06-유승협

대한응급의학회지제 25 권제 2 호 Volume 25, Number 2, April, 2014 원 저 Resuscitation 저체온치료를받은병원전심정지환자에서시간가중평균산소분압과예후와의관계 전남대학교의과대학응급의학교실 유승협 이병국 정경운 정용훈 이성민 이동훈 송경환 허 @ 탁 민용일 The Relation between Time-weighted Mean Oxygen Tension and Outcome in Out-of-hospital Cardiac Arrest Survivors Treated with Therapeutic Hypothermia Seung Hyup Ryu, M.D., Byung Kook Lee, M.D., Kyung Woon Jeung, M.D., Yong Hun Jung, M.D., Sung Min Lee, M.D., Dong Hun Lee, M.D., Kyung Hwan Song, M.D., Tag Heo, M.D., Yong Il Min, M.D. Purpose: Studies to determine the relation between oxygen tension and outcome in cardiac arrest survivors treated with therapeutic hypothermia (TH) are lacking. We investigated the relation of time-weighted mean oxygen tension (TWMO 2 ) and outcome in cardiac arrest survivors treated with TH. Methods: This was a retrospective observational study including 177 out-of-hospital cardiac arrest (OHCA) survivors. The patients were divided into four categories according to quartile values of TWMO 2. The primary outcome was neurologic outcome at discharge and the secondary outcome was all cause in-hospital mortality. We assessed neurologic outcome using the Cerebral Performance Categories (CPC) at hospital discharge. Neurologic outcome was dichotomised as either good neurologic outcome (CPC1 and CPC2) or poor neurologic outcome (CPC 3 to 5). The odds ratio with 95% confidence interval (CI) was estimated. Results: The median value of PaO 2 was 139(104.5-170.0) mmhg. Among a total of 1,239 PaO 2 values, 22(1.8%) values were hypoxia (<60 mmhg) and 16(1.3%) values were hyperoxia (>300 mmhg). Results of univariate logistic regression analysis showed a significantly low odds ratio for 책임저자 : 이병국광주광역시동구제봉로 42 전남대학교의과대학응급의학교실 Tel: 062) 220-6809, Fax: 062) 228-7417 E-mail: bbukkuk@hanmail.net 접수일 : 2013년 12월 12일, 1차교정일 : 2013년 12월 16일게재승인일 : 2014년 2월 4일 174 poor neurologic outcome [0.353(95% CI, 0.133-0.938) and 0.321(95% CI, 0.121-0.850), respectively] and for in-hospital mortality [0.338(95% CI, 0.132-0.870) and 0.387(95% CI, 0.154-0.975), respectively] for the third quartile and the fourth quartile. However, results of multivariate logistic regression analysis showed no significant relation between TWMO 2 and outcomes. Conclusion: In OHCA survivors treated with TH, timeweighted oxygen tension did not show an association with neurologic outcome and in-hospital mortality. Key Words: Cardiac arrest, Outcome, Oxygen, Hyperoxia Department of Emergency Medicine, School of Medicine, Chonnam National University, Gwangju, Korea Article Summary What is already known in the previous study Hyperoxia during the post resuscitation period is associated with poor clinical outcome in cardiac arrest survivors. The result of the association between hyperoxia and outcome in cardiac arrest survivors treated with therapeutic hypothermia is limited. What is new in the current study Under controlled oxygenation according to the guidelines, hyperoxia is not frequent and the time-weighted mean oxygen tension during post cardiac arrest care did not show an association with in-hospital mortality and neurologic outcome in cardiac arrest survivors treated with therapeutic hypothermia. 