21 µg/kg, respectively.

Table 2 Levels of total aflatoxins and aflatoxin B1.
Full size table

Table 3 illustrates the results of the analyzed samples based upon the type of sample and the occurrence and concentration of aflatoxins. The overall concentrations of AFT and AFB1 ranged from less than the limit of detection to 75.84 µg/kg and 116.41 µg/kg, respectively. Among the types of food samples, chili powder had the highest mean level of total aflatoxin (19.65 µg/kg), followed by nuts (14.06 µg/kg). Similarly, chili powder was the dominant food commodity, with the highest mean concentration of aflatoxin B1 detected (28.01 µg/kg), followed by nuts (22.96 µg/kg). Although split pulses had a higher mean, only one sample was detected with a high level of B1, which resulted in a higher overall mean. Notably, none of the rice samples tested had > LODs for AFT; hence, these samples were not subjected to further testing for AFB1.

Table 3 Levels of AFT and AFB1 in different food commodities in Bhutan.
Full size table

The concentration of chili products was significantly greater than that of all other food items (p value = 0.00104). Moreover, the concentration of aflatoxin B1 correlated well with the AFT concentration, ranging highest in chili powder, nuts and dried chili, at 27.42 ± 26.49, 23.41 ± 30.82 and 11.41 ± 4.27 µg/kg, respectively.

With respect to the mean AFT concentration in the food samples from the different districts, the highest concentration was detected in Monggar (24.19 µg/kg), followed by Wangdue Phodrang (21.7 µg/kg) and Chukkha (19.88 µg/kg). The lowest values were determined for the districts of Lhuentse, Sarpang and Trashiyantse. Similarly, aflatoxin B1 was also found to be the most abundant toxin in the abovementioned districts, at 37.30, 35.31 and 34.85 µg/kg, and the least abundant were from Sarpang, Tashiyantse and Gasa. However, there was no significant difference in the concentration of aflatoxins in the food samples (p value = 0.03–0.64). Figure 1 illustrates the findings.

Figure 1
figure 1

Burden of aflatoxins in different districts of Bhutan.

Full size image

Bhutan is divided into four regions and there is a difference in culture, food habits, ethnicity and climatic conditions. Hence, an attempt was made to report aflatoxin contamination in different regions of the country. The supplementary table shows the different levels of total aflatoxins detected in different regions of the country (see Supplementary Table S1). The highest percentage of positive samples was detected in the western region (56.84%), followed by the central region (56.67%). However, the southern region had the highest mean aflatoxin concentration (14.76 ± 16.15 µg kg−1), followed by the western region (12.93 ± 15.57 µg kg−1). One-way ANOVA was performed to compare the means of these different regions; however, no significant differences were detected in the mean level of aflatoxins.

The results from the analyzed samples were further compared with international guidelines elsewhere (CODEX, the EU and the US FDA). In total, 43.6% of the total samples had total aflatoxin levels above the European Union maximum level of 4 µg/kg, whereas 9.1% and 4.6% of the samples had total aflatoxin levels above the CODEX and US FDA maximum limits, respectively (Table 4). To ensure maximum food safety, foodstuffs were more specifically compared with the EU regulations of minimal aflatoxin contamination. The results showed that the chili powder had the highest level of aflatoxins above the regulatory limit of 94.87%, followed by dry chili (77.7%) and peanuts (60.6%). Supplementary Table S2 shows the data.

Table 4 Proportion of samples above the maximum regulatory limits.
Full size table

Human health risk characterization

Human health risk estimation was conducted on the most consumed food items in the country, which were chili products (chili powder and dried chili). The results are illustrated in \* MERGEFORMAT Table 5. The exposure estimates were calculated based on the estimated daily intake (EDI) of these food items. In the present study, the daily intake of chili powder and dry chili was estimated to be 2.1 g/day and 15.92 g/day, respectively, for an adult weighing an average body weight of 60 kg, which was determined from the Bhutan Living Standard Survey 202220. The EDIs for chili powder at the mean and 95th percentile exposure was 0.98 and 5.34 ng/kg−1 bw day−1, respectively. Similarly, for the dried chili, the mean exposure was 2.96, and for 95th percentile was 5.30 ng kg−1 bw day−1.

