Article Text


Cardiac tamponade: a case of kitchen floor thoracotomy
  1. K D Wright1,
  2. K Murphy2
  1. 1London Helicopter Emergency Medical Service, Helicopter Emergency Medical Service (HEMS) London, Royal London Hospital, London, UK
  2. 2London Ambulance Service, Helicopter Emergency Medical Service (HEMS) London
  1. Correspondence to:
 Mr K D Wright, Emergency Department, John Radcliffe Hospital, Headley Way, Oxford, UK;

Statistics from

Successful thoracotomy in the prehospital environment is becoming more widely accepted.1,2 Here we present the case of cardiac arrest secondary to penetrating chest injury and the successful prehospital thoracotomy that followed. The resuscitation was associated with the spontaneous return of motor activity and later, hospital discharge. The implication for the immediate need for anaesthesia and paralysis is discussed together with a description of the surgical technique.


A teenage male youth sustained a stab wound to the left chest, in the third intercostal space at the junction of the medial and middle thirds of the clavicle. This wound was part of a deliberate self harm attempt. On arrival of the medical team—15 minutes from 999 call—he was thrashing and taking a few agonal breaths, this rapidly deteriorated to cardiac arrest within the first few seconds of assessment.

The patient was placed on the floor of his first floor flat and endotracheal intubation was undertaken by the medical team paramedic; cannulation was achieved by a first responder paramedic and the medical team doctor undertook bilateral thoracostomies in the right and left 4th intercostal space, midaxillary line. This revealed a small haemothorax on the left side. The thoracostomies were joined by a skin incision using a 22 blade scalpel through skin and subcutaneous fat. Heavy duty shears were placed through the thoracostomy and used to cut through muscle and sternum thus making a large clamshell thoracotomy.

With the chest open the pericardium was visualised as a blue, tense sac. Mosquito forceps were used to tent the pericardium and it was incised with scissors and widely opened. A large clot was removed and the operator's right index finger used to occlude a hole in the posterior aspect of the left upper heart. As the hole was occluded the heart began to fill and beat, restoring a carotid pulse. There was no anterior wound.

The patient attempted to breathe and then localised both upper limbs towards the chest incision. He was rapidly sedated and paralysed with midazolam 10 mg and pancuronium 8 mg. The patient was then lifted down two flights of stairs and taken to the nearest accident and emergency/cardiothoracic centre still with digital occlusion of the hole. Treatment at scene lasted 18 minutes. Bleeding from the internal mammary vessels was controlled with mosquito forceps.

On arrival, (journey time four minutes) he was maintaining a heart rate of 100 beats per minute and a systolic blood pressure of between 60 and 90 mm Hg. A cardiothoracic response enabled haemostatic sutures to be placed while in the resuscitation room before transfer to theatre for definitive closure. He was then transferred to the intensive care unit.

The postoperative course was initially difficult requiring a second thoracotomy for intrathoracic bleeding. Sepsis caused a syndrome of renal failure requiring haemofiltration. However by day eight he was alert and appropriate neurologically, requiring no cardiovascular support and doing well. Subsequently he made a full recovery and was discharged from hospital with no neurological deficit.


Prehospital thoracotomy aims to treat one specific group of patients—those with penetrating chest injury leading to cardiac tamponade. The procedure aims to release that tamponade and restore cardiac output permitting the patient to be evacuated to hospital. It is extremely unlikely that prehospital thoracotomy would be undertaken by a cardiothoracic surgeon and most accident and emergency departments would be hard pressed to mount a cardiothoracic response as part of their trauma team. The technique therefore needs to be simple and easily learned, with the limited aims of release of tamponade, haemostasis from cardiac wounds, and perhaps aortic occlusion.

Advanced medical care at the scene is becoming increasingly available. The advent in the United Kingdom of physicians working in conjunction with the ambulance service as prehospital care providers and with medical/trauma systems such as the London Helicopter Emergency Medical Service and physician ride-alongs in the United States mean that there has been some experience with prehospital thoracotomy.

Battistella et al3 conducted a retrospective analysis of trauma patients who were pulseless at scene. Altogether 604 patients with traumatic cardiac arrest were studied, (304 from blunt injury and 300 from penetrating injury). Some 304 of the patients underwent EDT, 160 went to the operating room. Only 16 patients left hospital—seven with severe neurological impairment. There were no survivors among those whose initial rhythm was asystole. No patient survived to leave hospital if their initial cardiac electrical rate was less than 40 beats per minute. The study concluded those pulseless trauma patients in asystole or with an electrical activity rate of less than 40 beats per minute should be pronounced dead at the scene.

