Kaplan Lewis J, Kellum John A
Yale University School of Medicine, Department of Surgery, Section of Trauma, Critical Care, and Emergency General Surgery, USA.
Crit Care Med. 2004 May;32(5):1120-4. doi: 10.1097/01.ccm.0000125517.28517.74.
This study determines whether acid-base data obtained in the emergency department correlate with outcome from major vascular injury.
Observational, retrospective record review of trauma patients requiring vascular repair (torso or extremity, January 1988 to December 1997). Data included age, Injury Severity Score, injury mechanism, survival, laboratory profiling, calculated anion gap, strong ion difference, and strong ion gap. Patients were divided into survivors and nonsurvivors with comparison by Student's t-test; significance was assumed for p < or = .05. Multivariate logistic regression was used for further analysis of univariate predictors of mortality, and receiver operator characteristic curves were generated for mortality from each variable.
Urban level I trauma facility.
Trauma patients requiring vascular repair of torso or extremity injury.
None.
Both nonsurvivors (n = 64) and survivors (n = 218) were similar with respect to age (31 +/- 9 vs. 31.5 +/- 10.5, p = 0.15) and injury mechanics (81% penetrating in survivors vs. 83% penetrating in nonsurvivors, p = .71). Non-survivor Injury Severity Score exceeded that of survivors (27.5 +/- 7.8 vs. 12.4 +/- 9.4, p < .001). Nonsurvivor pH (7.06 +/- 0.15 vs. 7.34 +/- 0.08, p < .001) and apparent strong ion difference (31.38 +/- 4.39 vs. 37.53 +/- 3.86, p < .001) were significantly lower, whereas nonsurvivor standard base excess (-17.9 +/- 5.1 vs. -2.9 +/- 4.4 mEq/L, p < .001), lactate (11.1 +/- 3.6 vs. 3.6 +/- 1.5 mmol/L, p < .001), anion gap (28.2 +/- 4.1 vs. 15.6 +/- 3.1, p < .001), and strong ion gap (10.8 +/- 3.2 vs. 2.4 +/- 1.8, p < .001) were higher. All but one nonsurvivor had initial emergency department pH < or = 7.26, standard base excess < or = -7.3 mEq/L, lactate > or = 5 mmol/L, and strong ion gap > or = 5 mEq/L. All of the acid-base descriptors were strongly associated with outcome, but the strong ion gap discriminated most strongly with an area under the receiver operator characteristic of 0.991 (95% confidence interval, 0.972-0.998).
The initial emergency department acid-base variables of pH, base deficit, lactate, anion gap, apparent strong ion difference, and strong ion gap discriminate survivors from non-survivors of major vascular injury. The strong ion gap is most strongly predictive of mortality following major vascular trauma.
本研究旨在确定在急诊科获得的酸碱数据是否与严重血管损伤的预后相关。
对1988年1月至1997年12月期间需要进行血管修复(躯干或四肢)的创伤患者进行观察性、回顾性记录审查。数据包括年龄、损伤严重程度评分、损伤机制、生存情况、实验室检查结果、计算的阴离子间隙、强离子差和强离子间隙。将患者分为幸存者和非幸存者,采用学生t检验进行比较;p≤0.05被认为具有统计学意义。采用多变量逻辑回归对死亡率的单变量预测因素进行进一步分析,并为每个变量生成死亡的受试者工作特征曲线。
城市一级创伤中心。
需要对躯干或四肢损伤进行血管修复的创伤患者。
无。
非幸存者(n = 64)和幸存者(n = 218)在年龄(31±9岁 vs. 31.5±10.5岁,p = 0.15)和损伤机制(幸存者中81%为穿透伤,非幸存者中83%为穿透伤,p = 0.71)方面相似。非幸存者的损伤严重程度评分超过幸存者(27.5±7.8 vs. 12.4±9.4,p < 0.001)。非幸存者的pH值(7.06±0.15 vs. 7.34±0.08,p < 0.001)和表观强离子差(31.38±4.39 vs. 37.53±3.86,p < 0.001)显著较低,而非幸存者的标准碱剩余(-17.9±5.1 vs. -2.9±4.4 mEq/L,p < 0.001)、乳酸(11.1±3.6 vs. 3.6±1.5 mmol/L,p < 0.001)、阴离子间隙(28.2±4.1 vs. 15.6±3.1,p < 0.001)和强离子间隙(10.8±3.2 vs. 2.4±1.8,p < 0.001)较高。除一名非幸存者外,所有非幸存者的急诊科初始pH值≤7.26,标准碱剩余≤ -7.3 mEq/L,乳酸≥5 mmol/L,强离子间隙≥5 mEq/L。所有酸碱指标均与预后密切相关,但强离子间隙的区分能力最强,受试者工作特征曲线下面积为0.991(95%置信区间,0.972 - 0.998)。
急诊科初始的酸碱变量,即pH值、碱缺失、乳酸、阴离子间隙、表观强离子差和强离子间隙,可区分严重血管损伤的幸存者和非幸存者。强离子间隙对严重血管创伤后的死亡率预测能力最强。