Table 5 Human health risk estimation due to aflatoxin B1 exposure in Bhutan.
Full size table

Risk characterization was conducted based on the margin of exposure (MOE) and the liver cancer approach. MOE compares exposure levels to a toxicological threshold and is typically used to compare the health risk of various contaminants to prioritize risk management efforts21. The lower the MOE, the greater the risk to the consumers and a value of less than 10,000 is used to indicate health risk22. This is measured by the benchmark dose lower confidence limit (BMDL) for a benchmark response of 10% of 0.4 µg/kg body weight per day for the incidence of hepatocellular carcinoma (HCC) in male rats following AFB1 exposure as per the European Food Safety Authority23.

The range of MOEs for BMDL10 derived from the analysis ranged from 74.90 to 408.10. From these calculations, the MOE values were less than 10,000, indicating a public health risk due to AFB1 exposure through the consumption of chili products. Remarkably, if the 95th percentile values for large consumers are considered, there is a significant risk from exposure to aflatoxins to the general population. The incidence of HCC was estimated by the quantitative cancer risk approach and is expressed as cancer cases per 100,000 individuals per year. The values show that, for the maximum exposure scenario, the cancer risk was estimated to be 0.01 cancer cases per 100,000 people per year for the Bhutanese population.

Figure 2 shows the plot for exposure to AFB1 and the estimate of toxicity due to consumption of dried chili and chili powder in the Bhutanese population. The intersection of the estimates of toxicity and the estimates of exposure revealed the risk of AFB1. A higher risk and toxicity of AFB1 are illustrated by the closer point of intersection to the x-axis and farther from the y-axis. According to the risk 21 matrix, chili powder has a greater risk of daily exposure to AFB1 than does dry chili.

Figure 2
figure 2

Risk 21 plot for chili products.

Full size image

Discussion

This study reports the first findings on the status of aflatoxin contamination in major staple foods consumed in the region and its associated human health risk. The report revealed the presence of total aflatoxins and aflatoxin B1 in almost all of the samples included in the test. The MOE values demand a stringent public health measure.

An initial step for determining the true burden of aflatoxin contamination is the use of proper test assays24. For a specific test to be suitable for its intended use, the quality, reliability and consistency of the analytical results are integral. For this purpose, recovery analysis was performed to determine the actual concentration of an analyte in the samples, and we reported recoveries ranging from 87.19 to 112.02% for both AFT and AFB1. These percentages were within the acceptable range of the AOAC, CODEX and EU regulations of 60–120%25; therefore, the analytical method was reliable and in accordance with international guidelines.

The present study revealed a positive frequency of 48.85% for AFT and 79.35% for AFB1. This indicates that the Bhutanese population is exposed to the carcinogenic toxins that are present in the daily diet. Similar studies performed in Thailand reported an occurrence of 38.9%26; 27.1%, in Zhejiang Province, China27; 24.3%, in Vietnam28; 78.0%, in Nepal29; and 61.3%, in India30. The presence of these aflatoxins is inevitable, as factors such as changes in temperature and water availability related to actual climate changes (increased temperature, heavy rainfalls and droughts) all modulate mold growth and the production of mycotoxins31. This demands a safe and effective multifaceted approach for combating food and feed contamination. Although most developed countries have stringent rules for regularly monitoring these toxins in food, ensuring food insufficiency and security is a major challenge for developed and developing countries.

