Review
Injuries associated with resuscitation – An overview

https://doi.org/10.1016/j.jflm.2015.04.003Get rights and content

Highlights

  • In 1960, Kouwenhoven et al. developed a method of resuscitation using external cardiac massage.

  • Injuries from resuscitation may complicate forensic evaluation of cases.

  • Potential external, internal and multiple system findings are reviewed.

  • Injuries resulting from automated resuscitation devices are also considered.

Abstract

External cardiopulmonary resuscitation is a potentially lifesaving intervention aimed at preserving the cerebral function of a person in cardiac arrest. However, certain injuries can be caused by the various techniques employed. Although these are seldom consequential, they may complicate the forensic evaluation of cases. Fractures of the ribs and sternum are the most common internal injuries and are frequently acknowledged as a consequence of resuscitation. Nonethlesss, the recognition that less common fractures such as of the larynx or injuries involving the stomach, spleen, heart and liver can occur due to resuscitation will assist the forensic examiner assess the significance of these findings when they present in cases of sudden death.

Introduction

Cardiopulmonary resuscitation is an emergency procedure involving external chest compression and artificial ventilation that aims to restore the flow of oxygenated blood to the brain and heart in individuals in cardiac arrest.1 In 1960, Kouwenhoven et al. developed a method of external cardiac massage that did not require thoracotomy, which is now known as cardiopulmonary resuscitation.2 Its relative simplicity allowed resuscitation to be performed both in and out of hospital, by bystanders, as well as by trained medical personnel. The method has remained fundamentally unchanged.3, 4 Whilst potentially lifesaving, it was recognised from the early days that injuries or complications may arise.5 Complications from resuscitation may create unusual lesions or injuries6, 7 of which the pathologist should be aware to prevent findings being misinterpreted; a rare example being a 9 mm circular defect resembling a gunshot wound of the chest.8 In this paper the literature is reviewed to demonstrate the range injuries that may be encountered during the medicolegal autopsy when death has been declared following a period of resuscitation that may complicate the forensic evaluation of cases due to their rarity or potential significance in terms of determining the cause of death.

Injuries related to resuscitation may be observed on external examination. Bruises and abrasions may be seen on the scalp and face.9, 10, 11 Use of a mask for resuscitation can cause soft tissue damage.12 Pinching of the nose can result in injuries when mouth-to-mouth resuscitation has been performed and the action of maintaining an airway may result in fingernails causing abrasions under the jaw; the distribution and pattern is generally different to that encountered in cases of throttling, except for abrasions produced by forcefully feeling for a carotid pulse.9 Injuries to the face can follow the removal of vomit.9

Within the mouth mucosal injuries can occur involving the lips13 including necrosis reported from use of a Guedel airway.14 There may be damage to the teeth and lips from intubation10, 13 or insertion of an oropharyngeal airway.12 Use of nasopharyngeal airways have been associated with damage to the nasal passages, including damage to turbinates15 and intrusion into the cranial vault has been reported in head injured patients with base of skull fracture.16, 17 Oesophageal intubation may result in perforation of the oesophagus.18

From retrospective19 and prospective studies20 there is no convincing evidence that facial or conjunctival petechiae result from cardiopulmonary resuscitation.

Retinal haemorrhages in children may result from inflicted head injury; however, there has also been discussion as to whether they may occur as a complication of resuscitation.21, 22 Goetting and Sowa suggest that chest compression from any source increases intrathoracic pressure, which can directly increase retinal venous pressure resulting in haemorrhages.23 Thus, they warn that forensic pathologists should be wary of attributing retinal haemorrhages to inflicted injury if this is the only significant finding when cardiopulmonary resuscitation has been performed23; however, careful investigation of the circumstances of death is required if retinal haemorrhages are found.24 A study of 22 children who had received cardiopulmonary resuscitation revealed retinal haemorrhages in six; five had other factors (trauma or raised intracranial pressure) present, leaving only one with mild haemorrhages at the posterior pole that appeared attributable to resuscitation.22 A reported case in which retinal haemorrhages were documented to be absent before cardiopulmonary resuscitation, but were present within the retina, as well as in a subhyaloid location, from the posterior pole to the mid-periphery following resuscitation for 60 min adds to the suggestion that retinal haemorrhage may result from resuscitation.25

