Nail Analysis for Drugs: A Role in Workplace Testing? (2023)

Virginia A Hill,

Article history

Abstract

Introduction

Analysis of nails for detection of prior use of medications or drugs of abuse has become sufficiently widespread to warrant a recent review (1). Drugs can be found on or in nails from four sources: external contamination; drugs in sweat of a drug user; from circulation of drug in the blood during nail growth horizontally from the nail matrix or lunula; and from circulating drug in the nail bed feeding the ventral surface of the nail as it grows vertically, thickening the nail (1).

Nails may have some of the features of hair, such as trapping drug from blood in the follicle (for hair) or lunula (for nails) during growth of the tissue. With hair, this feature facilitates the accumulation of drug in the hair over time, which then results in the ability to approximate the amount of drug used during the time interval in which the tested hair was growing (2). Such estimations, however, depend on the removal of environmental and sweat-derived drug along with further criteria to determine that the drug found in the sample is a result of ingestion rather than external contamination (3–8). With hair, this determination also requires an ability to characterize the condition of the hair—i.e., whether it has been damaged by common or excessive cosmetic treatments (9).

With nails, to our knowledge, there are no published methods proven to distinguish externally derived drug from drug present due to ingestion. Common nail washing methods include short alcohol, water, acetone or dichloromethane rinses at room temperature (1).

This paper examines some aspects of nail analysis as they may, or may not, differ from hair analysis. These include swelling of nails in water; response of nails to environmental contamination in the form of soaking in aqueous drug solutions; and the ability of two different washing methods to distinguish drugs present in nails due to ingestion from drugs present due to sweat or environmental contamination.

Experimental

Nail samples

Nail samples used in these studies were anonymous stored samples from workplace testing, which had been stored at room temperature in their original collection envelopes. Samples that were negative for drugs or presumptive positive for cocaine (COC) and methamphetamine (METH) were identified by screening assays using FDA-cleared microplate enzyme immunoassay (EIA) (10). Confirmation of presumptive positives was by liquid chromatography mass spectrometry–mass spectrometry (LCMS–MS), described below. No morphine (MOR) or heroin positive samples were identified by EIA screening.

Nails that were used in soaking experiments were previously found to be negative for drugs by EIA screening. Nails were not pre-washed prior to soaking, so that they were representative of nails as they are received in the laboratory for workplace drug testing.

Methods

Water uptake by nails

Drug-negative fingernail clippings from five different subjects were weighed in the dry state and again after 15, 30 and 60 min in water at ambient temperature. At the 60-min time-point, when weight gains were at a plateau, they were removed, blotted and allowed to air-dry for 2 h, with weights taken at 30-min intervals until they had returned to their original dry weights after 2 h of drying.

Soaking nails in solutions containing drugs

Drug-negative fingernail clippings from nine different subjects were soaked for 1 h in an aqueous solution of 5 μg/mL of COC, METH and MOR. After soaking, the samples were rinsed with water for about 5 s, blotted and dried for 2 h. Nails were not piled on top of each other for drying, but were spread apart such that no nail touched another nail. The dried nails were weighed for the decontamination experiments, i.e., 6–12 mg of nails for extended buffer washing for each drug and, if available, another 6–12 mg for methanol washing for each drug. Washes and washed nails were analyzed as described below. In another experiment, drug-negative fingernail clippings from nine different subjects were soaked for 1 h in an aqueous solution containing 10 times higher concentrations of drugs—i.e., 50 μg/mL of METH and MOR. A different set of drug-negative nails from eight subjects were soaked for 1 h in an aqueous solution of 50 μg/mL of COC. After soaking, the samples were rinsed in water and dried for 2 h and then, for each drug, 6–12 mg placed in a tube for washing by the extended wash method and, in most cases, the methanol method. Washes and washed nails were analyzed as described below.

Washing procedures

The aqueous wash procedure and the application of the wash criterion in hair analysis used in this work have been presented previously (3–7, 9). This method was applied to nails as follows. Place 6–12 mg nail in a 12 × 75 mm polycarbonate round-bottom tube and add 2 mL of the initial non-swelling solvent, dry isopropanol; shake for 15-min in a 37°C shaking water bath (110 oscillations per minute (OPM)). This is followed by three 30-min washes and two 60-min washes in 2 mL of pH 6, 0.01 M phosphate buffer containing 0.1% bovine serum albumin (BSA), also in a 37°C 110-OPM shaking water bath. Washes were saved for analysis. For EIA analysis, the isopropanol wash was evaporated under nitrogen (after acidification with 20 μL 0.1 N HCl, in the case of METH), and then reconstituted in pH 6, 0.01 M phosphate buffer containing 0.1% BSA.

