원저 Lab Med Online Vol. 8, No. 3: 77-86, July 임상화학 개인용혈당측정시스템표준시험방법가이드라인에따른 GlucoDr.S 성능평가 Evaluation of t

원저 Lab Med Online Vol. 8, No. 3: 77-86, July 218 임상화학 개인용혈당측정시스템표준시험방법가이드라인에따른 GlucoDr.S 성능평가 Evaluation of the Self-Testing Blood Glucose Monitoring System GlucoDr.S According to ISO 15197:213 Guidelines 김남희 1,2 김보경 3 전선희 3 이경훈 1,3 오태정 4 송상훈 1,2 최성희 4 임수 4 송운흥 5 송정한 1,3 장학철 4 Namhee Kim, M.D. 1,2, Bo Gyung Kim, M.T. 3, Sun-Hee Jun, M.T. 3, Kyunghoon Lee, M.D. 1,3, Tae Jung Oh, M.D. 4, Sung Hee Choi, M.D. 4, Soo Lim, M.D. 4, Sang Hoon Song, M.D. 1,2, Woon Heung Song, Ph.D. 5, Junghan Song, M.D. 1,3, Hak Chul Jang, M.D. 4 서울대학교의과대학검사의학교실 1, 서울대학교병원진단검사의학과 2, 분당서울대학교병원진단검사의학과 3, 분당서울대학교병원내과 4, 신한대학교보건과학대학임상병리학과 5 Department of Laboratory Medicine 1, Seoul National University College of Medicine, Seoul; Department of Laboratory Medicine 2, Seoul National University Hospital, Seoul; Department of Laboratory Medicine 3, Seoul National University Bundang Hospital, Seongnam; Department of Internal Medicine 4, Seoul National University Bundang Hospital, Seongnam; Department of Biomedical Laboratory Science 5, Shinhan University, Uijeongbu, Korea Background: The performance of the self-monitoring of blood glucose in patients with diabetes should be properly evaluated to ensure strict glycemic control. This study evaluated the self-testing Blood Glucose Monitoring System GlucoDr.S (All Medicus Co., Ltd., Korea). Methods: This study recruited 12 patients. Use of the glucometer was evaluated according to ISO 15197:213 guidelines. The YSI 23 STAT PLUS Glucose Analyzer (YSI Life Sciences, USA) was used as the reference device. Results: The standard deviation and coefficients of variation ranges for measurement repeatability and intermediate measurement precision conducted with 1 meters and 3 reagent lots on the same day were 2.7 3.2 mg/dl (< mg/dl) and 3.4 3.7% ( mg/dl), respectively, and 3.7 mg/dl (< mg/dl) and 2.1 2.6% ( mg/dl), respectively. Each coefficient of determination (R 2 ) for linearity of the 3 reagent lots was >.99. The influence effect of hematocrit and the 24 interference agents was not significant, except for xylose. A system accuracy test was conducted with subjects taking duplicate measurements from each of the 3 reagent lots. When glucose levels were < mg/dl and mg/dl, >95% of the samples were within ±15 mg/dl and within ±15% of the average measured values of the reference measurement, respectively. In Consensus Error grid analysis, all results were distributed in zone A and B. The results of the user performance evaluation using 115 lay persons were also included in the acceptance range. Conclusions: The GlucoDr.S showed acceptable performance according to the ISO 15197:213 guidelines and could be a clinically useful selftesting glucometer. Key Words: POCT, GlucoDr.S, Blood Glucose Monitoring System, ISO15197:213 서론 Corresponding author: Kyunghoon Lee Department of Laboratory Medicine, Seoul National University Bundang Hospital, 82 Gumi-ro 173beon-gil, Bundang-gu, Seongnam 1362, Korea Tel: +82-31-787-7696, Fax: +82-31-787-415, E-mail: khlee5923@gmail.com Received: June 8, 217 Revision received: September 26, 217 Accepted: October 1, 217 This article is available from http://www.labmedonline.org 218, Laboratory Medicine Online This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4./) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. 당뇨병은고혈당을특징으로하는대사질환으로전세계적으로환자수가급증하고있다. 대한당뇨병학회와국민건강보험공단이국민건강정보자료를통해분석한결과에따르면, 26년부터 215년까지국내당뇨병유병률역시지속적인증가추세에있다. 213년을기점으로볼때, 3세이상성인의약 12% (35만명 ) 정도가당뇨병환자로파악되었는데이는 8 9명중 1명이당뇨병을앓고있다는의미이기도하다 [1, 2]. 당뇨병이적절히관리가되지않을경우, 만성합병증인당뇨병성신증, 신경병증, 망막증, 대혈관합병증으로진행될수있을뿐 eissn 293-6338 www.labmedonline.org 77

