An evaluation of mercury concentrations in three brands of canned tuna

Abstract

There is widespread concern over the presence of Hg in fish consumed by humans. While studies have been focused on determining the Hg concentration in sport fish and some commercial fish, little attention has been directed to canned tuna; it is widely held that concentrations are low. In the present study, the amount of Hg present in canned tuna purchased in Las Vegas, Nevada, USA, was examined, and the brand, temporal variation, type, and packaging medium impacts on Hg concentrations in tuna were explored. A significant (p < 0.001) brand difference was noted: Brand 3 contained higher Hg concentrations ( standard deviation (SD) (0.777 ± 0.320 ppm) than Brands 1 (0.541 ± 0.114 ppm) and 2 (0.550 ± 0.199 ppm). Chunk white tuna (0.619 ± 0.212 ppm) and solid white tuna (0.576 ± 0.178 ppm) were both significantly (p < 0.001) higher in mean Hg than chunk light tuna (0.137 ± 0.063 ppm). No significant temporal variation was noted, and packaging had no significant effect on Hg concentration. In total, 55% of all tuna examined was above the U.S. Environmental Protection Agency's (U.S. EPA) safety level for human consumption (0.5 ppm), and 5% of the tuna exceeded the action level established by the U.S. Food and Drug Administration (U.S. FDA) (1.0 ppm). These results indicate that stricter regulation of the canned tuna industry is necessary to ensure the safety of sensitive populations such as pregnant women, infants, and children. According to the U.S. EPA reference dose of 0.1 µg/kg body weight per day and a mean Hg value of 0.619 ppm, a 25-kg child may consume a meal (75 g) of canned chunk white tuna only once every 18.6 d. Continued monitoring of the industry and efforts to reduce Hg concentrations in canned tuna are recommended. Environ. Toxicol. Chem. 2010;29:237–242. © 2009 SETAC

INTRODUCTION

Canned tuna is a popular component of the human diet due to its convenience, affordability, taste, and health benefits 1. Fish provide a good source of dietary protein, omega-3 fatty acids, and vitamin D, and are relatively low in cholesterol 2, 3. Fish consumption has been linked to improved pregnancy outcomes and fetal growth 2. However, recent studies have established a link between heavy fish consumption and adverse health effects due to increased concentrations of mercury (Hg) and other contaminants in seafood 2, 4–6. Approximately 80% of the consumption advisories presently issued for fish are due to Hg 7. The developing fetus, infants, and children are particularly at risk due to the increased susceptibility of the developing nervous system to Hg 1, 2, 8.

ORIGIN/NATURE OF MERCURY

Mercury occurs naturally in the environment and is transported by natural processes as well as human activity. Natural sources of Hg include vapors from volcanic activity and the process of degassing 9, but these contribute relatively little to the overall environmental Hg burden. More significant are anthropogenic contributions of mercury to the atmosphere such as the burning of fossil fuels, waste incineration, mining and smelting, chlor-alkali production, and other industrial activities involving the use of Hg 7. The application of Hg-laden sewage sludge to crops and solid waste containing mercury to landfills contribute to sediment and soil burdens of Hg 7. Atmospheric and terrestrial Hg are mobilized to ground and surface water through rain, runoff, or leaching. Within aquatic systems, a complex set of chemical and biological processes convert inorganic Hg into the organic form, methylmercury 10. Methylmercury is of primary concern in aquatic food webs because of its tendency to bioaccumulate, or become concentrated in tissues. Studies have reported between 89% to 125% of Hg present in tuna to be in the organic form 1, 11.

Human exposure to mercury

Fish consumption is the primary route of exposure to methylmercury for humans 4, 12, 13. Several studies have demonstrated a positive correlation between increased consumption of fish and high Hg (most of which is in the methylated form) concentrations in human hair and blood 14–16. Health effects resulting from consumption of methylmercury for adults are considerable and may include central nervous system damage, ataxia, paresthesia, hearing loss, diminishing vision, loss of sensation to extremities, and loss of consciousness leading to death 4, 13, 17. Maternal fish consumption during and after pregnancy may also contribute to increased methylmercury concentrations in cord blood and breast milk 6, 13. Methylmercury readily crosses both the placenta and the blood–brain barrier and may confer neurological damage to developing fetuses, infants, and children. Due to the increased susceptibility of the developing nervous system, damages are more pronounced in fetuses, infants, and children and may include microcephaly, delayed development, impaired cognition, and gross neurological disorders 5, 12, 15, 18.

