| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Cardiology, University Hospital Jean Minjoz, Besançon, France.
Correspondence to: Nicolas Meneveau, MD, PhD, FESC, Department of Cardiology, University Hospital Jean-Minjoz, Blvd Fleming, 25030 Besançon Cedex, France; e-mail: nicolas.meneveau{at}ufc-chu.univ-fcomte.fr
Abstract
Background: The management of patients with acute massive pulmonary embolism (PE) who do not respond to fibrinolytic therapy remains unclear. We aimed to compare rescue surgical embolectomy and repeat thrombolysis in patients who did not respond to thrombolysis.
Methods: We conducted a prospective single-center registry of PE patients who underwent thrombolytic therapy. Lack of response to thrombolysis within the first 36 h was prospectively defined as both persistent clinical instability and residual echocardiographic right ventricular dysfunction. Patients underwent surgical embolectomy or repeat thrombolysis, at the discretion of the attending physician. The clinical end point was a combined end point including recurrent PE, bleeding complications, or PE-related death, which was defined as death from recurrent PE or cardiogenic shock. Long-term adverse outcomes included death, recurrent thromboembolic events, and congestive heart failure.
Results: From January 1995 to January 2005, 488 PE patients underwent thrombolysis, of whom 40 (8.2%) did not respond to thrombolysis. Fourteen patients were treated by rescue surgical embolectomy, and 26 were treated by repeat thrombolysis. There was no significant difference in baseline characteristics between the two groups. The in-hospital course was uneventful in 11 of the surgically treated patients (79%) and in 8 patients (31%) treated by repeat thrombolysis (p = 0.004). There was a trend for higher mortality in the medical group than in the surgical group (10 vs 1 deaths, respectively; p = 0.07). There were significantly more recurrent PEs (fatal and nonfatal) in the repeat-thrombolysis group (35% vs 0%, respectively; p = 0.015). While no significant difference was observed in number of major bleeding events, all bleeding events in the repeat-thrombolysis group were fatal. The rate of uneventful long-term evolution was the same in the two groups.
Conclusion: Rescue surgical embolectomy led to a better in-hospital course when compared with repeat thrombolysis in patients with massive PE who have not responded to thrombolysis. The transfer of patients who have not responded to thrombolysis to tertiary cardiac surgery centers could be considered as an alternative option.
Key Words: embolism • surgery • thrombolysis
The management of patients with acute massive pulmonary embolism (PE) who do not respond to fibrinolytic therapy remains unclear. This is partly due to the difficulty of defining "failed" thrombolysis in this setting, unlike the situation in patients with myocardial infarction, in whom the criteria and consequences of unsuccessful thrombolysis are well-established. In the setting of PE, the recovery of right ventricular (RV) function is an early marker of thrombolysis efficacy as well as a predictor of in-hospital course.12345 Furthermore, residual pulmonary vascular obstruction (> 30% at 10 days) after thrombolytic therapy is associated with adverse outcomes and increased long-term mortality.1 Despite this, there is no structured policy for the management of failed thrombolysis, and many physicians follow a conservative approach. However, optimizing early pulmonary revascularization could play a pivotal role in improving both the immediate and long-term evolution in patients with acute massive PE who do not respond to thrombolysis.
The therapeutic options in patients with persistent hemodynamic instability who do not respond to thrombolysis rely on two different strategies comprising either rescue surgical embolectomy or repeat thrombolysis. However, both therapeutic options entail potential risks, and to date the question of the most appropriate management strategy for PE patients who do not respond to thrombolysis has never been addressed. Until the past few years, pulmonary embolectomy was confined to clinically desperate circumstances, while repeat thrombolysis was reported to lead to increased bleeding risks, especially intracranial hemorrhage.67 The aim of our study was to compare rescue surgical embolectomy and repeat thrombolysis in patients who have not responded to thrombolysis, based on a prospective single-center registry of PE patients who had undergone thrombolytic therapy.
Materials and Methods
Selection of Patients
The study population was derived from a single-center registry of patients with confirmed massive and submassive PEs who had undergone thrombolytic treatment between January 1995 and January 2005. All patients gave their informed written consent for inclusion in the registry, and the protocol was approved by the local ethics committee.
