대한혈관외과학회지 : 제 23 권제 2 호 Vol. 23, No. 2, November, 2007 동맥증상이있는환자군에서동맥병변예측에대한동맥파형분석법의정확도 1 분당서울대학교병원외과, 2 서울대학교의과대학외과학교실 윤한미 1 ㆍ이태승 1,2 ㆍ배태석 1 ㆍ민승기 2 ㆍ하종원 2 ㆍ정중기 2 ㆍ김상준 2 The Accuracy of Arterial Waveform Analysis in Predicting Arterial Lesions Diagnosed by CT Angiography in Symptomatic Patients Han Mi Yun, R.V.T. 1, Taeseung Lee, M.D., R.P.V.I. 1,2, Taeseok Bae, M.D. 1, Seung Ki Min, M.D. 2, Jongwon Ha, M.D. 2, Joong Kee Chung, M.D. 2 and Sang Joon Kim, M.D. 2 1 Department of Surgery, Seoul National University Bundang Hospital, Seongnam, 2 Department of Surgery, Seoul National University College of Medicine, Seoul, Korea Purpose: CT angiography (CTA) is established as a standard method for the evaluation of patients with peripheral arterial disease. However, there are some drawbacks including overestimation of stenotic lesions as well as dye toxicity and allergic reactions and renal function impairment. Arterial waveform analysis (AWA) is widely accepted as a diagnostic as well as a screening tool in the vascular laboratory. The purpose of this study was to determine the diagnostic accuracy of the AWA compared to the CTA. Method: One hundred thirty-eight AWA procedures were performed among symptomatic patients in our laboratory between October 2004 and February 2007. Sixty patients were entered into the study; they were eligible to have AWA and CTA synchronously. There were 5 women and 55 men with an average age of 64 years. The disease entities were atherosclerosis in 53, Burger's disease in three, popliteal entrapment in 2, arterial embolism in 1 and vascular trauma in one. Continuous-wave Doppler velocity waveforms were recorded at the common femoral, popliteal and dorsal pedal and posterior tibial arterial levels with compression cuffs. Four hundred and eighty segmental interpretations were analyzed and compared with the CTA findings. Result: The sensitivity of the AWA to detect iliac, femoropopliteal and tibial lesions were 93.8%, 64.4% and 88.4% respectively. The specificity of the AWA for the iliac, femoropopliteal and tibial lesions were 87.3%, 93.4% and 95.6% respectively. Overall, the accuracy of the AWA was 88.9% compared to the CT findings. Additional exercise AWA improved the results from 82.8 % to 88.6% for the overall sensitivity of the AWA. The 20 false positives included technical problems in 14 and reference errors (CTA) in six that were due to calcifications. There were 33 false negatives mostly from cases with stenosis and good collaterals. Conclusion: The results of this study showed that the AWA was a valuable method