ORIGINAL ARTICLE http://dx.doi.org/10.1563/jlmqa.016.38.3.19 Verification of the Comparability of Laboratory Results from Two Instruments within One Health Care System According to Clinical and Laboratory Standard Institute EP31-A-IR Eun Jin Lee, Eunyup Lee, Miyoung Kim, Han-Sung Kim, Young Kyung Lee, and Hee Jung Kang Department of Laboratory Medicine, Hallym University Sacred Heart Hospital, Hallym University College of Medicine, Anyang, Korea Corresponding author: Hee Jung Kang Department of Laboratory Medicine, Hallym University Sacred Heart Hospital, Hallym University College of Medicine, Gwanpyeong-ro 170beongil, Dongan-gu, Anyang 14068, Korea Tel: +8-31-380-399 Fax: +8-31-380-3934 E-mail: kangheejung@hallym. ac.kr Background: For convenience, multiple instruments can be used to measure the same laboratory results within one health care system. However, the laboratory must verify the comparability of the results. In this study, we evaluated a method for verifying the comparability of patient results obtained from two instruments within one health care system, EP31-A-IR, proposed by the Clinical and Laboratory Standards Institute. Methods: Using the range test proposed by the EP31-A-IR, we evaluated the comparability of 17 clinical chemistry test results from the HITACHII/MODULAR system (Roche Diagnostics, Switzerland) and the TOSHIBA/00FR system (Toshiba Medical Systems Co., Japan). The 0.33 biological variability, allowable total error, and standards of the Clinical Laboratory Improvement Amendments were used to determine the acceptance criteria. Results: Among 16 test parameters, the differences of means between the two instruments were less than their range rejection limit in 15 tests, and so the comparability between the two instruments was considered acceptable. Creatinine was not evaluated using this protocol because its range rejection limit was not deducible from the EP31-A-IR statistics table. Conclusions: The EP31-A-IR guideline is useful for verifying the comparability of results between two instruments. However, not all parameters are covered by the guideline. With consideration of the characteristics of each test parameter, each laboratory should devise its own method for evaluating comparability. (J Lab Med Qual Assur 016;38:19-136) Key Words: Comparability, Quality control, Range test pissn: 384-458 eissn: 88-761 Received March 10, 016, Revision received May 7, 016, Accepted June 15, 016 서론 단일의료기관내에서도여러대의검사장비로동일한검사를시행하는경우가흔하다. 한정된검체처리량을극복하고더욱신속한검사결과를얻기위해추가적으로장비를사용하거나진료의편의를위해응급실이나수술실등에서현장검사 를시행하는등의다양한형태로동일한검사를수행하게된다. 그러나서로다른장비에서나타나는측정값은상이한장비간측정방법, 보정이나비정밀도등의차이에의하여동등하지않은결과를보일수있다 [1]. 이러한검사결과의차이는진단과치료에중대한영향을미칠수있기때문에검사실에서는각각의검사항목에서서로다른장비를이용해측정된검 Copyright 016 Korean Association of External Quality Assessment Service 19