서 심정지후증후군은심정지때문에발생하는허혈과자발순환회복후에발생하는재관류에의한병리적과정을통해 론

유승협외 : 저체온치료를받은병원전심정지환자에서시간가중평균산소분압과예후와의관계 / 175 주로발생한다 1). 심정지후증후군환자에게산소를공급하는것은매우기본적이며필수적인조치이지만, 동물실험을통해서허혈손상을입은뇌에과량의산소를공급하는것은활성산소를형성함으로써재관류손상을심화시키는것으로밝혀졌다 2,3). 그렇기때문에미국심장협회 (American heart association, AHA) 2010 지침은심정지후증후군환자에게 94% 이상의산소포화도를유지할수있는가장낮은농도의산소를공급하도록권고하고있는데, 이러한권고안역시임상적인근거가명확하지는않다 4). 중환자실에입원한비외상심정지환자들을대상으로산소분압과예후와의관계를규명하고자한후향적연구들이최근에발표되었다 5-7). Kilgannon 등 5,6) 에의하면, 60 mmhg 미만의저산소혈증뿐만아니라 300 mmhg 이상의고산소혈증은병원내사망과유의한관계가있는것으로나타났고, 60 mmhg 미만의저산소혈증환자들을제외하고분석한이들의또다른연구에서는산소분압이 100 mmhg 증가할때마다병원내사망률이 24% 씩증가한다고보고하였다. 하지만, Bellomo 등 7) 은고산소혈증과병원내사망의연관성은약하며, 질환의중증도와같은공변량을감안한다면, 고산소혈증자체는심정지후증후군환자에게확고부동한예후관련인자가아닌것으로보고하였고, 소아심정지환자를대상으로산소분압과병원내사망의관계를분석한연구에서도유의한관계는나타나지않았다 8). 최근의임상연구결과를종합해보면고산소혈증과예후의연관성에대해서는이견이있음을알수있다. 앞선연구들은다기관에서시행되었고대규모의환자들을대상으로했다는장점이있기는하지만, 심정지후증후군의표준치료법인저체온치료를적용한환자의비율이너무낮았다는단점이있다. 심정지후저체온치료를시행한환자를대상으로고산소혈증과병원내사망이나신경학적예후와의관계를규명하고자한연구는매우적다 9). 또한, 저체온치료를적용한환자만을대상으로했던연구라고할지라도 Kilgannon 등 5,6) 이나 Bellomo 등 7) 에의한연구와마찬가지로중환자실에입원한후에측정한단한개의산소분압값만을이용하여예후와의관련성을분석하였다 5-7,9). 하지만, 자발순환이회복되더라도심정지후증후군은수시간이상지속되고환자의임상적상황은시시각각변하기때문에, 여러시점에서측정한산소분압값을이용하여분석하는것이타당하리라추정한다 10). 따라서, 본연구에서는병원전심정지후자발순환이회복되어저체온치료를받은환자들을대상으로자발순환이후부터저체온치료가종료되는시점까지여러시점에서측정한산소분압을이용하여시간을반영한산소분압값을구하고예후와의관련성여부를확인하고자하였다. 대상과방법본연구는전남대학교병원연구윤리심의위원회의허가 (CNUH-2013-101) 를받고진행되었다. 1. 연구대상 2008년 1월부터 2012년 12월까지전남대학교병원응급의료센터로내원한환자중병원전심정지후자발순환이회복되어저체온치료를포함한소생후치료를받았던 16세이상의환자들을대상으로하였다. 자발순환회복후의식이회복되거나, 말기질환자, 혈역학적으로불안정한환자, 위장관출혈이있는환자, 그리고보호자가저체온치료를거부한환자를제외하고저체온치료를시행하였다. 저체온치료치료도중치명적인부정맥이발생하거나혈역학적으로불안정하여저체온치료가중단된경우, 체외순환보조 (extracorporeal membrane oxygenation, ECMO) 를받았던경우, 저체온치료유지기간이나목표체온이달랐던경우 ( 목표체온이 <32 C 이거나 >34 C, 유지기간이 48시간혹은 72시간 ), 동맥혈가스분석자료가누락된경우는연구대상에서제외되었다. 2. 저체온치료 33 C 의목표체온에도달하기위하여 4 생리식염수의정주와얼음팩등의고식적인방법과함께온도조절냉각요 (Blanketrol II, Cincinnati Subzero Products, USA), 냉각용중심정맥카테터 (COOLGARD 3000 Thermal Regulation System, Alsius Corporation, USA), 또는온도조절패드 (Artic Sun Energy Transfer Pads TM, Medivance Corp, Louisville, USA) 를이용하여저체온치료유도를시작하였다. 목표체온에도달하면 24시간동안 32~34 C 의저체온치료를유지하고, 이후에는시간당 0.25~0.5 C 의속도로재가온을실시하였다. 저체온치료를실시하는동안진정과떨림의방지를위해 fentanyl 혹은 remifentanyl과, midazolam을투여하였다. 떨림을조절하기위해필요에따라 atracurium도투여하였다. 산소분압을적정하기위해여러차례동맥혈가스분석을시행하였다. 채혈은자발순환회복후, 저체온치료의시작, 목표체온에도달하였을때, 목표체온도달후매6시간마다시행하였고, 임상의의판단에따라추가적으로실시하였다. 3. 연구방법의무기록지를후향적으로조사하였다. 연령, 성별, 기저질환 ( 관상동맥질환, 심부전, 고혈압, 당뇨, 폐질환, 신질환, 뇌혈관질환, 간질환 ) 의여부, 심정지의목격여부, 초기심