The wash procedure was followed by analysis of the washes and calculation of a “wash criterion.” As with hair, the wash criterion for COC and MOR was calculated as follows: if the amount of drug in the last wash multiplied by 5 and then subtracted from the amount of drug in the nail results in a nail value below the cutoff (5 ng COC/10 mg; 2 ng MOR/10 mg), this indicates that the sample may be contaminated or porous. This calculation is essentially a mathematical estimate of washing for another 5 h, as an indication of whether or not all the drug would be washed out in that amount of washing; this is, of course, an overestimate of drug removal since the wash kinetic curve is not linear but tends to plateau; the method errs on the side of safety in protecting the subject from being reported as positive from environmental contamination (8, 10). For METH, the wash criterion was calculated by multiplying the last wash by 3.5 and the result subtracted from the amount of drug in the nail (5). A result less than the cutoff (5 ng/10 mg sample) indicates that the sample may be contaminated (3, 6).

Rinsing nails with methanol was performed as described in (11). The method described was to place 100 mg nails in a 13 × 100 mm tube, add 3 mL methanol and vortex mix for 15 s. This was repeated twice more, and the washes saved. Our application of this method consisted of washing 6–12 mg of nails in 1 mL of methanol in a polycarbonate round-bottom tube, and vortexing for 15 s, three times. The methanol wash solutions were evaporated (for METH samples, after acidification with 10 μL 0.1 N HCl added per mL methanol wash solution), and reconstituted in pH 6, 0.01 M phosphate buffer containing 0.1% BSA.

The total amount of drug taken up by each nail was determined by adding the drug in all the washes and the nail. The percent of drug remaining in the nail after washing was calculated by dividing the amount of drug in the nail after washing by the total drug taken up by the nail.

Analysis of washes

Washes from the in vitro experiments were analyzed by quantitative chemiluminescent enzyme immunoassay. The assays were modified FDA-cleared enzyme immunoassays using microplates coated with BSA-conjugated antigens (10). Fifty μL of the 2-mL wash solutions were added to the microplate wells, followed by antibody against either COC, MOR or METH. After incubation, the wells were washed and second antibody labeled with horseradish peroxidase was added. After incubation, the wells were washed and chemiluminescent substrate (ThermoScientific Pierce, Product 37074, Rockford, IL) was added and incubated. Plates were read immediately in a Biotek Luminescence Reader (Winooski, VT 05404, USA).

Characteristics and validation results for the EIA assays are shown in Tables Ia and Ib. Cross-reactivities of analytes related to the parent analytes are shown in Table Ia, as percent cross-reactivity relative to the parent; only those analytes that showed significant cross-reactivity are shown. The ranges of the assay curves (Table Ib) were COC, 0–1,000 pg; METH, 0–420 pg; MOR, 0–125 pg. These values (in pg/50 μL) have the following values for a 2.0 mL wash solution of a 10-mg sample: COC, 0–40 ng/10 mg; METH, 0–16.8 ng/10 mg; MOR, 0–5 ng/10 mg. Limits of detection (LOD) for each assay were COC, 20 pg/50 μL or 0.8 ng/10 mg nail; METH, 8.5 pg/50 μL or 0.34 ng/10 mg; MOR, 6.25 pg/50 μL or 0.25 ng/10 mg. Intra-assay precision (% CV at each calibrator value, 10 replicates per value) for each assay over the ranges of the curves was COC, 4.1–7.3; METH, 2.4–10.4; MOR, 3.6–6.0. Inter-assay precision (% CV at each calibrator value, 10 replicates per value), performed over 3 days, was COC, 3.8–10.6; METH, 2.2–9.9; MOR, 4.9–9.3. Accuracy was determined by comparing EIA quantitations of washes with their LCMS–MS quantitations and performing a regression analysis to determine agreement. The R-squared values were COC, 0.990 (n = 99); METH, 0.994 (n = 313); MOR, 0.963 (n = 8).

Extraction of COC, opiates and METH from nails

COC and MOR for LC/MS/MS confirmation were extracted from 6 to 12 mg nails in 1.1 mL of a methanol and trifluoroacetic acid (90:10 by volume) solution plus 100 μL of internal standard solution in glass tubes subjected to 60 min in a microwave oven at 90°C. The internal standard (I.S.) solution for the COC assay contained 5 ng/100 μL cocaine-d3, benzoylecgonine-d3, cocaethylene-d3, norcocaine-d3, meta-hydroxycocaine-d3, and 0.4 ng/100 μL para-hydroxycocaine-d3, and ortho-hydroxycocaine-d3. Although routinely included in the I.S. mixture, the hydroxycocaine isomers were not performed in these experiments. The I.S. solution for opiates contained 5 ng/100 μL codeine-d6, morphine-d6, 6-monoacetylmorphine-d6, oxycodone-d6, hydrocodone-d6, hydromorphone-d6 and oxymorphone-d3. Internal standards were obtained from Cerilliant (Round Rock, TX 78665, USA), with the exception of hydroxycocaines which were obtained from ElSohly Laboratories (Oxford, MI 38655, USA).