Studies across the region have reported AFB1 concentrations ranging from 48–383 µg/kg in corn samples in India, 0.04–21.30 µg/kg in rice in Pakistan and 1–34.80 µg/kg in corn in Vietnam32. In the present study, the overall AFT and AFB1 concentrations ranged from < LOD to 75.84 µg/kg to 116.4 µg/kg for all the food commodities. This is very alarming since getting exposed to such a level of toxins leads to severe health consequences. A study by Angele et al. revealed that a concentration of 0.099 µg/kg in eggs and 0.525 µg/L in milk are risk factors that could increase the incidence and prevalence of malnutrition and cancer in Cameroon33. A similar study in India by Asim et al. revealed a threefold increase in the risk of hepatocellular cancer with respect to urinary AFB1 and a higher incidence of aflatoxin contamination in the region34. Currently, 27.5% of the children in the country are stunted, almost half (42.6%) of the stunted children are severely shunted35, and the incidence of liver cancer is 5.3%36. The association between exposure to aflatoxins and the probability of stunting and the development of liver cancer should be considered. The question arises whether these health implications are attributed to issues related to the contamination of aflatoxin in the food that the population consumes daily. An exposure assessment study using appropriate biomarkers for aflatoxin exposure involving the recruitment of liver cancer patients and shunted children is necessary.

Notably, there was a clear difference in the concentrations of both AFT and AFB1 among the different districts of the country. This may be attributed to the geographical locations of the districts. For instance, the Monggar district, which falls in the eastern region of the country, had an overall mean concentration of 24.19 µg/kg, and the Chukkha district, which lies in the southern region, had 19.88 µg/kg. The climate conditions vary among these regions; thus, differences were observed. A study by Battilani et al. reported that in Europe, aflatoxin B1 contamination in maize has increased due to climate change37. This finding is further supported by the findings of Benigni et al. and Sarah et al., whose findings correlated well with the current findings38,39.

The most prevalent foods contaminated with aflatoxins reported in this study are chili products, followed by nuts. Contamination of chili powders by aflatoxins has been described by several surveys from studies in the region. In India, almost 66.7% of the samples had values above 10 µg/kg40; 86.7%, in Thailand41; and 100% of the samples > 20 µg/kg, in Bangladesh42. Similarly, in the present study, 84.6% of the chili powder samples had a concentration greater than 10 µg/kg, which is in concordance with the findings of other studies. Chili powder is produced locally using traditional methods by sun drying the chilies and grinding it to powder. In this process, there is a maximum likelihood of mold formation. The occurrence of aflatoxin contamination is determined by factors such as pre-harvest, post-harvest, storage conditions, temperature and humidity15,32. Therefore, prevention measures such as treating fungal contamination in crops, appropriately handling food items, ensuring effective food regulation systems and providing mass advice are essential for minimizing aflatoxin contamination in food.

The estimated daily intake of AFB1 via the consumption of chili products was determined to be 0.98 and 2.96 ng/kg−1 bw day−1 in the present study. Studies elsewhere in Ethiopia reported an EDI of 0.13–15.8 ng/kg−1 bw day−143, and those in Malaysia were 0.21–1.32 ng/kg−1 bw day−144. This discrepancy might be due to differences in the concentrations and consumption data collected during the particular study period. Nevertheless, the findings reveal that the risk of aflatoxin B1 exposure originates from both the high occurrence of aflatoxin B1 contamination and the high consumption of chili products by the population. Taken together, these findings indicate that the current regulatory limits of 20 µg/kg may still not be low enough to eliminate these concerns.

Based on the current findings, it is imperative that the intake of contaminated chili products is of great importance for public health since all the MOE values are below the safe threshold value of 10,000. The MOE values of AFB1 recorded in the present study ranged from 74.90–408.10, which are comparable to those reported for chili products from Sri Lanka ranging from 45–78 and 997–909 in Turkey45,46, exposing the respective consumers.

Conclusion

This study illustrates the presence of total and aflatoxin B1 in selected food commodities from the country, providing the first insights into their occurrence and human health risk assessment. The incidence of both total and aflatoxin B1 contamination in food samples was greater than that in half of the samples. The margin of exposure was less than the safety limit of less than 10,000, indicating a serious public health concern. In addition, the liver cancer risk was estimated to be 0.01 and 0.007 cancers per year per 100,000 population resulting from the consumption of chili products. The data presented here provide evidence that addressing aflatoxin contamination and exposure to aflatoxins is necessary and may cause severe health consequences for the population.

Methods

Study area and site

The Bhutan region is a subtropical region and has twenty administrative districts. The study site was located in each district, and samples were collected from each site. Figure 3 represents the districts of the country and the number of samples collected from each district.

