1 Introduction

Heavy metals are naturally existing elements widely distributed in our living environment and therefore can be absorbed into the human body through external sources, such as water, food, air, industrial waste, and chemical pollutants. Some toxic heavy metals (THMs), including lead (Pb), arsenic (As), cadmium (Cd) and mercury (Hg), can cause harmful health effects with extremely small amounts. As recommended by the Joint FAO/WHO Expert Committee for Food Additives (JECFA), the provisional tolerable weekly intake (PTWI) of Pb, inorganic As, Cd and Hg were set as 25 μg/kg body weight (BW) [1], 15 μg/kg BW [2], 7 μg/kg BW [3], and 5 μg/kg BW [1], respectively. The maximum tolerable intake for a 60-kg healthy adult is 214 μg/d for Pb, 129 μg/d for inorganic As, 60 μg/d for Cd and 43 μg/d for Hg by calculation. THMs contamination in Chinese materia medica (CMM) can be attributed to its own properties, agricultural environment, and the processing method. For example, Cinnabaris and Realgar are red mercuric sulfate and arsenic sulfide mineral containing Hg and As, respectively. In addition, contamination in soil and water and improper use of pesticides and fertilizers during cultivation as well as the use of containers or additives containing THMs during the processing step contributed to THMs contamination.

Several analytical techniques, including flame atomic absorption spectrometry, graphite furnace atomic absorption spectrometry, cold vapor atomic absorption spectrometry, atomic fluorescence spectrometry, high-performance liquid chromatography, inductively coupled plasma (ICP) optical emission spectrometry or mass spectrometry (MS), X-ray fluorescence spectrometry, differential pulse polarography, neutron activation analysis, and anodic stripping voltammetry, have been used to determine heavy metals content in natural products or herbal products [4]. ICP-MS is one of the most sophisticated analytical instrumentation with the advantage of multi-elemental character, low limits of detection, wide dynamic range, etc. Several national pharmacopeias, including Taiwan Herbal Pharmacopeia (THP), Hong Kong Chinese Materia Medica Standards (HKCMMS) and Chinese Pharmacopoeia (ChP), have recommended ICP-MS as the analytical method to determine THMs content in CMM for ensuring the safety of pharmaceutical products. In this study, THMs content, including Pb, As, Cd, and Hg, was determined in 1887 samples for 27 selected types of CMM using nitric acid and microwave digestion followed by a validated ICP-MS method. The microwave was used for the CMM samples digestion because it exhibited better accuracy in relation to both time and recovery rate [5]. Based on the ICH Q2(R1), the validation items included linearity, accuracy (recovery rate), inter-day and intra-day precision, method detection limit (MDL) and limit of quantitation (LOQ) [6].

2 Materials and methods

2.1 CMM samples collection

A total of 1,887 samples for the 27 types of selected CMM were purchased from different provinces of China and chosen according to the top 20 imported CMM between 2016 and 2018. The imported quantities of the CMM were retrieved by the department of Chinese Medicine and Pharmacy, Ministry of Health and Welfare (Taipei, Taiwan), as indicated in Table 1. It should be noted that the sum of imported quantities of Astragali radix (AR) and Hedysari radix (HR) is ranked second. In addition, the sum of imported quantities of seven CMM, including Raphani semen (RS), Cuscutae semen (CUS), Kochiae fructus (KF), Arctii fructus (AF), Cassiae semen (CAS), Plantaginis semen (PS) and Benincasae semen (BS), is ranked ninth.

Table 1 Toxic heavy metals contamination in 27 selected Chinese materia medica in our factory
Full size table

2.2 Preparation of analytical sample solution

All CMM were ground into powder using pulverizer followed by passing through a 40-mesh filter. Briefly, 0.2 g of each sample powder was placed into a microwave digestion tube with 6 mL of concentrated nitric acid and 1.5 mL of hydrogen peroxide solution for 20 min. The sample solution was digested using MARS microwave-accelerated reaction system (MARS Xpress, CEM, USA). Digestion parameters for the microwave system were set as 1,600 W for microwave, 30 min for ramp time, 205 °C for temperature, and 20 min for hold time. After digestion, 1 mL of internal stock standard solution was added to deionized water to produce a final volume of 100 mL followed by passing through a 0.45-μm filter to obtain the analytical sample solution.