아니라, 일시적으로저혈당증으로빠질수있어환자들의지속적인혈당감시와이를바탕으로한당뇨병관리의필요성이강조되어왔다 [3-5]. 이에병원이나가정에서환자들의혈당상태를간편하게측정할수있는개인용혈당측정기가여러회사에서개발되어시판되고있다. 당뇨병환자에게혈당측정기의측정결과는정확하고재현성이좋아야한다. 따라서표준화된방법으로혈당측정기의성능이평가되어야한다 [6, 7]. 개인용혈당측정시스템의유효성을평가하는대표적인기준은 International Organization for Standardization (ISO) 15197 [8, 9] 과 Clinical and Laboratory Standards Institute (CLSI) 지침 [1-14] 을꼽을수있다. ISO 15197 의경우, 23년 1판이발행되었고 [8], 213년일부개정되어 ISO 15197:213 2판이발간되었다 [9]. 개정된 ISO15197: 213은기존규정에비해더욱강화되었다. 가장큰변화는정확도의기준농도가 75 mg/dl에서 mg/dl로변경되었으며, 기존허용오차범위 ±15 mg/dl (75 mg/dl 미만 ) 또는 ±2% (75 mg/dl 이상 ) 에서 ±15 mg/dl ( mg/dl 미만 ) 또는 ±15% ( mg/dl 이상 ) 로강화된것과내 외인성간섭물질의영향평가에대한항목이도입된것이다 [15]. 본연구에서는 ISO 15197:213 에맞추어국내식품의약품안전처 (Ministry of Food and Drug Safety, MFDS) 식품의약품안전평가원에서제시한지침 [16] 에따라성능이개선된개인용혈당측정시스템인 GlucoDr.S (All Medicus Co., Ltd., Anyang, Korea) 의성능을평가하고자하였다. 대상및방법 2. 대상검체시스템정확도및사용자성능평가는 216년 3월 21일부터 6월 24일까지분당서울대학교병원내과에내원하여본연구참여에동의한환자 12명을대상으로하였다. 모세관혈액은손가락끝에서채혈하고, 채혈후즉시검사지에혈액을묻혀혈당을측정하였다. 정맥혈액은상완정맥에서채혈하여 EDTA가포함된용기에담은후, 즉시혈당과적혈구용적률을측정하고잔여검체는냉장보관하였다. 측정반복성, 직선성, 간섭물질에대한영향평가는잔여검체중적합한검체를선택하여진행하였다. 연구를위해사용한잔여검체의전처리는지침 [16] 에따라준비하였다. 잔여검체는채혈후 36시간안에모두사용하였으며, 이중저농도구간의검체는실온에하룻밤동안방치하여확보하였고, 고농도구간의검체는일정량의포도당용액을지원자혈액에섞어확보하였다. 모든평가전후에 YSI 23 STAT PLUS Glucose Analyzer로각각 2회씩측정하여단기안정성을확인하였으며검사의기준농도는해당값들의평균을사용하였다. 본연구는분당서울대학교병원 IRB 심의를통과한후진행하였다 ( 심의번호 E-1511-322-1). 3. 평가방법모든평가항목에서필요한검체종류, 측정기수, 검사지수, 피험자및검체수, 당농도분포, 적혈구용적률, 측정횟수, 검체량, 간섭물질의종류와농도, 시험환경, 검체의안정성등은의료기기기준규격별표 66. 개인용혈당측정시스템 (ISO 15197:213) 규정 [16] 및 CLSI 지침 [1-14] 에서규정한바에따라진행하였다. 1. 대상장비 GlucoDr.S 개인용혈당측정시스템의평가를위해기본모델인 AGM-513S로시험을진행하였다. 이기기는플라빈아데닌디뉴클레오티드 (Flavin adenine dinucleotide, FAD) 를조효소로하고포도당탈수소효소 (glucose dehydrogenase) 방법을활용하여전기화학적으로혈당을측정하는방식을취한다 [17]. 검사에사용된검사지 3종관리번호 ( 로트 ) 는 E16A11213, E16A13213, E16A3213이고용기와검사지, 측정기는임의로추출하였다. 평가에사용된기준장비는 YSI 23 STAT PLUS Glucose Analyzer (YSI Life Sciences, Yellow Springs, OH, USA) 를이용하였다. 적혈구용적률 (hematocrit) 은진단검사의학과에서진행한시험의경우, XE-2 Hematology Analyzer (Sysmex Corp., Kobe, Japan) 로, 내과에서진행한시험의경우, HemoCue (HemoCue Inc., Mission Viejo, CA, USA) 로측정하였다. 1) 정밀도정밀도 (precision) 는 ISO 15197:213 에서제시한측정반복성 (measurement repeatability) 과중간측정정밀도 (intermediate measurement precision), 두가지로평가하였다. 측정반복성평가는적혈구용적률이 35 5% 이며저농도에서고농도까지 5단계농도구간 (level 1, 3 5 mg/dl; level 2, 51 11 mg/dl; level 3, 111 15 mg/dl; level 4, 151 25 mg/dl; level 5, 251 4 mg/dl) 을대표하는잔여정맥혈액검체를대상으로, 하루내에동일측정자가측정기 1대에 3종로트에걸쳐평가하였다. 각로트당 1개의측정기에 1회씩측정하여 5단계농도구간에대한평균, 95% 신뢰구간표준편차, 변이계수를계산하였다. 중간측정정밀도평가를위해서, All Medicus 사에서제공한 4가지농도의대조물질 (level 1, 3 5 mg/dl; level 2, 96 144 mg/dl; level 3, 28 42 mg/dl; level 4, 425 55 mg/dl) 을 3종로트의혈당측정검사지를사용하여각각하루 1회씩총 1일에걸쳐매번측정기 1대를이용해검사하였다. 대조물질은 dextrose water 78 www.labmedonline.org