Mild basal subarachnoid haemorrhage produced by resuscitation has been observed by one of the authors (NEIL) and has also been reported.26

Deep haemorrhage may be seen in the neck from the insertion of central venous lines. Mucosal injuries of the larynx can occur10, 13 as well as contusions from the base of the tongue to the oropharynx such that resuscitative intubation can cause artifactual injury that may simulate inflicted injuries caused by neck compression.27 In the 1960's Sellick28 described the use of cricoid pressure in resuscitation to prevent the reflux of gastric contents into the airways.29 It is, however, difficult to judge the correct amount of force to apply, with experienced staff exerting a variable amount of pressure ranging from 10.8 to 120.6 N and fracture of the cricoid cartilage has been reported as a result.30 Fractures of the larynx (hyoid bone and thyroid cartilage) appear to be a rare complication of resuscitation and intubation, but have been reported.13, 31 Haemorrhages9 and haematomas within the larynx and pharynx may be observed following intubation.13, 32 Ulceration or laceration can occur.33 Oedema of pharyngeal structures may result from use of a laryngeal mask.12 Haemorrhage can be an immediate complication of emergency tracheostomy.34, 35

Parchmented abrasions/bruising of the central chest may mark areas where external compression has been applied.10, 36, 37 Abrasion, electrical burns or parchmenting from defibrillation may also be present13, 38, 39; in addition, burns can occur from external pacing.40 Also on the chest there may be incisions from drain insertion or puncture marks from the use of cannulae to decompress pneumothoraces37, 41 or from emergency pericardiocentesis.42

Fractures of the ribs and sternum are the most commonly encountered internal resuscitation injuries13, 36; however the range of their reported incidence43 may relate to the method of their detection. Rib fractures become more likely with age and are encountered anterior to the mid-axillary line.37, 44 It is unclear whether prolonged and/or untrained resuscitation is associated with a greater number of rib fractures. Pneumothoraces, haemothoraces and/or subcutaneous emphysema may result.44 Rib fractures may contribute to death by causing pneumothoraces or cardiac lacerations.13, 45, 46 However, when a pneumothorax is found consideration has to be given to it being the cause of the arrest.47 Similarly, haemothorax has been observed due to a tear of the azygos vein resulting from external cardiac massage,48 but spontaneous haemothorax can be a rare cause for cardiac arrest.49 Distinguishing between primary (arrest causing) and secondary (caused by resuscitation) pathology can be difficult and reliance may have to be placed on the history in addition to the findings. The more recent squeezing, rather than compressive, type of resuscitation that is being employed in infants50, 51 may be associated with an increased number of rib fractures. Fractures of the spine are extremely rare.52, 53

Epicardial petechial haemorrhages are not uncommon and are more likely to be seen on the posterior surface.11 Epicardial contusions may also be present.54 Myocardial haemorrhages attributed to the effects of compression of the heart may be present in the septum11 and in the region of the bundle of His.54, 55 Haemorrhage may also be seen in the adventitia of the aorta.13 ‘Fractures’ of the coronary arteries have been described as disruption of the intima, media or adventitia associated with evidence of haemorrhage; such lesions are reportedly more common in the left anterior coronary artery, at branch points and in areas of calcified atherosclerosis.54 Histological examination of the myocardium may reveal the presence of contraction bands and areas of myocyte necrosis.37, 56 Although this could be a direct consequence of blunt trauma57 from chest compressions, the finding is more likely to be attributable to the effects of adrenaline.58