for the prediction of hemodynamically significant arterial lesions. The addition of highly skilled operators and a protocol including a high thigh cuff application might improve the accuracy of this diagnostic method. Key Words: Arterial waveform analysis, CT angiography, Atherosclerosis 중심단어 : 동맥파형검사, CT 동맥조영술, 동맥경화증 책임저자 : 이태승, 경기도성남시분당구구미동 300 우 463-707, 분당서울대학교병원외과 Tel: 031-787-7092, Fax: 031-787-4016 E-mail: tslee@snubh.org 이논문의요지는대한혈관외과학회제 46 차추계학술대회에서발표하였음. 서론조영제의부작용 (allergic reaction, renal toxicity) 없이동맥질환을선별할수있고그에더하여혈역학적정보를제공해주는비침습적혈관검사법들이다양한방법으로발전되어 99
100 대한혈관외과학회지 : 제 23 권제 2 호 2007 왔으며 ABI, Segemental pressure test, Pulse volume recording, Arterial waveform analysis, Duplex scan등이포함된다 (1,2). 최근 duplex ultrasound를이용한진단법이혈관조영술을대체할수있는진단법으로주목받아왔다. 그러나 duplex scan은고가의장비를필요로하며, 검사자간편차가크고, 하지의모든혈관을선별검사하기에는많은시간소모가단점으로지적된다. 또한가장보편화된또다른검사법으로 Ankle brachial index (ABI) 가있다. ABI는재현성이높고정확도도우수하나병변의위치를감별할수없고석회침착이심한동맥에적용할수없다 (3,4). Pulse volume recording (PVR) 은쉽고간편하고전체적인혈류상태를알수있으나동맥분절 (arterial segment) 을분리할수없고진전 (tremor) 이나사지움직임 (limb motion) 이있는환자에서측정이어렵다 (5). 한편동맥병변에의해혈류장애발생시병변의정도에따라도플러동맥파형이변화한다는것이객관적으로증명된바있다 (6). 이러한파형을분석한동맥파형분석법은각동맥분절별 혈역학적변화를볼수있어동맥병변의위치를추측할수있고, 연속적인협착 (sequential stenosis) 이있는경우전반적인혈역학적변화를볼수있으며 duplex scan 에비해검사시간이짧고조작이간편하므로경제적이득이있다. 또한석회침착에도영향을받지않고금식등의전처치가필요없는장점이있다. 현재국내다수의검사실에서동맥파형분석법을사용하고있지만그정확도에대한국내보고가없었다. 이에저자들은하지증상이있었던환자들에대하여, 동맥의해부학적병변을 3차원으로구성ㆍ진단할수있는비교적정확한검사법인 CT 혈관조영술 (7) 을동맥병변진단의기준검사로잡고서, 동맥파형분석법의정확도를비교연구하였다. 1) 대상 대상과방법 2004 년 10 월부터 2007 년 2 월까지 2 년 4 개월간동맥증상 Fig. 1. Pattern of waveform. (A) Triphasic waveform, (B) tri to biphasic waveform, (C) biphasic waveform, (D) bi to monophasic waveform, (E) monophasic waveform, (F) flat waveform, (G) absence. Fig. 2. (A) Damped waveforms at right CFA and both POP A. AWA indicated significant right inflow and both outflow disease, (B) There is significant stenosis at right distal EIA (long arrow), occlusion of both SFA (short arrows) in CTA.
윤한미외 : 동맥증상이있는환자군에서동맥병변예측에대한동맥파형분석법의정확도 101 Table 1. Comparison of arterial waveform analysis and CT angiography for lower extremity arterial occlusive disease Assesment Sensitivity Specificity PPV* NPV Accuracy Waveform analysis Overall 82.2% 93.2% 88.4% 89.2% 88.9% Aortoiliac 93.8% 87.3% 83.6% 95.3% 90.1% Femoropopliteal 64.4% 93.4% 90.4% 73.1% 79.1% Infrapopliteal 88.4% 95.6% 90.7% 94.5% 93.3% Exercise test Pre 82.8% 75.9% 61.1% 90.7% 78.1% Post 88.5% 58.4% 49.2% 91.8% 67.8% *PPV = positive predictive value; NPV = negative predictive value ( 파행증. 휴식기통증 ) 으로내원한환자들중동맥파형분석과 CT 동맥조영술를함께시행한환자 60명을대상으로하였다. 남자가 55명, 여자 5명이었고평균연령은 64 세이었다. 