(DW) 를용매로하고 EDTA, 염색제 (dye), 안정제 (stabilizer) 를섞어만들었고, 식약처품목허가진행시, 개봉후 4개월안정성유효기간을허가받아유효기간내의대조물질을사용하였다. 해당대조물질들은각각 3회씩측정하여균질성을확인하였고평균, 95% 신뢰구간표준편차, 변이계수를계산하였다. 정밀도평가는 ISO 15197:213 규정 [16] 에따라, mg/dl 미만구간은표준편차를, mg/dl 이상구간은변이계수를판단의지표로삼았다. 2) 직선성평가직선성 (linearity) 평가는적혈구용적률이 35 5% 이며 11단계혈당농도구간을대표하는잔여정맥혈액검체를대상으로, 하루내에동일측정자가측정기 9대를이용해 3종로트에걸쳐평가하였다. 각로트에대해 YSI 23 STAT PLUS Glucose Analyzer로측정한기준농도와 GlucoDr.S 측정농도의상관관계그래프를그리고, 기울기와결정계수 (R 2 ) 를계산하였다. 3) 간섭물질에대한영향평가혈당측정값에영향을주는간섭물질들이 GlucoDr.S 으로측정할경우어느정도의오차를발생시키는지확인하기위하여, 적혈구용적률에따른검사와기타 24종의간섭물질 (interference) 에대한평가를진행하였다. 적혈구용적률에따른영향평가는 ISO 15197:213[16] 에따라정상수준 (42±2%) 을포함한 5단계적혈구용적률구간 (2±2%, 3 ±2%, 42±2%, 5±2%, 65±2%) 을대표하는잔여정맥혈액검체를대상으로하였다. 각적혈구용적률단계에서 3단계 (3 5 mg/ dl, 96 144 mg/dl, 28 42 mg/dl) 당농도구간시료에대해혈당측정을하였다. 측정기 1대에 3종로트에걸쳐측정한후, 평균혈당값을구하고, 해당구간평균혈당값과정상수준 (42±2%) 적혈구용적률에서의평균혈당값의차이를평가하였다. 다양한내외인성간섭물질에의한영향평가를위해 ISO 15197: 213에서명시하고있는 24종의간섭물질에대해서 US FDA (The Food and Drug Administration) 지침, CLSI EP7-A2 및기타관련자료들을참고하여 24종간섭물질의농도를결정하였다 [18-2]. 적혈구용적률이 35 5% 인잔여검체를대상으로, 2단계 (5 mg/ dl, 25 35 mg/dl) 당농도구간에서실시하였고시험시료는해당검체에간섭물질을첨가하고, 대조시료에는첨가된간섭물질의부피만큼동일부피의용질을첨가하였다. 측정기 1대, 3종로트에걸쳐각각평가하고, 각구간에서의통합평균을계산한후간섭물질이포함된시험시료와대조시료간의평균차이를구하였다. ISO 15197:213 에서언급한허용오차범위를초과할경우, 처음검사한간섭물질농도를최대값으로하여해당간섭물질을 5단계 로희석한후용량반응관계 (dose-response evaluation) 실험을추가하였다. 4) 모세관혈액및정맥혈액을이용한시스템정확도평가시스템정확도평가는 12명의모세관혈액과정맥혈액을모두채혈하고적혈구용적률이 2 65% 인범위에해당되는 개의검체에대하여진행하였다. 각혈당농도구간의검체수는규정 [16] 에따라 5 mg/dl 이하 5개 ( 총분석검체의 5%), 5 8 mg/dl 15 개 (15%), 8 12 mg/dl 2개 (2%), 12 2 mg/dl 3개 (3%), 2 3 mg/dl 15개 (15%), 3 4 mg/dl 1개 (1%), 4 mg/dl 초과 5개 (5%) 가되도록조정하였다. 평가는선택한 개의검체를각각측정기 2대와로트 3종에걸쳐총 6회를평가하였다. YSI 23 STAT PLUS Glucose Analyzer로측정한기준농도를기준으로측정농도와의오차를도식화하고 Consensus Error Grid (CEG) 일치오류격자중어느구역에존재하는지확인하였다. 5) 사용자성능평가사용자성능평가 (user performance evaluation) 는시험지원자에게사용설명서를읽게하고, 처음부터끝까지스스로모세관혈액을이용하여검사를진행하게하였다. 적혈구용적률이 2 65% 에해당하는사람만대상으로하였고, 최종분석에사용되는혈당농도분포및비율은시스템정확도평가와동일하게구성하였다. 검사과정상의오류로측정이실패할경우반복해서 3회까지시험을수행할수있으며, 실패사유를기록하도록하였다. 4. 통계분석정밀도평가를위한평균, 표준편차, 변이계수, 직선성평가를위한상관관계그래프및결정계수, 시스템정확도평가및사용자성능평가를위한 CEG 오류격자등을포함한모든통계자료분석은 Microsoft Excel 213 (Microsoft Corporation, Redmond, WA, USA) 을이용하였다. 결과 1. 정밀도 : 측정반복성, 중간측정정밀도평가 측정반복성과중간측정정밀도평가결과, 각농도구간별 3종로트의통합평균, 표준편차 (95% 신뢰구간 ), 그리고변이계수를 Table 1에나타내었다. 측정반복성평가의표준편차혹은변이계수는농도구간 1부터 5에대하여각각 2.7 mg/dl, 3.2 mg/dl, 3.7%, 3.4%, 3.7% 이었다. 1일에걸친중간측정정밀도평가결과의표준편차혹은변이계수는농도구간 1부터 4에대하여각각 1.4 mg/dl, 2.2%, 2.1%, 2.6% 의값을보였다 (Table 1). www.labmedonline.org 79

Table 1. Data analysis and presentation for precision of GlucoDr.S Measurement repeatability* Intermediate measurement precision Level 1 Level 2 Level 3 Level 4 Level 5 Level 1 Level 2 Level 3 Level 4 Standard concentration, YSI (mg/dl) 39.1 85.5 121. 25. 352. 37.2 118. 353. 52. Mean (mg/dl) 42.3 81.9 113. 185. 315. 38. 119. 353. 514. SD (mg/dl) 2.7 3.2 4.2 6.3 11.7 1.4 2.6 7.4 13.3 CV (%) 6.3 3.9 3.7 3.4 3.7 3.7 2.2 2.1 2.6 *Measurement repeatability test was conducted with 1 meters, 3 reagent lots, and 5 sample levels (Level 1, 3 5 mg/dl; Level 2, 51 11 mg/dl; Level 3, 111 15 mg/dl; Level 4, 151 25 mg/dl; Level 5, 251 4 mg/dl) with glucose concentrations representing hyperglycemic, euglycemic and hypoglycemic conditions. Ten measurements were performed with each combination of meter, reagent lot and sample. In other words, each sample (level) was measured 3 times (1 meters 3 reagent lots 1 measurements); Intermediated measurement precision test was performed with control materials. The evaluation was conducted with one measurement of each sample per day and 1 meters, 3 reagent lots, and 4 sample levels (Level 1, 3 5 mg/dl; Level 2, 96 144 mg/dl; Level 3, 28 42 mg/dl; Level 4, 425 55 mg/dl) with glucose concentrations representing hyperglycemic, euglycemic and hypoglycemic conditions over 1 days. In other words, each sample (level) was measured 3 times (1 meters 3 reagent lots 1 measurement/day 1 days). Measurement value, GlucoDr.S TM (mg/dl) 7 6 5 4 3 2 2. 