ML18153A901
| ML18153A901 | |
| Person / Time | |
|---|---|
| Site: | Perry, Surry, Shoreham  |
| Issue date: | 03/14/1994 |
| From: | Bush L Office of Nuclear Reactor Regulation |
| To: | Bush D External (Affiliation Not Assigned) |
| References | |
| NUDOCS 9403290302 | |
| Download: ML18153A901 (7) | |
Text
..,.
/
e e
UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20555--0001 March 14, 1994 Dr. Donna Bush Chief, Drug Testing Section Division of Applied Research National Institute on Drug Abuse 5600 Fishers Lane (Room 9-A-54)
Rockville, Maryland 20857
Dear Dr. Bush:
Forwarded for your action in accordance with Section 2.8(e) of Appendix A to 10 CFR Part 26 are summarie~ of 17 reports of errors by HHS-certified testing laboratories and other relevant incidents involving drug testing.
Pertinent documents are attached as enclosures.
- 1.
Virginia Electric and Power Company's (VEPCO) letter dated June 2, 1993, concerns a blind performance test in which a spiked sample (~ 750 ng/ml methaqualone) was screened positive at the'5,t:Jrry Power Station's onsite testing facility and reported as negative (~ 750 ng/ml) by the Roche C.ompuChem laboratory.
In response to VEPCO' s request for a quantitative value of the specimen, CompuChem reported that a GC/MS test confirmed 607 ng/ml of methaqualone.
CompuChem said that test profile 725200 uses a methaqualone screening cutoff of 750 ng/ml, so a 607 ng/ml sample would screen negative.
Concurrently, another NIDA-certified laboratory tested an aliquot from the batch from which the blind sample had been taken, finding that the aliquot screened positive for methaqualone using an 800 ng/ml standard and confirmed positive at 1020 ng/ml by GC/MS.
(Enclosure 1)
- 2.
Southern Nuclear Operating Company's letter of June 28, 1993, concerns a known positive blind sample that was reported as negative by SmithKline Beecham Clinical Laboratories (SKBL), Atlanta, Georgia.
The cause was a suppression in discrimination above the cut-off for the THC-100 assay used by SKBL, which was subsequently isolated to channel one on the SKBL Olympus analyzer.
The corrective actions included switching channels on the Olympus analyzer and requesting that the manufacturer repair the machine.
(Enclosure 2.
NOTE: PART OF THIS ENCLOSURE IS, OR IS CLAIMED TO BE, PROPRIETARY MATERIAL.)
- 3.
Shoreham Nuclear Power Station's letter (Long Island Power Authority) of June 30, 1993, concerns a known positive blind sample being reported as negative by Roche Biomedical Laboratories, Raritan, New Jersey.
Roche's response stated that a retest indicated the presence of "an amphetamine compound" and that "the technologist performing the secondary FPIA rescreening handled the incorrect aliquot."
(Enclosure 3)
SOME OF THE ENCLOSED DOCUMENTS CONTAIN, OR ARE CLAIMED TO CONTAIN, PROPRIETARY INFORMATION
- -----~
9403290302 940314 ___ _
PDR ADOCK 05000280 P
PDR tst>>O --, rp
I
'\\
Dr. Donna Bush 4.
Georgia Power Company's letter of July 2, 1993, concerns a positive blind performance sample being reported negative by SmithKline Beecham Clinical Laboratories (SKBL), Atlanta, Georgia.
The investigation found that no error had occurred.
An interference on the M/Z 91 amphetamine fragment could not be resolved, so the test was reported as negative (SKBL uses the 91, 118, and 140 ions).
The blind specimen was validated to contain amphetamine, but was not specifically validated for the 91 ion fragment.
Methamphetamine was detected at the appropriate level, and amphetamine detected at the 118 and 140 ion fragments.
However, the 91 ion fragment could not be determined due to a compound (possibly a phenylamine) that interfered with the test.
(Enclosure 4.
NOTE: PART OF THIS ENCLOSURE IS, OR IS CLAIMED TO BE, PROPRIETARY MATERIAL.)
- 5.
Entergy Operations' letter of July 23, 1993, concerns D and L isomers of a blind performance sample test by Abuse Control Associates.
The sample was spiked at 1572 ng/ml amphetamine and 261 ng/ml methamphetamine, as confirmed by GC/MS.
The drugs used were DL amphetamine sulfate and methamphetamine HCl.
Abuse Control Associates reported that "any screening technique excluding one of the isomers in the process will exclude half of the value for amphetamine.
If this is coupled with either a low bias in machine calibration or a small degradation in spike level, the opportunity for conflicting results is enhanced.
To preclude this from becoming a regular occurrence, we have adjusted the spike targets and now use only D-amphetamine in formulation."
(Enclosure 5)
- 6.