The U.S. Food and Drug Administration (U.S. FDA) is responsible for regulating the amount of Hg present in commercially sold fish. The U.S. FDA's action level for Hg in fish is 1.0 ppm (19; http://www.fda.gov/Food/GuidanceCompliance RegulatoryInformation/GuidanceDocuments/ChemicalContaminantsandPesticides/ucm077969.htm). This value represents the concentration at or above which the U.S. FDA may take legal action to remove products from the marketplace. The U.S. Environmental Protection Agency (U.S. EPA) is responsible for monitoring the amount of Hg present in recreationally caught fish from local waters. The agency works with states to issue consumption advisories when Hg concentrations are high enough to be detrimental to human health. The U.S. EPA and the National Academy of Sciences have established a reference dose of 0.1 µg methylmercury/kg body wt/d for human health 7. This reference dose, combined with consumption rate and estimates of average adult body weight, lead to the U.S. EPA's value of 0.5 ppm as an acceptable standard of Hg in fish for human consumption. Although this value is legally applicable only to recreationally caught fish, many states and nations have adopted this more conservative standard for health and safety reasons 1.

Approximately one billion pounds of canned tuna are consumed each year by Americans, accounting for 25 to 35% of all seafood consumption in the United States 20. Canned tuna's affordability and convenience contribute to its appeal and allow for its widespread use as a source of low fat protein and omega-3 fatty acids. The purchase of canned tuna is subsidized by the U.S. federal government's Women, Infants, and Children's program (WIC), which provides supplemental nutrition to low-income mothers and their children 20. While this program is instrumental in fighting childhood hunger, women of childbearing age and infants are known to be the most susceptible population with respect to methylmercury exposure. Unfortunately, highly sensitive populations may be consuming disproportionate amounts of canned tuna due to the subsidy.

For this reason, it is important to distinguish which types of tuna are safest to consume and to distribute this information to the populations most at risk. The variables examined in the present study included three popular national brands of tuna, two packaging methods (oil and water), four months during which canned tuna was purchased, and three variations of tuna type (solid white, chunk white, and chunk light). The U.S. FDA has established guidelines for these designations based upon species and a darkness gradient 1. White tuna must only contain the species Thunnus germo (albacore), whereas light tuna consists mainly of Katsuwonus pelamis (skipjack), a smaller and lower trophic level species than albacore. Solid tuna must contain only “loins cut in transverse segments to which no free fragments are attached.” However, chunk tuna consists of “a mixture of pieces in which the original muscle structure is retained” 21.

METHODS

Initially, a pilot study was completed sampling 155 cans of tuna from three national brands: Brand 1 (n = 54), Brand 2 (n = 46), and Brand 3 (n = 55). Four variables were examined in the pilot study: brand of tuna, tuna type (chunk white vs chunk light), seasonal variation, and packaging medium (oil vs water). Canned tuna was purchased from one central grocery store in Las Vegas, Nevada, USA, during each month from November 2005 to February 2006. Cans were purchased throughout this four-month span from a variety of lot numbers to identify any temporal variation in Hg concentrations as well as to provide a sufficient representation of each brand. The cans were identified using lot number and expiration date. The initial study revealed that there was a significant difference in Hg concentration among brand of tuna and tuna type. A follow-up study was then completed. In the second study, the brand containing the highest Hg concentration (Brand 3) was oversampled (n = 147) during May 2006, this time accounting for three types of tuna: solid white, chunk white, and chunk light. Sampling for the follow-up study was completed similarly to the first at the same grocery store.

Mercury analysis

Tuna was homogenized using a Polytron® PT 6100 homogenizer (Kinematica). A pilot study was completed that determined 2 min was sufficient to ensure thorough homogenization of canned tuna (data not shown). All Hg analyses were performed on an AMA 254 atomic absorption Hg analyzer (Leco® Corporation). This instrument measures both organic and inorganic Hg and reports them as a single total Hg concentration. Tissue samples weighing between 0.05 and 0.12 g were analyzed in nickel sample boats with drying, decomposition (550°C), and waiting times of 120, 300, and 45 s for all samples and certified reference materials. Ultra pure oxygen was used as a carrier gas with an inlet pressure of 250 kPa and a flow rate of 200 ml/min. The AMA 254 has a detection limit of 0.01 ng Hg and a linear range from 0.05 to 40 ng. All samples analyzed were within the calibration range. Total Hg concentration was recorded in units of wet weight parts per million (ppm) throughout the study.