176 / 대한응급의학회지 : 제 25 권제 2 호 2014 율동, 심정지의원인 ( 심인성혹은비심인성 ), 심정지로부터자발순환회복까지의시간, 자발순환회복으로부터저체온치료유도까지의시간, 저체온치료의시작으로부터목표체온까지도달시간, 재가온기간, 퇴원시생존여부를조사하였다. 질환의중증도를반영하기위하여내원후첫 24 시간동안의 sequential organ failure assessment (SOFA) 점수를측정하였다 11). 또한퇴원시의신경학적예후판정을위하여 cerebral performance category scale (CPC) 를측정하였다 12). CPC 1, 2점은예후우량군으로 CPC 3~5점은예후불량군으로정의하였다. 일차결과 (primary outcome) 는신경학적예후로이차결과 (secondary outcome) 는병원내사망으로정의하였다. 여러차례동맥혈가스분석이시행되었기때문에, 각환자마다시간을반영한산소분압을제시하기위하여시간가중평균산소분압을계산하였다. 각환자마다총 7회 ( 자발순환회복후, 저체온치료시작, 저체온치료유지기의시작, 유지기시작후 6시간째, 12시간째, 18시간째, 24시간째 ) 의동맥혈가스분석결과와자발순환회복후부터재가온이완료되는저체온치료의종료까지총 7구간의시간간격을이용하였다. 매동맥혈가스분석시점의사이시간과가스분압을곱한총합을전체시간으로나누어각각의시간가중평균산소분압으로정의하였다. 4. 통계분석 명목변수들은빈도 ( 백분율 ) 로표현하였고, 두군간의비교를위해서조건에따라카이제곱검정이나 Fisher의정확검정을이용하였다. 연속변수는정규성분포검정결과모두비정규분포를보였기때문에중앙값 ( 사분위값 ) 으로표현하였고, 두군의비교를위해서 Mann-Whitney U 검정법을이용하였다. 시간가중평균산소분압은사분위값을기준으로구분한후단변량로지스틱분석을이용하여승산비 (95% 신뢰구간 ) 를구하였다. 신경학적예후와병원내사망에관련이있는독립적인변수를찾기위해이분형다변량로지스틱회귀분석을이용하였다. 다변량로지스틱회귀분석에는단변량분석에서 p<0.10의결과를나타냈던변수들만을포함하였고, 변수들간의다중공선성의여부를진단하여포함여부를결정하였다. 통계분석은 PASW/ SPSS TM software, version 18(IBM Inc., Chicago, USA) 를사용하였으며, p값이 0.05 미만인경우를통계학적으로유의하게판정하였다. 결 1. 연구대상의일반적인특성및일차목적과이차목적에따른비교 과 연구기간동안에 476 명의환자가병원전심정지후자 Fig. 1. Flow diagram included patients. A total of 476 out-of-hospital cardiac arrest (OHCA) patients achieved spontaneous circulation. Of them, 261 OHCA survivors treated with therapeutic hypothermia. Finally, 177 patients were included in the present study.

유승협외 : 저체온치료를받은병원전심정지환자에서시간가중평균산소분압과예후와의관계 / 177 Table 1. Patient demographics and clinical characteristics. p Neurologic outcome In-hospital mortality p Favourable (n=59) Unfavourable (n=118) Survivors (n=127) Nonsurvivors (n=50) Age, yr, median (IQR) 48.0 (37.0-57.0) 63.0 (48.0-72.0) <0.001 53.0 (43.0-68.0) 63.0 (50.3-71.0) <0.042 Male gender, n (%) 44 (74.6) 083 (70.3) <0.555 <94 (74.0) 33 (66.0) <0.286 Pre-existing illness, n (%) Coronary artery disease 09 (15.3) 019 (16.1) <0.884 <20 (15.7) <8 (16.0) <0.967 Heart failure 05 (08.5) 011 (09.3) <0.853 <11 (<8.7) <5 (10.0) <0.775 Hypertension 12 (20.3) 055 (46.6) <0.001 <43 (33.9) 24 (48.0) <0.081 Diabetes 07 (11.9) 041 (34.7) <0.001 <34 (26.8) 14 928.0) <0.869 Pulmonary disease 02 (03.4) 014 (11.9) <0.064 <11 (<8.7) <5 (10.0) <0.775 Renal impairment 00 (00.0) 012 (10.2) <0.011 <<8 (<6.3) <4 (<8.0) <0.742 Cerebrovascular accident 02 (03.4) 012 (10.2) <0.146 <11 (<8.7) <3 (<6.0) <0.760 Hepatic disease 01 (01.7) 004 (03.4) <0.666 <<4 (<3.1) <1 (<2.0) <1.000 Witness of collapse, n (%) 51 (86.4) 085 (72.0) <0.032 104 (81.9) 32 (64.0) <0.011 First monitored rhythm, n (%) <0.001 <0.001 Shockable 36 (61.0) 013 (11.0) <45 (35.4) <4 (<8.0) Non-shockable 23 (39.0) 105 (89.0) <82 (64.6) 46 (92.0) Etiology, n (%) <0.001 <0.015 Cardiac 52 (88.1) 058 (49.2) <86 (67.7) 24 (48.0) Non-cardiac 07 (11.9) 060 (50.8) <41 (32.3) 26 (52.0) Time from collapse to ROSC, min, median (IQR) 25.0 (15.0-34.0) 32.0 (25.0-41.5) <0.001 29.0 (20.0-40.0) 35.0 (24.5-45.0 ) <0.043 Time from ROSC to initiation of TH, min, median (IQR) 213 (160-300). 