Samples were tightly capped and microwaved for 60 min at 90°C. The extracts were placed on cation exchange DPX tips (DPX Industries, Columbia, SC, USA), which were preconditioned with methanol.

For the COC extraction, the sample was applied to the DPX tip followed by two 600 μL washes with 0.1 M HCl and then two methanol washes followed by elution with 550 μL dichloromethane, isopropanol and ammonium hydroxide (80:20:3 by volume) solution. Eluted extracts were dried for 10 min at 40°C under nitrogen gas and reconstituted in 300 μL of a water, acetonitrile and formic acid (80:20:0.08 by volume) solution.

For the MOR extraction, the sample was applied to a cation exchange DPX tip followed by two 600 μL washes of 0.1 M acetate buffer pH 4, one wash with methanol and aqueous 0.1 M HCl (40:60 by volume) mixture, and then one methanol wash. The nail extracts were then eluted with 550 μL dichloromethane, isopropanol and ammonium hydroxide (80:20:3 by volume) solution. Eluted extracts were dried for 10 min at 40°C under nitrogen gas and reconstituted with 225 μL of a 0.1% formic acid in water solution.

Amphetamines were recovered from nails by incubating 6–12 mg of nails in 0.9 mL of a water, methanol and aqueous 0.1 M HCl (70:20:10 by volume) solution and 100 μL of internal standard solution in glass tubes. The I.S. solution for amphetamines contained 5 ng/100 μL amphetamine-d8, methamphetamine-d11, 3,4-methylenedioxyamphetamine-d5, 3,4-methylenedioxy-N-methamphetamine-d5 and 3,4-methylenedioxy-N-ethylamphetamine-d6. Internal standards were obtained from Cerilliant (Round Rock, TX, USA). Samples were tightly capped and microwaved for 30 min at 90°C. Cooled extracts were added to 200 μL of 0.5 M sodium hydroxide in new 12 × 75 mm tubes. The reverse phase DPX tips were preconditioned with a water and methanol (70:30 by volume) solution. The nail extracts were placed on the reverse phase DPX tip and washed with 600 μL water. Samples were eluted with 550 μL of 0.01 M HCl in methanol. Fifty μL of 0.1 M HCl in methanol was added to each eluted extract before it was dried for 15 min at 45°C under nitrogen gas. The extracts were reconstituted with 150 μL of a water and methanol (90:10 by volume) solution.

Confirmation by LC/MS/MS

All confirmations were performed by LC/MS/MS on an instrument consisting of a Shimadzu LC-20AD binary pump system (Shimadzu, Columbia, MD, USA) coupled to an API 3200 (Sciex, Framingham, MA, USA) and utilizing a LEAP PALHTC-xt autosampler system (LEAP Technologies, Carrboro, NC, USA).

For COC confirmations, chromatographic separation was performed on a Phenomenex Kinetex® 5 μm Biphenyl 100 Å 50 mm × 3.0 mm column (Phenomenex, Torrance, CA, USA) at a flow rate of 0.500 mL/min with isocratic elution and a 2.0 min cycle time. The injection volume was 10 μL. The mobile phase consisted of 33% acetonitrile with 0.1% formic acid and 67% water with 0.1% formic acid. The instrument was operated in positive electrospray, multiple reaction monitoring mode (Table IIa). Mass resolution on Q1 and Q3 was set to unit resolution. The interface heater was on, the ion spray voltage was 5,500 V and the source temperature was 450°C. The curtain gas was set to 45 psi, and ion source Gases 1 and 2 were set to 50 and 40 psi, respectively. Hydroxycocaine isomers were not analyzed in this study.