Quality assurance/quality control

Standard reference materials and blanks were analyzed at the beginning and end of each session as well as every five samples. Two certified reference materials were used: a National Research Council of Canada DORM-2 Dogfish Muscle Standard for Trace Metals (http://www.nrc-cnrc.gc.ca/obj/inms-ienm/doc/crm-mrc/eng/DORM-2_e.pdf) as well as a National Institute of Standards and Technology Standard Reference Material® NIST 1633b Trace Elements in Coal Fly Ash (https://www-s.nist.gov/srmors/certificates/view_cert2gif.cfm?certificate=1633B). Duplicate measurements were taken for every 10th sample. All certified reference materials and duplicate measurements fell within 80 to 120% recovery of the expected values.

Statistical analyses

All data were analyzed using SPSS® version 16.0 (SPSS). A Shapiro-Wilk test was used to ensure normality in all data. Data sets that were normal included the comparison between chunk white and chunk light tuna, and the comparison between packaging in oil and in water. Data sets that were not normal (p > 0.05) were transformed using a log¹⁰ transformation. Data sets that required transformation included the comparison between brands, the comparison by month, and the comparison between Brand 3 solid white, chunk white, and chunk light tuna. Once the data sets were transformed, standard parametric tests were used to perform analyses.

Brand of tuna

The Hg concentrations in chunk white tuna from November through February were compiled for each of the three brands. Brand 1 (n = 29) had a mean Hg concentration of ( ± SD) 0.541 ± 0.114 ppm. Brand 2 (n = 46) had a mean concentration of 0.550 ± 0.199 ppm. Brand 3's (n = 55) mean Hg concentration was the highest at 0.777 ± 0.320 ppm (Table 1). An analysis of variance demonstrated that there was a significant overall difference among the three brands of tuna (F = 13.62; p < 0.001). A post hoc Bonferroni test indicated that Brand 3 had significantly higher Hg concentrations (p < 0.05), while the Hg concentrations of Brand 1 and Brand 2 were statistically similar (p > 0.05).

Table 1. Mercury concentrations in three brands of canned tuna expressed as ppm or standard deviation (SD)

Brand	n	Median	Mean	SD	Max
a Post hoc Bonferroni indicated a significant difference (p < 0.05) between mean values of Brands 1 and 3. b Post hoc Bonferroni indicated a significant difference (p < 0.05) between mean values of Brands 2 and 3.
Brand 1a	29	0.566	0.541	0.114	0.869
Brand 2b	46	0.502	0.550	0.199	1.144
Brand 3ab	55	0.777	0.714	0.320	1.666

Temporal variation

The variation in Hg concentrations over 4 months of the study was then examined. Samples (n = 130) were divided into categories relating the brand of tuna and month collected. The canning data were confidential. Therefore, the exact dates of packaging could not be determined. For this reason, expiration date was used as an indicator of packaging date and the assumption was made that a regular stock replacement of tuna existed in the grocery store from which cans were purchased. Mean values of Hg concentration by month are reported in Figure 1. An analysis of variance determined that there was no significant difference (p > 0.05) between Hg concentration and month collected for any of the brands (Fig. 1).

Figure 1. Mercury concentration (ppm) in canned tuna by month. A comparison of each brand's chunk white tuna over four months of collection. Data are expressed as means ± standard deviation. Mercury concentration did not differ significantly among the four months (p > 0.05). Brand 1 has no data for November due to an error in data collection. Diamond = Brand 1; square = Brand 2; triangle = Brand 3.

Tuna type

Tuna type was initially analyzed by comparing data from 10 cans of Brand 1 chunk white tuna with 15 cans of chunk light tuna (n = 25) of the same brand. The different sample sizes for each type were taken into account, and results were weighted accordingly. Analysis indicated a significant difference (p < 0.05) between chunk white tuna with a mean Hg concentration of 0.502 ± 0.086 and chunk light with a mean concentration of 0.264 ± 0.066 (Table 2).