210 (160-300). <0.833 230 (163-300). 185 (150-300). <0.377 Time from initiation of TH to achieving target temperature, h, 3.5 (2.0-5.5)0 2.0 (1.4-4.0)0 <0.001 3.0 (2.0-5.0)< 2.0 (1.0-3.1)< <0.001 median (IQR) Rewarming duration, h, median (IQR) 10.5 (6.5-12.0)0 11.0 (7.0-13.0)0 <0.658 11.0 (7.0-12.0)< 10.0 (5.5-13.0)< <0.723 SOFA score 8 (6-10)0. 10 (8-12)00. <0.001 9 (7-11)<. 10 (9-12)<<. <0.005 TWMO2, mmhg, median (IQR) 0149.1 (129.0-160.6) 0135.1 (117.8-155.5) <0.039 <143.9 (125.0-160.0)..126.9 (106.8-155.4) <0.023 IQR: interquartile range, ROSC: restoration of spontaneous circulation, TH: therapeutic hypothermia, SOFA: sequential organ failure assessment, TWMO2: time-weighted mean oxygen tension, TWMCO2: time-weighted mean carbon dioxide tension

178 / 대한응급의학회지 : 제 25 권제 2 호 2014 발순환을회복하였으며이들중저체온치료를시도한환자는 261례였지만, 저체온치료의방법이표준프로토콜 ( 목표체온은 33±1 C, 유지기간은 24시간 ) 과달랐던 57례, 저체온치료가중지된 19례를제외한 185례가저체온치료를완료하였다. 또한체외순환보조요법 3례, 동맥혈가스분석결과가누락된 5례를제외하여, 177례가분석에포함되었다 (Fig. 1). 생존여부와신경학적예후에따른두군의비교는 Table 1에기술되었다. 59(33.3%) 례가예후우량군에해당하였으며, 50(28.2%) 례가병원내에서사망하였다. 예후불량군에비해예후우량군의고혈압, 당뇨, 콩팥질환의빈도가유의하게낮았다 (Table 1). 예후우량군과생존군은예후불량군과사망군과각각비교하였을때, 연령, 심정지목격의여부, 초기심율동, 심정지의원인, 목표체온까지의시간, SOFA 점수에서유의한차이를나타냈다 (Table 1). mmhg (44 례 ), 159.3 mmhg(44 례 ) 의분포를보였다. 3. 시간가중평균산소분압의신경학적예후와병원내사망과의관계 연속변수인시간가중평균산소분압을단변량로지스틱회귀분석을이용하여예후불량군과사망군에대해승산비를구하면, 각각 0.992(95% 신뢰구간, 0.982-1.002; p=0.118) 와 0.990(95% 신뢰구간, 0.979-1.001; 2. 동맥혈산소분압의분포와시간가중평균산소분압 전체 1,239개산소분압값의중앙값은 139(104.5-170.0) mmhg의분포를나타냈고, 60 mmhg 미만의산소분압은 22(1.8%) 개, 300 mmhg를초과하는산소분압은 16(1.3%) 개였다 (Fig. 2). 각시간별산소분압은자발순환이후에비해시간이지날수록그사분위값의범위가점차줄어드는양상으로나타났다 (Fig. 3). 시간가중평균산소분압전체의중앙값은 140.0(122.1-159.3) mmhg의분포를보였으며, 예후우량군과생존군에서는각각 149.1(129.0-160.6) mmhg, 143.9(125.0-160.0) mmhg이었고, 예후불량군과사망군에서는각각 135.1(117.8-155.5) mmhg, 126.9(106.8-155.4) mmhg로유의한차이를보였다 (Table 1). 시간가중평균산소분압을사분위값을기준으로나누면 <122.1 mmhg (45례), 122.1~140.0 mmhg(44례 ), 140.0~159.3 Fig. 2. Distribution of partial oxygen tension. Among a total of 1,239 PaO 2 values, 1,201(96.9%) values were within the range of 60~300 mmhg. Only 22(1.8%) values and 16(1.3%) values were <60 mmhg and >300 mmhg, respectively. Fig. 3. Median with interquartile range of partial oxygen tension of each time points. ROSC: return of spontaneous circulation, TH: therapeutic hypothermia, M0: initiation of maintenance phase, M6: 6 hr after maintenance, M12: 12 hr after maintenance, M18: 18 hr after maintenance, M24: 24 hr after maintenance (initiation of rewarming), TWMO 2: time-weighted mean oxygen tension.