사값이동등함을검증하여보고하는것이중요하다. 최근에 Clinical and Laboratory Standards Institute (CLSI) 는단일의료기관내에서정량검사결과들의비교검증방법에대한지침을발간하여환자검체를이용한검사방법간결과의동등성검증을위한가이드라인을제시하고있다 [1]. 이방법은최대 10개의장비간동등성을검증할수있으며정도관리데이터를이용하여간편하게수행할수있도록고안되었다. 이동등성검증은 (1) 주기적검증을위해서, () 보정시약제품번호와시약제품번호가변경될때, (3) 보정과문제해결 ( 또는동등성에관한문제가해결된이후추적관찰 ) 등의목적으로광범위하게이용될수있다. 이에본연구에서는한의료기관임상화학검사실에서실제로이용중인서로다른두대의장비에서측정된검사결과간동등성을 CLSI EP31-A-IR에서제시한동등성평가지침을활용하여항목별로검증하고그경험을공유해보고자한다. 재료및방법 1. 분석장비진단검사의학과에서임상화학검사에이용하고있는자 동화검사기기인 HITACHII/MODULAR system (Roche Diagnostics, Basel, Switzerland) 과 TOSHIBA/00FR system (Toshiba Medical Systems Co., Tokyo, Japan) 에서얻어진결과간의동등성을평가하였다.. 검증항목알부민 (albumin), 알칼리인산분해효소 (alkaline phosphatase), 알라닌아미노전달효소 (alanine transaminase), 아스파르테이트아미노전달효소 (aspartate aminotransferase), 칼슘 (calcium), 콜레스테롤 (cholesterol), 염소 (chloride), 크레아티닌 (creatinine), 직접빌리루빈 (direct bilirubin), 감마글루타밀전이효소 (gamma-glutamyltranspeptidase), 포도당 (glucose), 칼륨 (potassium), 젖산탈수소효소 (lactate dehydrogenase), 나트륨 (sodium), 단백질 (protein), 총빌리루빈 (total bilirubin), 요산 (uric acid) 의 17항목에대하여검증하였다. 3. 동등성검증이연구는한림대학교임상연구심사위원회 / 기관생명윤리위원회로부터심의면제 (016-I057) 를승인받았으며 CLSI Table 1. QC data of chemistry parameters using MODULAR and 00FR: pooled mean, SR, ST, and analyte concentration for the comparability test Parameters Pooled mean SR ST Analyte concentration* Albumin (g/dl) 4.57 0.06 0.07 3.7 5.5 Alkaline phosphatase (IU/L) 59.64 0.88 0.99 47.7 71.6 Alanine transaminase (IU/L) 13.85 0.48 0.63 11.1 16.6 Aspartate aminotransferase (IU/L) 0.71 0.9 1.3 16.6 4.9 Calcium (mg/dl) 9.44 0.11 0.16 7.6 11.3 Cholesterol (mg/dl) 183.8.3.8 147.0 0.6 Chloride (mg/dl) 103.5 0.7 0.81 8.8 14. Creatinine (mg/dl) 0.79 0.03 0.08 0.63 0.95 Direct bilirubin (mg/dl) 0.3 0.01 0.01 0.18 0.8 Gamma-glutamyltranspeptidase (IU/L) 0.88 0.81 1.35 16.7 5.1 Glucose (mg/dl) 9.40 1.18 1.4 73.9 110.9 Potassium (mg/dl) 4.35 0.04 0.05 3.5 5. Lactate dehydrogenase (IU/L) 301.75 9.8 13.09 41.4 36.1 Sodium (mg/dl) 143.01 0.73 0.89 114.4 171.6 Total protein (mg/dl) 7.5 0.17 0.07 6.0 9.0 Total bilirubin (mg/dl) 0.73 0.03 0.04 0.58 0.88 Uric acid (mg/dl) 4.61 0.06 0.06 3.7 5.5 Abbreviations: SR, within-run SD; ST, between-run SD. *Analyte concentration was determined within 0% of the test sample target value (pooled mean) according to Clinical and Laboratory Standards Institute EP31-A-IR. 130 J Lab Med Qual Assur 016;38:19-136 www.jlmqa.org