For amphetamines confirmations, chromatographic separation was performed on a Phenomenex Kinetex® 5 μm Biphenyl 100 Å 100 mm × 3.0 mm column at a flow rate of 0.700 mL/min with gradient elution. The injection volume was 10 μL. The mobile phases were 5 mM ammonium formate in water (MPA) and methanol with 0.1% formic acid (MPB). The gradient started at 10% MPB from 0 to 0.5 min, adjusting to 40% MPB from 0.5 to 0.7 min and remaining at 40% MPB until 2.0 min. The MPB concentration then increases to 70% MPB from 2.0 to 2.2 min, remaining at 70% MPB until 2.99 min. The MPB percentage then drops to 10% from 2.99 to 3.0 min and remains at that concentration until the end of the cycle at 3.75 min. The instrument was operated in positive electrospray, multiple reaction monitoring mode (Table IIa). Mass resolution on Q1 and Q3 was set to unit resolution. The interface heater was on, the spray voltage was 2,500 V and the source temperature was 550°C. The curtain gas was set to 40 psi, and ion source Gases 1 and 2 were set to 50 and 60 psi, respectively.

For opiates confirmations, chromatographic separation was performed on a Phenomenex Kinetex® 5 μm Biphenyl 100 Å 100 mm × 3.0 mm column with a variable flow rate and gradient elution. The injection volume was 20 μL. The mobile phases were water with 0.1% formic acid (MPA) and methanol with 0.1% formic acid (MPB). The initial flow rate was 0.43 mL/min until 3.2 min increasing to 0.55 mL/min from 3.2 to 3.5 min, then decreasing to 0.43 mL/min from 3.5 to 3.65 min and maintaining that flow rate until the termination of the cycle at 4.3 min. The gradient started at 25% MPB until 0.25 min, increasing to 70% MPB between 0.25 and 3.19 min. It then increased to 85% MPB between 3.19 and 3.2 min and maintained that composition until 3.35 min. From 3.35 to 3.36 min, the percentage of MPB dropped to 25% and maintained that composition until the end of the cycle at 4.3 min. The instrument was operated in positive electrospray, multiple reaction monitoring mode (Table IIb). Mass resolution on Q1 and Q3 was set to unit resolution. The interface heater was on, the spray voltage was 2,500 V and source temperature was 500°C. The curtain gas was set to 40 psi, and ion source Gases 1 and 2 were set to 45 and 50 psi, respectively.

Validation of extraction and confirmation procedures

Validation experiments were performed for the assays for COC, benzoylecgonine (BE), cocaethylene (CE) and norcocaine NOR) (Table III); Opiates, including MOR, codeine (COD), oxymorphone (OXYM), oxycodone (OXYC), hydromorphone (HYDM), Hydrocodone (HYDC) (Table IV); and amphetamines, including METH, amphetamine (AMP), 3,4-methylenedioxy-amphetamine (MDA), 3,4-methylenedioxy-N-ethylamphetamine (MDEA) and 3,4-methylenedioxymethamphetamine (MDMA) (Table V). The validation experiments were performed according to the Scientific Working Group for Forensic Toxicology (SWGTOX) guidelines (12). The following performance characteristics were evaluated: ionization suppression/enhancement (matrix effect), carryover, LOD, limits of quantitation (LOQ), accuracy and precision. The LOD and LOQ were experimentally determined by analyzing five sets of calibrators and controls. The LOQ was chosen as the lowest calibrator that could be determined with accuracy (within ± 20%); the LOD was administratively set as the lowest calibrator.

Results

Water uptake by nails

Weight gain by nails when placed in water was about 15–30% of the original weight of the nails (Figure 1). These data concur with estimates from literature, where weight gain of nails immersed in water for 2 h was 22% (13). Nail thickness has been reported to be a primary source of variability in nail permeation (12), such as between thumbnails and fingernails (14). Weight gain in water, as well as drying times, has been reported to be similar for hair and nails (15). After 2 h of drying, nail weights were –2.0 to 4.5% of the original weights.

The 15–30% weight gain by the nails translates to 1.5–3 mg (or μL) of water/10 mg nail. If the water had contained 5 μg/mL of drug, the amount of drug that would be expected to be carried into the nail would be 7.5–15 ng/10 mg nail. And, of course, the more drug in the solution, the more diffusion of drug into the nail; at 50 μg/mL, the drug in the nail would be at least 75–150 ng/10 mg nail. These estimates of drug uptake relative to water uptake inform expected results of the following experiments of soaking nails in 10-fold different concentrations of drugs.

