| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Cardiothoracic Surgery, Papworth Hospital, Papworth Everard, Cambridge, United Kingdom
b Department of Anesthesia and Intensive Care, Papworth Hospital, Papworth Everard, Cambridge, United Kingdom
c Cambridge Perfusion Services, Papworth Hospital, Papworth Everard, Cambridge, United Kingdom
Accepted for publication June 10, 2008.
* Address correspondence to Dr Jenkins, Papworth Hospital, Cambridge, CB23 3RE, United Kingdom (Email: david.jenkins{at}papworth.nhs.uk).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentReferences |
|---|
Methods: We conducted a retrospective review of all patients undergoing PTE from a single national referral center between August 2005 and August 2007.
Results: One hundred twenty-seven consecutive patients underwent PTE surgery. Seven patients (5.5%) had extreme cardiorespiratory compromise in the immediate postoperative period and required venoarterial ECMO support. Their mean age was 51.3 years with 3 males. When compared with patients not requiring ECMO, these patients had significantly poorer hemodynamic indices before the operation with mean pulmonary artery pressure of 62 mm Hg versus 51 mm Hg (p = 0.02) and pulmonary vascular resistance of 907 dynes/sec/cm–5 versus 724 dynes/s–1/cm–5 (p < 0.02). Mean duration of support was 119 hours (49 to 359 hours). Five patients were successfully weaned from ECMO support (73%) and 4 left the hospital alive, giving a salvage rate of 57%. For those who did not require ECMO support, hospital mortality was 4.2%.
Conclusions: Early venoarterial ECMO support has a role as rescue therapy post-PTE in patients with severe compromise who would probably otherwise die.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Potential complications from this surgery include reperfusion injury with poor gas exchange, persistent pulmonary hypertension, and right ventricular failure. Patients suffering these complications early after surgery account for many of the in-hospital deaths. Because right ventricular function, PVR gas exchange, and cardiac output are intimately related, venoarterial extracorporeal membrane oxygenation (VA-ECMO) is necessary to prevent the vicious cycle of hypoxic vasoconstriction increasing PVR, with reduction in cardiac output. Institution of VA-ECMO has the advantage of decreasing right heart volume and allowing recovery of ventricular function and optimizing oxygen transport by improving cardiac output and oxygen content.
We believe that patients with the correct diagnosis and an adequate technical result at operation, as demonstrated by a satisfactory endarterectomy specimen, should be supported through potentially reversible life threatening complications by mechanical means when conventional treatment has failed. This technique gives patients with residual pulmonary hypertension after PTE a chance of recovery in situations that were previously impossible to treat and were usually fatal. Only two other groups have reported ECMO use in patients after PTE surgery [10, 11]. We report our own experience with ECMO in this patient population over the last 2 years.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Surgical Techniques
Surgery was performed through a median sternotomy with hypothermic cardiopulmonary bypass using techniques similar to those established by Jamieson and Kapelanski [12]. We have previously reported that complete circulatory arrest is not necessary in all patients [13].
The ECMO Circuit
The decision for ECMO institution was made for each case individually. The ECMO circuit was designed to allow patients to be supported within 10 minutes of the clinical decision being made. The circuit consists of 3/8 inch Jostra Bioline heparin bonded tubing, a Quadrox true membrane oxygenator, and a Rotaflow centrifugal pump (Jostra Medizintechnik AG, Hirrlingen, Germany). Intraventricular spikes on the arterial and venous limbs allow the circuit to be gravity primed and recirculated as a closed circuit by a 1 L bag of Hartmann's solution.
Cannulation
Cannulation for central venoarterial ECMO was similar to that for CPB. Venous drainage was obtained by wire reinforced two-stage venous cannula inserted through the right atrium that avoids kinking (Medtronic 32/40 Fr MC2; Medtronic Inc, Minneapolis, MN). A 24 French Sarns (13030; Terumo Corp, Ann Arbor, MI) straight arterial cannula was used for arterial return. The cannulae were double snugged and secured with spigots. The cannulae are positioned so that they exit the chest at the caudal end of the sternotomy wound without tunneling, as support is usually short term. In most cases the chest was lightly packed with gauze and the sternum left open with the wound covered by a sterile membrane or simple skin closure.
ECMO Control
Pulsatile flow was allowed by reducing preload by 50%, allowing the left ventricular to eject with ECMO blood flow set at approximately 3 L/minute (depending on body surface area). This allowed some blood flow (approximately 1 L) through the heart and lungs to reduce the risk of thrombus formation. Flow was adjusted according to pulmonary artery catheter readings with the aim to keep a systolic to diastolic difference of greater than 10 mm Hg, but to avoid further damaging pulmonary hypertension. Hemoglobin was kept greater than 10 g/dL to optimize oxygen transfer with perfusion assessed by arterial oxygen saturation (SaO
2), saturated venous oxygen (SvO
2), acid base status, and urine output. The patient temperature was kept at 36°C to 37°C, and ECMO blood flow was adjusted in response to native cardiac output. Protamine was administered to reduce the activated clotting time to 200 seconds. Anticoagulation with heparin was commenced typically within 24 hours to achieve activated clotting times (ACT) of 180 ± 20 seconds, unless there was on-going coagulopathy with hemorrhage. Coagulation parameters were monitored as required in addition to repeated platelet counts. Platelet administration was considered in patients with active bleeding or platelet count less than 100,000 platelets/mL. Ventilator management included reduction of tidal volume (3 to 5 mL/kg) and peak airways pressures (maintained <25 mm Hg) to minimize volutrauma and barotraumas to lungs. Inspired oxygen concentrations (FIO
2) were weaned empirically (<0.7) to reduce the potential of oxygen toxicity.
