ML20043B700
| ML20043B700 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 05/25/1990 |
| From: | Walt T PORTLAND GENERAL ELECTRIC CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9005310155 | |
| Download: ML20043B700 (4) | |
Text
.. - -
4.
(
m W M B0ddC C0f14MN1y i
g i
i n
I f-i May 25, 1990 j
Trojan Nuclear plant Docket 50-344 License NPF-1 I
i U.S. Nuclear Regulatory Commission Attn Document Control Desk 1
Washington DC 20555.
3 Dear Sirst j
1
+
Attached please find a report describing an incident concerning unsatisfactory performance testing that occurred as part of the Trojan Nuclear plant Fitness for Duty program. The report.is submitted in l-
<accordance with the-requirements of Title 10, code of Federal Regulations, part 26.-Appendix A.
The incident involved the contract drug testing laboratory reporting a certified positive blind perforrance test specimen
-l as being negative.
l Sincerely, f
T. D. Walt Acting Vice president, Nuclear Attachment ct Mr. John B. Martin "I
Regional-Administrator, Region V U.S. Nuclear Regulatory Commission l
L Hr. David Stewart-Smith State of Oregon
- Department of Energy Mr. R. C. Barr NRC Resident Inspector Trojan Nuclear Plant d,'MRR D Rig RS 6 B 9005310155 900525 l
PDR. ADOCK 05000344 pg[
l
'S-PDC 121 SW Sairron Street Pumand. Oregon 97204
// 7
/..
H
l Trojan Nucioar plant Document Control Desk Docket 50-344 May 25, 1990 License NpF-1 Attachment page 1 of 3 Report of Unsatisfactory performance Testing t! Contract Drum Testing Laboratory Description of Occurrence On April 26, 1990, a shipment of nine urine specimens was made to the contract laboratory for portland General Electric (pGE) Company's Fitness for Duty (FFD) program. Tour of these represented genuine specimens submitted as part of PGE's random screening program and one genuine specimen submitted as part of the preemployment screening program. The remaining four specimens were blind performance test specimens, two of which were negative, one of which was positive for Delta-9 tetrahydrocannabinol (THC), and a fourth certified to be positive for benzoylectonine (cocaine). The specimens were analyzed by the laboratory and results reported to pGE's medical review officer (MRO) on April 27, 1990. The MRO notified the FFD coordinator and the contract laboratory that the blind performance test specimen certified to contain benzoylocgonine at 483 ng/mi was reported as negative. pGE's screening cut-off levels for benzoyleesonine are 300 ng/ml for initial screening and 150 ng/ml for confirmatory screening.
pGE's medical technologist immediately requested an investigation of the false negative interpretation of the specimen by the laboratory.
In addition, the laboratory was requested to send an aliquot of the specimen to an alternate National Institute of Drug Abuse (NIDA) laboratory for retesting by gas chromatography / mass spectrometry (GC/MS). On this same date, the medical technologist submitted a blind specimen certified to be identical in content to the benzoyleegonine specimen in question. On April 30, 1990, the laboratory correctly reported this subsequent specimen as positiva.
In addition to requesting the contract laboratory to investigate the incident, pGE also requested the contract blind specimen supplier to submit certifying information regarding the spiked specimen. The specimen was analyzed by CC/MS on March 3, 1990 by the supplier and was certified to contain 483 ng/ml of benzoylocgonine. The supplier sent three other specimens from this batch to pCE and approximately 40 other specimens from this batch to other nuclear facilities.. No other incidents of falso negative reporting from this batch have occurred to date.
Because the spe:imen was shipped to pGE on April 18, 1990, refrigerated by PGE, and submitted to the laboratory on April 26, 1990, sample degradation is unlikely.
This is supported by the fact that GC/MS analysis by both the contract laboratory and the second NIDA lab confirmed the concentration at $22 ng/ml and 528 ng/m1, respectively.
The laboratory incident report to pCE states that the initial screching detected the presence of benzoyleegonine at a borderline level. Because the laboratory's standard practice is to rerun samples giving borderline results, the sample was again screened. The result was reported as negative.
?
i
{
I e
o i
-Trojan Nuclear plant Document Control Desk Docket 50-344 May 25, 1990 License NpF-1 Attachment L
page 2 of 3 After the laboratory was notified of the false negative by pCE, the original specimen was ratested on both Monarch and Hitachl 717 instruments.
The Monarch instrument gave a borderline result. The Hitachi 717'gave a positive result just above the cut-off level uf 300 ng/ml.
Five additional tests on the Monarch analyzer all resulted in instrument readings less than the cut-off level; however, when the specimen was analyzed by GC/MS at the laboratory, it gave a benzoyleesonine concentration of 522 ng/ml. The GC/MS screening by the i
p second NIDA laboratory agreed with this result, giving a reading of $28 ng/ml.
Cause of Occurrence A review of all available data rules out samplo degradation as a possible cause based on (1) extensive certifying information provided by the blind specimen supplier, (2) correct reporting as positive by the laboratory of an identical blind specimen exposed to identical environmental conditions subsequently submitted by pCE, and (3) GC/MS screoning of the specimen by two NIDA laboratories. Thorough investigation and numerous retcsts through various methodologies by the laboratory indicate a sample-unique problem
'j rather than a process problem.
It appears the particular composition of this urine sample adversely affected the immunoassay reaction such that benzoyleegonine concentration could not be reliably determined.
In other words, it is likely that normal physiological metabolites unique to this sample caused interference with immunoassay screening instru.;.ents.
This did not affect CC/MS technology, as the laboratory confirmed the sample positive by this method at 522 ng/ml.
l Corrective Actions Because the root cause of this incident appears not to indicate a process problem, no further actions are deemed necessary at this time. pCE will continue its_ blind performance testing program with the laboratory. This is the first incidence of false negative reporting by this laboratory since PGE began submitting specimens on March 16, 1990. Given the industry acceptability rate for falso negativo reporting of up to 10 percent, the laboratory is well within the acceptable range.
Of 425 blind samples submitted to the laboratory as of May 16, 1990, this is the only falso negative incident to have occurred.
In addition, the laboratory has not had any falso negative incidents with specimens containing benzoyleegonine in the proficiency testing program administered by NIDA.
Significance of Occurrence This event did not at'fect plant safety or security. The incident involved a spiked test specimen. Genuine samples submitted to the laboratory at the same time were deemed to be correctly interpreted and reported. The laboratory problem was identified by pGE's FFD Quality Assurance program.
Blind test specimens are routinely submitted to the laboratory to ensure integrity of the testing and reporting process.
o U"
Trojan Nuclear plant Document Control Desk Docket 50-344 May 25, 1990 License NpF-1 Attachment page 3 of 3 No press releases have been made by pCE, nor are any contemplated.
previous Similar Events Since the implementation of pGE's FFD program on January 3, 1990 as required by 10 CFR 26, this is the first occurrence of false negative reporting of a known positive quality control specimen at this laboratory. An earlier incident involving false negative reporting of a specimen known to contain Delta-9 THC occurred at pCE's backup laboratory. The cause of this earlier incident was tentatively determined to be a phenomenon of THC mets.bolite adhering to the plastic sample bottic or pipette, thus reducing metabolite concentration. Because CC/MS screenings of the specimen in this incident verified the benzoylecgonine concentration that was certified by the blind supplier, it is unlikely that this incident occurred due to the metabolite adhering to the plastic sample bottle or pipette.
4846W i