EP31-A-IR 지침에서제시한범위검정방법에따라동등성을검정하였다. 검증방법은아래에기술하였다. (1) 015년 5월한달간건강검진을위해방문한성인 500 명의일반화학검사후잔여혈청을무작위로모아서혼합한후분주하여 -70 o C에서냉동보관한냉동혼주혈청을 MODULAR와 00FR 장비간의동등성검증에이용하였다. 이혼주혈청을각장비에서매일오전 회오후 회반복측정한항목별결과값을활용하여, 17가지임상화학검사항목에대해 015년 6월부터 11월까지의 6개월간의평균 (mean), within-run SD (SR) 와 between-run SD (ST) 를장비별로계산하였다 []. 그리고두장비의통합평균및통합 (pooled) SR과 ST를아래와같은식으로계산하여 Table 1에나타내었다. Pooled mean=(mean 1 +mean )/ Pooled SR=([SR 1 +SR ]/) 1/ Pooled ST=([ST 1 +ST ]/) 1/ 이때두장비간 ST의차이가 배이하일경우에만동등성평가를위해범위검정방법을이용할수있으므로 [1] 장비간 ST의차이를확인한다. () 분석을위한검체는 (1) 에서구한통합평균의 ±0% 이내농도를가진환자의잔여혈청검체또는혼주혈청으로선정하였다. 동등성검증을위해필요한검체의농도범위는 Table 1에함께제시하였다. (3) 동등성검증에적용할적합성기준을설정하였다. 검증대상인 17개항목에대한생물학적변이계수 (within-subject biological variability, CV I ) 등고려할수있는적합성기준들을 Table 에제시하였다. CLSI 지침에따르면전문가로서평가자의의학적판단에따라다양한적합성기준을선택할수있다 [1]. 이에본연구에서는임상적중요성및항목별평균측정치와 6개월동안두대의장비간의결과의치우침, 바이어스를종합적으로고려하여각각의적합성기준을선정하였다. 선택된적합성기준을토대로동등성검증에적용할임계차이 (critical difference, CD) 를아래와같은식으로계산하였다. CD=initial measured value selected acceptance criteria (4) 범위거절한계에대한 CLSI EP31-A-IR의부록 (appendix) B에제시된표 (tables of range rejection limit) 를사용하여항목별로범위거절한계값과측정반복횟수 (number of replicates) 를결정한다. 이표에서는측정횟수 (run) 가 1일 Table. Multiple acceptance criteria that can be applied to the comparability test Parameters CV I (%) CV G (%) I (%) B (%) TE (%) CLIA Selected acceptance criteria Albumin (g/dl) 3. 4.75 1.6 1.43 4.07 10% TE Alkaline phosphatase (IU/L) 6.45 6.1 3.3 6.7 1.04 30% 0.33 CV I Alanine transaminase (IU/L) 19.4 41.6 9.7 11.48 7.48 0% TE Aspartate aminotransferase (IU/L) 1.3 3.1 6.15 6.54 16.69 0% TE Calcium (mg/dl).1.5 1.05 0.8.55 1.0 mg/dl CLIA Cholesterol (mg/dl) 5.95 15.3.98 4.1 9.01 10% TE Chloride (mg/dl) 1. 1.5 0.6 0.5 1.5 5% TE Creatinine (mg/dl) 5.95 14.7.98 3.96 8.87 15% CLIA Direct bilirubin (mg/dl) 36.8 43. 18.4 14. 44.5 - TE Gamma-glutamyltranspeptidase (IU/L) 13.4 4.15 6.7 11.06.11 - TE Glucose (mg/dl) 5.6 7.5.8.34 6.96 10% TE Potassium (mg/dl) 4.6 5.6.3 1.81 5.61 0.5 mmol/l TE Lactate dehydrogenase (IU/L) 8.6 14.7 4.3 4.3 11.4 0% TE Sodium (mg/dl) 0.6 0.7 0.3 0.3 0.73 4 mg/dl CLIA Total protein (mg/dl).75 4.7 1.38 1.36 3.63 10% CLIA Total bilirubin (mg/dl) 1.8 8.4 10.9 8.95 6.94 0.4 mg/dl 0.33 CV I Uric acid (mg/dl) 8.6 17.5 4.3 4.87 11.97 17% 0.33 CV I Abbreviations: CV I, within-subject biological variability; CV G, between-subject biological variability; I, desirable specification for imprecision; B, desirable specification for inaccuracy; TE, desirable specification for allowable total error; CLIA, Clinical Laboratory Improvement Amendments. www.jlmqa.org J Lab Med Qual Assur 016;38:19-136 131
