Selenium toxicity from a misformulated dietary supplement, adverse health effects, and the temporal response in the nail biologic monitor

Abstract

Use of dietary supplements in the U.S. has increased steadily over the last 25 years. While misformulation is uncommon, the consequences can be serious. A March 2008 voluntary market recall removed supplement products responsible for the most serious selenium toxicity outbreak that has occurred in the U.S. We quantified selenium concentrations in the misformulated supplement products, measured the temporal response in the nail biologic monitor, and associated exposure to self-reported selenosis symptoms. Subjects recruited through state health departments and referrals provided samples of the misformulated supplement products, exposure information, monthly toenail and or fingernail clippings or onycholysitic nail fragments, and listed their newly onset adverse health effects attributed to selenium toxicity. Ninety-seven subjects enrolled and submitted at least one test sample. Peak selenium concentrations (up to 18.3 and 44.1 μg/g for toenails and fingernails, respectively) were measured. Multiple samples (52 total) of all six recalled supplement lots were analyzed ranging from 22,300 to 32,200 μg selenium per daily dose. Average consumption was 30.9 ± 13.9 doses; 73 subjects provided follow-up data on selenosis symptoms at 2.50 ± 0.14 years. Nail samples accurately reflect exposure in this selenium toxicity outbreak, which resulted in long-term/permanent adverse health effects.

Figure 1

Se concentrations (μg/dose) in the six recalled TBF product lots. Se concentrations are given in micrograms per 1-ounce recommended dose for the 36 (of 54) samples submitted with corresponding lot numbers. The dose means (s.d.) range from 22,319 (151) to 32,184 (301) μg/dose over the 6 lots. The intended dose for all six lots was 200 μg/dose. Each box plot indexes the 10th, 25th, 50th (median), 75th, and 90th percentiles. An additional 16 samples without lot numbers, but obviously distributed among the 6 recalled lots, and 2 samples from TBF lots produced prior to September 2007, were also submitted and analyzed. Results for all 54 samples are summarized in Table 2.

Figure 2

Dose responses measured in the toenail biologic monitor relating peak (A) and restored-baseline (B) Se concentrations (μg/g) to total Se exposure (total milligrams Se consumed). The peak TNSe concentration (μg/g) could be measured in 29 subjects who submitted dose-consumption information and either monthly samples or onycholysitic nail fragments that could be segmented to give a chronological profile. As expected the peak TNSe concentration was directly correlated with Se exposure (A). The negatively-correlated temporal response (B) for the restored-baseline TNSe concentration relative to exposure was measured in 45 subjects including the initial group and those subjects who had either joined the study after their TNSe concentration peak had passed or were unable to collect monthly samples in the early phase of the study but were able to submit “new growth” samples as their nail growth was restored.

Figure 3

Temporal response in toenail Se concentration (μg/g) from monthly toenail clipping samples relative to Se exposure (6 to 37 doses; 149 to 926 mg Se) from misformulated TBF supplement products. The temporal response in toenail Se (TNSe) concentration for 5 representative subjects ranging in Se exposure from 6 to 37 doses (149 to 926 mg Se) who were able to submit monthly samples is somewhat variable in the breadth of the elimination period, the number of peaks, and the lapsed time from last use of the misformulated product to the peak TNSe concentration (Δ days). The peak TNSe concentration is highly correlated with total Se intake from the TBF products (r = 0.96, p = 0.01) in these 5 subjects. The median lapsed times spanning the last exposure to the TNSe peak concentration and the restored-baseline concentration, were 237 and 411 days, respectively.

Figure 4

Temporal response in fingernail Se concentration (μg/g) from segmented onycholysitic fragments from all ten fingernails in two subjects who consumed 44 (Subject A) and 32 (Subject B) doses of misformulated TBF supplement products. These two subjects lost fragments from all 10 FNs that were approximately half the length of the entire nail. In the graph legends, the nails are identified as L (left), R (right) and 1 through 5 as thumbnail, index FN, middle FN, ring FN and little FN, respectively. These fragments were subdivided along the arc defined by the distal end of the fragment. Each segment was approximately 1.5 mm and was analyzed separately for Se. Each subject reported the TBF product lot numbers, total number of 1-ounce doses consumed, and the use period. From these data we compute that Subject A consumed 1,104,340 μg of Se over a 50 day period and Subject B consumed 831,456 μg over a 34 day period. The nail Se concentrations for the two subjects reflect their exposures (Subject A > Subject B; p = 0.0007). For both subjects, the highest Se concentrations were found in the ring FN (LR4). Then the Se concentrations followed LR5 > LR3 > LR1 > LR2 and LR3 > LR5 > LR1 > LR2 for Subjects A and B, respectively.

Figure 5

Post-exposure, post-peak temporal response in new-growth fingernail and toenail Se concentrations (μg/g) in Subject A (A) and in comparison to 2 other unexposed male subjects who do not use Se supplements (B). The post-exposure restored baseline for Subject A has been measured in new-growth FN and TN samples collected twice-a-month for FNs and once-a-month for TNs over approximately 4 years. Each FN and TN data point in (A) is the mean of 10 individual FN and TN clipping samples taken individually from each nail and analyzed separately. Done in this way each data point integrates dietary Se intake over an estimated 2 weeks for FNs and 4 weeks for TNs in the context of the temporal response profiles. For both FNs and TNs the post-exposure restored baselines cycle with a near annual period while remaining below the normal Se concentration expected for this subject (0.85 μg/g). As expected, the TN monitor lags somewhat behind the FN monitor. The post-exposure restored baselines for Subject A (male, Caucasian, non-smoker) are compared to 2 male subjects of comparable age, race, diets, and BMI in (B). The 2 comparison subjects have, for over 20 years, participated in a Missouri study of Se status in which they provide FN and TN samples on a quarterly schedule. The quarterly FN and TN samples include clippings from all FNs in the FN sample and all TNs in the TN sample. Representative aliquots for analysis are prepared from the combined FN and TN quarterly samples. Neither Caucasian comparison-subject smokes cigarettes or takes a Se supplement. There has been a gradual increase in both FNSe and TNSe concentrations observed in both comparison subjects over the 20-year study period attributable to increasing dietary Se intakes.

Figure 6

Initial and continuing symptoms of Se toxicity. Seventy-three subjects returned an August 2010 questionnaire inquiring about the impact of their consumption of the misformulated TBF supplement products on their hair and nails and their persistent Se toxicity symptoms. At 2.50 ± 0.14 years follow-up, these 73 subjects reported 1 to 27 persistent symptoms they attributed to Se toxicity; mean = 11.5 ± 6.5; median = 10.