Table 2. Mercury concentration by type: Brand 1 in water expressed as ppm or standard deviation (SD)

Type	n	Median	Mean	SD	Max
a Statistical analysis indicated a significant difference (p < 0.05) between mean values of chunk white and chunk light tuna types.
Chunk whitea	10	0.502	0.502	0.086	0.617
Chunk lighta	15	0.277	0.264	0.066	0.342

The brand with the highest Hg concentration was further investigated with additional sampling. Three types of Brand 3 tuna (n = 147) were analyzed: solid white (n = 49), chunk white (n = 48), and chunk light (n = 50). An analysis of variance revealed a significant overall difference between the three tuna types (F = 290.37; p < 0.001). A post hoc Bonferroni test revealed significant differences between solid white and chunk light (p < 0.001) and between chunk white and chunk light (p < 0.001). The mean difference was not statistically significant (p > 0.05) between solid white and chunk white (Table 3).

Table 3. Mercury concentration (ppm) by type: Brand 3 in water expressed as ppm or standard deviation (SD)

Type	n	Median	Mean	SD	Max
a Post hoc Bonferroni indicated a significant difference (p < 0.05) between mean values of solid white and chunk light tuna types. b Post hoc Bonferroni indicated significant difference (p < 0.05) between mean values of chunk white and chunk light tuna types.
Solid whitea	49	0.543	0.576	0.178	0.988
Chunk whiteb	48	0.561	0.619	0.212	1.159
Chunk lightab	50	0.116	0.137	0.063	0.310

Packaging medium

Concentrations of Hg were compared between 10 cans of Brand 3 chunk white tuna in water and 10 cans of the same brand and type packaged in oil (n = 20). The tuna packaged in oil had a mean Hg concentration of 0.807 ± 0.298 ppm, while the tuna in water had a mean concentration of 0.579 ±  0.330 ppm (Table 4). Although the mean values were quite different, a t test revealed no significant difference between the two types of packaging (p = 0.122) due to a large amount of variance within the two groups (Table 4).

Table 4. Mercury concentration by packaging: Brand 3 chunk white tuna in oil and water expressed as ppm or standard deviation (SD)a

Packaging	n	Median	Mean	SD	Max
a Mean values did not differ significantly (p = 0.122).
Oil	10	0.7470	0.8068	0.2978	1.367
Water	10	0.4976	0.5788	0.3297	1.481

A summary of the data is provided in Table 5, which displays the percentage of cans for each brand, type, and packaging material that were found to have Hg concentrations higher than the U.S. EPA Hg standard and the U.S. FDA Hg standard, respectively.

Table 5. Frequency of canned tuna samples that tested above federal standards for mercury; the U.S. Environmental Protection Agency standard is 0.5 ppm, the U.S. Food and Drug Administration standard is 1.0 ppm

Brand	n cans tested	% over 0.50 ppm	% over 1.0 ppm
NA, not applicable.
Brand 1
Total	54	39	0
Chunk white	29	69	0
Chunk light	25	4	0
Oil	10	1	0
Water	44	45	0
Brand 2
Total	46	52	4
Chunk white	46	52	4
Chunk light	0	NA	NA
Oil	0	NA	NA
Water	46	52	4
Brand 3
Total	202	53	7
Solid white	49	65	0
Chunk white	103	73	14
Chunk light	50	0	0
Oil	10	90	20
Water	192	51	6

DISCUSSION

Brand of tuna

The results of the present study demonstrated that there was a significant difference in Hg concentrations between brands collected during the initial sampling period. This difference could be due to several factors. One possible explanation is the location from which the tuna were caught. Mercury concentrations in aquatic organisms are dependent on the availability of Hg in their surrounding environment (which can be affected by nearby industrial activity or runoff), resulting in variable Hg concentrations in fish. The locations used by companies are confidential and not available to the consumer, making spatial comparisons difficult, although most canned tuna supplied to the United States is exported from Thailand, Ecuador, Indonesia, or the Philippines (http://www.globefish.org/dynamisk.php4?id=4583). Although light tuna consists primarily of the species Katsuwonus pelamis, or skipjack tuna, other species such as bluefin or yellowfin tuna may be included. Different feeding regimens and trophic levels of these species can impact the concentration of Hg in tuna. Studies have also reported positive correlations between both age 17 and size 10, 22 of fish and Hg concentrations.