유승협외 : 저체온치료를받은병원전심정지환자에서시간가중평균산소분압과예후와의관계 / 179 p=0.071) 로유의하지않았다. 하지만, 시간가중평균산소분압의사분위값을기준으로네군으로구분한후범주형변수로적용하면 Table 2와같다. 첫번째사분위군을기준으로비교했을때, 세번째, 네번째사분위군들은첫번째사분위군에비하여, 예후불량군과사망군에대한승산비가유의하게낮게나타났다. 단변량분석을통해 p<0.10으로나타난변수들과사분위군으로구분한시간가중평균산소분압을다변량로지스틱회귀분석을이용하여예후와의관계를분석하였다. 신경학적예후와독립적인관계를나타낸변수는연령, 심정지의원인, 초기심율동, 자발순환회복까지의시간이었고, 병원내사망과독립적인관계를나타낸변수는심정지목격의여부, 초기심율동, SOFA 점수였으며, 시간가중평균산소분압은신경학적예후와병원내사망두가지모두와유의하지않았다 (Table 3). 각시간별산소분압과최대산소분압그리고최소산소분압을이용하여단변량로지스틱회귀분석과다변량로지스틱회귀분석결과자발순환회복후산소분압만이신경학적예후와독립적인관계가있는것으로나타났다 (Table 4). 또한각시간별산소분압과최대산소분압, 그리고최소산소분압을각각의사분위값을기준으로사분위군으로나누어예후와의관계를조사하였다. 그결과자발순환회복후산소분압에서만첫번째사분위군에비해네번째사분위군이예후불량군이될승산비가 5.404(95% 신뢰구간, 1.134-25.749, p=0.034) 로유의하게나타났다. 고찰심정지후저체온치료를시행받은환자들에서산소분압 Table 2. Univariate association between time-weighted mean oxygen tension and outcomes. Poor neurologic outcome OR (95% CI) p In-hospital Mortality OR (95% CI) p Quartiles of TWMO 2 1 st quartile, <122.1 mmhg reference (1.000) reference (1.000) 2 nd quartile, 122.1~140.0 mmhg 0.403 (0.151-1.073) 0.069 0.478 (0.196-1.165) 0.105 3 rd quartile, 140.0~159.3 mmhg 0.353 (0.133-0.938) 0.037 0.338 (0.132-0.870) 0.025 4 th quartile, 159.3 mmhg 0.321 (0.121-0.850) 0.022 0.387 (0.154-0.975) 0.044 TWMO 2: time-weighted mean oxygen tension, OR: odds ratio, CI: confidence interval Table 3. Multivariate logistic regression for outcomes. Poor neurologic outcome OR (95% CI) p Age 1.078 (1.045-1.112)0 <0.001 Non-cardiac etiology 7.164 (2.108-24.344) <0.002 Shockable rhythm 0.095 (0.031-0.290)0 <0.001 Time from collapse to ROSC 1.070 (1.033-1.107)0 <0.001 Quartiles of TWMO 2 1 st quartile, <122.1 mmhg reference (1.000) 2 nd quartile, 122.1~140.0 mmhg 0.527 (0.123-2.249)0 <0.387 3 rd quartile, 140.0~159.3 mmhg 1.829 (0.353-9.461)0 <0.472 4 th quartile, 159.3 mmhg 1.004 (0.193-5.234)0 <0.996 In-hospital Mortality OR (95% CI) p Witness 0.385 (0.172-0.861)0 <0.020 Shockable rhythm 0.208 (0.068-0.634)0 <0.006 SOFA 1.230 (1.073-1.410)0 <0.003 Quartiles of TWMO 2 1 st quartile, <122.1 mmhg reference (1.000) 2 nd quartile, 122.1~140.0 mmhg 0.542 (0.198-1.482)0 <0.233 3 rd quartile, 140.0~159.3 mmhg 0.476 (0.157-1.445)0 <0.190 4 th quartile, 159.3 mmhg 0.465 (0.149-1.448)0 <0.186 ROSC: restoration of spontaneous circulation, TWMO 2: time weighted mean oxygen tension, SOFA: sequential organ failure assessment, OR: odds ratio, CI: confidence interval