대상질환은동맥경화증 53예, 버거씨질환 3 예, 슬동맥포착증 2예, 동맥색전증1예, 동맥외상 1예이었다. 2) 검사방법 모든검사는시작전앙와위자세에서 5분간휴식후 8 MHz continuous wave ultrasound transducer (VasoGuard, Nicolet vascular Inc.) 를이용하여양측총대퇴동맥, 슬동맥, 전방경골동맥, 후방경골동맥에서기록하였다. 총대퇴동맥은서혜부에서, 슬동맥은슬와정중앙에서, 전방경골동맥은배족부에서, 후방경골동맥은내측과에서기록하였다. 동맥파형은 triphasic waveform, tri to biphasic waveform, biphasic waveform, bi to monophasic waveform, monophasic waveform, flat waveform, absence로나누어해석하였으며 (Fig. 1), 총대퇴동맥에서는 triphasic waveform이외의모든파형을병변이있는것으로, 슬동맥이하에서는직상방분절의파형에비하여분명한변화가있으면의미있는병변이있는것으로판독하였다 (Fig. 2). 분절압검사를위하여상완부, 대퇴부, 발목에 10 cm과 12 cm 커프를감고 automated cuff inflator (VasoGuard, Nicolet vascular Inc.) 를사용하여수축기혈압을측정하였고하지의혈압을상완혈압으로나눈지수를구하였다. 지수가 0.96 미만일경우유의한병변이있는것으로판독하였다. 또한대상환자 60명중 28명에서는운동부하검사를실시하였다. 검사방법은 12 o 의경사에서평균 3.5 km/h의속도로보행을하여하지통증이유발된후보행이더이상어려울때까지혹은증상이없는경우 10분간운동을시행하였으며, 운동직후침대에다시누워최대한빨리동맥파형과상하지혈압을측정하였다. 파형및압력치를종합하여병변여부를판단하였고이를 CT 혈관조영술판독과비교하여결과를통계적분석을시행하였다. Table 2. False positive and false negative finding in waveform analysis False positive or false negative False positive 20 CT angiography error (calcification) 6 Technical error 14 Turbulence flow at CFA* 7 Angle error 7 False negative 33 Mild stenosis 11 Severe stenosis with good collateral 20 Technical error (angle error) 2 *CFA = common femoral artery 3) 통계 60명의환자양측 480 동맥분절에대하여 CT 혈관조영술을기준으로동맥파형분석법에대한민감도, 특이도, 양성예측율, 음성예측율, 정확도를계산하였으며, 운동부하를시행한 28명 224 동맥분절에대한운동부하전 후를분석하였다. 통계처리는통계프로그램 SPSS for window (version 15.0) 를이용하였다. 결 전체 60명에서측정된총 480부위의말초동맥분절에대하여 CT 혈관조영술과비교한결과동맥파형검사의민감도는 82.2%, 특이도는 93.2% 이었고양성예측율 88.4%, 음성예측율 89.2%, 정확도는 88.9% 이었다. 동맥부위별정확도는장골동맥부위는 90.1%, 대퇴-슬동맥부위에서 79.1%, 슬동맥하방부위는 93.3% 였다 (Table 1). 위양성군은 20례를분석하였을떄 6례 (30%) 는석회침착등으로인한 CT 혈관조영술의판독오류였으며, 이를보정하면동맥파형분석 과 N
102 대한혈관외과학회지 : 제 23 권제 2 호 2007 법의정확도는 90.2% 였다. 위양성군의나머지 14례 (70%) 는와류가형성되는지점에서의잘못된위치선정이나, 잘못된입사각으로인해파고에오류가생기는술기적인문제로판단되었다. 위음성군은 33례로, 30 50% 경도협착이 11 례 (33%) 였고, 20례 (60%) 는측부순환이매우잘발달된고도협착 (severe stenosis with good collateral) 인경우로분석되었다 (Table 2). 상기이유로위음성을보였던 31례중 12례 에서운동부하를실시하였으며, 그중 4례에서파형및혈압의변화가있었다. 운동부하시행한 28명을분석한결과운동부하전, 후검사의민감도가운동부하전 82.8% 에서부하후 88.5% 로증가되었으며운동부하후저항성감소로인해위양성군비율이높아지면서특이도가다소감소하였다 (Table 1). Fig. 3. (A) Example of error in CT angiography due to calcification, (B) Damped waveform at right CFA, (C) Conventional angiography demonstrated mild stenotic lesion in external iliac artery, PTA was done. (D) Post- PTA AWA showed complete recovery at right CFA. Fig. 4. (A) No definitive lesions on CTA. (B) Monophasic waveform at right axillary artery. (C) Conventional angiography demonstrated tight stenosis of right subclavian artery.