직선성평가 y=.8894x.6834 R 2 =.9993 y=.9332x 3.9339 R 2 =.999 y=.917x+.4833 R 2 =.9997 직선성평가그래프 (Fig. 1) 를각로트별로살펴보면, 로트 1 은기 울기.8894, y 절편 -.6834, R 2 =.8894, 로트 2 는기울기.9332, y 절편 -3.9339, R 2 =.999, 그리고로트 3 은기울기.917, y 절편.4843, R 2 =.9997 값을갖는것으로나타났다. 즉, 최저혈당 21 mg/dl 에서최고혈당 613 mg/dl 까지세로트의기울기는각각.8894,.9332,.917 으로모두결정계수 (R 2 ) 가.99 이상인높은 직선성을보인다 (Fig. 1). 2 3 4 5 6 7 Fig. 1. Linearity results of the three lots of GlucoDr.S. Linearity test was conducted with 9 m, 3 reagent lots, and 11 sample levels. One tester was performed with each combination of reagent lot and sample within one day. Good linearity was indicated at R 2.99. Table 2. Paired difference interference testing* of GlucoDr.S Substance Test concentration (mg/dl) Mean difference in bias from control Bias for glucose levels, 5- mg/dl (mg/dl) Bias for glucose levels, 25-35 mg/dl (%) Acetaminophen 2 3. 1.6 Ascorbic acid 6 5.7 2.1 Bilirubin 25 1.2 1.3 Cholesterol 55 5.6 2.4 Creatinine 5 1.5 1.7 Dopamine 1.5 6.9 3.4 EDTA 36 1.9.7 Galactose 15 4.3. Gentisic acid 1.8 4.2.4 Glutathione 93 1.1.4 Hemoglobin 2, 7.7 4. Heparin 2.143.7 1.6 Ibuprofen 5 1. 1. Icodextrin 46 3.2 1.2 L-DOPA.5 2.7.2 Maltose 2.3.5 Methyl-DOPA 1.5 6.2 3.4 Pralidoxime iodide (PAM) 8 21.3 26. Salicylic acid 6.5 2. Tolazamide 1 2.9 1. Tolbutamide 64 2..6 Triglycerides 3,333 5.9 1.6 Uric acid 24 23.8 NT Xylose 2 18. 5.2 *The interference effects were described in the instructions for use if they met either of the following performance criteria. For glucose concentrations < mg/dl, the average difference between the test sample and the control sample exceeds 1 mg/dl. For glucose concentrations mg/dl, the average difference between the test sample and the control sample exceeds 1%. Abbreviations: L-DOPA, L-3,4-dihydroxyphenylalanin; NT, Not tested. 3. 간섭물질에대한영향평가 적혈구용적률에따른영향평가결과, 각적혈구용적률구간에 서, 혈당 mg/dl 미만인경우, 평균측정값이정상적혈구용적 률 (42.9%) 에서구한측정값기준으로오차 ±1 mg/dl 이내였다. 8 www.labmedonline.org

Difference (mg/dl) Difference (%) Difference (%) 15 1 5 5 1 15 15 1 5 5 1 15 15 1 5 5 1 15 Hematocrit (%) 2±2 3±2 42±2 5±2 65±2 2±2 3±2 42±2 5±2 65±2 Hematocrit (%) Hematocrit (%) 2±2 3±2 42±2 5±2 65±2 Fig. 2. Packed cell volume test. (A) glucose level: 3 5 mg/dl; (B) 96 144 mg/dl; (C) 28 42 mg/dl. 혈당 mg/dl 이상에서는평균오차가 ±1% 미만이었다 (Fig. 2). 따라서적혈구용적률에따른영향평가결과는허용기준 ( 혈당 mg/dl 미만에서검체와기준장비간의차이 ±1 mg/dl 이 내, 혈당 mg/dl 이상에서검체와기준장비간의차이 ±1% 이내 ) 범위내에있으므로허용 (acceptable) 판정을내릴수있었다. 간섭물질 24 종에대하여실험한결과 (Table 2), 21 종의경우대 조시료와비교하여, 혈당 mg/dl 미만 (5 mg/dl 구간 ) 에 서평균측정값이오차 ±1 mg/dl 미만이었으며, 혈당 mg/ dl 이상 (25 35 mg/dl 구간 ) 에서는오차 ±1% 미만차이를보 A B C Table 3. Dose response interference testing of GlucoDr.S Substance* 였다. 즉, 21 종은허용기준 ( 혈당 mg/dl 미만에서시험시료와 대조시료간의차이 ±1 mg/dl 이내, 혈당 mg/dl 이상에서 시험시료와대조시료간의차이 ±1% 이내 ) 범위내에있으므로 허용판정을내릴수있다. 하지만 PAM (pralidoxime iodide), uric acid, xylose 는간섭물질기준농도내에서판정기준을벗어나서 추가적으로용량반응관계실험을진행하였다 (Table 3). 