때와 일때에각각다른측정반복횟수를제시하고있는데, 측정횟수가 1일때보다 일때더적은수의측정반복횟수로도동일한통계적검정능력을갖는점 [1] 을고려하여모든항목의측정횟수는 로통일하였다. (5) 두장비를이용하여각각의항목마다결정된반복횟수만큼측정하여결과값의평균을구하고장비별평균의차이를이용하여아래와같이범위를구하였다. Range=maximum observed mean-minimum observed mean (6) (5) 에서구한범위값이 (4) 에서설정한범위거절한계값보다작다면두장비간의동등성은적합하다고판정하였고만약범위값이범위거절한계값보다크다면분석적문제가있다고확인하였다. 결과 1. 적합성기준의선정본연구에서는임상적중요성및검증항목각각의정도관리 데이터를고려하여항목별로적절하게적합성기준을선택하였다. 17개의동등성검사항목모두에서두장비로부터얻어진 ST가 배이하의차이를보여범위검정방법을적용할수있었다. CLSI에서제시한적합성기준중에서 Weastgard QC database [3] 에근거한 CV I 에 0.33을곱한값, Weastgard QC database의총오차 (total error, TE), 또는 Clinical Laboratory Improvement Amendments 지침 [4] 에따른기준을적합성기준으로조사하였으며각항목의임상적특성을고려하여이중한가지를항목별로선정하였다. 조사한적합성기준과본연구에서선정한기준을 Table 에제시하였다.. 적합성평가결과범위검정의시행결과 17개의검사항목중 16개의항목에서범위거절한계값보다두장비의측정값평균의차이가작게나타나장비간측정결과의동등성이있는것으로평가되었다. 그러나크레아티닌에서는 CLSI EP31-A-IR 지침에따른범위검정을시행할수없었는데, Table 의여러적합성기준중어떠한기준을선택하여도측정반복횟수및범위거절한계값 Table 3. Results of the range test for the comparability test Parameters Level of analyte CD/ST SR/ST Run (N) Replication (N) Range rejection limit Range test (mean A-mean B) Albumin (g/dl) 4.8.6 0.78 0.13 0.08=4.67 4.59 Alkaline phosphatase (IU/L) 63 1.35 0.88 5 0.94 0.7=6.3 61.6 Alanine transaminase (IU/L) 14 6.13 0.77 1 1.89 0=14 14 Aspartate aminotransferase (IU/L).83 0.71.53 1.=4.8 3.6 Calcium (mg/dl) 10.1 6.43 0.73 1 0.49 0.16=9.66 9.5 Cholesterol (mg/dl) 179 5.77 0.8 1 7.9 =17 170 Chloride (mg/dl) 103 1.91 0.9 4 1.04 0.5=103.63 103.38 Creatinine (mg/dl) 0.8 1.47 0.3 NA* NA* NA* NA* Direct bilirubin (mg/dl) 0. 5.96 0.67 1 0.04 0.03=0.0 0.3 Gamma-glutamyltranspeptidase (IU/L) 19 3.1 0.6 1.77 1=0 19 Glucose (mg/dl) 96 4.70 0.83 1 3.7 =93 91 Potassium (mg/dl) 4.4 5.03 0.8 1 0.1 0.05=4.45 4.40 Lactate dehydrogenase (IU/L) 306.67 0.75 3.36 4.5=99.0 94.5 Sodium (mg/dl) 144 4.48 0.8 1 1.96 1=143 14 Total protein (mg/dl) 7.5 1.56 0.38 1 0.37 0.1=5.5 5.4 Total bilirubin (mg/dl) 0.76 1.54 0.89 8 0.04 0.01=0.74 0.73 Uric acid (mg/dl) 4.7.1 1.01 0.09 0.05=4.68 4.63 If the calculated ranges are less than or equal to the range rejection limit, the comparability of the methods is acceptable. Abbreviations: CD, critical difference; ST, between-run SD; SR, within-run SD; NA, not available. *Comparability testing for creatinine was not applicable because the appropriate replication number and range rejection limit were not obtainable from the table of range rejection limits in Clinical and Laboratory Standards Institute EP31-A-IR. 13 J Lab Med Qual Assur 016;38:19-136 www.jlmqa.org