Laceration of the heart and pericardial sac may result from fractures of the ribs.46 Cardiac rupture during resuscitation is, nonetheless, very rare and is postulated to be due to a hydrostatic effect from increased pressure from attempted compression of an incompressible fluid (blood)59; it would be expected ruptures would be more likely to occur in regions with a thinner wall, such as the right ventricle or atria.60 Thus, rupture of the right ventricle of the heart is believed to occur when there is increased pressure in the right side of the heart (for example due to outflow obstruction by pulmonary thromboembolus) or as a consequence of the application of an incorrect technique.60 It is considered that rupture of infarcted regions of the left ventricle does not occur due to resuscitation, but that the rupture of the infarct with resulting haemopericardium precedes application of chest compressions.61 However, rupture of areas of non-infarcted heart has been recorded46 and cases with small volumes (less than 200 ml) of non-clotted blood in the pericardial sac in association with a ruptured myocardial infarction have been encountered at this centre that suggest the rupture was caused by the external cardiac massage. Haemopericardium can also result from a tear of the inferior vena cava within the pericardial sac.45

Laceration of the pleural surface of the lungs can result from fractures of the ribs.45

Haematomas may be encountered in the mesocolon, pancreas62 and omentum.13 Perinephric haematoma has been reported.62 Haemoperitoneum may be encountered for which the only apparent cause may be laceration of small vessels.13

Incorrect placement of the hands over the xiphoid process of the sternum during cardiopulmonary resuscitation can increase the likelihood of visceral injuries.45 Abnormalities of the spleen or liver may contribute to an increased risk of laceration.62 Ossification of a thoracotomy scar has caused a resuscitation laceration of the liver.63

Mouth to mouth resuscitation or the use of a bag and mask may create gastric distension and promote gastric rupture64, 65 that must be differentiated from gastromalacia.66, 67 Gastric rupture from resuscitation typically occurs along, or parallel to, the lesser curve below the gastro-oesophageal junction. A ‘lightening bolt’ appearance to the tears may be seen.68 An absence of haemorrhage is an indication that such findings, including radial tears of the gastric mucosa, occurred after death and can be attributed to gastric distension during resuscitation.69

Tears of the spleen and splenic artery have been encountered.62, 70

Injury to the liver was reported by Adler and colleagues.71 Tears may occur deep to the xiphoid process of the sternum72 or in either lobe,73 but it has been suggested that damage to the left lobe is more common due to misplacement of the hands of those attempting resuscitation.73 Children may be more vulnerable to hepatic injuries during resuscitation due to the relatively small size of the paediatric chest.74

Bruising around intravenous puncture wounds in the cubital fossae or dorsa of the hands and feet are common and substantial associated haematomas may form.10 Ambulance personnel may circle these wounds with marker pens to help examining pathologists to differentiate them from pre-existing needle punctures.

Pulmonary fat embolism was found histologically in 46/57 (81%) of patients who received external cardiac massage terminally; cerebral fat embolism was found in 4/16 of cases. The significance of the finding of fat emboli was unclear.64 Around two thirds of the cases with pulmonary fat emboli did not have rib fractures,64 a phenomenon that has been observed by others.75 It is postulated that expulsion of fat results from deformation of the ribs associated with chest compression.64 Bone marrow emboli within vessels of the lungs may be found histologically following both bone trauma and cardiopulmonary resuscitation.75

Cardiopulmonary resuscitation rarely results in cerebral air embolism. It is though that small pulmonary vessels may rupture from positive pressure due to chest compression and/or artificial respiration. Findings at autopsy may include air bubbles present in the epicardial and cerebral vessels and frothy blood in the right ventricle.76 The presence of air embolism may be more effectively demonstrated by computed tomography (CT) radiology, but care has to be taken to exclude gas from decomposition.77

There are now three different major categories of resuscitation: traditional/manual, device-assisted manual (i.e. with a hand pump to improve blood flow), and automated device-assisted (i.e. piston- or band-containing devices to produce chest compressions).78 An assessment of resuscitation methods determined that two alternative techniques and four devices could provide a benefit over manual cardiopulmonary resuscitation.1 The alternative techniques comprised abdominal counterpulsation and active compression/decompression; the devices included a vest, a mechanism to perform active compression/decompression, an automated valve with active compression/decompression and a mechanism to perform compressions.1, 79

In a comparison of standard cardiopulmonary resuscitation with active compression/decompression there were more subjects with bruising over the sternum in the active compression/decompression group; one patient who had standard cardiopulmonary resuscitation suffered a splenic rupture; there were more chest complications (rib fractures, haemothorax, pneumothorax or rib fractures) in the standard cardiopulmonary resuscitation group, but this was not regarded as statistically significant.80