Drug uptake by nails soaked in solutions containing drugs

Nails soaked in 5 μg/mL of COC, METH, MOR

The nails soaked in 5 μg/mL of drugs and then washed by the extended buffer wash and the methanol wash were analyzed by LC/MS/MS. The total drug taken up by the nails and the amounts remaining in the nails after extended buffer and methanol washing are shown in Table VI. Nearly all of the drug taken up by nails during the soaking was washed out by the extended buffer wash, leaving averages of 0.9, 0.7 and 0.7 ng/10 mg nail, or 1.4%, 2.2% and 2.3%, of METH, MOR and COC, respectively. The methanol-washed nails, where there were fewer samples, contained averages of 15.4, 12.1 and 15.4 ng/10 mg nail, or 33.8%, 48.1% and 52.2% of METH, MOR and COC uptake, respectively. The uniformly low amounts of drug in the buffer washed samples occurred in spite of the wide range of total drug absorbed by the nails: 12.8–144, 4.6–47.3 and 4.1–243 ng/10 mg nail for METH, MOR and COC, respectively. Some of the amounts of drug far exceeded the amounts predicted by just the amount of drug in the 15–30% water absorbed by the nails shown in Figure 1. All but one (#3 for METH) of the methanol-washed samples would be positive, while all but one of the buffer-washed samples (#5 for MOR) were below the cutoffs used for these studies (5 ng/10 mg hair for COC and METH, and 2 ng/10 mg for MOR). Sample #5 for MOR would be indicated as contaminated using the wash criterion: 2.1 – (5 × 0.8) = −1.9 (last wash values are not shown in Table VI). No BE was detected in any of the nails after either wash method.

Table VII shows the drug uptake and effects of extended buffer and methanol washing of nail samples soaked at 10 times higher concentrations than those in Table VI. As in the lower concentration soaking, the nails showed large variations in the total amount of drug absorbed during the soaking, ranging from 284 to 4,455 ng/10 mg nail for METH; 198 to 2,707 for MOR; and 382 to 2,710 for COC for the extended buffer-washed samples. With nails, the cause of increased drug uptake is not obvious. With Samples 11–19, there tends to be more variation among different subjects than among different aliquots of the same subject’s nails. For example, all aliquots of Subjects #13 and #17 showed thousands of nanograms total uptake, and all aliquots of #s 12, 14 and 15 showed uptake in hundreds of nanograms. Exceptions occur, however—for example, with #11 one aliquot took up 2,315 ng/10 mg, while the other three aliquots took up in the hundreds of nanograms. As stated above, nail thickness has been reported to be a primary source of variability in nail permeation (12). In our work, we did not evaluate thickness. The average amount of drug left in the extended wash samples was 2.1%, 3.2% and 2.7% for COC, METH and MOR, respectively. No BE was detected in the nails. Regardless of the degree of uptake, the soaked nails washed by the extended buffer wash were decontaminated to below the cutoffs or identified as contaminated by the wash criterion described in Methods.

Drug uptake and results of methanol washing of nails soaked in 50 μg/mL of drug solutions are also shown in Table VII. Since these nails were from the same donors and were soaked simultaneously in the same solutions as those washed with the extended buffer wash, the uptakes are similar, although not identical. The drug uptake of these samples was 225–2,341 ng/10 mg nail for METH; 131–2,652 for MOR; and 328–1,266 for COC. The average amount of drug left in the methanol rinsed samples was 57%, 48% and 70% of the total uptake for COC, METH and MOR, respectively, with no BE detected in the COC samples. All of these samples of the methanol-washed samples would be identified as positive, even though they are known to be contaminated.

Washing of authentic workplace presumptive positive samples

Results of extended aqueous washing of unknown workplace nail samples containing COC (Table VIII) and METH (Table IX) showed higher percentages of total drugs remaining in the nails in some cases after extended washing than in the controlled contamination studies described above. If the wash criterion is applied in these cases, all of these samples would be reported as contaminated. Since these were unknown samples, unlike in the controlled in vitro soaking studies, it is completely unknown if, how long, or at what concentrations such samples may have been exposed to external drug. While BE was present in these cases, ranging from 11.6 to 248% of COC, no other metabolites were detected. It has been suggested that BE or BE/COC ratios in hair are not useful as metabolic indicators due to in vitro formation of BE in COC-contaminated samples where there was no ingested COC (8, 16) as well as hair from known COC users not containing 5% BE (4). The same is likely true with nails; also the BE in Samples 40–46 is highly variable, as might be expected from in vitro formation.

In the five presumptive METH-positive cases, the nails contained higher percentages of the total METH in the samples after extended washing (from 17.7 to 55.6% of total drug) as compared with the in vitro contaminated samples (average of 3.2%). If the wash criterion is applied to these samples, all but one (#58) would be reported as contaminated. The AMP metabolite was identified in all of the samples, ranging from 6.6 to 15.9% of METH. Because AMP may be present in consumed or “street” METH (17, 18), AMP can only be used as a potential metabolic indicator of ingestion when the analysis is performed on samples that have been washed and evaluated by the extended wash procedure and wash criterion. For comparison, the amount of AMP as percent of METH in 38 hair samples from known METH users averaged 9.2 (range 1.4–17.9; median 8.5) (5).