이 CLSI 지침에서제공하는표에제시되지않았기때문이었다. 항목별범위검증시행결과는 Table 3에, 알칼리인산분해효소의예시로살펴본범위검증절차와검증결과를 Fig. 1에나타내었다. 고찰 여러대의장비로측정한검사항목에대하여동일성이검증되지않은결과를보고하는것은진단및치료에혼선을주어임상적으로중대한영향을미칠수있다. 따라서대한진단검사 Modular Mean 60.1 SR 0.66 ST 0.87 00FR Mean 59. SR 1.05 ST 1.10 Precision Pooled mean 59.6 Pooled SR 0.88 Pooled ST 0.99 Analyte concentration: 47.71 71.57 Acceptance criteria: CV (6.45%) 0.33 CD: 63 (initial value) * 0.33* 6.45%=1.34 CD/pooled ST: 1.34/0.99=1.35 Pooled SR/pooled ST: 0.88 I * Runs CD/ST SR/ST 1.5 1.5 1.5 1.5 1.00 0.95 0.90 0.85 1.00 0.95 0.90 0.85 0.80 Nof replicates 6 9 90-4 5 9 79 - Range rejection limit 0.65 CD 0.70 CD 0.70 CD - 0.66 CD 0.69 CD 0.70 CD 0.70 CD - CLSI EP31-A-IR Appendix B. *Interpolation: CD/ST (x) 1/n=1/n 0+(x x) 0 *[n0 n 1]/n0n 1* (x1 x 0) 1/L=1/L +(x x) *[L L ]/L L *(x x ) 0 0 0 1 0 1 1 0 1/n=1/90+(1.35 1.5) *(90 9)/90* 9 *(1.5 1.5) n=4.7 1/L=1/0.7+(1.35 1.5) *(0.7 0.7)/0.7 *0.7*(1.5 1.5) L=0.7 Replication number: 5 Coefficient for rejection limit: 0.7 Measurement Modular Run1: 63, 6, 6, 6, 63 Run: 63, 6, 63, 6, 61 Mean: 6.3 00FR Run1: 6, 64, 61, 61, 6 Run: 6, 60, 60, 6, 6 Mean: 61.6 Range rejection limit: 0.7* CD (1.34)=0.94 Range: 6.3-61.6=0.7 Status: pass Fig. 1. One example of the procedure used for verifying comparability: alkaline phosphatase. Abbreviations: SR, withinrun SD; ST, between-run SD; CD, critical difference; CV I, within-subject biological variability. www.jlmqa.org J Lab Med Qual Assur 016;38:19-136 133
의학회와진단검사의학재단의우수검사실신임인증평가에서는동일검사항목에대하여서로다른두대이상의장비로검사를시행하는경우, 각장비간일치도에대한기준설정여부와 6개월마다검사결과간비교를실시하는지를평가항목으로두고있다 [5]. CLSI에서제시하는동등성을평가하는지침을이용하면임상검사실에서정도관리데이터를이용하여간편하게검사결과간일치도를평가할수있다. Kim과 Cha [6] 는 CLSI EP31-A-IR의이전버전인 C54-A 를토대로두대의동일한장비에서측정되는결과의동등성을검증하였고, 그결과평가항목모두에서두장비간의동등성을보였다고보고하였다. 그러나이렇게동일제조사의장비로진행한동등성검증은다른제조사의장비를이용하고있는검사실의검증과그결과에차이가있을수있다. 실제임상검사실에서는서로다른제조사의장비를사용하는경우가흔한데, 본연구는서로다른제조사의장비를대상으로동등성을검증한경우이다. 이에본연구에서제시한서로다른두장비간의동등성검증경험은다양한임상검사실에서동등성검증에실제적으로유용한도움을줄수있을것으로생각한다. CLSI 지침에따른검증결과를살펴보면, 본검사실에서이용중인두장비간임상화학검사항목은동등성이대체로잘유지되고있는것으로나타났다. 그러나검사항목중 CLSI 지침으로평가할수없는항목이존재하였다. 이러한항목의경우최소 0개이상의다양한농도를가진검체를이용해선형회귀분석등의다른통계적기법을이용한바이어스분석방법을통해서검증가능하다 [,7,8]. EP31-A-IR 지침은범위검정방법외에다른동등성검증절차를각검사실에서개발하여사용할수있으며, CLSI의지침이이를배제하기위함이아님을언급하고있다. 따라서실제임상검사실에서동등성검증을시행할때본연구의경우처럼 EP31-A-IR 지침이적용되지않는경우가존재한다면이지침보다는다소복잡하겠지만다른절차를이용하는것이가능하다고판단된다. 범위검정방법에따른동등성평가에서적합성판정기준은검사결과의동등성을판정하는데결정적인영향을미칠수있다. CLSI 지침에서는이기준으로잘설계된임상결과연구 (well-designed clinical outcome study), 임상의를대상으로한설문조사, 생물학적변이를이용한설정, 전문가집단에의해출판된추천지침이나승인된기관들에의해정해진목표, 내부비정밀도데이터 (internal imprecision data) 또는외부신빙도조사데이터 (external proficiency testing data) 에근거한일반적인허용범위등을제시하고있다. 