윤한미외 : 동맥증상이있는환자군에서동맥병변예측에대한동맥파형분석법의정확도 103 고 서론에서언급하였듯이동맥파형분석법은유용한검사법이나정확도를더욱높이기위하여위양성및위음성결과에대한분석이필요하다. 위양성에서의기술적인오류로나타난 14례중 7례는총대퇴동맥측정시와류로인하여파형에왜곡이있었던경우였고, 3례는슬동맥협착시병변보다근위부를기록하게되면슬하방경골동맥협착으로오인하였던것으로분석되었다. 이러한기술적인오류를피하기위해서는기록부위에서이상이있을시기록지점보다몇 cm 근위부및원위부에서의파형을확인하거나, Segmental pressure test를병행하는경우슬동맥협착감별을위하여 3 cuff 이상으로적용하여야하겠다. 나머지 4 례는입사각을잘못주어파고가실제보다낮게측정되어파형해석에오류가생긴경우였다. 한편 CT 혈관조영술에서오류를보였던 case 중일부는동맥파형분석법의진단에따른기준으로치료하여증상이호전되었다 (Fig. 3, 4). 수술을요하는환자군에서의동맥파형분석법은정확한해부학적병변의위치를알기어려운제한점이있는것은사실이나, 본연구에서경험한바동맥파형분석법으로진단된 single isolated lesion에대한중재적시술을고려하는경우 CT angiography를생략하고 C-arm & angiography를이용한진단과치료가가능하리라생각한다. 본연구에서는운동부하검사대상의대부분이운동부하전검사에서이미혈역학적변화가나타났었기때문에운동부하전후로유의한차이는없었으나, Stein 등 (8) 의연구에따르면휴식시 normal ABI 였던 84명의환자중운동부하후 ABI 가 0.9이하로떨어진경우가 26명 (31%) 이었다. 한편운동부하를실시하였던 28명중혈관과신경문제를동시에가지고있었으며환자의주증상이나타날때까지운동하였으나동맥파형이나압력의변화가없었던 3 (10%) 례는신경병증이원인으로추측되어신경학적치료를받았으며, 결과증상이호전되었다. 이러한점들을미루어볼때운동부하검사가동맥파형검사의위음성율을낮추고민감도를향상시킬뿐아니라증상의원인감별에도도움이된다는것을알수있었다. 지금까지살펴본바와같이여러가지술기를개선함으로위양성율과위음성율을낮출수있다. 그러나본연구에서분류한동맥파형의유형은주관 찰 적으로해석될수있을뿐아니라정성적인것으로치료효과판정이나추적관찰에서객관적인비교가어렵고측부순환을포함한전체적인혈류상태를볼수없으며 duplex 보다덜하기는하나검사자간편차가있다는제한점이있다. 따라서동맥파형에대한객관적인지표와측부순환을함께볼수있으며검사자간편차를줄일수있는보조적인검사가필요하다. 우선객관적인지표로는 PI (pulsatility index), RI (resistance index), AT (acceleration time) 등이있다 (9-11). De Morais Filho 등 (10) 에따르면 PI를적용했을경우정확도 82% 이상, RI를적용했을경우 90% 이상의정확도를보여주며, AT를적용하였을때혈역학적으로 insignificant group (50% 이하 stenosis) 에서 96% 이상, hemodynamically significant stenosis group (50% 이상 stenosis) 에서 60% 의정확도를보여준다. 그러나이러한지표의사용은다소번거롭기때문에, 현재동맥파형에정량적인정보를더함과동시에검사자간편차를줄일수있는 ABI test가흔히사용되고있다. 그러나 pressure test는 uncompressible vessel 에적용할수없고동맥압력이상의압력을환자에게가하기때문에때때로환자들에게고통을줄수있다. 한편 uncompressible vessel 에적용할수있고측부순환까지포함한검사법으로 PVR 있으나이방법역시정성적인검사법이다. 이러한단점들을극복한검사법으로 Transfer Function Index (TFI) 가최근소개되었다. TFI는 PVR wave를 averaging하는방법으로 averaging된파형은 fast Fourier transform analyser에의해이산적분된 frequency로변환되고, TFI는 inflow frequency에대한 outflow frequency의비로나타내진다 (TF = FTd(f) / FTp(f)). 따라서 transfer function이내려가면 1.0 이하, 높아지면 1.0 이상의지수를나타낸다 (12). Doverspike 등 (12) 에따르면 TFI 0.94를기준으로하였을때말초동맥질환을발견하는데있어민감도 92%, 특이도 90% 로서높은유용성을보여준다. 또한 transit time이 TFI와함께기록되기때문에 collateral flow에대한간접적인정보를알수있다. 더욱이환자에게고통없이 10분이내의짧은시간에자동으로측정되기때문에동맥파형분석법을보완할수있는유용한검사법으로사용될수있으리라사료된다 (13,14). 따라서향후객관적인지표를사용한동맥파형분석법에대한연구와동맥파형의정확도와유용성을높여줄수있는 TFI 검사법추가에대한연구가필요할것이다. 결 동맥파형분석은간편하고, 신속하며임상적유용성이높음을알수있었다. 검사의정확도를더욱높이기위해서는검사자의숙련도및검사술기개선이필수적이다. 향후파형분석의객관적지표인 PI, RI, AT 분석이나 PVR 이나 TFI 등의측정법추가에대한연구가필요하리라생각된다. 론 REFERENCES 1) 정석인. Intermittent claudication. 대한혈관외과학회지 2002; 18(1):174-178. 2) Steven C. Noninvasive vascular laboratory for evaluation of peripheral arterial occlusive disease: Part I hemodynamic principles and tools of the trade. J Vasc Interv Radiol 2000; 11(9):1107-1114.
