Optimizing prone cardiopulmonary resuscitation: Identifying the vertebral level correlating with the largest left ventricle cross-sectional area via computed tomography scan

Abstract

BACKGROUND: Placing the patient in the prone position frequently is required for some surgical procedures. If cardiac arrest occurs and the patient cannot be safely turned supine, cardiopulmonary resuscitation (CPR) may need to be performed with the patient in the prone position. Although clear landmarks have been defined for supine CPR, the optimal hand position for CPR in the prone position has not been clearly determined. The purpose of this study was to determine anatomically the optimal hand position for CPR in the prone position. METHODS: We reviewed retrospectively the chest computed tomography images of 100 patients taken in the prone position. The vertebral body levels crossing the medial angle of the scapula, the inferior angle of the scapula, and the spinous process of the vertebral body connected to the most inferior rib were identified, and we selected the image level at which the left ventricular (LV) cross-sectional area was the largest. This level was defined as the optimal compression level and correlated to surface anatomical landmarks. We calculated the ratio of the distance from the C7 spinous process to the level of the largest LV cross-sectional area divided by the distance from the C7 spinous process to the spinous process of the vertebral body connected with the most inferior rib. RESULTS: The level of the largest LV cross-sectional area in the prone position was 1 vertebral segment below the inferior angle of the scapula in 45% (99% confidence interval [CI], 33-58) of patients and 0 to 2 vertebral segments below that in 95% (99% CI, 86-98) of patients. The mean (SD) ratio of the distance from the C7 spinous process to the level of the largest LV cross-sectional area divided by the distance from the C7 spinous process to T12 spinous process was 67% ± 7% (99% CI, 65-69). CONCLUSIONS: When the patient is positioned prone, the largest LV cross-sectional area is 0 to 2 vertebral segments below the inferior angle of the scapula in at least 86% of patients. Further studies are needed to determine whether this position is optimal for chest compressions in the prone position.