180 / 대한응급의학회지 : 제 25 권제 2 호 2014 의중앙값은 139(104.5-170.0) mmhg였고, 60 mmhg 미만의저산소혈증이나 300 mmhg 이상의고산소혈증의빈도가낮았다. 시간가중평균산소분압을사분위값으로구분하여비교하였을때, 단변량분석에서는첫번째사분위군을기준으로세번째, 네번째사분위군이예후불량군과병원내사망군에대한승산비가유의하게낮았지만, 다변량분석결과시간가중평균산소분압은신경학적예후나병원내사망두가지모두와독립적인관계를보이지는않았다. 본연구에서전체산소분압중 60 mmhg 미만의산소분압은 22(1.8%) 개, 300 mmhg를초과하는산소분압은 16(1.3%) 개로매우낮은수준을나타냈다. 기존의연구들에서는한환자당한번의산소분압값을이용하였기때문에직접적으로본연구와비교하기는어렵지만, Kilgannon 등 5) 에의한연구에서는 63% 의환자에서고산소혈증이나타났으며, 이들의또다른연구에서도산소분압의중앙값은 231(149-349) mmhg로본연구에비해높게나타났다 6). 이에반해 Bellomo 등 7) 에의한연구에서는고산소혈증의빈도가 10.6% 로 Kilgannon 등 5,6) 에비해서는훨씬낮았는데, 이렇게차이를보이는이유는각연구에서산소분압을선택하는기준이달랐기때문이다. 본연구와기존연구사이에산소분압이크게차이가났던이유는치료프로토콜이서로달랐기때문으로생각된다. Kilgannon 등 5,6) 의두연구에포함된연구대상들은 2001부터 2005년사이에치료를받았던환자들이었고 Bellomo 등 7) 의연구기간은 2000년부터 2009년까지였는데, AHA 2000 지침 과 2005년지침에는심정지후증후군환자에대한산소투여목표나세부적인방법에대한권고안이아직마련되지않았었다 13,14). 고산소혈증을피하면서투여산소량을적정하는방법의가장큰합병증은저산소혈증의발생이라고할수있는데, 본연구의결과로는기존연구들에비해낮은산소분압을유지하면서도저산소혈증의빈도가높지않은것으로나타나서실제적으로어렵지않게목표산소분압을유지할수있을것으로생각한다. 저체온치료는 2002년에발표된무작위대조군연구들을통해심정지후신경학적예후향상에효과적인것으로입증되어 ILCOR 2010 지침에서는표준치료법으로권장하고있다 4,15,16). 하지만, 고산소혈증과예후와의관계를연구한다기관임상연구들의연구대상은 2000년부터포함되었기때문에매우소수의환자들만체온조절치료를받았다 5-7). 저체온치료는조직으로의산소공급, 대사, 세포호흡및활성산소형성에영향을미친다. 저체온치료로인해대사가감소하면조직에서산소의이용이줄어들어고산소혈증이발생할수있지만, 반면에세포죽음과밀접한관련이있는활성산소의생성이억제되는효과가있다 17,18). 저체온치료가산소의이용과대사에대한직접혹은간접적으로미치는영향때문에저체온치료를받은환자들에대한산소분압과예후와의관계는추가로연구되어야한다. 따라서본연구에서는저체온치료를받은환자들만을대상으로하였다. 저체온치료를받은심정지환자들만을대상으로한 Janz 등 9) 의연구는 Kilgannon 등 6) 의연구와마찬가 Table 4. Univariate and adjusted odds ratio of each PaO 2 for outcomes. Poor neurologic outcome Unadjusted OR (95% CI) p Adjusted OR (95% CI) p After ROSC 1.008 (1.003-1.013) 0.003 1.010 (1.002-1.018) 0.010 Initiation of TH 1.000 (0.996-1.004) 0.980 1.003 (0.996-1.010) 0.375 Start of maintenance 0.998 (0.991-1.004) 0.477 1.008 (0.995-1.021) 0.234 6 h after maintenance 0.987 (0.978-0.996) 0.003 0.994 (0.980-1.008) 0.428 12 h after maintenance 0.995 (0.987-1.002) 0.155 0.995 (0.981-1.008) 0.431 18 h after maintenance 0.993 (0.985-1.001) 0.103 0.999 (0.986-1.013) 0.914 24 h after maintenance 0.991 (0.982-1.000) 0.055 0.999 (0.983-1.015) 0.900 Minimum PaO 2 1.000 (0.989-1.012) 0.986 1.008 (0.988-1.028) 0.434 Maximum PaO 2 1.001 (0.996-1.006) 0.590 1.005 (0.996-1.014) 0.318 In-hospital mortality After ROSC 1.002 (0.997-1.007) 0.381 1.001 (0.995-1.006) 0.792 Initiation of TH 1.000 (0.995-1.004) 0.855 1.001 (0.997-1.006) 0.542 Start of maintenance 0.999 (0.993-1.006) 0.847 1.003 (0.995-1.011) 0.473 6 h after maintenance 0.993 (0.985-1.002) 0.145 1.000 (0.989-1.010) 0.933 12 h after maintenance 0.992 (0.983-1.001) 0.066 0.992 (0.983-1.002) 0.111 18 h after maintenance 0.990 (0.981-0.999) 0.035 0.993 (0.983-1.003) 0.198 24 h after maintenance 0.989 (0.979-0.999) 0.032 0.992 (0.981-1.003) 0.163 Minimum PaO 2 1.001 (0.989-1.013) 0.891 1.004 (0.990-1.018) 0.599 Maximum PaO 2 0.995 (0.990-1.001) 0.114 0.996 (0.990-1.002) 0.160 OR: odds ratio, CI: confidence interval, ROSC: return of spontaneous circulation, TH: therapeutic hypothermia PaO 2: arterial partial pressure of oxygen