Patients with proven recent PE (symptom onset, < 15 days) and no contraindication to thrombolytic therapy were included in the registry if they met at least one of the following criteria: (1) cardiogenic shock defined as systolic BP 
90 mm Hg associated with clinical signs of organ hypoperfusion and hypoxia; (2) syncope; (3) pulmonary vascular obstruction of > 50%; and (4) mean pulmonary artery pressure of >20 mm Hg by right heart catheterization, plus at least one echocardiographic finding indicating RV dysfunction (ie, RV/left ventricular end-diastolic diameter ratio, 
1 in the four-chamber view; paradoxical septal systolic motion and/or pulmonary hypertension [PH], defined as a RV/atrial gradient > 30 mm Hg). Repeat echocardiographic examination was systematically performed 24 to 36 h after thrombolytic therapy, and RV dysfunction criteria were recorded.
Patients from the registry were selected for inclusion in this study if they had massive PE and had not responded to the initial thrombolysis. Unsuccessful thrombolysis within the first 36 h was defined as both persistent clinical instability and residual echocardiographic RV dysfunction. Persistent clinical instability was prospectively defined as the presence of at least two of the following criteria: refractory cardiogenic shock; systemic arterial hypotension (defined as systolic BP of 90 mm Hg or a pressure drop of 40 mm Hg for > 15 min if not caused by new-onset arrhythmia, hypovolemia, or sepsis)8; severe hypoxemia (ie, room-air pulse oximetry of 90% or PaO2 without oxygen therapy of 55 mm Hg); or tachycardia (heart rate, 110 beats/min). Residual echocardiographic RV dysfunction was defined as the persistence of at least two initial RV dysfunction criteria.
In addition, a lack of improvement in pulmonary vascular obstruction was documented in most patients with repeat spiral CT scan or pulmonary angiogram performed within the first 36 h. The therapeutic approach was left to the discretion of the attending physician following repeat spiral CT scan or pulmonary angiography.
Management Strategies and Medication
Surgical Embolectomy:
Surgical embolectomy was performed within 72 h of the initial thrombolysis. After median sternotomy and pericardiotomy, patients were heparinized and cannulated for cardiopulmonary bypass. The arterial cannula was placed in the ascending aorta with either a bicaval or a single venous cannula placed through the right atrium. The procedure was performed under normothermia without cardiac arrest and with vacuum-assisted venous drainage. The clot was extracted through a longitudinal arteriotomy in the main pulmonary artery, under direct vision using forceps. In all patients, an inferior vena caval filter was inserted perioperatively prior to sternal closure.
Repeat Thrombolysis: Repeat thrombolysis was started at least 24 h after initial thrombolysis and consisted of the administration of streptokinase in patients previously treated with alteplase, while patients who received streptokinase initially were subsequently treated with alteplase. Streptokinase was administered as an infusion of 1.5 million IU over 2 h. Alteplase was infused at a dose of 100 mg over 2 h. Both thrombolytics were only administered once fibrinogen level rose above 1 g/L.
Postintervention Treatment: Therapy with IV unfractionated heparin was started immediately after surgery or at the end of thrombolytic infusion, was maintained at a dose of 1,000 IU/h, and was adapted thereafter to achieve an activated partial thromboplastin time ratio of two to three times the control value. Oral anticoagulant therapy was introduced within 3 to 5 days and was continued for at least 6 months, but was adjusted to maintain the international normalized ratio between 2 and 3.
Definition of Clinical End Point
The clinical end point of the in-hospital course was a combined end point including recurrent PE, bleeding complications, or PE-related death, which was defined as death from recurrent PE or cardiogenic shock. Major bleeding complications were prospectively defined as any bleeding event that required blood transfusion, surgical control, and discontinuation of thrombolytic or anticoagulant treatment; hemorrhagic stroke confirmed by CT scan or autopsy; or any bleeding causing death or defined as a fall of 15% in hematocrit. Other important bleeding events, which was defined as a fall of 10% in hematocrit, were also recorded.8 Patients with symptoms suggesting PE and with new filling defects seen on spiral CT scan or pulmonary angiogram were interpreted as having recurrent PE.