Weaning of ECMO
Weaning timing was variable and patient dependent. Weaning from the ECMO circuit was commenced at the earliest possible opportunity when there was agreement that recovery had progressed sufficiently so that the patient was at less risk with their native circulation than with the on-going risks and potential complications with ECMO. In practice this was after at least 48 hours support. After this time, patients were evaluated regularly for hemodynamic and gas exchange improvement assessing the possibility of weaning. While weaning, a bolus of 5,000 units of heparin was administered. Flow rate was reduced stepwise by increments of 0.5 L/minute under continuous monitoring of hemodynamic and respiratory variables. Ventricular function was assessed using transesophageal echocardiography. Inotropic and vasodilator support was increased as required and nebulized iloprost was continued. Acceptable parameters included central venous pressure of less than 15 mm Hg and mean systemic arterial pressures of greater than 60 mm Hg, and stable left and right ventricular function with adequate gas exchange. At this stage actual pulmonary artery (PA) pressure was less relevant. Weaning took place in stages on the intensive care unit. Final weaning and decannulation was performed in the operating theatre. After sustained stability for 1 hour the cannulae were removed and the circuit discarded. In one patient peripheral venous cannulae were placed after a further hour for venovenous (VV)-ECMO because of borderline gas exchange with stable circulation.
Statistical Analysis
Continuous variables are expressed as mean ± standard deviation. Univariate comparisons between patients were performed by using a 
2 test or Fisher exact test, where appropriate for dichotomous variables, and by the two-tailed Student t test or Mann-Whitney U test for continuous variables, absent variables and variables calculated from them were excluded from any analysis. Statistical tests were two-sided and assumed a 5% level of significance. Software used for the data and statistical analyses was Statsdirect 2.0.0 (Statsdirect, Cheshire, UK).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
The postmortem findings in the remaining 2 patients who died while on ECMO support showed evidence of multiorgan failure with pulmonary hemorrhage in both patients, and additional bowel ischemia in one. Mean duration of support was 119 hours (49 to 359 hours). For those who did not require ECMO support, hospital mortality was 4.2%.
The ECMO group had poorer hemodynamics pre-CBP with pulmonary artery systolic pressures of 98 mm Hg versus 83 mm Hg (p = 0.06) and mean pulmonary artery pressure of 62 mm Hg versus 51 mm Hg (p = 0.02) (Table 1). In 5 patients, ECMO was commenced in the first hour after weaning from bypass due to evolving cardiorespiratory arrest. Initiating factors were raising PA pressure (4), severe partial pressure of oxygen/partial pressure of carbon dioxide (PO 2/PCO 2) derangement (2), and low systemic blood pressure (1). Patients who required VA-ECMO in the postoperative period had significantly higher postoperative mean PA pressures immediately post-CPB (prior to deterioration) than those who did not require VA-ECMO (39 vs 26 mmHg, p <0.0001) and higher immediate post-CPB pulmonary vascular resistance (479 vs 273 dynes/s–1/cm–5, p = 0.0003).
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Residual pulmonary hypertension and reperfusion injury account for the majority of postoperative deaths [5]. The pathophysiology of persistent pulmonary hypertension after pulmonary thromboendarterectomy is probably multifactorial, and involves reversible and irreversible factors. Although small vessel vasculopathy or "distal disease" is probably the most important single contributor, the damaging effects of CPB and the trauma of surgery compound the problem. In extreme cases the resultant vicious cycle of increasing hypoxia, worsening PVR, and falling cardiac output can only be interrupted by full cardiorespiratory support. Otherwise, patients can deteriorate very suddenly and conventional resuscitation is usually ineffectual.
Reperfusion injury is probably caused by derangement at the alveolar-capillary interface and is recognized after lung transplantation and PTE, but the exact cause is not understood. It may occur in up to 20% of patients after PTE, but the majority responds to aggressive diuresis and continued positive pressure ventilation [11]. In rare cases when conventional treatment is failing, ECMO may be the only method of supporting the patient and often veno-venous is sufficient [11].
In general, the results of ECMO support in an adult population have been disappointing and in most centers it is reserved as salvage therapy. The results of the largest randomized trial (conventional ventilation or ECMO for severe adult respiratory failure (CESAR) trial of ECMO versus conventional ventilation in adult respiratory distress syndrome) are awaited [14]. A retrospective analysis of 202 adult patients who underwent venoarterial ECMO for cardiac failure secondary to numerous causes, including failure to separate from CPB postcoronary arterial grafting [15], showed a 30-day survival after initiation of ECMO of 38%; our results in this patient population with a 30-day survival of 71.4% and survival to discharge of 57% appear favorable despite the small numbers. Higher success post-PTE may be due to reduced myocardial injury and end organ pathology compared with patients with primary cardiac pathology.