유승협외 : 저체온치료를받은병원전심정지환자에서시간가중평균산소분압과예후와의관계 / 181 지로고산소혈증은사망뿐만아니라신경학적예후와유의한관계가있음을보여주었다. 하지만, 본연구에서는다변량분석결과산소분압과예후와의유의한관계는나타나지않았다. Janz 등 9) 은중환자실에입원한첫 24시간이내에가장높은산소분압단한개의값을이용하여본연구와는차이가있었으며, 가장높은산소분압의값을이용하였기때문에산소분압의중앙값이 226.6(162.7-309) mmhg로본연구의중앙값인 139(104.5-170.0) mmhg보다훨씬높아서연구에포함된산소분압의범위가매우다른분포를보였다. 산소분압의분포가다르기때문에본연구의결과에서산소분압과예후와의관계가나타나지않음이기존연구와상반된결과라고말하기는어렵다. 즉, 기존연구들과는다르게훨씬낮은범위의산소분압은예후와의관련성이높지않다고할수도있다. 환자마다저체온치료가끝날때까지산소에노출되는정도와시간이다르기때문에본연구에서는시간을반영한산소분압값을제시하기위하여시간가중평균산소분압값을이용하였는데, 시간가중평균산소분압값이대표값으로적절한지에대한연구는찾을수없었다. 다만각시간별산소분압과최대산소분압그리고최소산소분압의예후와의연관성을추가로분석하였는데, 자발순환회복후의산소분압만이신경학적예후와유의한연관관계가있는것으로나타났으며, 나머지시간대의산소분압과최대및최소산소분압모두예후와유의한관계를보이지는않았다. 그러므로본연구에서시간가중평균산소분압을이용한것은어느정도타당성이있는것으로판단할수있다. 또, 본연구에서단순비교결과시간가중평균산소분압이예후와관련이있는것으로나타났지만, 시간가중평균산소분압을연속변수로로지스틱회귀분석에투입하면예후와유의한관계가나타나지않았다. 이러한결과는시간가중평균산소분압의특정구간에서만예후와관련이있을수있음을의미하는데, 특정구간에대한기준을설정하기어렵기때문에사분위수를이용하였다. 기존의다른연구에서도병원전심정지환자를대상으로적혈구크기분포의사분위구간에따라사망에대한위험이다름을제시하였다 19). 기존연구들은고산소혈증이예후와관련이있기때문에산소분압을적정하는것이필요함을주장한다 5,6,9). 또한, 대부분의임상의들은고산소혈증의악영향에대해인지하고있기때문에산소분압을적정하며치료하려할것이다. 하지만, 어느정도의산소분압을유지하는것이예후에악영향을미치지않으면서충분한산소를공급할수있는지에대해서는연구가부족한실정이다. 이러한물음에대한해답을찾기위해서는산소분압을달리하는환자군을설정하여무작위대조군연구가필요하지만, 이와같은연구는윤리적인면이나임상시험을수행하는면에서매우큰어려움이있기때문에현실적으로실현되기어려울것이다. 따라서, 향후에는기존연구들보다산소분압의적정이잘이 루어진더많은수의환자군을대상으로산소분압과예후와의관계를분석하여, 심정지환자치료에서있어서예후에영향을미치지않는산소분압의허용구간을찾기위한연구들이필요하겠다. 본연구의제한점으로는첫째, 본연구는일개의료기관에서적은수의환자를대상으로시행되었기때문에연구결과의적용여부나통계적인분석력에있어서제한점이있다. 또한후향적연구이기때문에고산소혈증과예후의인과관계라기보다는단순관계의여부만을파악할수있었다. 향후다기관에서저체온치료의치료프로토콜을동일화하여이에대한추가연구가필요하다. 둘째, 환자들은모두저체온치료를받았지만, 실제환자의체온을동맥혈가스분석에적용하지않았고, 임상의들또한환자의체온에따른교정없이기계환기를시행하였다. 하지만, 체온에따라교정하였더라도 ph나이산화탄소분압의변화에비해서산소분압에대한변화는크지않았을것으로추정한다. 셋째, 본연구에서는허혈-재관류손상기간을최대한반영하기위해단일시점의산소분압보다는저체온치료의종료까지시간을연구변수에포함하여여러차례의측정된산소분압을이용하였지만, 산소분압은시간에따라지속적으로변동을보이기때문에이를충분히반영했다고보기는어렵다. 넷째, 심정지환자들의예후에영향을미치지만연구자들이기대하지못한혼란변수혹은교란변수들이다변량분석에서누락되었을가능성이있다. 또한, 일차결과에대한시간가중평균산소분압의검정력분석결과는 45% 의검정력를보여검정력이충분치않았음을나타냈다. 추후더많은증례를포함한분석을통해산소분압과예후와의관계를분석함이필요하다고하겠다. 다섯째, 산소분압값을제시함에있어서흡기산소분획 (fraction of inspired oxygen, FiO 2) 이반영되지않았다. 흡기산소분획을반영한산소분압값이더정확하겠지만, 이를해석하는데에도어려움이있다. 하지만, SOFA 점수를계산할때산소분압 / 흡기산소분획의값이반영되므로그영향은크지않을것으로생각한다. 결론저체온치료를받은심정지후증후군환자에서산소분압의중앙값은 139(104.5-170.0) mmhg의분포를보였고, 저산소혈증 (60 mmh 미만 ) 이나고산소혈증 (300 mmhg이상 ) 의빈도는흔하지않았다. 이러한분포를보이는산소분압값을대상으로시간가중평균산소분압을사분위값으로구분하여비교하였을때, 시간가중평균산소분압은신경학적예후나병원내사망두가지모두와독립적인관계를보이지는않았다. 향후심정지후치료프로토콜을동일화하여다기관에서연구를통해안전한산소분압의범위를규명하기위한연구가필요하다.