Discussion

Uptake of water and drug in solution by nails

It is well known that water readily permeates the nail (13), along with small hydrophilic molecules dissolved in water (13). High individual variability in nail permeation is also known, and also great variations among locations, such as thumb vs. long finger vs. toenail (14). Some of these variations may relate to nail thickness (12). Permeation may also depend on pH in some cases, as well as temperature (13). Our demonstration of water uptake showed 15–30% weight gain after 1 h soaking (most of the increase had occurred at 30 min), far less variation than seen in the drug uptake results. If the only drug to be absorbed by the nail were the content of 1.5–3 μL water (15–30% of 10 mg nail), the total amount of drugs in the nail would be 7.5–15 ng for the 5 μg/mL soaking and 75–150 ng from the 50 μg/mL. Yet the range of drug taken up by the nails in the 5 μg/mL soaking study was from 4.1 ng/10 mg nail to as high as 243 ng/10 mg nail for all drugs, METH, MOR and COC. The range of drug uptake by the nails in the 50 μg/mL study was from 131 ng/10 mg nail to 4,455 ng/10 mg nail for all drugs, METH, MOR and COC. Visual appearance of the nails did not reveal any obvious causes of the variations (see online supplementary material for photos of some of the nails). From this exercise that showed a fairly tight range of water uptake (15–30% of nail weight) but very wide ranges of drug uptake in the nails under conditions of two concentrations, it becomes apparent that effects of exposure of nails to drugs in the environment or to drugs in the sweat of a user can be severe and unpredictable. Overall, the pattern in Table VII is that nails of a particular subject tended to perform similarly with different drugs—e.g., at the low end of the range, Nail #12 absorbed 392 ng METH/10 mg nail and 391 ng MOR/10 mg nail. At the high end of the range, Nail #17 absorbed 4,455 METH/10 mg nail and 2,969 ng MOR/10 mg nail. (Nails used for the 50 μg/mL study of METH and MOR were from the same subjects, as indicated by the sample numbers. Samples in the 50 μg/mL COC study were from different subjects.) However, there were exceptions to this uniform behavior of nails from the same subject (e.g., #18 Buffer washed MOR in Table VII).

Previous research has shown large differences between nail clippings of PCP users taken from left and right hands or feet (19). For the same subject, the right foot nail was 1.55 ng PCP/mg and the left foot nail was 9.74 ng/mg, a 6-fold difference. For another subject, the right-hand nail was 21.2 ng/mg and the left foot nail was 147.9 ng/mg, a 7-fold difference. In a paper on COC in nails by Garside et al. (20), one subject’s foot and hand nails contained 2.83 and 0.51 ng COC/mg, respectively. Another subject’s foot and hand nails contained 2.13 and >10 ng/mg, respectively, and a third subject had 1.18 and 13.1 ng/mg in foot and hand, respectively. These are differences between foot and hand of 5.5, >4.7 and 11.1-fold, sometimes hand greater than foot and sometimes the opposite. Another publication (21) reported a mean COC concentration in hand nails from a group of subjects of 198 ng COC/mg (right) and 145 ng/mg (left), while the mean concentrations in foot nails were 6.3 ng/mg (right) and 8.7 ng/mg (left)—a large discrepancy in the same people. In the same paper, the values for MOR concentrations in hand nails were 47.7 ng MOR/mg (right) and 43.5 ng/mg (left), while the mean concentrations in foot nails were 1.4 ng/mg (right) and 1.6 ng/mg (left)—also a large discrepancy. The wash procedures used for these various clippings comparisons were minimal, essentially rinsing (e.g., 15 s) with water, mixed water and methanol, or methanol. The methanol wash method applied in this work, since it did not decontaminate known contaminated samples, could result in such disparities.

That the disparities described above were more likely due to external contamination or unknown factors other than nail growth characteristics is supported by studies that show nail growth rates at different sites to differ only about 2-fold. One study showed: (1) average fingernail growth rate is about twice that of toenails (3.47 vs. 1.62 mm/month, P < 0.01) and (2) no significant difference between right and left fingernail/toenail growth rates (22). The high likelihood for fingernail contamination by people who handle the drugs they consume is obvious. The difficulty of removing contamination from fingernails is exemplified by an example of microbes lodged between the overhanging free nail and the skin adjacent to it (the hyponychium), a reservoir for microbes (23), and by extension, other contaminants. It was found that after 10 min of scrubbing the fingers with povidone iodine and then culturing the nail clippings for bacteria, yeasts and molds, 19 of 20 subjects showed presence of staphylococcus epidermidis; seven had an additional bacteria, eight had molds and three had yeasts.