추천되는잘설계된임상결과연구를적용하는방법으로대표적인항목에는 hemoglobin A1c가있다 [9]. 하지만이러한임상결과연구는 적용가능한항목이매우제한적이다. 다음으로는임상의를대상으로한설문조사방법등이있으나현실적으로어려우며, 임상의마다다른견해차이가존재하는경우에는객관적이지않을수있는문제가있다. 생물학적변이에의한허용가능바이어스인 0.33 CV I 는만족할만한분석비정밀도를가진두장비에서낸차이가개인에서시간에따라보일수있는결과값의차이보다작도록목표하는것이다. 그러나개인내변동성이크지않은검사항목, 예컨대전해질농도나단백농도등은분석에서포함하는변이가개인내변동성에비하여상대적으로커서모든항목에이기준을적용하기에는현실적으로무리가있다 [10]. 또한 EP31-A-IR에서제시하는범위검정을수행하기위해서는우선 CD가 ST보다커야하는데적합성판정기준을 0.33 CV I 로정한경우많은항목에서 ST가 CD보다컸다. 즉장비의분석능이이러한목표에비하여많이부족하다는것을보여주었다. 따라서본연구에서는검사항목의 CV I 및 desirable TE, 실제검사실장비의비정밀도등을참고하여항목별로적합성판정기준을결정하였다. 이처럼검증방법과장비에따라검사실마다실정에맞는다른기준이선택될수있을것으로생각한다. 이연구에서나타난범위검정의또다른한계는 SR/ST가 CLSI EP31-A-IR 부록 B의표에서제시된 SR/ST보다훨씬더작으면범위검정을적용할수없다는것이다. 크레아티닌의경우 CD/ST값이 1.47, SR/ST가 0.3으로 CLSI EP31-A-IR 부록의표에서 CD/ST값이 1.5일때측정반복횟수와범위거절한계값을제시하고있는 SR/ST 최소값이 0.85인것에비해상대적으로매우작은 SR/ST값을보여범위검정에필요한측정반복횟수를얻을수없었다. 본검사실에서크레아티닌검사에이용중인 Jaffe 법은시간의흐름에따라색소침착이일어나시약이진해지는간섭현상이존재한다 [11]. 이로인해 within-run SD (pooled SR, 0.03) 값에비해 between-run SD (pooled ST, 0.008) 값이상대적으로크다고판단된다. 통계적정도관리를위해서는해당검사종목의중요임상적판단범위가포함되도록 가지이상농도의정도관리물질을사용한다 [1]. 본연구는 CLSI EP31의범위검정방법을실제검사실의결과에적용해봄으로써해당프로토콜이임상적으로유용하게적용가능한지여부를평가하고문제점을파악하는것이일차적인목적이었으므로연구의편의를위해한가지농도의검체만을이용하였다. 그러나실제검사실에서정도관리를목적으로동등성평가에이연구의프로토콜을적용하고자할때는임상적판단에중요한레벨에서모두동등성평가를하는것이바람직하겠고그러려면임상적판단이중요한각레벨에서각장비의 SR과 ST 조사가선행되어야할것으로판단 134 J Lab Med Qual Assur 016;38:19-136 www.jlmqa.org
된다. 현실적으로는정도관리물질을서로다른장비에서공통으로사용한다면쉽게필요한자료를수집할수있으리라여겨진다. 결론적으로 CLSI EP31-A-IR 지침은검사실의정도관리데이터를동등성검증에활용할수있고비교적간편한절차를통해동등성정도를평가가능하여임상적으로유용하게이용될수있음을확인하였다. 그러나본연구결과에서나타난대로 CLSI EP31-A-IR 지침만으로동등성검증을시행하는과정에는검증이불가능한항목이관찰되었으며 CD값을설정하는과정의어려움도존재하였다. 따라서본연구에서제시한동등성검증과정의예시를참고하여각검사실에서는실정에맞는동등성검증방법을고안하여평가에사용하는것이필요하겠다. REFERENCES 1. Clinical and Laboratory Standards Institute. Verification of comparability of patient results within one health care system: approved guideline (interim revision). EP31- A-IR. Wayne (PA): Clinical and Laboratory Standards Institute, 01.. Clinical and Laboratory Standards Institute. Evaluation of precision of quantitative measurement procedures: approved guideline. EP5-A3. Wayne (PA): Clinical and Laboratory Standards Institute, 014. 3. Westgard JO. Desirable specifications for total error, imprecision, and bias, derived from intra- and interindividual biologic variation. http://www.westgard.com/ biodatabase1.htm (Accessed March 3, 016). 4. Clinical Laboratory Improvement Amendments. Recommended total allowable error limits. http://wwwn. cdc.gov/clia/regs/subpart_i.aspx (Accessed March 3, 016). 