Extracorporeal membrane oxygenation has also been reported to be effective in selected patients with graft failure after lung transplantation [16, 17] and with posttransplant pulmonary hypertension [18], whose postoperative recoverable physiology may be similar to the PTE patient. More recently Aigner and colleagues [19] have expanded the role of ECMO in lung transplantation, favoring it over CPB for intraoperative support and extending the use into the early postoperative period with 147 of 306 patients receiving support at some stage of their treatment. Wigfield and colleagues [20] reported ECMO use as a bridge to recovery in patients with primary graft dysfunction after lung transplantation, and highlighted the need for early support before the onset of multiorgan failure. The same group compared with pro and cons of VV-ECMO versus VA-ECMO, suggesting that the VA-ECMO when instituted early can provide a higher oxygen delivery capacity and additional circulatory support to facilitate early recovery. Reducing pulmonary vascular flow potentially modulates the endothelial activation and aggravation of pulmonary edema secondary to reperfusion injury.
Different centers have favored different cannulation strategies with the main distinction between central and peripheral. We believe that in patients with a recent sternotomy the ease of access to central vessels has advantages for short-term VA-ECMO use. The risk factors of peripheral VA-ECMO were emphasized by Zimper and colleagues [21]. In 174 patients requiring ECMO after cardiac surgery or lung transplantation, 57% survived to discharge but there was a 20% infection rate, neurologic events in 22%, and 28% limb complication rate. In survivors, 12% experienced late vascular complications secondary to the femoral artery cannulation site. They found that the two most important independent factors for the development of late vascular complication were technical problems during ECMO explantation and a history of peripheral vascular disease. However, peripheral VV-ECMO has less bleeding complications [22].
Compared with the previous reports of ECMO use after PTE surgery [10, 11] our series has some similarities, but some important differences. At UCSD [11], ECMO use was restricted to a small subset of patients (1.2%) with stable hemodynamics and severe reperfusion injury using peripheral veno-venous support. Whereas Ogino and colleagues at Osaka [10], like us, attempted salvage with ECMO more frequently in 8 out of 88 patients (9.1%) of their series and also preferred veno-arterial full hemodynamic support, but with peripheral percutaneous cannulation rather than our central technique. As with our supported patients, the UCSD patients [11] had higher than average preoperative PA systolic pressure (mean 87 mm Hg) and PVR (1,148). In the UCSD experience, ECMO was instituted relatively late, at a mean of 86 hours postsurgery, and patients were supported for a mean of 123 hours with only 30% surviving to discharge. At Osaka, ECMO was used early in patients who had difficulty weaning from CPB and patients were supported for a mean of 110 hours, with a weaning success of 50% and hospital survival of 33%. With increasing experience patients were supported earlier, had shorter duration of support (48 hours), and better survival. A major difference in our approach is that we have demonstrated the potential for full central VA-ECMO support in patients with extreme hemodynamic and respiratory compromise. At UCSD [11], ECMO is used more selectively and only for those patients exhibiting a good hemodynamic response to endarterectomy with a residual PVR of less than 600.
The high weaning success from ECMO post-PTE in this patient population is multifactorial and, in our opinion, depends on the following factors: (1) early institution of ECMO with simple ECMO circuit immediately available that can be quickly primed; (2) maintenance of adequate body perfusion and oxygenation, decompression of the right ventricle, and prevention of further lung injury; (3) adequate hemostasis to minimize the blood product requirement, reducing further risk of pulmonary injury and infection; and (4) a potentially reversible process that can be improved by support.
The main cause of death in our population was multiorgan failure and pulmonary hemorrhage. Some of these events may in fact be contributed to by the need for anticoagulation in the setting of surgical trauma and coagulopathy in the postoperative period, and all interventions affecting coagulation must balance the risk of bleeding with the risk of thrombosis, often difficult in these patients.
Limitations of this study include the retrospective nature of analysis and the relatively low number of patients. However, our recent experience increases the number of patients reported and proves that this therapy has a useful place in the treatment of a very sick patient population with over half surviving to discharge. A further reason to publicize this experience is that patients who survive to hospital discharge after PTE have been shown to have excellent midterm survival with 93% of our own patient group (1997 to 2006) being alive at 5 years [23].
In conclusion, early ECMO placement has a role as rescue therapy post-PTE in patients who would probably otherwise die. The success for adult ECMO in this group is relatively high and we would recommend early use of venoarterial support for any patient unstable with residual pulmonary hypertension and reperfusion injury after PTE.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  M. Berman, E. Pavlushkov, E. Abraham, J. Dunning, S. Tsui, R. Hall, A. Klein, and D. P. Jenkins Pulmonary endarterectomy - an example of treatment of right ventricular after load failure MMCTS, June 12, 2009; 2009(0612): 3491. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Jamieson and M. Madani Invited Commentary Ann. Thorac. Surg., October 1, 2008; 86(4): 1267 - 1267. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