Distinguishing external contamination from ingestion-derived drug is possible for hair at least partly because of the two-compartment structure of the hair. First, the hair cuticle, in itself a complex laminal structure consisting of four sub-layers (s-layer, exocuticle, endocuticle and cell–membrane complex, which is further composed of four more layers), forms an outer barrier to both penetration of external drugs into the hair and loss of drugs during an extended wash procedure (24). The cuticle surrounds the second major compartment of the hair fiber, the inner cortex, with its major components of macrofibrils which are in turn composed of intermediate filaments in a protein matrix (25). Some analogies between hair and nail are offered in (23), all of which recognize only one compartment in nails. A diagram of the nail structure is shown in Figure 2. One model suggests that the nail unit is comparable in some respects to a hair follicle, sectioned longitudinally and laid on its side, the hair bulb considered analogous to the intermediate nail matrix and the cortex to the nail plate. Another suggestion is that the nail unit could be seen as an unfolded form of the hair follicle, producing a hair with no cortex, just cuticle. Scanning electron microscopy of the nail confirms that its structure is more similar to compacted cuticular cells than cortical fibers. A third model could represent the nail unit as a form of follicle abbreviated on one side, providing a modified form of outer root sheath to mold and direct nail growth. With only one compartment, a nail can be compared to a hair with its cuticle entirely destroyed, as evidenced by the ready uptake and washout of drug by nails that resemble the effects of soaking and washing damaged hair in (4). In the washing experiments performed in these studies of authentic presumptive positive nail samples, there was no evidence of a compartmental barrier to penetration of wash medium which would allow distinguishing of external from ingested drug.

In this work, with methanol rinsing, about half of the contaminating drug remaining in all of the contaminated nails, for all drugs, at low and high soaking concentrations, which, therefore, rules out methanol rinsing as applied in this work as a means of decontaminating, since all contaminated samples were above the cutoffs. The extended buffer washing method and wash criterion used for hair analysis to identify external contamination was evaluated for its utility with nails. Decontamination of in vitro contaminated samples with this method was very successful, identifying all samples as contaminated using the wash criterion. However, all of the presumptive positive authentic samples, seven COC and five METH samples, also were identified as contaminated by the wash criterion. Either all of these samples were merely contaminated, or nails that are positive from ingestion cannot be distinguished from those that are externally contaminated.

This difference between hair and nails can be illustrated by comparing the washing of a COC-user hair (Figure 3) and an in vitro contaminated hair (Figure 4) (3, 26–28) with a (possible) user nail (Figure 5) and an in vitro contaminated nail (Figure 6). Figure 3 vs. 4 shows the presence of ingestion vs contamination compartments in hair. Figure 2 charts a typical washing result for an intact, COC-positive hair that contained 154 ng COC/10 mg hair after washing away 28.6 ng/10 mg hair by the six washes (3.75 h) of the extended wash procedure. After washing for 3.75 h, there remains the large amount of drug that would not wash out even if the washing continued for another 5 h. Applying the last lash criterion (subtracting 5 times the 5.7 ng COC/10 mg in the last wash from the 154 ng COC/10 mg in the hair (154 – (5 × 5.7) = 125.5 ng/10 mg hair) indicates the positive result is not likely due to contamination. On the other hand, Figure 3 charts the results of washing an in vitro COC-contaminated sample (3). In this case 632 ng COC/10 mg hair was removed by washing, leaving only 15.4 ng COC/10 mg in the washed hair. Applying the wash criterion gives 15.4 ng/10 mg hair – (5 × 6 ng/10 mg hair) = −14.6, which indicates the result is due to contamination.

Figure 5 charts the results of the extended washing and analysis of one of the seven workplace presumptive COC-positive nail samples shown in Table VIII (#41). In this case 73.5 ng COC/10 mg nail was removed by washing, leaving 5.9 ng COC/10 mg in the washed nail. Applying the wash criterion gives 5.9 ng/10 mg nail – (5 × 3.4 ng/10 mg nail) = −11.1, which if it were hair, would indicate that the result is very likely due to contamination. All seven COC presumptive positive nails would be designated contaminated by the wash criterion. Figure 6 charts the washing results of an in vitro COC contaminated nail sample (#22, Table VII), which would be determined negative by the hair wash criterion, as would all of the in vitro-COC-contaminated nails. The washout curves of all seven of the workplace presumptive-COC-positive samples, using the hair extended wash procedure, are similar to those of the contaminated samples, demonstrating that the extended wash method developed for hair analysis is not a means of discriminating contamination from ingestion in nails, assuming that at least some of the presumptive positive nails in the study were from drug users.