혈당 mg/dl 미만구간에서는 PAM 2 mg/dl, uric acid 8 mg/dl, xylose 11 mg/dl 이하부터각각대조시료와실험시료의평균차이가 1 mg/dl 미만임을확인하였다. 혈당 mg/dl 이상구간에서는 PAM 1 mg/dl, uric acid 24 mg/dl, xylose 2 mg/dl 이하부터각 각평균차이가 1% 미만이었다. 따라서 ISO 15197:213 기준에의 거하여 PAM, uric acid, xylose 는각각 1 mg/dl, 8 mg/dl, 11 mg/ dl 까지는혈당측정에영향을주지않음을확인하였다. 4. 시스템정확도평가 모세관혈액을이용한시스템정확도평가에서는, 혈당 mg/ dl 미만인 156 개의모세관혈액시료중 153 개의시료 (98.1%) 가기 준값의 ±15 mg/dl 이내에분포하고, 혈당 mg/dl 이상인 444 개의모세관혈액시료중기준값의 436 개시료 (98.2%) 가기준값 의 ±15% 이내에분포하고있었다 (Fig. 3A). 따라서, 전체 6 개의 모세관혈액시료중 589 개시료 (98.2%) 는허용가능한최소성능 기준내에있으며모두일치오류격자 A 영역에분포하는것을확 인하였다 (Fig. 3B). Maximum test concentration (mg/dl) Upper concentration limit at which there is no interference (mg/dl) Glucose levels, 5 mg/dl Glucose levels, 25 35 mg/dl Pralidoxime iodide (PAM) 8 2 1 Uric acid 24 8 24 Xylose 2 11 2 *PAM, uric acid, and xylose exceeded the acceptability criteria in interference testing (Table 2). 정맥혈액으로평가한결과, 혈당 mg/dl 미만인 156 개의정 맥혈액시료모두기준값의 ±15 mg/dl 이내에분포하고, 혈당 mg/dl 이상인 444 개의모세관혈액시료는기준값 ±15% 이 내에 438 개시료 (98.6%) 가분포하고있었다 (Fig. 3C). 따라서, 전체 6 개의정맥혈액시료중 594 개시료 (99.%) 는허용가능한최소 성능기준내에있으며, 일치오류격자분석시, A 영역에 599 개 (99.8%), B 영역에 1 개 (.2%) 가분포하는것을확인하였다 (Fig. 3D). 시스템정확도평가의허용기준은혈당 mg/dl 미만에서검 체와기준장비간의차이 ±15 mg/dl 이내 (95% 신뢰구간 ), 혈당 mg/dl 이상에서검체와기준장비간의차이 ±15% 이내 (95% www.labmedonline.org 81

Difference (mg/dl) 8 6 4 2 2 4 6 8 5 15 2 25 3 35 4 45 5 55 Measurement value, GlucoDr.S TM (mg/dl) 55 5 45 4 35 3 25 2 15 5 5 15 2 25 3 35 4 45 5 55 A B 55 Difference (mg/dl) 8 6 4 2 2 4 6 8 5 15 2 25 3 35 4 45 5 55 Measurement value, GlucoDr.S TM (mg/dl) 5 45 4 35 3 25 2 15 5 5 15 2 25 3 35 4 45 5 55 C D Fig. 3. System accuracy of GlucoDr.S. (A) Difference-plots of capillary blood samples; (B) Clarke error grid analysis of capillary blood samples; (C) Difference-plots of venous blood samples; (D) Clarke error grid analysis of venous blood samples. A zone: No effect on clinical action, B zone: Little or no effect on clinical outcome. 신뢰구간 ), 99% 측정값이일치오류격자의 A, B구간이내에있는것이므로모세관혈액검사와정맥혈액검사모두허용판정을내릴수있었다. 5. 사용자성능평가전체 115개의시료에서혈당 mg/dl 미만인 24개의모세관혈액시료는기준값의 ±15 mg/dl 이내에모두분포하고, 혈당 mg/dl 이상인 91개의모세관혈액시료는기준값의 ±15% 이내에 87개 (95.6%) 의시료가분포하고있었다 (Fig. 4A). 따라서, 전체 115 개의모세관혈액시료중 111개시료 (96.5%) 는허용가능한최소성능기준내에있으며, 모두일치오류격자 A영역에분포하는것을확인하였다 (Fig. 4B). 사용자성능평가의허용기준은혈당 mg/dl 미만에서비전문가가측정한 GlucoDr.S 결과와전문가가측정한기준장비결과간의차이 ±15 mg/dl 이내 (95% 신뢰구간 ), 혈당 mg/dl 이상에서차이 ±15% 이내 (95% 신뢰구간 ) 이므로허용판정을내릴수있었다. 고찰 정밀도평가결과 (Table 1) 를참고하면, 변이계수값은모두 CLIA 88 (Clinical Laboratory Improvement Amendments of 1988) 에서제시된기준인 1% 이내를충족시키는값이다 [21]. 측정반복성평가는 level 1 5, 5개의구간으로, 중간측정정밀도평가는 level 1 4, 82 www.labmedonline.org

55 8 5 Difference (mg/dl) 6 4 2 2 4 6 8 Measurement value, GlucoDr.S TM (mg/dl) 45 4 35 3 25 2 15 5 5 15 2 25 3 35 4 45 5 55 5 15 2 25 3 35 4 45 5 55 A B Fig. 4. User performance test. (A) Difference-plots; (B) Clarke error grid analysis. A zone: No effect on clinical action, B zone: Little or no effect on clinical outcome. 4개의구간으로나누어분석하였는데두평가에서모두가장낮은당농도구간인 level 1의변이계수가가장큰값을보였다. 이는기준값의절대치가낮기때문이다. 그런데 level 1 (3 5 mg/dl) 에해당하는환자는임상적으로다양한저혈당증세가나타나고중추신경계까지영향을미쳐의식장애를보이는응급상황에처하게되므로 [22], 혈당평가가전문의료기관을통해실시될가능성이많다. 즉, 일상생활에서간단히개인용혈당측정기를흔히사용하는군 (level 2 이상 ) 에서는변이계수가모두 4% 미만으로 GlucoDr.S 의정밀도가훨씬더높아진다고볼수있다. National Health and Nutrition Examination Survey [23] 에서는혈장포도당의분석방법변이계수 (analytical method coefficient of variation, CV a) 허용범위를 1.5% 로언급하고있다. 