104 대한혈관외과학회지 : 제 23 권제 2 호 2007 3) Baker JD, Dix D. Variability of Doppler ankle pressures with arterial occlusive disease: an evaluation of ankle index and brachial-ankle pressure gradient. Surgery 1981;89(1):134-137. 4) Quigley FG, Faris IB, Duncan HJ. A comparison of Doppler ankle pressures and skin perfusion pressure in subjects with and without diabetes. Clin Physiol 1991;11(1):21 25. 5) Diagle R. Techniques in Noninvasive Vascular Diagnosis. 2nd ed. Littleton:Summer Pulishing;2002. 6) 이홍기, 염명결. 하지동맥의도플러파형에대한푸리에분석. 대한혈관외과학회지 2001;17(1):56-62. 7) Catalano C, Fraioli F, Laghi A, Napoli A, Bezzi M, Pediconi F, et al. Infrarenal aortic and lower-extremity arterial disease: diagnostic performance of multi-detector row CT angiography. Radiol 2004;231(2):555-563. 8) Stein R, Hriljac I, Halperin JL, Gustavson SM, Teodorescu V, Olin JW. Limitation of the resting ankle-brachial index in symptomatic patients with peripheral arterial disease. Vasc Medicine 2006;11(1):29-33. 9) Bagi P, Sillesen H, Bitsch K, Hansen HJ. Doppler waveform analysis in evaluation of occlusive arterial disease in the lower limb: comparison with distal blood pressure measurement and arteriography. Eur J Vasc Surg 1990;4(3):305-311. 10) De Morais Filho D, Miranda F Jr, Del Carmen Janeiro Peres M, Barros N Jr, Buriham E, Salles-Cunha SX. Segmental Waveform analysis in the diagnosis of peripheral arterial occlusive diseases. Ann Vasc Surg 2004;18(6):714-724. 11) Bagi P, Sillesen H, Hansen HJ. Quantitative Doppler ultrasound evaluation of occlusive arterial disease in the lower limb. Eur J Vasc Surg 1988;2(6):409-415. 12) Doverspike GA, Bendick PJ, Brown D, Hubbard J, Shelpman D, Shanley CJ. Evaluating the lower extremities for peripheral arterial obstructive disease using the transfer function index. J Vasc Ultrasound 2004;28(1):9-12. 13) Ihlberg LH, Mätzke S, Albäck NA, Roth WD, Sovijärvi AR, Lepäntalo M. Transfer function index of pulse volume recordings: a new method for vein graft surveillance. J Vasc Surg 2001;33(3):546-553. 14) Tan CP, Civil I. Transfer function index: is it a reliable method for vein graft surveillance? ANZ J Surg 2003;73(12): 1032-1035.