Greater than half of the tuna analyzed, in two of the three brands, contained Hg concentrations higher than that considered safe for consumption by the U.S. EPA (0.5 ppm). Additionally, 4 to 7% of the tuna from brands 2 and 3 exceeded Hg concentrations above the U.S. FDA standard (1.0 ppm) for safe consumption (Table 5). The discrepancy between agencies regarding Hg concentrations considered safe for human consumption is cause for confusion among consumers. Many states and nations have adopted the more conservative standard of 0.5 ppm for health safety reasons 1. All three brands analyzed contained Hg in tuna above the U.S. EPA standard for safe consumption, and two of the three brands contained concentrations higher than the U.S. FDA standard (Table 5). Therefore, regulatory agencies should implement stricter controls over the canned tuna industry. Screening of tuna for Hg content should take place on a regular basis before entering the market. Tuna age, size, and origin should be regulated if Hg concentrations exceed federal standards, as all of these factors may play a role in increased mercury concentrations. Furthermore, the tuna industry should be required to provide detailed consumer information regarding Hg content of its products.

Temporal variation

No significant temporal variation was observed in the present study, presumably because the span of the study was too short to observe any seasonal or yearly trends. Some speculate that the overfishing of the tuna population would require a younger and smaller target over time so that mercury would decrease as a result of smaller and younger fish being caught 23. To explore this idea, we compare the results of the present study with four similar studies spanning a period of 15 years (Fig. 2). Yess's study took place in 1991 24, Burger and Gochfeld's over a period from 1998 through 2003 1, Forsyth et al. in 2003 25, Shim et al. in 2004 3, and the data for the present study were collected between 2005 and 2006. These combined data do not suggest a decreasing trend but rather reflect a slight increase in overall Hg concentrations over the past 15 years. However, the methodologies throughout these studies were not designed to be comparable and more consistent data are needed to make any concrete conclusions.

Figure 2. Results compared to other studies of canned tuna. Data are expressed as means ± standard deviation (when available). Dark bars = white tuna; white bars = light tuna.

Tuna type

White tuna has consistently been reported to contain higher concentrations of Hg than light tuna in previous studies 1, 24. The findings of the present study indicate a mean of 0.502 ± 0.086 ppm in chunk white and a mean of 0.576 ±  0.025 ppm in solid white tuna. These values were more than three times higher than the mean for chunk light tuna at 0.137 ± 0.008 ppm. White tuna contains only the albacore species, which is a typically larger and higher trophic level species with an average weight between 25 and 45 pounds. Light tuna consists mainly of the species skipjack which averages between 6 and 12 pounds (http://www.atuna.com/species/species_datasheets.htm). This trophic discrepancy makes albacore prone to higher Hg concentrations by means of bioaccumulation. White tuna makes up 29% of the market share of canned tuna. Therefore, consumers should be made aware of the higher Hg concentrations in white tuna and be provided with the necessary information to make educated dietary choices with regard to canned tuna.

Packaging medium

No significant variations between the packaging of tuna in oil versus water were found. This is consistent with the findings of Burger and Gochfeld 1, although Yess reported a significantly higher concentration in water-packed tuna than in oil-packed tuna 24. However, Burger and Gochfeld 1 called attention to the fact that the oil in Yess's study was mixed into the sample, rather than being drained off, which would result in a dilution of total Hg. Although other health effects may be related to packaging in oil or water, the present study revealed no difference for exposure to mercury in tuna.

CONCLUSIONS

In the present study, average Hg concentrations in all three brands routinely exceeded the designated safe standard set by the U.S. EPA of 0.5 ppm. Two of the brands analyzed were found to have individual cans of tuna that exceeded the U.S. FDA's action level for Hg in tuna of 1.0 ppm. Using the most conservative safety standard for fish (0.5 ppm, as determined by the U.S. EPA) consider this scenario: If a meal of tuna is assumed to be 75 g and contains 0.5 ppm Hg, a 25-kg child would only be allowed to consume canned tuna once every 15 d. This would keep the average daily intake below the established reference dose. The results of the present study show that one brand of tuna was significantly higher in Hg than the others. If the above calculation is performed using the mean Hg value in chunk white tuna from Brand 3 (0.777 ppm) the same child should eat tuna only every 23.3 d. The findings of the present study were consistent with data found in the literature, where white tuna contained the highest Hg concentrations. The results of the present study indicate that stricter regulation of Hg in the canned tuna industry is necessary. Locations from which tuna are caught could provide insight into the elevated concentrations of Hg. Furthermore, the designations from various federal agencies regarding safe concentrations of Hg in edible tissues should be consistent and clarified. For at-risk populations such as women, infants, and children, these guidelines should be modified and explicitly stated. The variation in data both in the literature and throughout the present study suggest the need for a long-term monitoring program to ensure the safety of tuna that we consume.