그러나본실험의정밀도는 Table 1을보면, 중간측정정밀도평가의 level 2와 level 3 을제외하고모든변이계수가 1.5% 를벗어난다. 즉, 허용범위를약간상회하는것이다. 하지만현재사용되고있는다른개인용혈당측정시스템의정밀도변이계수역시이러한양상을보이며 [24-26], GlucoDr.S 의경우, 측정반복성평가의최저농도구간만제외하고는변이계수가 4% 미만인상대적으로우수한성적을보인다. 따라서일상생활에서 GlucoDr.S 을유용하게사용할수있을것으로예상된다. 개인용혈당측정기는대부분가정에서일반인들이사용하고있기때문에, 전문인력에의해행해지는평가지침인 CLSI 지침과는다른표준화방법이필요하다. CLSI EP5-A3 정량검사방법의정밀도평가 에서상황에따라제안하는검사디자인은다양하나, 기본적으로 2 2 2 디자인, 2일간하루에 2번 run을하고한번 run할경우반복측정을실시하는것을권고한다 [11]. 반면 ISO 15197은하루내에 1대기기와 3종로트검사지를가지고검사하는측정반복성검사와, 1일에걸쳐하루 1번만검사하는중간측정정밀도검사로정밀도를확인한다. 따라서 ISO 15197은정밀도면에서, CLSI EP5-A3보다간소화되었다고할수있다. 또한 CLSI POCT6 포도당측정을위한샘플종류에따른영향 에서는정확한혈당측정이이루어지는지보기위하여식후혈당변동, 샘플종류에따른측정값차이, 적혈구용적률의변화에따른효과등을모두포함한다 [1]. 하지만 ISO 15197은실제혈당이얼마나정확한지를파악하기위해모세관혈액과정맥혈액의상호관계를평가하는단계가배제되어있다. 즉, 개인용혈당측정시스템이얼마나참값을나타내는지에대한검증은 CLSI보다부족하다고할수있다. 혈장포도당의임계값 (critical value) 은검사실마다약간의기준차이가있으나대다수의검사실이최소임계값은 4 59 mg/dl, 최고임계값은 4 599 mg/dl로잡고있다 [27]. GlucoDr.S 는측정가능범위 (AMR) 가 2 6 mg/dl이며, 이번연구에서최소및최고임계범위를포함하는 21 613 mg/dl 에서높은직선성을보이므로임상적으로환자의관리및치료를위한혈당범위를충분히포함하는것으로볼수있다 [4]. 시험에사용된 24종의간섭물질은혈중참고범위 ( 혹은치료범위 ) 농도이상이며, 대부분은독성을보인다고알려진농도에서간섭효과여부를관찰하였다 (Table 2). 따라서간섭물질판정기준에벗어나지않는 21종의물질은실제체내에서시험농도보다낮은농도로존재할것이기때문에임상적으로사용시측정값에의미있는영향을주지않을것이다. 또한기준을벗어나는 3종중 PAM 과 uric acid는추가로실시한용량반응관계를참고할때, 혈당측정에영향을주지않을최대농도가독성농도근방이므로, 이두 www.labmedonline.org 83

물질도실제사용시에문제를일으키지않을것으로보인다. 단, xylose 경우용량반응관계검사결과확실히 11 mg/dl까지영향을주지않으나혈중에그이상농도로존재할수있기때문에해당기기로혈당을측정할경우이에대한간섭효과의가능성을염두에두어야한다 (Table 3). 정확도평가는모세관혈액과정맥혈액, 이렇게두종류의검체를대상으로각각분석을시행하였다. 모세관혈액검체의 %, 정맥혈액검체의 99.8% 가일치오류격자분석에서 A구역 ( 임상조치에미치는영향없음 ) 에위치하므로 GlucoDr.S 의정확도는매우높음을확인할수있었다. 또한이렇게두종류의검체로정확도를분석함으로써, 일상생활에서모세관혈액으로혈당을측정하는것을반영할뿐만아니라의료기관등에서채혈된정맥혈액을현장검사 (POCT) 로측정하게되는상황을고려하게되었다. 사용자성능평가는시스템정확도평가와분석방법, 평가기준등이동일하나, 모세관혈액만을대상으로하며, 의료인혹은기타전문가가개입하지않고비전문가인사용자가처음부터끝까지스스로모든과정을수행하는것을분석하는데에의의가있다. 본실험에참여한비전문가사용자의연령은 2대에서 7대이상까지분포되어있고학력은초등학교중퇴자부터대학원졸업자까지다양하여검사자의보편성을뒷받침한다고할수있다. 평가후모든사용자들을대상으로검사에대한설문을실시하였는데, 9% 이상의사용자가측정기에나타나는측정결과는이해하기쉽다고답하였으며, 6% 이상의사용자가사용설명서는읽기편하며, 설명서를따라직접측정기를다루기쉬웠고측정기에뜨는오류메시지에대한대처방법이잘되어있다고긍정적으로평가하여, GlucoDr.S 가일상생활에서사용하기쉬운측정기임을유추할수있었다. 본연구에서는기준장비로 YSI 23 STAT PLUS Glucose Analyzer를사용하였는데, 이는국내에사용자가많지않고오래된검사법이라는단점이있다. 그러나적은양의검체 (25 μl) 로검사할수있는비교적정확도가높은장비로검증되었으며 [28], 이전의여러논문 [29, 3] 에서해당장비를기준장비로삼아다양한개인용혈당측정기를평가하였기에, 모세관혈액도대상으로하는본연구에서는유용한기준장비로사용되었다. 연구를진행하면서몇가지애로사항이있었는데, 첫째, 검체량이적어검사실중앙장비와비교가불가능했다는점, 둘째, 시스템정확도평가를진행할때피검사자의모세관혈액과정맥혈액을동시에얻기가어려웠던점, 마지막으로매우낮은혈당혹은매우높은혈당을대표하는검체는환자혈액자체가아니라조작을가해만들었다는점이다. 이번연구는국내최초로 ISO 15197:213 지침을완벽하게따라개인용혈당측정기의성능을평가하였다. GlucoDr.S 은정밀도, 직선성, 간섭물질에대한영향, 시스템정확도, 사용자성능평가 면에서모두우수한수행능력을보이며 ISO 15197:213 지침의기준을충족시켰다. 또한이미발표된다양한혈당측정기평가연구들 [6, 2, 26, 29-31] 과비교해볼때, GlucoDr.S 의성능은다른측정기기에뒤지지않는다. 궁극적으로 GlucoDr.S 개인용혈당측정시스템은효과적인당뇨병관리와치료계획및결정에도움을줄개인용혈당측정기로사용될것으로기대된다. 요약 배경 : 당뇨환자의엄격한혈당관리를위하여개인용혈당측정기의성능은적합하게평가되어야한다. 저자들은본연구에서개인용혈당측정시스템인 GlucoDr.S (All Medicus Co., Ltd., Korea) 의성능을평가하고자하였다. 방법 : 12명의분당서울대병원내과환자의모세관혈액과정맥혈액을대상으로, 개정된 ISO 15197:213 에맞추어국내식품의약품안전처식품의약품안전평가원에서제시한지침에따라 GlucoDr. S 을평가하였다. 평가항목은정밀도 ( 측정반복성, 중간측정정밀도 ), 직선성, 적혈구용적률및기타간섭물질에대한영향, 시스템정확도 ( 모세관혈액및정맥혈액 ) 그리고사용자성능평가로이루어져있다. 평가에사용된기준장비는 YSI 23 STAT PLUS Glucose Analyzer (YSI Life Sciences, USA) 이다. 결과 : GlucoDr.S 의측정반복성평가결과, 당농도 mg/dl 미만구간에서표준편차는 2.7 3.2 mg/dl, mg/dl 이상구간에서변이계수는 3.4 3.7% 였고, 중간측정정밀도평가결과, 당농도 mg/dl 미만구간에서표준편차는 3.7 mg/dl, mg/dl 이상구간에서변이계수는 2.1 2.6% 였다. 직선성평가에서는결정계수 (R 2 ) 가.99 이상이었다. 적혈구용적률및 24종의간섭물질영향평가에서는 xylose를제외한모든간섭물질이참고범위혹은치료범위농도에서측정값에의미있는영향을주지않을것으로나타났다. 시스템정확도평가에서는 95% 이상의시료가, 혈당 mg/dl 미만에서는오차 ±15 mg/dl 이내, 혈당 mg/dl 이상에서는오차 ±15% 이내에포함되었다. 이모든결과는일치오류격자 A영역혹은 B영역에분포한다. 일반사용자가직접처음부터끝까지스스로기기를사용하여검사한사용자성능평가결과역시허용가능범위에포함되었다. 결론 : 이번연구는 ISO 15197:213 지침을완벽하게따라개인용혈당측정시스템 GlucoDr.S 의기능을평가하고결과를발표한것에의의가있다. GlucoDr.S 개인용혈당측정시스템은지침의기준을충족시켰으며, 효과적인개인용혈당측정기로사용될수있음을확인하였다. 84 www.labmedonline.org