5. The Korean Society for Laboratory Medicine, Laboratory Medicine Foundation. Laboratory accreditation program checklist: clinical chemistry. Seoul: The Korean Society for Laboratory Medicine, 01. 6. Kim BH, Cha YJ. Verification of comparability among quantitative results obtained using multiple instruments within a health care system. Lab Med Online 014;4:65-70. 7. Clinical and Laboratory Standards Institute. User verification of performance for precision and trueness: approved guideline. CLSI document EP15-A. nd ed. Wayne (PA): Clinical and Laboratory Standards Institute, 005. 8. Johnson R. Assessment of bias with emphasis on method comparison. Clin Biochem Rev 008;9 Suppl 1:S37-4. 9. Larsen ML, Fraser CG, Petersen PH. A comparison of analytical goals for haemoglobin A1c assays derived using different strategies. Ann Clin Biochem 1991;8(Pt 3):7-8. 10. Asberg A, Odsæter IH, Carlsen SM, Mikkelsen G. Using the likelihood ratio to evaluate allowable total error: an example with glycated hemoglobin (HbA1c). Clin Chem Lab Med 015;53:1459-64. 11. Peake M, Whiting M. Measurement of serum creatinine: current status and future goals. Clin Biochem Rev 006; 7:173-84. 1. Clinical and Laboratory Standards Institute. Statistical quality control for quantitative: measurement procedures: principles and definitions: approved guideline third edition. CLSI document C4-A3. Wayne (PA): Clinical and Laboratory Standards Institute, 006. www.jlmqa.org J Lab Med Qual Assur 016;38:19-136 135
Clinical and Laboratory Standards Institute EP31-A-IR 에따른단일의료기관내두대의자동임상화학검사기기간결과값의동등성검증이은진 이은엽 김미영 김한성 이영경 강희정한림대학교의과대학한림대학교성심병원진단검사의학과 배경 : 단일임상검사실에서신속한검사결과를얻기위해여러대의검사장비를이용하는경우가흔하게존재한다. 검사실은이러한검사결과간의동등성을검증하여야한다. 이연구에서는 Clinical and Laboratory Standards Institute 에서제안한동등성검증가이드라인인 EP31-A-IR 을이용하여두대의서로다른장비에서도출되는임상화학검사결과의동등성을검증해보았다. 방법 : EP31-A-IR 가이드라인에서제시하는범위검정방법을이용하여 HITACHII/MODULAR system (Roche Diagnostics, Switzerland) 과 TOSHIBA/00FR system (Toshiba Medical Systems Co., Japan) 에서도출되는 17 가지임상화학검사결과의동등성을비교하였다. 결과 : 생물학적변동성에 0.33 을곱한값, 총오차, Clinical Laboratory Improvement Amendments 지침에의한기준을임계차이기준으로선정하였다. 범위검정결과검사항목중 16 가지항목에서두장비간평균의차가범위거절한계값보다작게나타나두장비에서도출된결과에동등성이있는것으로확인되었다. 그러나크레아티닌의경우 EP31-A-IR 가이드라인에제시된표를이용하여범위거절한계값을구할수없어평가가불가능하였다. 결론 : EP31-A-IR 가이드라인은다수의검사기기에서도출되는임상화학검사결과의동등성을평가하는데유용하나모든항목을이가이드라인에적용할수없었다. 각검사실에서는항목에따른특성을고려하여검사실마다의동등성평가방법을고안하여야한다. (J Lab Med Qual Assur 016;38:19-136) 교신저자 : 강희정우 )14068 경기도안양시동안구관평로 170 번길, 한림대학교의과대학한림대학교성심병원진단검사의학과 Tel: 031)380-399, Fax: 031)380-3934, E-mail: kangheejung@hallym.ac.kr 136 J Lab Med Qual Assur 016;38:19-136 www.jlmqa.org