A comparison of manual cardiopulmonary resuscitation with interposed abdominal compression and high impulse compression (a method employing rapid chest compressions) using an animal model revealed no difference in the trauma scores between the two methods; there was no correlation between the type of resuscitation and injuries that occurred, which included fractured ribs, myocardial contusion, pulmonary contusion, pulmonary oedema and liver laceration.78

A study comparing traditional/manual cardiopulmonary resuscitation with an AutoPulse™ automated band using device found rib fractures were statistically significantly more common in patients who had undergone manual resuscitation, but it was noted pattern of injury was also different with posterior fractures being more common in the AutoPulse™ group. Visceral injuries seen three cases from the 88 in the AutoPulse™ group comprised laceration of the right lobe of the liver, mesenteric laceration and laceration of the spleen. Skin abrasions were more common in the AutoPulse™ group and were distributed in the axillary, shoulder, chest and arm regions compared to traditional/manual cardiopulmonary resuscitation in which abrasions occurred on the anterior chest.81 Abrasions of the lateral chest/breast region have also been reported by others.7

In contrast a comparison of manual resuscitation with a mechanical chest device using a piston to compress the chest (LUCAS™) revealed more rib fractures in the mechanically resuscitated group. Ruptures of the aorta (thoracic and abdominal) and adventitia of a pulmonary artery were reported in the mechanical resuscitation group, but were not encountered in the manual group. Liver injuries were more common in the mechanical group. It was the opinion of the pathologists involved that the injuries found did not directly cause death. Vertebral fracture occurred in two cases following mechanical resuscitation and a collarbone fracture was found in the manual resuscitated group.43

Section snippets

Conclusion

The list of complications that can arise from resuscitation is potentially long and has been reviewed by others.13, 32, 44, 82, 83, 84, 85 This review has focussed on injuries and physical that may be found when life has been pronounced extinct following resuscitation as the subject is worthy of consideration as the presence of injuries produced by resuscitation can create difficulties when interpreting the significance of findings at post-mortem examination.10, 86, 87

Conflict of interest

None.

Funding

None.

Ethical approval

Forensic Science South Australia.

References (87)

  • C.J. Black et al.

    Chest wall injuries following cardiopulmonary resuscitation

    Resuscitation

    (2004)
  • L. Danielsen et al.

    Skin changes following defibrillation the effect of high voltage direct current

    Forensic Sci Int

    (2003)
  • J.A. Reyes et al.

    Increased incidence of CPR-related rib fractures in infants — Is it related to changes in CPR technique?

    Resuscitation

    (2011)
  • R.M. Goldberg et al.

    Thoracic vertebral fracture as a complication of cardiopulmonary resuscitation

    J Emerg Med

    (1988)
  • A. Takada et al.

    Cardiopulmonary resuscitation does not cause left ventricular rupture of the heart with acute myocardial infarction: a pathological analysis of 77 autopsy cases

    Leg Med

    (2003)
  • M.D. Ziegenfuss et al.

    Traumatic liver injury complicating cardio-pulmonary resuscitation. The value of a major intensive care facility: a report of two cases

    Crit Care Resusc

    (2004)
  • G. Meron et al.

    Cardiopulmonary resuscitation-associated major liver injury

    Resuscitation

    (2007)
  • J. Dzięcioł et al.

    Widespread myocardial and pulmonary bone marrow embolism following cardiac massage

    Forensic Sci Int

    (1992)
  • K.B. Kern et al.

    Cpr-induced trauma: comparison of three manual methods in an experimental model

    Ann Emerg Med

    (1986)
  • K. Lurie

    Mechanical devices for cardiopulmonary resuscitation: an update

    Emerg Med Clin North Am

    (2002)
  • D. Mauer et al.

    Active compression-decompression resuscitation: a prospective, randomized study in a two-tiered EMS system with physicians in the field

    Resuscitation

    (1996)
  • S.G. Atcheson et al.

    Complications of cardiac resuscitation

    Am Heart J

    (1975)
  • C.M. Bush et al.

    Pediatric injuries from cardiopulmonary resuscitation

    Ann Emerg Med

    (1996)
  • K. Lurie et al.

