Pulmonary embolism is mainly caused by deep venous thrombosis in the lower extremities, mostly due to longer immobilisation and/or restricted venous backflow. Thus, most patients have a typical history; however, symptoms may be misinterpreted and therapy delayed. While smaller or fragmented thrombi in the peripheral segments of the pulmonary arteries may only cause dyspnoea and hardly any cardiovascular impairment, large thrombi in the pulmonary trunk or the origin of the main pulmonary arteries may present with severe dyspnoea and haemodynamic instability, often necessitating cardiopulmonary resuscitation. Severe pulmonary embolism is considered to be one of the main reasons for sudden death of unknown origin, not only after long immobilisation due to illness but also due to long flights or bus journeys.
Laboratory tests for pulmonary embolism concentrate mainly on blood gas analysis and D-dimers as typical markers of venous thrombosis. A computed tomography scan is considered the first choice among the imaging procedures, while in experienced hands transoesophageal echocardiography may also be appropriate, especially in emergency situations.
The established therapeutic options today are heparinisation, preferably with low-molecular-weight heparins, systemic thrombolysis and transcatheter embolectomy. In most emergency units, in accordance with recent literature, surgery is considered to be the last choice and only in cases of severe haemodynamic instability or in patients under cardiopulmonary resuscitation.
Surgical Therapy
The first pulmonary embolectomy was performed by Trendelenburg in 1908. In brief, the chest is opened by a median sternotomy and, after making an incision in the pericardium, both caval veins are encircled. Two sutures are placed close to the pulmonary valve and in the middle of the pulmonary bifurcation; these sutures serve as stay sutures during embolectomy and may later be used for closure of the arterial incision. If necessary, cardiopulmonary resuscitation is continued throughout the preparations. After tying the two caval snares, the pulmonary artery is incised longitudinally between the two stay sutures and the thrombus is extracted under direct vision by special forceps with the assistance of suction. Both lungs are manually compressed after liberal opening of the pleural spaces to mobilise the remaining smaller thrombi out of the peripheral branches. After this procedure, the inflow occlusion is stopped and the pulmonary incision is closed with the aid of a side-biting clamp to avoid major blood loss. This operation can be performed anywhere and by any general surgeon without the back-up of a cardiac surgical unit.
After the introduction of extracorporeal circulation as a routine technique, more and more pulmonary embolectomies were performed by cardiac surgeons with the aid of a heart–lung machine.1,2 One major reason for using this technique was the longer time available for extraction of the thrombi and inspection of the pulmonary arteries. It also became clear that relief of the right heart under total cardiopulmonary bypass would improve the recovery of the right ventricle and, consequently, patient outcome.3,4 Successful embolectomies under extracorporeal circulation have been reported even after long helicopter transport of patients under continuous resuscitation, underlining the importance of right ventricular evacuation (Kirchoff PG, personal communication).
The question to be answered, which is a matter of controversial discussion in various articles and guidelines, is whether any of the older or more recent surgical options, as described above, still play a role in the treatment of pumonary embolism alongside anticoagulation, thrombolysis and interventional procedures and, if so, what the indications for surgery are.5,6 The discrepancies have been well described by Samoukovic,7 who stated “sufficient evidence exists to extend the criteria for surgical embolectomy from strictly rescue therapy to include hemodynamically stable patients with RV dysfunction” and, in contrast, by Meneveau,8 who still considers “thrombolytic therapy… the first line therapy in patients with high-risk pulmonary embolism presenting with cardiogenic shock and/or persistent arterial hypotension”.9
Surgery is seen as an alternative only after conservative treatment has failed. Unfortunately, there are no prospective or even randomised studies that could answer these questions in an evidence-based manner.
Anatomical Approach
If we look at the indications for the different conservative and surgical therapies based on anatomical location, peripheral emboli cannot easily be approached using surgical techniques and are generally considered to be an indication for anticoagulation or lysis. One has to be aware that the anatomical location does not say anything about the clinical severity or the prognosis. A diffuse occlusion of many peripheral vessels may have the same deleterious haemodynamic effects as a more central embolus and will probably have an even worse prognosis. The aim of any therapy should be the reopening of as much vasculature as possible, thus lytic therapy should be the first-line option, if not contraindicated.
We have seen a young girl after cardiac transplantation who was referred to hospital with symptoms of dyspnoea and haemodynamic instability. Echocardiography revealed a hypokinetic left ventricle and a dilated and hypokinetic right ventricle. The patient’s clinical status deteriorated rapidly and resuscitation became necessary. Haemodynamic stabilisation could not be accomplished and the patient died. The autopsy revealed complete occlusion of practically all of the peripheral pulmonary vessels, which has the same haemodynamic effects as a central embolus with maximal dilatation of the right ventricle. It is probable that in this patient a larger embolus, originating from the superior vena cava after multiple myocardial biopsies, became fragmented and then spread into the periphery.