182 / 대한응급의학회지 : 제 25 권제 2 호 2014 참고문헌 01. White BC, Grossman LI, Krause GS. Brain injury by global ischemia and reperfusion: a theoretical perspective on membrane damage and repair. Neurology. 1993;43:1656-65. 02. Douzinas EE, Andrianakis I, Pitaridis MT, Karmpaliotis DJ, Kypridades EM, Betsou A, et al. The effect of hypoxemic reperfusion on cerebral protection after a severe global ischemic brain insult. Intensive Care Med. 2001; 27:269-75. 03. Richards EM, Fiskum G, Rosenthal RE, Hopkins I, McKenna MC. Hyperoxic reperfusion after global ischemia decreases hippocampal energy metabolism. Stroke. 2007;38:1578-84. 04. Peberdy MA, Callaway CW, Neumar RW, Geocadin RG, Zimmerman JL, Donnino M, et al. Part 9: post-cardiac arrest care: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2010;122:S768-86. 05. Kilgannon JH, Jones AE, Shapiro NI, Angelos MG, Milcarek B, Hunter K, et al. Association between arterial hyperoxia following resuscitation from cardiac arrest and in-hospital mortality. JAMA. 2010;303:2165-71. 06. Kilgannon JH, Jones AE, Parrillo JE, Dellinger RP, Milcarek B, Hunter K, et al. Relationship between supranormal oxygen tension and outcome after resuscitation from cardiac arrest. Circulation. 2011;123:2717-22. 07. Bellomo R, Bailey M, Eastwood GM, Nichol A, Pilcher D, Hart GK, et al. Arterial hyperoxia and in-hospital mortality after resuscitation from cardiac arrest. Crit Care. 2011;15:R90. 08. Bennett KS, Clark AE, Meert KL, Topjian AA, Schleien CL, Shaffner DH, et al. Early oxygenation and ventilation measurements after pediatric cardiac arrest: Lack of association with outcome. Crit Care Med. 2013;41:1534-42. 09. Janz DR, Hollenbeck RD, Pollock JS, McPherson JA, Rice TW. Hyperoxia is associated with increased mortality in patients treated with mild therapeutic hypothermia after sudden cardiac arrest. Crit Care Med. 2012;40:3135-9. 10. Polderman KH. Induced hypothermia and fever control for prevention and treatment of neurological injuries. Lancet. 2008;371:1955-69. 11. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonca A, Bruining H, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996;22:707-10. 12. Booth CM, Boone RH, Tomlinson G, Detsky AS. Is this patient dead, vegetative, or severely neurologically impaired? Assessing outcome for comatose survivors of cardiac arrest. JAMA. 2004;291:870-9. 13. Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Part 6: advanced cardiovascular life support: section 8: postresuscitation care. The American Heart Association in collaboration with the International Liaison Committee on Resuscitation. Circulation. 2000;102:I166-71. 14. Ecc Committee S, Task Forces of the American Heart A. 2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2005;112:IV1-203. 15. Bernard SA, Gray TW, Buist MD, Jones BM, Silvester W, Gutteridge G, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med. 2002;346:557-63. 16. Hypothermia after cardiac arrest study group. Mild therapeutic hypothermia to improve neurologic otucome after cardiac arrest. N Engl J Med. 2002;346:549-56. 17. Lin JS, Chen YS, Chiang HS, Ma MC. Hypoxic preconditioning protects rat hearts against ischaemia-reperfusion injury: role of erythropoietin on progenitor cell mobilization. J physiol. 2008;586:5757-69. 18. Ostadal P, Micek M, Kruger A, Horakova S, Skabradova M, Holy F, et al. Mild therapeutic hypothermia is superior to controlled normothermia for the maintenance of blood pressure and cerebral oxygenation, prevention of organ damage and suppression of oxidative stress after cardiac arrest in a porcine model. J Transl Med. 2013;11:124. 19. Kim J, Kim K, Lee JH, Jo YH, Rhee JE, Kim TY, et al. Red blood cell distribution width as an independent predictor of all-cause mortality in out of hospital cardiac arrest. Resuscitation. 2012;83:1248-52.