    Mechanical advances in cardiopulmonary resuscitation

    Curr Opin Crit Care

    (2001)
  • W.B. Kouwenhoven et al.

    Closed-chest cardiac massage

    J Am Med Assoc

    (1960)
  • R.A. Berg et al.

    Part 5: adult basic life support 2010 American heart association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care

    Circulation

    (2010)
  • A.H. Travers et al.

    Part 4: CPR overview 2010 American heart association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care

    Circulation

    (2010)
  • J.R. Baringer et al.

    External cardiac massage

    N Engl J Med

    (1961)
  • Complications of cardiac massage

    Br Med J

    (1969)
  • A.P. Hart et al.

    Autopsy artifact created by the revivant AutoPulse™ resuscitation device

    J Forensic Sci

    (2005)
  • J.A. Prahlow et al.

    Lesions that simulate gunshot wounds

    J Clin Forensic Med

    (1997)
  • M.P. Ryan et al.

    Do resuscitation attempts in children who die, cause injury?

    Emerg Med J

    (2003)
  • A. Bhardwaj et al.

    Necrosis of lower lip due to Guedel's airway in a case of head injury

    J Anaesthesiol Clin Pharmacol

    (2014)
  • M. Zwank

    Middle turbinectomy as a complication of nasopharyngeal airway placement

    Am J Emerg Med

    (2009)
  • J.E. Martin et al.

    Intracranial insertion of a nasopharnygeal airway in a patient with craniofacial trauma

    Mil Med

    (2004)
  • D.Y. Ellis et al.

    Intracranial placement of nasopharyngeal airways: is it all that rare?

    Emerg Med J

    (2006)
  • J. Jougona et al.

    Esophageal perforation: life threatening complication of endotracheal intubation

    Eur J Cardiothorac Surg

    (2001)
  • A. Odom et al.

    Prevalence of retinal hemorrhages in pediatric patients after in-hospital cardiopulmonary resuscitation: a prospective study

    Pediatrics

    (1997)
  • H. Pham et al.

    Retinal hemorrhage after cardiopulmonary resuscitation with chest compressions

    Am J Forensic Med Pathol

    (2013)
  • M.G. Goetting et al.

    Retinal haemorrhage after cardiopulmonary resuscitation in children: an etiologic reevaluation

    Pediatrics

    (1990)
  • M.G.F. Gilliand et al.

    Are retinal hemorrhages found after resuscitation attempts? A study of the eyes of 159 children

    Am J Forensic Med Pathol

    (1993)
  • K. Kramer et al.

    Retinal hemorrhages following cardiopulmonary resuscitation

    Clin Pediatr

    (1993)
  • G. Lau

    A case of sudden maternal death associated with resuscitative liver injury

    Forensic Sci Int

    (1994)
  • Cited by (58)

    • EMS in Crime Scene: Role of Medical Emergency Teams in Forensic Cases

      2021, EMS in Crime Scene: Role of Medical Emergency Teams in Forensic Cases
    • Complicated central catheterization in atypical celphos poisoning simulating throttling homicide: Review of neck resuscitation artefacts

      2020, Journal of Forensic and Legal Medicine
      Citation Excerpt :

      Resuscitation artefacts related to neck injuries are a small but important subset of neck artefacts at autopsy in addition to much described Prinsloo-Gordon and post-mortem hypostatic haemorrhages.4 There are three main types of resuscitation-related neck injuries which overlap strangulation.5 Most common is intra-muscular haemorrhage in sternomastoid by central venous catheterization in the internal jugular with a needle puncture, unapparent in most cases.

    • Comparison of different resuscitation methods with regard to injury patterns in cardiac arrest survivors based on computer tomography

      2020, European Journal of Radiology
      Citation Excerpt :

      Load-distributing bands or piston systems with or without suction cup are the most common devices [10]. Recent studies on CPR-related injuries mostly focused on patients with failed CPR and were based on autopsies [11,12]. However, data from survivors are of even more importance for clinical practice as these data could indicate significant changes in treatment strategies in regard to post-cardiac arrest care.

    View all citing articles on Scopus
    View full text