2.3 Preparation of standard solution

One thousand μg/mL of Pb, As, Cd, Hg, rhodium (Rh), and gold (Au) standard solution was purchased from Sigma-Aldrich (St. Louis MO, USA). The external standard stock solution I containing 10 μg/mL of Pb and As and 1 μg/mL of Cd and Hg was prepared by adding 1 mL of 1,000 μg/mL of Pb and As standard solution and 0.1 mL of 1,000 μg/mL of Cd and Hg standard solution to 1% (v/v) nitric acid solution to produce a final volume of 100 mL. The external standard stock solution II containing 1 μg/mL of Pb and As and 0.1 μg/mL of Cd and Hg was prepared by adding 1 mL of external stock standard solution I to 1% (v/v) nitric acid solution to produce a final volume of 10 mL. The internal standard stock solution consisting 1 μg/mL of Rh and 10 μg/mL of Au was prepared by combining 0.1 mL of 1000 μg/mL of Rh and 1 mL of 1000 μg/mL of Au standard solution to 1% (v/v) nitric acid solution to produce a final volume of 100 mL. Standard working solution containing 0.2–50 μg/L of Pb and As, 0.02–5 μg/L of Cd and Hg, 10 μg/L of Rh and 100 μg/L of Au was made by combining different volumes (10, 50, 250, 1000 and 2500 μL) of the external standard stock solution II to 500 μL of internal standard stock solution diluted in 1% (v/v) nitric acid to a final volume of 50 mL.

2.4 ICP-MS condition

A PerkinElmerNexIONTM300X ICP-MS (MA, USA) was used for simultaneous detection of Pb, As, Cd and Hg content in CMM. The setting of the ICP-MS condition was established as the followings: plasma power 1500 W, carrier gas Ar, plasma gas flow 18 L/min, auxiliary gas flow 1.2 L/min and nebulizer gas flow 1.1 L/min. The m/z ratios for the four THMs were set as 208Pb, 75As, 111Cd and 202Hg.

2.5 Method validation

The linearity of the calibration curves for the four THMs was established using five different concentrations (0.2–50 μg/L for Pb and As and 0.02–5 μg/L for Cd and Hg) in triplicates for each concentration. The intercept, slope of the regression line and the correlation coefficient were recorded. The values of correlation coefficient for the four THMs should be higher than 0.995. For accuracy, free-THMs samples were spiked with low (4.0 μg/mL for Pb and As and 0.4 μg/mL for Cd and Hg), medium (5.0 μg/mL for Pb and As and 0.5 μg/mL for Cd and Hg) and high (6.0 μg/mL for Pb and As and 0.6 μg/mL for Cd and Hg) concentrations of the four THMs followed by nitric acid and microwave digestion in triplicates. The recovery rate was calculated as (detection value ÷ spike value) × 100%, and the detection value has been subtracted the elemental background from matrix. An acceptable quality level of recovery rate should be ranged in 75–120%. For inter-day and intra-day precision, free-THMs samples were spiked with 5.0 μg/mL for Pb and As and 0.5 μg/mL for Cd and Hg of the four THMs followed by nitric acid and microwave digestion in six times. The percentage of relative standard deviation (RSD) was calculated as follows: (SD value ÷ mean value) × 100%, and the calculated value should be lower than 10%. For determining MDL and LOQ, the calibration curves for the four THMs were established using relative low different concentrations (0.01–1.0 μg/L for Pb, 0.01–5.0 μg/L for As and 0.01–2.0 μg/L for Cd and Hg). Test solution containing 0.4 μg/L for Pb, 1 μg/L for As and 0.1 μg/L for Cd and Hg was carried out in seven times. The slope of calibration curves for the four THMs was recorded. The standard deviation (SD) in ratio of intensities between standard and internal standard for seven times determination was calculated. The MDL and LOQ were calculated as 3.3 × SD ÷ slope and 10 × SD ÷ slope, respectively.

3 Results and discussion

3.1 Method validation

The definition of validation is the action of checking or proving the validity or accuracy of developed experimental method. Evaluation by ICP-MS method was validated by assessing the correlation coefficient, accuracy (recovery rate), inter-day, and intra-day precision, MDL, and LOQ. The correlation coefficient of the calibration curves for the four THMs were all higher than 0.995, indicating that the linearity is fine (Table 2). THMs determination by ICP-MS is known to be interfered by spectral and non-spectral interferences. This interference might affect the low recovery of THMs. For example, the determination of As (m/z 75) by ICP-MS can be interfered by ArCl+ (m/z 75), polyatomic species. Herein, we used dynamic reaction cell (DRC) technology to solve this interference. DRC technology features a scanning quadrupole that removes all interferences created by reactions in the Universal Cell, which enables any reactive gas to be used. By optimizing the cell’s quadrupole conditions, only the element of interest is allowed to pass through to the analyzing quadrupole. It is well recognized that ion-molecule reaction chemistry offers the best performance for the reduction of complex polyatomic spectral interference [7]. The recovery rate was in the range of 101.4–105.4% for Pb, 82.6–89.7% for As, 84.7–87.8% for Cd and 91.7–97.2% for Hg which meet the acceptable quality level of 75–120% (Table 2). The precision is the condition where the results from independent test are acquired with the same method on the same sample in the same or different day by the same or different operator for several replicates using the same analytical instrument [7]. The calculated RSD (%) in inter-day and intra-day precision for the four THMs in spiked samples was all lower than 10% (Table 2). The LOQ in test solution for Pb, As, Cd and Hg was 0.212, 0.500, 0.038, 0.044 μg/L, respectively, and in solid samples was 0.106, 0.250, 0.019, 0.022 μg/g, respectively (Table 2). The MDL in test solution for Pb, As, Cd and Hg was 0.070, 0.165, 0.013, 0.014 μg/L, respectively, and in solid samples was 0.035, 0.0825, 0.0065, 0.007 μg/g, respectively (Table 2). Results obtained from LOQ and MDL indicated that this validated ICP-MS method exhibited excellent sensitivity.

