BACKGROUND Open chest cardiac massage (OCCM) is a commonly performed procedure after traumatic cardiac arrest (TCA). OCCM has been reported to be superior to closed chest compressions (CCC) in animal models and in non-TCA. The purpose of this study is to prospectively compare OCCM versus CCC in TCA using end-tidal carbon dioxide (ETco(2)), the criterion standard for determining the effectiveness of chest compressions and detection of return of spontaneous circulation (ROSC), as the surrogate for cardiac output and marker for adequacy of resuscitation. METHODS This prospective observational study enrolled patients over a 9-month period directly presenting to a level 1 trauma center after TCA. Continuous high-resolution ETco(2) measurements were collected every 6 seconds for periods of CCC and OCCM, respectively. Patients receiving CCC only were compared with patients receiving CCC followed by OCCM. Student's t tests were used to compare ETco(2) within and between groups. RESULTS Thirty-three patients were enrolled (16 OCCM, 17 CCC-only). Mean time of CCC before OCCM was 66 seconds. Within the OCCM group, final, peak, mean, and median ETco(2) levels significantly increased when comparing the initial CCC period to the OCCM interval. Using a time-matched comparison, significant increases were observed in the final and peak but not mean and median values when comparing the first minute of CCC to the remaining time in the CCC-only group. However, when periods of OCCM were compared with equivalent periods of CCC-only, there were no differences in the initial, final, peak, mean, or median ETco(2) values. Correspondingly, no difference in rates of ROSC was observed between groups (OCCM 23.5% vs. CCC 38.9%; p = 0.53). CONCLUSION Although we could not control for confounders, we found no significant improvement in ETco(2) or ROSC with OCCM. With newer endovascular techniques for aortic occlusion, thoracotomy solely for performing OCCM provides no benefit over CCC. LEVEL OF EVIDENCE Therapeutic study, level III.