Four of the five METH-presumptive workplace samples washed by the extended wash method would be determined contaminated by the wash criterion. Only one sample (#58) would be positive by the wash criterion: 15.1 ng/10 mg nail – (3.5 × 2.7 ng/10 mg nail) = 5.7 ng/10 mg nail, positive at either a three or 5 ng/10 mg cutoff. While the extended wash method determined all of the METH-contaminated samples to be contaminated, it also called four out of five workplace presumptive-METH-positive samples contaminated.

The one METH nail sample that would be called positive by the extended wash is so by use of criteria for hair (cutoff of 3 or 5 ng/10 mg using wash criterion of (drug in nail minus (3.5 × drug in last wash). Hair cutoff criteria may be inappropriate, however, since slower nail growth rates relative to hair may double the amount of drug deposited in a given weight of nail (15, 22, 29). If the cutoff for nail clippings, for example, should be 6 or 10 ng/10 mg nail, as the slower growth rate from the lunula might indicate, then this sample would be negative (contaminated) by the wash criterion. On the other hand, incorporation of COC and COD in unwashed nail shavings has been shown to be much less (per mg) than incorporation into unwashed hair. In a controlled dose study with COC and codeine, where only nail shavings were studied and growth from the lunula was not sampled, hair incorporated 5–30 times more drug than the shavings (30). These results suggest that the main source of ingestion-deposited drug in clippings may be the lunula. After washing with one 15-min isopropanol and three 30-min buffer washes, the hair in this controlled dose study retained about 80% of the drug and the nails nearly zero, consistent with the washing results in the present study.

Aside from the difficulty of distinguishing between contamination and ingestion, relating nail results to time of use is less determinate than for hair. A fingernail clipping can only indicate a wide ranging time of use, since nails come in many lengths, even on the same subject. When a clipping arrives in the lab, the length of the nail from which it was clipped is unknown. If the drug present is from ingestion, it is unknown whether it was deposited from the nail bed or from the lunula. If it was deposited from the lunula, the nail growth rate of 3 mm/month suggests that it could have been deposited 3–5 months previous. Or if deposited from the nail bed, it may derive from more recent ingestion (2, 11, 30).

A number of authors have stated that an absence of melanin in nails is an advantage of nails over hair. It is true that, except under pathological conditions, Caucasians do not have melanin in their nails (23). On the other hand, melanonychia (defined as the presence of melanin in nails) occurs in darkly pigmented individuals such as Black, Asians, Hispanics and Middle Easterners, and increases with age (23, 31). Specifically, it occurs in 77% of Black people by the age of 20, in almost 100% by the age of 50 and it occurs in 10–20% of Japanese by the age of 20 (23, 27). With extended washing, this laboratory has reported no influence of varying amounts of melanin in large populations of workplace hair testing subjects (32), and the same may well be true of nails. All of the contaminated nail samples in the current study were successfully identified as contaminated by the extended wash criterion regardless of color. However, the washing procedures we applied could not distinguish between known contaminated samples and authentic presumptive positive samples.

Conclusion

What is the role of nail analysis in testing for drugs of abuse? The studies in this report failed to identify a washing method for nails that distinguishes contamination from ingestion: the methanol rinse method used in this study failed to remove over half of contaminating drug, and the extended aqueous wash method used for hair removed drug similarly from contaminated and presumptive positive nails. Contributions from sweat-derived drug cannot be minimized, since extended washing required to do so removed nearly all drug whether ingested or external, from authentic METH or COC presumptive positives samples. Time of use is difficult to estimate without accurate measurement of distance from the lunula. Useful applications of nail analysis may be where contamination is highly unlikely—e.g., consumption of tablets or capsules as medication (if the fingers do not become contaminated) or where the question of time or quantity of use rather than the fact of use is being investigated. Even if a method is developed to distinguish between exposure and ingestion, or if the nail analysis is applied where contamination is unlikely, cutoffs for nails need to be evaluated. At this time, until further research demonstrates a method for identifying external contamination, a negative result is lack of evidence of exposure or ingestion, while a positive result may only be reported as exposure to drug, either by ingestion or from external contamination.

Supplementary Data

Supplementary data are available at Journal of Analytical Toxicology online.

References

© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Issue Section:

Article

Email alerts

• Web of Science
• Google Scholar

Citing articles via

More from Oxford Academic