AUTHORS DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST No potential conflicts of interest relevant to this article were reported. ACKNOWLEDGEMENTS This work was supported by a grant from the Seoul National University Bundang Hospital (6-216-35). REFERENCES 1. Korean Diabetes Association. Diabetes fact sheet in Korea 213. http:// www.diabetes.or.kr/pro/news/admin.php?code=admin&mode=view &number=1222 (Updated on Nov 215). 2. Korean Diabetes Association. Korean diabetes fact sheet 215. http:// www.diabetes.or.kr/pro/news/admin.php?code=admin&mode=view &number=1223 (Updated on Oct 216). 3. Mazze RS, Shamoon H, Pasmantier R, Lucido D, Murphy J, Hartmann K, et al. Reliability of blood glucose monitoring by patients with diabetes mellitus. Am J Med 1984;77:211-7. 4. Korean Diabetes Association. Treatment guideline for diabetes. J Korean Diabetes 211;12(S1):S1-S244. 5. Diabetes Control and Complications Trial Research Group, Nathan DM, Genuth S, Lachin J, Cleary P, Crofford O, et al. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977-86. 6. Mahoney J and Ellison J. Assessing the quality of glucose monitor studies: a critical evaluation of published reports. Clin Chem 27;53:1122-8. 7. Yoo EH and Lee SY. Glucose biosensors: an overview of use in clinical practice. Sensors (Basel) 21;1:4558-76. 8. International Organization for Standardization. In vitro diagnostic test systems Requirements for blood glucose monitoring systems for selftesting in managing diabetes mellitus. International Organization for Standardization: Geneva, Switzerland, 23. 9. International Organization for Standardization. In vitro diagnostic test systems Requirements for blood-glucose monitoring systems for selftesting in managing diabetes mellitus. International Organization for Standardization: Geneva, Switzerland, 213. 1. Clinical and Laboratory Standards Institute. Effects of different sample types on glucose measurements. 1st ed. POCT6. Wayne, PA: Clinical and Laboratory Standards Institute, 215. 11. Clinical and Laboratory Standards Institute. Evaluation of precision of quantitative measurement procedures; approved guideline. 3rd ed. EP5-A3. Wayne, PA: Clinical and Laboratory Standards Institute, 214. 12. Clinical and Laboratory Standards Institute. Evaluation of the linearity of quantitative measurement procedures: a statistical approach; approved guideline. EP6-A. Wayne, PA: Clinical and Laboratory Standards Institute, 23. 13. Clinical and Laboratory Standards Institute. Interference testing in clinical chemistry; approved guideline. 2nd ed. EP7-A2. Wayne, PA: Clinical Laboratory Standards Institute, 25. 14. Clinical and Laboratory Standards Institute. Measurement procedure comparision and bias estimation using patient samples; approved guideline. 