Figure 3
figure 3

Sample collection sites. The map was created with qgis version 3.26 (https://www.qgis.org/).

Full size image

Sampling

A total of 10 different categories of food commodities were randomly sampled and purchased between September and October 2022 from grocery and supermarket shelves. Each sample was homogenized by grinding, and an aliquot of 20 g was kept in zip lock bags and stored at − 20 °C until analysis. Based on the literature on the highest likelihood of aflatoxin contamination, the following types of samples were collected and subjected to chemical analysis (Table 6).

Table 6 Types of food commodities included in the study.
Full size table

Chemicals

Commercially available detection kits (AgraQuant, Romer labs, Inc., Newark, DE, USA) for both total aflatoxin (AFB1, AFB2, AFG1 and AFG2) and AFB1 were used for detection and quantification in the present study. The methanol used for extraction was of HPLC grade, and all the other chemicals used were of ACS grade or higher.

Sample analysis

Briefly, 20 g of representative sample was meshed and extracted with 100 ml of 70% methanol. The solution was vortexed and settled, and the top aqueous layers were filtered using a Whatman No. 1 filter. The extracted samples were subsequently analyzed using direct competitive enzyme linked immuno-sorbent assay (ELISA) to determine the total aflatoxin AFB1, AFB2, AFG1 and AFG2 concentrations.

Health risk estimation

Estimated daily intake

The estimated daily intake (EDI) of aflatoxin via oral ingestion was calculated by using the mean and 95th percentile levels of total aflatoxins and aflatoxins B1 determined from the food samples and the daily intake by the population divided by the average body weight. The following equation was used for the calculation of the EDI:

$${\text{EDI(ng}}\,{\text{kg}}^{\text{-}1} {\text{day}}^{\text {-}1} {) = }\frac{{{\text{Daily}}\,{\text{Intake}} \times {\text{Mean}}\,{\text{Concentration}}}}{{{\text{Average}}\,{\text{Body}}\,{\text{Weight}}}}$$

Margin of Exposure (MOE)

For the derivation of the MOE, the point of departure or reference point from the animal data was divided by the estimated dietary exposure.

$$\text {MOE}=\frac {\text {BMDL}_{10}\; (\text {ng/kg.bw/day})} {\text {Estimated\,Daily\,Intake}\;(\text {ng/kg.bw/day})}$$

Cancer risk

To estimate the potency of aflatoxin B1, we adopted the proposed JECFA formula to evaluate the carcinogenic risk in humans. Therefore, 0.3 cancer/year/100,000 population ng/kg bw/day for HBsAg-positive individuals and 0.01 cancer/year/100,000 population ng/kg bw/day for HBsAg-negative individuals were considered to determine the average potency. Moreover, a HBsAg-positivity prevalence of 5.9% was adopted for Bhutan47, and the HBsAg-negative group was extrapolated to 92.26%. The following equation was used for the calculation:

$${\text{Average}}\,{\text{Potency}} = (0.3 \times {\text{HBsAg}}{-}{\text{ve}}\,{\text{prevalence}}) + (0.01 \times {\text{HBsAg}} + {\text{ve}}\,{\text{prevalence}})$$
$${\text{Cancer risk = Exposure/EDI}} \times {\text{Average potency }}$$

Data analysis

The concentrations of AFT and AFB1 were calculated via regression analysis from the standard calibration curves generated from standard reference materials of both AFT and AFB1 using the standard spreadsheet provided by the manufacturer (Romerlabs). The data are presented as the mean, standard deviation and 50th to 95th percentiles. The significance differences among the food sample types and regions were analyzed using ANOVA tests in SPSS version 22 (IBM, Chicago, IL, USA). For the risk assessment models, estimated daily intake, MOE values, average potency and cancer risk were all used as per the guidelines of the EFSA and WHO23,48,49.

Ethical issues

No ethical approval was needed because there were no human participants involved. However, permission for a waiver of ethical clearance was sought from the Research Ethics Board of Health, Bhutan (Ref. No. REBH/PO/2022/029). Verbal informed consent was obtained from the respective shopkeepers and vendors during sample collection.