Florida Power Corporation's letter dated August 9, 1993, concerns a blind sample containing methadone being reported as negative by the Doctors and Physicians Laboratory.
The follow-up investigation found that a random pipetting error failed to pick up the correct amount of urine specimen, resulting in a very low level of creatinine (3 mg/dl) and a negative test report. A requested retest correctly identified methadone at 735 ng/ml (42 mg/dl creatinine).
The error appears random, since all quality control specimens within the batch were determined to be positive or negative, as applicable.
(Enclosure 6)
- 7.
Pennsylvania Power and Light Company's letter dated August 10, 1993, concerns a positive blind specimen being reported as negative by the POLA testing laboratory.
Nichols Institute (the specimen supplier) confirmed the specimen as 458 ng/ml cocaine metabolite benzoylecgonine.
POLA replied that: (1) they detected the benzoylecgonine but could not report a positive because the sample failed their mass ratio criteria; (2) that they monitor ions 243 and 364 as quantifier and qualifier for internal standard and ion 259 as second qualifier for internal standard; and (3) that_ ion 259 is preferable to ion 85 because ion 85 is found in many compounds.
When POLA reanalyzed the sample by monitoring ions 243, 364, and 85, they found the sample to be positive.
For future ion ratio discrepancies on the internal standard 03 benzoylecgonine, POLA will reanalyze by monitoring ion 85 instead of ion 259.
(Enclosure 7)
SOME OF THE ENCLOSED DOCUMENTS CONTAIN, OR ARE CLAIMED TO CONTAIN, PROPRIETARY INFORMATION
-i Dr. Donna Bush 8.
Perry Nuclear Power Plant's (PNPP) letter of August 16, 1993, concerns a blind proficiency sample known to be positive for D-amphetamine (later verified by GC/MS) but reported as negative by CPF MetPath Laboratories, Cleveland, Ohio.
Investigation found MetPath had earlier changed from EMIT dau monoclonal to EMIT II monoclonal screening kits, without realizing this would affect the initial screening capability for amphetamine.
PNPP checked with three other labs that use the same kit as MetPath, but use different analytical equipment.
While the other labs had no problem with initial screening for amphetamine, they did have a concern about the ability to screen amphetamine at 1000 ng/ml.
PNPP concluded that the false negative blind sample result was due to a combination of the EMIT II Screening Assay Kit and MetPath's particular analytical equipment (the other labs used different equipment).
MetPath reported that GC/MS certified solutions of D-amphetamine DO NOT screen positive until concentrations of 1500 to 2000 ng/ml are present.
PNPP began using another HHS-certified lab and informed MetPath that their detection capability using EMIT II Screening Assay Kit was unacceptable.
MetPath stated they would go back to the EMIT dau monoclonal kit and perform testing to ensure that amphetamine would be initially screened at 1000 ng/ml.
PNPP will audit MetPath to verify amphetamine detection capability before reestablishing testing at MetPath.
(Enclosure 8)
- 9.
Commonwealth Edison's letter of August 20, 1993, reported test results for six nuclear power stations and Corporate Support personnel.
1,164 on-site test negatives were randomly selected fron 11,488 on-site tests (random, pre-access, and for-cause) and sent to the SAMHSA lab (MEDTOX Bioanalytical, Chicago, IL) as quality control checks.
One of the 1,164 specimens was found positive (cocaine).
Possible causes of the on-site false negative are variability of reagents, on-site test conditions, and differences in test equipment.
In the same period, 70 blind performance specimens were sent to MEDTOX, with all results as expected except for two spiked samples.. One lot of four samples was certified at 1394 ng/ml amphetamine and 1667 ng/ml methamphetamine.
The first two samples were tested by MEDTOX in December, 1992, and correctly found positive for amphetamine/methamphetamine.
In February, 1993, the third sample was reported positive by MEDTOX for amphetamine/methamphetamine and morphine.
The fourth sample was then screened on-site, found to be positive for amphetamine/methamphetamine and morphine, sent to MEDTOX, and confirmed positive for amphetamine/methamphetamine and morphine.
Spiked samples are prepared in the vendor's laboratory using in-house secured blank urines certified as negative.
The vendor reported that samples spiked with amphetamine/methamphetamine and morphine are not prepared unless specifically requested, that no such samples had been prepared in the last five years, and that none of the vendor's other clients had reported a problem with this same lot.
MEDTOX retested the third sample in March, 1993, and got the same results as before.
The electrolyte profiles of the first and third samples were then compared by MEDTOX and found dissimilar; the samples do not appear to have been prepared in the same urine matrix.
Four additional samples from this SOME OF THE ENCLOSED DOCUMENTS CONTAIN, OR ARE CLAIMED TO CONTAIN, PROPRIETARY INFORMATION
- . ~
Dr. Donna Bush same lot were then obtained from the vendor, tested by MEDTOX, and found positive for amphetamine/methamphetamine - with no morphine.