3rd ed. EP9-A3. Wayne, PA: Clinical Laboratory Standards Institute, 213. 15. Freckmann G, Schmid C, Baumstark A, Rutschmann M, Haug C, Heinemann L. Analytical performance requirements for systems for self-monitoring of blood glucose with focus on system accuracy relevant differences among ISO 15197: 23, ISO 15197: 213, and current FDA recommendations. J Diabetes Sci Technol 215;9:885-94. 16. Ministry of Food and Drug Safety of Korea. Evaluation of self-testing glucometer according to ISO 15197:213 guidelines. http://nifds.go.kr/ nifds/3_info/sub_2.jsp?mode=view&board_no=167&article_no=957 (Updated on Jul 216). 17. Ko DH, Lim S, Song SH, Choi SH, Park YJ, Park KU, et al. Performance evaluation of the GlucoDr plus glucometer. Diabetes Technol Ther 21;12:37-12. 18. Food and Drug Administration. Self-monitoring blood glucose test systems for over-the-counter use draft guidance for industry and food and drug administration staff. Food and Drug Administration, 214. 19. Clinical and Laboratory Standards Institute. Interference testing in clinical chemistry; approved guideline. 2nd ed. EP7-A2. Wayne, PA: Clinical and Laboratory Standards Institute, 21. 2. MacRury S, Srinivasan A, Mahoney JJ. Performance of a new meter designed for assisted monitoring of blood glucose and point-of-care testing. J Diabetes Sci Technol 213;7:389-98. 21. Medicare, Medicaid and CLIA programs; regulations implementing the Clinical Laboratory Improvement Amendments of 1988 (CLIA)--HCFA. Final rule with comment period. Fed Regist 1992;57:72-186. 22. Cryer PE, Davis SN, Shamoon H. Hypoglycemia in diabetes. Diabetes www.labmedonline.org 85

Care 23;26:192-12. 23. Lacher DA, Hughes JP, Carroll MD. Biological variation of laboratory analytes based on the 1999-22 National Health and Nutrition Examination Survey: US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics; 21. 24. Kim YB, Seo JY, Lee SY, Park HD. Performance evaluation of glucometer Barozen H based on ISO 15197 standards. Lab Med Online 215; 5:6-14. 25. Kuo CY, Hsu CT, Ho CS, Su TE, Wu MH, Wang CJ. Accuracy and precision evaluation of seven self-monitoring blood glucose systems. Diabetes Technol Ther 211;13:596-6. 26. An D, Chung HJ, Lee HW, Lee W, Chun S, Min WK. Analytical performance evaluation of glucose monitoring system following ISO15197. Korean J Lab Med 29;29:423-9. 27. Schifman RB, Howanitz PJ, Souers RJ. Point-of-care glucose critical values: a Q-probes study involving 5 health care facilities and 2,349 critical results. Arch Pathol Lab Med 216;14:119-24. 28. Burmeister JJ and Arnold MA. Accuracy of the YSI stat plus analyzer for glucose and lactate. Analytical Letters 1995;28:581-92. 29. Tack C, Pohlmeier H, Behnke T, Schmid V, Grenningloh M, Forst T, et al. Accuracy evaluation of five blood glucose monitoring systems obtained from the pharmacy: a European multicenter study with 453 subjects. Diabetes Technol Ther 212;14:33-7. 3. Pfützner A, Mitri M, Musholt PB, Sachsenheimer D, Borchert M, Yap A, et al. Clinical assessment of the accuracy of blood glucose measurement devices. Current Med Res Opin 212;28:525-31. 31. Rajendran R and Rayman G. Point-of-care blood glucose testing for diabetes care in hospitalized patients: an evidence-based review. J Diabetes Sci Technol 214;8:181-9. 86 www.labmedonline.org