Table 2 Method validation
Full size table

3.2 Monitoring of THMs contamination in selected CMM

CMM has been regarded as a potential source of THMs toxicity, including Pb, As, Cd and Hg, to human and animals [8]. WHO has also recommended that CMM and its products should be determined for the presence of THMs [9]. Herein, an ICP-MS method was used to determine THMs content in a total of 1,887 samples for the 27 selected types of CMM. The maximal, average and median values of all CMM for four THMs are shown in Table 1. Pb was detected in 1,145 samples out of the total samples with the contaminated rate of 61% (Table 1). Among these Pb-contaminated CMM, Cinnamomi ramulus (CIR), RS, CUS, KF, Chuanxiong rhizome (CHR), Citri reticulatae pericarpium (CRP), Eucommiae cortex (EC) and Bupleuri radix (BR) are most commonly contaminated by Pb with the contaminated rate higher than 80% (Table 1). A total of 7 (6.4%) and 4 (3.5%) samples in CHR and BR, respectively, exceeded the PL of 5.0 μg/g for Pb as established by THP and HKCMMS (Table 4). In addition to CHR and BR, a total of 27 (32%) and 4 (18%) samples in CIR and EC, respectively, exceeded the PL of 5.0 μg/g for Pb as established by HKCMMS (Table 4). However, only 1 (1.2%) and 2 (9.1%) samples in CIR and EC, respectively, exceeded the PL of 10.0 μg/g for Pb as established by WHO and ISO18664:2015 (Table 4).

Moreover, As was detected in 143 samples out of the total samples with the contaminated rate of 7.6% (Table 1). RS, AF and BR are seldom contaminated by As with the contaminated higher than 30% (Table 1). None of the samples exceeded the PL of 3.0 or 5.0 μg/g as set by THP (Table 4). Similarly, none of the samples exceeded the PL of 2.0 or 5.0 μg/g as set by Japanese Pharmacopeia (JP) (Table 4). Only 3 (2.4%), 2 (1.6%) and 1 (0.8%) samples in Rehmanniae radix (RR) exceeded the PL of 2.0 μg/g and 4.0 μg/g, respectively, as set by HKCMMS and ISO18664:2015, respectively (Table 4).

Furthermore, Cd was detected in 1,257 samples out of the total samples with the contaminated rate of 67% (Table 1). Nearly half of all CMM, including HR, CIR, RS, KF, AF, PS, CHR, Paeoniae alba radix (PAR), Ginseng radix (GR), Codonopsis radix (COR), EC, Armeniacae amarum semen (AMR) and BR, are most commonly contaminated by Cd with contaminated rate higher than 80% (Table 1). Only 1 (2.6%) sample in HR and 1 (1.1%) sample in PAR exceeded the PL of 0.3 μg/g for Cd as formulated by THP and HKCMMS (Table 4). Among the 27 selected CMM, the ChP only regulated the PL for THMs in AR, Glycyrrhizae radix et rhizoma (GRER) and PAR (Table 3). Similarly, only 1 (1.1%) sample in PAR exceeded the PL of 0.3 μg/g for Cd as formulated by ChP (Table 4). A total of 2 (2.7%), 19 (17%), 1 (0.9%) and 2 (1.8%) samples in COR, CHR, AMR and BR, respectively, exceeded the PL of 1.0 μg/g for Cd as formulated by THP and HKCMMS (Table 4). Under the PL of 0.3 μg/g for Cd as formulated by WHO, a total of 74 and 108 samples in CIR and CHR, respectively, with the exceeding rate was 88% and 98%, respectively (Table 4). Also, the exceeding rate in RS, EC, AMR and BR was ranged from 10% to 30% (Table 4). In addition, the exceeding rate in other Cd-contaminated CMM, including Jujubae fructus (JF), HR, PS, PAR, GR, COR and Scutellariae radix (SR), was lower than 10% (Table 4). Only 1 (0.9%) sample in CHR and 1 (0.9%) sample in BR exceeded the PL of 2.0 μg/g for Cd as formulated by ISO 18664:2015 (Table 4).