The unique morphine results are not considered by Commonwealth Edison as laboratory false-positive results but, instead, as technological or methodological errors.
(Commonwealth Edison advises that MEDTOX Bioanalytical of Chicago has recently closed and consolidated into MEDTOX Laboratories of St. Paul, MN.)
(Enclosure 9)
- 10.
Georgia Power Company's letter dated August 26, 1993, concerned five unsatisfactory blind performance tests done by SmithKline Beecham Clinical Laboratories (SKBL) of Atlanta, Georgia, over two months.
After the first false negative result, SKBL made various changes to their Olympus analyzer in an attempt to identify the error source, had an Olympus Company engineer perform a sample probe alignment, and added a special precision check after every calibration to verify probe alignment and accuracy.
SKBL also changed from a DB-5 to a DB-1 column in the gas chromatograph to separate amphetamine and B-phenethylamine.
The other four false negative results occurred during this investigation period (a second false negative amphetamine plus three false negative THC tests).
Comprehensive reports and data were provided by Georgia Power and Southern Nuclear Operating Company (the parent company acting as FFD contractor) and addressed false negative test results for both Vogtle and Farley nuclear plants.
(Enclosure 10.
NOTE: PART OF THIS ENCLOSURE IS, OR IS CLAIMED TO BE, PROPRIETARY MATERIAL.)
Southern Nuclear Operating Company submitted a letter dated August 30, 1993, which concerns the same false negative test results as reported for the Farley plant in the Georgia Power Company's letter of August 26, 1993 (above).
Both letters forward the same documentation package, so the Southern Nuclear package is not enclosed.
- 11.
Carolina Power and Light Company's letter of October 1, 1993, concerns a random test sample that initially screened negative.
The low creatinine level (8 mg/dl) prompted special Limit of Quantification (LOQ) GC/MS processing by Roche CompuChem, Research Triangle Park, NC.
The LOQ test was positive (782 ng/ml cocaine).
Two Level of Detection (LOO) analyses done by POLA in Princeton, NJ, were negative.
Two additional LOQ tests by CompuChem were also negative.
The false positive test result was for a sample which was analyzed by a special testing protocol that extends beyond the methods and practices approved by HHS and the basic testing requirements of the NRC FFD regulations.
Roche found four possible causes of the false positive: (1) improper use of Pasteur pipette, (2) not maintaining a clean extraction area for GC/MS processing, (3) an improper use of the extraction box, and (4) switching of specimens.
Roche concluded the fourth cause was least likely, because all of the samples in the batch were positive for cocaine.
Roche initiated corrective actions.
(Enclosure 11)
SOME OF THE ENCLOSED DOCUMENTS CONTAIN, OR ARE CLAIMED TO CONTAIN, PROPRIETARY INFORMATION
e Dr. Donna Bush
- 5
- 12.
Pacific Gas and Electric Company's letter of October 26, 1993, pertains to five unsatisfactory performance test results over four months.
All five specimens were supplied by Bensinger, DuPont, and Associates (BOA),
and tested by Roche CompuChem Laboratories, Research Triangle Park, NC.
CompuChem reported the following probable causes of incorrect results:
(1) Amphetamine false negative due to a change in the cross-reactivity of the test kit towards methamphetamine; (2) Cocaine false negative cause unknown; (3) Cocaine false negative caused by insufficient volume being pipetted by the Olympus analyzer; (4) Amphetamine false negative cause unknown; and (5) Cocaine false negative caused by On-Line Reagents coating the cells of the screening instrument.
As a check of corrective actions, CompuChem was sent 250 blind specimens during the period from April through September 1993.
All results were correctly reported by CompuChem.
(Enclosure 12)
- 13.
Public Service Electric and Gas Company's letter of December 1, 1993, concerns an unsatisfactory performance test result. A sample certified as 2054 ng/ml amphetamine and 1982 ng/ml methamphetamine was sent to two contracted laboratories.
One lab correctly reported a positive test at 1931 ng/ml amphetamine and >2000 ng/ml methamphetamine.
The other lab (POLA, South Plainfield, NJ) incorrectly reported the test as negative, with 374 ng/ml d-amphetamine and 366 ng/ml methamphetamine being below the 500 ng/ml cut-off level.
Investigation found that POLA had diluted the sample five-to-one to run the test. Thus, the results should have been 1870 ng/ml amphetamine and 1830 ng/ml methamphetamine.
POLA has implemented corrective actions to preclude recurrence.
(Enclosure 13)
- 14.
Houston Lighting and Power Company's letter dated December 21, 1993, concerns a blind specimen (certified positive at 199 ng/ml THC) being reported as negative by SmithKline Beecham Laboratories (SKBL).
The false negative was investigated and found to be due to a substance which was unique to the blind specimen (i.e., not due to the analytical process) and which caused additive interference with the 488 ion.