In the face of such adverse statistics it becomes important to justify undertaking such a procedure. In the past 10 years the author is aware of six successful on scene thoracotomies. In all cases the patient left hospital neurologically intact. Only endotracheal intubation has been validated to improve outcome in patients in extremis, if transport time is to be delayed then advanced surgical care at the scene can restore cardiac activity. This is supported by Freezza and Mezghebe4 who determined that 30 minutes is the optimum period from injury to EDT. Thus if transport is delayed then onscene thoracotomy is both reasonable and perhaps a standard of care when applied to a set patient condition. This is penetrating chest injury associated with cardiac arrest. Penetrating injury can be thought of as high, medium, or low energy transfer—rifles, handguns, and knives respectively. High and medium energy transfer injuries are associated with the phenomenon of temporary and permanent cavitation thus the damage pattern may be much more severe than outward signs suggest. This means that the limited range of surgical options available in the field could be insufficient to cope with the injuries found and rapid transfer to the emergency department may be more appropriate.

Ivatury5 and his colleagues studied a series of 100 patients in extremis and requiring EDT. Patients were in two groups. Group I received stabilisation and group II underwent rapid transfer to the emergency department. A higher proportion of group II patients reached the emergency room with signs of life than group I and overall survival was higher in the rapid transfer group. The anatomical injury severity and mode of injury was similar in the two groups. Prehospital thoracotomy was not available to these groups as a prehospital stabilisation method. Thus in a non-surgical option setting rapid transfer seems to confer a better outcome. This message was echoed by Honigan and colleagues.6 Seventy consecutive patients with cardiac injury were studied. On scene time by paramedics averaged 10.7 minutes and these patients were intubated and cannulated. It was concluded that paramedics can perform these interventions without prolonging the time spent in the prehospital phase thus delivering them to hospital for advanced surgical care. Early thoracotomy seems to be fundamental to the survival of these patients. Delayed thoracotomy significantly raised the mortality from injury in 228 patients studied.7 Thoracotomy on scene must therefore be the standard of care that is applicable to these patients.

The restoration of a circulation may well lead to an improvement in the patients conscious level with dramatic effect. As the arrested patient will have been intubated without anaesthesia then rapid paralysis and sedation must be available. Previous reported cases have improved neurologically at hospital but not in the immediate resuscitative phase. The choice of sedating agents and the use of paralysing drugs will depend on the operator's own experience. Many physicians will be familiar with agents such as the benzodiazepines and opioids for sedation. Most sedating drugs will tend to lower blood pressure. It would be prudent for the physician to use the lowest dose that achieves clinical effect. The danger for the non-anaesthetist using paralysing drugs normally lies in the failed intubation scenario. Under the circumstances described advanced airway management should be undertaken before thoracotomy. Failure to intubate may represent an indication for immediate transfer to hospital rather than undertake thoracotomy. The implication is that use of these drugs facilitates the transfer of the resuscitated patient. Those who envisage using this procedure should familiarise themselves with the available agents.

The technique chosen for the thoracotomy is simple and uses only instruments that are familiar to all emergency physicians. No specific cardiothoracic instruments are used. If the operator is familiar with the technique of chest drain insertion and thoracostomy then they will be able to extend this to perform a “clamshell thoracotomy”. As the procedure is being undertaken in far from ideal circumstances, this familiarity will boost confidence in the operator. Complex cardiothoracic repair should not be attempted. Haemostasis either by digital occlusion or suture is all that is required. Digital occlusion should be placed over but not in cardiac wounds to avoid enlarging the defect. The “clamshell” technique permits good visualisation of structures and allows for aortic occlusion is required. This technique has been shown to be effective when used by anaesthetists and emergency physicians in a recent series.8 If the operator is not able to carry out thoracotomy then needle pericardiocentesis may be life saving. However, traumatic tamponade is often clotted and pericardiocentesis fails yet thoracotomy allows for clot removal. If the skills are available then thoracotomy should be used as this is the definitive end point and avoids using precious time in a procedure likely to fail.


Prehospital thoracotomy is an aggressive treatment that should be reserved for those patients likely to have cardiac tamponade. If applied promptly and judiciously it can be successful and lead to a neurologically intact survivor. If a prehospital provider is not familiar with the technique then rapid transfer to the nearest institution capable of providing resuscitative thoracotomy should be undertaken. Practitioners should become familiar with sedative and paralysing agents for use in the event of a recovery of spontaneous circulation and awareness. Patients in asystole or with a downtime of greater than 30 minutes should not undergo thoracotomy.


Both Kelvin Wright and Ken Murphy attended the patient. Kelvin Wright performed the surgical procedure while Ken Murphy undertook airway management. The paper was written by Kelvin Wright who also acts as guarantor for the content.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.