Rajajee Venkatakrishna, Soroushmehr Reza, Williamson Craig A, Najarian Kayvan, Ward Kevin, Tiba Hakam
Department of Neurosurgery, University of Michigan, Ann Arbor, MI.
Department of Neurology, University of Michigan, Ann Arbor, MI.
Crit Care Explor. 2023 Aug 25;5(9):e0953. doi: 10.1097/CCE.0000000000000953. eCollection 2023 Sep.
Transcranial Doppler (TCD) has been evaluated as a noninvasive intracranial pressure (ICP) assessment tool. Correction for insonation angle, a potential source of error, with transcranial color-coded sonography (TCCS) has not previously been reported while evaluating ICP with TCD. Our objective was to study the accuracy of TCCS for detection of ICP elevation, with and without the use of angle correction.
Prospective study of diagnostic accuracy.
Academic neurocritical care unit.
Consecutive adults with invasive ICP monitors.
Ultrasound assessment with TCCS.
End-diastolic velocity (EDV), time-averaged peak velocity (TAPV), and pulsatility index (PI) were measured in the bilateral middle cerebral arteries with and without angle correction. Concomitant mean arterial pressure (MAP) and ICP were recorded. Estimated cerebral perfusion pressure (CPP) was calculated as estimated CPP (CPPe) = MAP × (EDV/TAPV) + 14, and estimated ICP (ICPe) = MAP-CPPe. Sixty patients were enrolled and 55 underwent TCCS. Receiver operating characteristic curve analysis of ICPe for detection of invasive ICP greater than 22 mm Hg revealed area under the curve (AUC) 0.51 (0.37-0.64) without angle correction and 0.73 (0.58-0.84) with angle correction. The optimal threshold without angle correction was ICPe greater than 18 mm Hg with sensitivity 71% (29-96%) and specificity 28% (16-43%). With angle correction, the optimal threshold was ICPe greater than 21 mm Hg with sensitivity 100% (54-100%) and specificity 30% (17-46%). The AUC for PI was 0.61 (0.47-0.74) without angle correction and 0.70 (0.55-0.92) with angle correction.
Angle correction improved the accuracy of TCCS for detection of elevated ICP. Sensitivity was high, as appropriate for a screening tool, but specificity remained low.
经颅多普勒（TCD）已被评估为一种无创颅内压（ICP）评估工具。在使用TCD评估ICP时，经颅彩色编码超声（TCCS）对潜在误差来源——声束入射角的校正此前尚未见报道。我们的目的是研究使用和不使用角度校正时TCCS检测ICP升高的准确性。
诊断准确性的前瞻性研究。
学术性神经重症监护病房。
连续纳入的有创ICP监测的成年人。
使用TCCS进行超声评估。
在双侧大脑中动脉测量舒张末期速度（EDV）、时间平均峰值速度（TAPV）和搏动指数（PI），测量时使用和不使用角度校正。同时记录平均动脉压（MAP）和ICP。计算估计脑灌注压（CPP）为估计CPP（CPPe）=MAP×（EDV/TAPV）+14，估计ICP（ICPe）=MAP - CPPe。纳入60例患者，55例接受了TCCS检查。对检测有创ICP大于22 mmHg时的ICPe进行受试者操作特征曲线分析，结果显示未进行角度校正时曲线下面积（AUC）为0.51（0.37 - 0.64），进行角度校正时为0.73（0.58 - 0.84）。未进行角度校正时的最佳阈值为ICPe大于18 mmHg，灵敏度为71%（29 - 96%），特异度为28%（16 - 43%）。进行角度校正时，最佳阈值为ICPe大于21 mmHg，灵敏度为100%（54 - 100%），特异度为3%（17 - 46%）。PI的AUC在未进行角度校正时为0.61（0.47 - 0.74），进行角度校正时为0.70（0.55 - 0.92）。
角度校正提高了TCCS检测ICP升高的准确性。作为一种筛查工具，灵敏度较高，但特异度仍然较低。