We should keep this example in mind when looking at more centrally located emboli. While the central vessels may be reopened by lytic therapy, no study has systematically looked at the periphery and at whether parts of the embolus have become fragmented, especially after accompanying cardiopulmonary resuscitation. Surgery, if applied early and as first-line therapy, may retrieve the whole thrombus or at least major parts of it and may lead to an improved prognosis. A prerequisite is an experienced surgical team that is familiar with the special cannulation and instrumentation techniques required. Occlusions of peripheral vessels not only may lead to an impaired gas exchange but also may be the cause of chronic pulmonary hypertension; this is also true for lost thrombus fragments under surgical therapy.
Haemodynamic Approach
Smaller emboli in peripheral vessels rarely cause major haemodynamic alterations. In these cases, anticoagulation and, if possible, thrombolysis are the first-line therapeutic options. Most of these emboli are not readily accessible by surgical techniques and, consequently, surgical embolectomy is indicated only in rare cases.
Central emboli, especially if located in the pulmonary trunk or the two main pulmonary arteries, will always cause haemodynamic problems as a result of two pathophysiological effects: antegradely, blockage of the transpulmonary blood flow leads to insufficient filling of the left ventricle; retrogradely, the blockage causes acute dilatation of the right ventricle with so-called structural dilatation of the heart muscle and venous congestion.
Of note, the structural dilatation of the muscle takes place on the microscopic level and is not acutely reversible. In such cases, anticoagulation alone will not be sufficient, with the exception of elderly patients who present with numerous co-morbidities. Thrombolytic therapy may sufficiently reopen one or both major branches and may be the therapy of first choice, especially under stable haemodynamic conditions.8 However, one has to keep in mind that further embolisation of smaller fragments into the peripheral vessels will lead to pulmonary hypertension of a different degree and to reduced exercise capacity, especially in younger patients and over a longer follow-up period. Consequently, pulmonary embolectomy should be considered an alternative, especially in younger patients and in those with a massive thrombus load. This is even more true for haemodynamically unstable patients, as cardiopulmonary resuscitation may further contribute to fragmentation of the thrombus and the acute dilatation of the right ventricle is not adequately adressed by the thrombolysis alone. Two studies have demonstrated improved results of surgical thrombectomy and justify an early indication in cases of massive emboli in central parts and in those with haemodynamic instability.2,4,6
On the other hand, recent publications and our own experiences have shown that delayed surgical therapy after failed thrombolysis has a worse prognosis. The prolonged periods of right ventricular dialtation, impaired pulmonary perfusion and a severely compromised coagulation system mainly result in a need for a high degree of post-operative catecholamine support, impaired gas exchange and, ultimately, severe reperfusion oedema and major blood loss.
A further example demonstrates the capacity of surgical embolectomy in such complex cases. A 77-year-old female patient suffered a massive pulmonary embolism and was in cardiogenic shock at arrival in our hospital. Computed tomography performed immediately after arrival in our emergency department revealed a central pulmonary embolism located in the pulmonary trunk and the two major branches. She was directly transferred to the operating theatre and, after typical median sternotomy, extracorporeal circulation was established and an extensive embolectomy performed, as described above. During weaning from cardiopulmonary bypass, a massive reperfusion oedema was seen, associated with severe impairment of the gas exchange.
We therefore decided to connect an extracorporeal membrane oxygenation (ECMO) system to the cannulae already already in position and, after temporary closure of her chest, the patient was transferred to our intensive care unit. Under additional nitric oxide ventilation, the gas exchange continuously improved. After four days, the patient could be weaned from the ECMO system and the sternotomy was closed definitively. After a prolonged period of mechanical ventilation, the patient could finally be weaned and decannulated, and she was transferred to a rehabilitation unit. A control computed tomography scan after six months showed no visible thrombus in the pulmonary vessels.
Of course, surgical therapy is primarily indicated in cases of central and surgically accessible emboli and if there are contraindications to thrombolytic therapy in any patient with a reasonable risk profile.
Summary
In many recent articles and guidelines, surgical therapy of pulmonary embolism is considered to be only a rescue option after failed thrombolytic therapy. However, there are still indications for primary surgical intervention. Newer extracorporeal circulation techniques that provide more complete extraction of the emboli and immediate relief of the right ventricular distension have demonstrated improved results compared with earlier approaches such as the Trendelenburg operation. The temporary use of an ECMO system to bridge the critical period of reperfusion oedema may further improve the prognosis. While there are clearly defined indications for anticoagulation alone and also for thrombolytic therapy, perhaps augmented with catheter devices, surgical embolectomy should be considered in any case of severe central emboli and of haemodynamic instability, if a surgical procedure is not contraindicated by the patient’s age or condition. Also, surgical interventions after failed thrombolysis have a significantly worse prognosis, even in younger patients, so the general recommendation to consider a surgical intervention only after failure of conservative therapy is no longer justified. As in many other fields of cardiovascular medicine, only comprehensive approaches will lead to optimal results for the individual patient and only trained cardiovascular surgeons can offer optimal results.