Table 3 The permissible limit of Pb, As, Cd, and Hg for 27 selected types of Chinese materia medica in different countries/organizations
Full size table
Table 4 The percentage of Chinese materia medica exceeding permissible limits in different countries/organizations
Full size table

Likewise, Hg was detected in 17 (0.9%) samples out of the total samples (Table 1). The contaminated rate for Hg in all selected CMM was not higher than 10% (Table 1). Only 1 (0.7%) sample in GRER exceeded the PL of 0.2 μg/g for Hg as regulated by THP, ChP and HKCMMS (Table 4). Similarly, only 1 (2.6%) sample in HR exceeded the PL of 0.2 μg/g for Hg as regulated by THP and HKCMMS (Table 4).

3.3 Comparison of PL among different countries/organizations

As shown in Table 3, we further summarized the PL set by different countries/organizations for the four THMs of the all selected CMM. The Taiwan authority has renewed the PL for the four THMs in 355 types of CMM in 2016 and brought it into the THP [10]. The general rules of the PL in THP were established as 5.0, 3.0, 1.0, and 0.2 μg/g for Pb, As, Cd, and Hg, respectively, in most CMM [10]. For specific CMM, the PL was 10.0 μg/g in CUS or 15.0 μg/g in CIR and EC for Pb, 5.0 μg/g in ASR, RR, Poria (P), CHR, AMR, BR and SR for As, 0.3 μg/g in AR, HR, GRER and PAR or 1.5 μg/g for Cd and 0.5 μg/g for Hg (Table 3) [10]. The ChP has drawn up the PL of 5.0, 2.0, 0.3 and 0.2 μg/g for Pb, As, Cd and Hg, respectively, in eleven types of CMM, whereas, the PLs were varied in some specific CMM [11]. Herein, only AR, GRER and PAR were included (Table 3). The JP only proposed the PL of 2.0 or 5.0 μg/g for As in certain CMM [12]. The PL for As was 2.0 μg/g in GR and 5.0 μg/g in AR, GRER, ASR, RR, P, BR and SR (Table 3). The nine volumes of HKCMMS covering standards for a total of 299 CMM have been published. The PL of 5.0, 2.0, 1.0 and 0.2 μg/g for Pb, As, Cd and Hg, respectively, as a general rule in most CMM [13]. The WHO only recommended the PL of 10.0 μg/g for Pb and 0.3 μg/g for Cd in dried plant materials [14]. The International Organization for Standardization (ISO) also released ISO 18,664:2015 and published the PL of 10.0, 4.0, 2.0, and 3.0 μg/g for Pb, As, Cd, and Hg, respectively, in all CMM except minerals [15].

Taking CIR for example, the PL of 15.0, 5.0 and 10 μg/g for Pb has been established by THP, HKCMMS and WHO, respectively, with the exceeding rate of 0%, 32% and 1.2%, respectively (Table 4). The PL of 1.0 and 0.3 μg/g for Cd in CIR has been set by THP and WHO, respectively, with the exceeding rate of 0% and 88% (Table 4). Another example is CHR. Similarly, the PL of 1.0 and 0.3 μg/g for Cd has been drawn up by THP and WHO, respectively, with the exceeding rate of 27% and 98%, respectively (Table 4). The diverse PL for the four THMs set by various countries/organizations is the major reason for the difference in the exceeding rate in CMM. As compared to other countries/organizations, the THP did set a more practical and reasonable PL. The THP established the PL for the four THMs based on the inspection data collected from several certificated Chinese herbal medicine pharmaceutical factories in Taiwan and on the 80th percentile values of the Taiwan Food and Drug Administration (TFDA) surveys from 2006 to 2013 [16,17,18,19,20,21,22,23] as well as referred to the PL established by other countries/organizations. There is an increasing number of people worldwide using CMM as remedy, health food and dietary supplements, therefore, it is necessary to establish the unifying PL for THMs in CMM. One limitation existed that no clear correlation was found between the levels of THMs in soil and collection site in CMM because our CMM samples were collected from herbal markets or imported from trading agent in different provinces of China. It should also be noted that the present results did not represent the exact quality of the CMM sold in Taiwan and our company only imported the qualified CMM.

4 Conclusions

This study provides insights into THMs, including Pb, As, Cd and Hg, contamination in the imported 27 types of selected CMM. Overall, Pb and Cd are the most abundant contaminant in selected CMM. In diverse permissible limits (PL) established by different countries/organizations, CIR and EC usually exceeded the PL of 0.3 μg/g for Cd and occasionally exceeded the PL of 5.0 μg/g for Pb. CHR seldom exceeded the PL of 1.0 μg/g for Cd. EC, RS, AMR, and BR occasionally exceeded the PL of 0.3 μg/g for Cd. We hope to raise awareness of THMs contamination in CMM that affects safe medication use among farmers, manufacturer and consumers.