Computed tomographic findings of chest injuries following cardiopulmonary resuscitation: More complications for prolonged chest compressions?

Chest injuries are common and inevitable complications of chest compressions during cardiopulmonary resuscitation (CPR). This study aimed to investigate lung parenchymal and thoracic skeletal injuries after CPR by using computed tomography (CT) and to analyze the correlation between the duration of CPR and related complications. We examined 43 non-traumatic cardiac arrest patients who were successfully resuscitated after CPR and had chest CT scans within 24hours of CPR. Lung parenchymal injuries were assessed by quantifying the lung contusion score (LCS) on the CT images, and each skeletal injury was investigated by classifying the location and the distribution. Other CPR-related chest injuries were also described, such as pleural effusion/hemothorax, pneumothorax, and retrosternal hematoma. Statistical analysis was conducted to determine whether the duration of CPR was correlated with each complication. Lung contusions were found in all of the patients (mean LCS: 22, range: 5-47). The distribution of lung contusions were predominantly in the bilateral dependent portions of the lungs (41 patients). All of the rib fractures occurred in the anterior arc (43 patients), and the sternal fractures occurred predominantly in the mid-sternal body (31 patients). In patients younger than 70 years old, the number of rib fractures significantly increased among those who underwent CPR for more than 25 minutes compared to those who received CPR for less than 25 minutes (median 4.5 vs 9; mean 8.3 vs 5.6 per person, respectively; P=.035). The risk of sternal fracture tended to be higher for patients who received CPR for more than 10 minutes compared to those who received CPR for less than 10 minutes (odds ratio: 3.60; 95% confidence interval: 0.86-15.06; P=.079). However, there was no statistically significant correlation between the duration of CPR and LCS or other CPR-related chest injuries. The duration of CPR was associated with the number of rib fractures and the occurrence of sternal fractures, but it did not affect the extent of CPR-related lung contusions or other CPR-related chest injuries. All of the rib fractures occurred in the anterior arc, while the sternal fractures occurred predominantly in the mid-sternal body. However, since this study was conducted in a single institution, the number of patients included was relatively small, thus limiting the statistical analysis.