The current standard for cardiopulmonary resuscitation (CPR) was first described in the early 1960s [1]. CPR is a series of maneuvers aimed to restore the flow of oxygenated blood to the brain and heart when the patient is in cardiac arrest [2, 3]. Despite the many changes in international guidelines for CPR, chest compression remains the cornerstone of successful resuscitation. However, chest compression can lead to extensive traumatic injuries due to the force and violence necessary [4, 5]. The spectrum of reported injuries ranges from 21% to 97% [3, 4, 5, 6] with an array of clinical patterns which range from skeletal chest fractures to severe life-threatening injuries such as cardiac tamponade, cardiac rupture or liver laceration [7].

In the current literature there are many studies that focus primarily on skeletal lesions, few on intrathoracic injuries (ITI) and to our knowledge none that focus exclusively on cardiovascular damages related to cardiac massage [8, 9, 10, 11, 12, 13].

The study based on post- mortem autopsy assessed the incidence of non-skeletal CPR related injuries with particular attention to cardiovascular injuries directly related to chest compression.

Autopsy reports showed cardiac lesions in 19 patients (group A). Nineteen patients without cardiac injury were controls (group B). The average age was 65.7 years (69.05 in group A and 66.5 in group B), median weight was 75.2 Kg and was significantly higher in group B (p = 0.01). The two groups were similar in terms of general demographics, admission and autopsy diagnosis as summarized in Table 1.

A total of 84.2% of the patients (17 from group A and 15 of group B) had at least one comorbidity. The most common were hypertension and coronary artery disease. Ischemic disease, confirmed by autopsy and histopathology, was the most frequent cause of death (50% of cases), followed by pulmonary embolism (18.4%).

Table 2 shows CPR and first treatment data. Three patients in group A had cardiac lesions not compatible with life (2 with atrial lacerations and 1 with ventricle rupture). Pericardial lesions were present in 6 patients in group A and 2 in group B respectively (p = 0.2 ns). Hemotransfusion was similar in both groups as was CPR starting time, need of catecholamines, troponin T values and serum levels of potassium.

Autopsy findings showed skeletal lesions in 78.9% of patients (17 from group A and 13 from group B) (p = 0.6) (Table 3). The incidence of visceral injuries was 60.5% (89% in group A and 42% in group B; p = 0.18 ns). The injuries included rib fractures, sternal and vertebral lesions, liver and lung injuries, pericardial lesions, heart contusions and spleen lacerations. The average number of rib fractures was higher in group A (p = 0.04) but no other significant differences were seen between the two groups. The multivariate analysis showed that a higher median weight and male sex have a protective effect on visceral injuries.

Table 4 outlines the pattern of cardiac CPR related injuries with heart injuries in 50% of patients with a high prevalence (52.6%) of left ventricle injuries.

This is an autopsy-based study focused on the incidence and severity of non-skeletal injuries following CPR. Patients recruited for this study were assigned to two groups: those with and without cardiac lesions (19 patients in each group). Most patients (84.2%) had at least one comorbidity including hypertension, coronary artery diseases, diabetes, arteriosclerosis, cirrhosis, lung disease or osteoporosis.

We report that lacerations of visceral organs are common in autopsies after CPR. The pericardium, heart and liver are the most frequently involved organs. Contusions of the lungs and heart are the most common non skeletal injuries. Left ventricular injuries were seen in 52.6% and cardiac contusion in 84.2% of autopsies following CPR. The presence of visceral injuries was not associated with age, height or diagnosis at admission. Female gender was not an independent prognostic factor for visceral injuries as reported in previous studies. Conversely, a higher median weight seemed to have a protective effect on the onset of visceral injuries (Table 1). A possible explanation of this difference could be the lack of the thickness of tissues resulting in decreased protection to visceral organs during CPR.

Hemotransfusion, CPR starting time, need of catecholamines, troponin T values and serum levels of potassium were not independent prognostic factors for CPR related-cardiac lesions. The total incidence of CPR associated thoracic injuries in this study population was (78.9%). Other authors showed similar results in term of the injurie rates after CPR [3, 5, 7, 8, 10].

The average number of rib fractures was significantly higher in patients with cardiac lesions at autopsy. The mechanisms of skeletal injuries are well known and described in the medical literature. The high rate of visceral lesions (60.5% of patients of which 89% had cardiac lesions) may be a result of our study population composed of only non survivors after CPR. Therefore, it seems logical to attribute this data to a more aggressive resuscitation technique and to a prolonged cardiac massage performed to save a patient’s life.

Moreover, our study is based on autopsy not imaging findings that may be more accurate in lesion detection.

The highest percentages of heart injuries (50% of patients with a high prevalence and 52.6% of left ventricle damage) reported in our study compared to those previously [6, 9, 10, 14] described could be similarly explained.

In our series pericardial and cardiac injuries were detected by autopsy (21.05% and 50% respectively) with high prevalence of left ventricle lesions (26.3%). Miller et al. [6], in a recent analysis of CPR-associated injuries, reported rates of pericardial injury of 8.9% and of cardiac damage rate up to 4.4% respectively.

The spectrum of cardiac injuries (Table 4) detected in our series are consistent with those previously described [15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26]. Both contusions of the epi-, myo-, or endocardium, lacerations or chamber ruptures involving the right atrium, right ventricle and left ventricle were detected by autopsy. It is important to highlight that there are cases in which cardiac damage is not associated with thoracic wall injuries and/or myocardium weakened by ischemia. Abdominal organs are shown, in previous studies, to be less affected by CPR related trauma. Percentage of liver and spleen lesions CPR related injuries varies through different series, from 0.6 % by Meron et al. [27] to 2.1% by Krischer et al. [28]. In our series a liver contusion was detected in one patient (Table 4).

CPR related non-skeletal and cardiac injuries may be underestimated and clinically misunderstood. In survivors after CPR this could lead to a rapid deterioration or impact on the patient’s quality of life. Further comparative studies based on post- mortem autopsy are required to strengthen the body of evidence in order to assess the real incidence of non-skeletal CPR related injuries with particular attention to cardiovascular lesions directly related to chest compression. Cardiovascular injuries related to CPR represents a risk that may be often overlooked.

PG—data collection, data analysis, writing, review, last versions control. AR—data collection, data analysis, writing, review, last versions control. VO—writing, review, last versions control. VH—data collection, last versions control. IK—review, last versions control.

Not applicable.

The authors would like to thank Ms. A. Hodja for proofreading the article.

This research received no external funding.

The authors declare no conflict of interest.