This interference has resulted in initial ion ratio failures in several instances for blind specimens of this lot.
The South Texas plant no longer has any specimens from the lot that contained the interfering substance.
(Enclosure 14)
- 15.
Nebraska Public Power District's letter of December 27, 1993, pertains to a false positive report on a random drug test of an individual.
This false positive did not result from a testing laboratory error.
- Instead, the error involved a unique convergence of prescription drugs, over-the-counter (OTC) medications, and the individual's habit of drinking about 20 cans of soda pop daily.
In July 1993, the individual was selected for random testing, provided a specimen, and listed MAPOP (over-the-counter Tylenol generic), Tylenol, Advil, and PSEDDO 61 P51 caplets for SOME OF THE ENCLOSED DOCUMENTS CONTAIN, OR ARE CLAIMED TO CONTAIN, PROPRIETARY INFORMATION
Dr. Donna Bush Zephrez LA (prescription drug) on the consent form.
The specimen was negative, but below acceptance limits for both specific gravity and creatinine, so a second specimen was taken.
The consent form listed the same drugs as the first consent form.
The second specimen also screened negative and tested low for creatinine. A third specimen was taken as a for-cause test and, therefore, was a witnessed collection.
On the third consent form, the individual listed the same drugs as before, less MAPOP, and added Ventolin Z-Max, Pheneygan E Codeine, and PS ERCB 61 PSI caplets (all three are prescription medicines).
The third specimen screened negative and was low for specific gravity and creatinine.
The MRO directed a GC/MS test of all three samples.
The first specimen had detectable levels of Levorphenol (an opiate family drug), the second specimen had detectable levels of amphetamine and methamphetamine (at about 67 and 7 times below established cut-off levels, respectively),
and the third specimen tested negative for all drugs at any level.
The investigation found that a metabolite of dextromethorphan (common in OTC cough medicines) has a GC/MS "footprint" almost identical to Levorphenol (i.e., graphically identical). Tests then confirmed dextromethorphan in the sample.
The MRO confirmed that the individual mentioned taking such a medication, but had failed to list it on the consent form.
Regarding the amphetamine, a follow-up test on the second specimen found that the amphetamine was the L-type (common in OTC decongestant inhalers). This case contains valuable "lessons-learned" which both HHS and NRC should consider.
The dextromethorphan/Levorphenol GC/MS "footprints" and the L-amphetamine in OTC decongestants are particularly noteworthy points for testing labs, and the ingestion of unusually large amounts of any liquid should be noted by MROs.
(Enclosure 15)
- 16.
Northeast Utilities' letter of February 1, 1994, pertains to four false negative test results by SmithKline Beecham Clinical Laboratories (SKBL) of Norristown, PA.
The investigation concluded that, due to the elapsed time after the initial false negative reports, only one specimen still remained in a stable enough condition to allow a valid confirmation test. This one specimen was retested by GC/MS and resulted in detection of 432 ng/ml methaqualone.
The details of these four false negative cases and the subsequent investigation results are given in the attached documents, which SKBL has certified as being proprietary information.
(Enclosure 16.
NOTE: PART OF THIS ENCLOSURE IS, OR IS CLAIMED TO BE, PROPRIETARY MATERIAL.)
- 17.
Northern States Power Company's letter of February 10, 1994, concerns a known positive blind performance specimen that was reported by MEDTOX Laboratories, Saint Paul, Minnesota, as being positive for opiates at levels which were approximately double the known values.
The known values were 596 ng/ml codeine and 617 ng/ml morphine.
The initial test report gave levels of 1244 ng/ml codeine and 1322 ng/ml morphine.
The retest showed 656 ng/ml codeine and 638 ng/ml morphine, which were very close to the known values.
An investigation concluded that MEDTOX had failed to dilute the sample by a factor of two, as was annotated on the SOME OF THE ENCLOSED DOCUMENTS CONTAIN, OR ARE CLAIMED TO CONTAIN, PROPRIETARY INFORMATION
Q
,l. -
Dr. Donna Bush GC/MS confirmation cover sheet. A MEDTOX analyst, believing the sample had been diluted by a factor of two, then calculated the test results by doubling them - and caused the erroneous initial report.
(Enclosure 17)
If you have any questions, please call me at 301-504-2944.
Sincerely, en L. Bush Reactor Safeguards Branch Office of Nuclear Reactor Regulation
Enclosures:
As stated DISTRIBUTION W/Q./ENCLOSURES:
Cent\\ay*iles ;/
DR88,ff PDR L. Bush C. Hendren PSGB r/f P. McKee OFC PSGB:NRR PSGB:NRR BC:PSGB:NRR NAME CHHend SOME OF THE ENCLOSED DOCUMENTS CONTAIN, OR ARE CLAIMED TO CONTAIN, PROPRIETARY INFORMATION