Long-term Follow-up
Follow-up data were obtained during hospital readmission, during scheduled patient consultations to the department, or from a standardized questionnaire sent to the attending physician and/or cardiologist. Adverse outcomes included death, recurrent phlebitis, recurrent PE, development of congestive heart failure (CHF), or change of New York Heart Association (NYHA) functional class to class III or IV. Hospital records and the death certificates of patients who died during the follow-up period were also reviewed. Follow-up was concluded in March 2005.
Statistical Analysis
Continuous variables are expressed as the mean ± SD; categoric variables are expressed as a percentage. Nonparametric Wilcoxon test and Fisher exact test were used for the comparison of continuous and categoric variables, respectively. Associations between variables and adverse clinical events were expressed as odds ratios (ORs) and 95% confidence intervals (CIs). All tests were two-sided, and a p value of < 0.05 was considered to be significant. Analyses were performed with a statistical software package (BMDP; BMDP Statistical Software, Inc; Los Angeles, CA).
Results
From January 1995 to January 2005, 1,876 consecutive patients were referred to the cardiology department with confirmed PE, of whom 488 (26%) were treated with thrombolytic therapy. At 24 to 36 h, thrombolysis was considered to be unsuccessful in 40 (8.2%) patients, of whom 37 had both persistent clinical instability and echocardiographic findings of RV dysfunction, and 3 patients experienced early echocardiographic deterioration following initial thrombolysis. Two eligible patients were excluded from the analysis because they experienced fatal cardiac arrest following the initial thrombolysis and had received repeat thrombolysis only as a treatment of last resort in clinically desperate circumstances.
Clinical Presentation on Hospital Admission
The clinical characteristics of patients at hospital admission are reported in Table 1
. The study population was made up of 15 men (38%) and 25 women (62%), with a mean age of 62 ± 18 years (age range, 26 to 84 years). Fourteen patients underwent rescue embolectomy, and 26 underwent repeat thrombolysis. There were no significant differences between the two treatment groups in terms of baseline demographic and clinical characteristics. PE was initially diagnosed by pulmonary angiogram in 12 patients (30%), by high-probability lung scan in 7 patients (17%), and by spiral CT scan in 21 patients (53%). The onset of symptoms dated from < 6 days in about 60% of patients, with no significant difference between the two groups.
|
|
Initial Thrombolysis
Most patients (73%) initially received streptokinase therapy in accordance with the policy of our institution. However, the distribution of both thrombolytic agents was similar between the two treatment groups.
Twenty-three patients underwent control spiral CT scanning, and 10 patients underwent control pulmonary angiogram between 24 and 36 h. Seven patients who were in very unstable condition did not undergo repeat spiral CT scan or pulmonary angiogram (three patients underwent rescue surgical embolectomy; four patients underwent repeat thrombolysis). Three patients who had undergone rescue surgical embolectomy underwent additional pulmonary angiography before surgery according to the surgeon’s requirement. Persistent proximal thrombi were reported in all 33 patients who had undergone control spiral CT scan or pulmonary angiogram. Thrombi were bilateral in 24 patients (73%) and unilateral in 9 patients (27%). There was no significant difference between the two groups in terms of thrombi location.
In-Hospital Course
The mean duration of hospital stay was 12 ± 5 days. Of the 26 patients who underwent repeat thrombolysis, 19 (73%) were initially treated with streptokinase and subsequently received tissue plasminogen activator (t-PA), while 7 (27%) who had initially been treated with t-PA were given streptokinase. All patients in the surgically treated group received a vena caval filter perioperatively, compared with only 8 of 26 patients in the repeat-thrombolysis group (percutaneous femoral or jugular approach) [100% vs 31%, respectively; p < 0.0001].
The in-hospital clinical course was uneventful in 11 patients (79%) who had undergone rescue embolectomy, compared with 8 patients (31%) who had undergone repeat thrombolysis (p = 0.004). This unfavorable evolution in the condition of patients in the repeat-thrombolysis group was the result of a higher mortality rate, which was associated with more recurrent PE. A total of 10 patients (38%) died in the repeat-thrombolysis group, of whom 3 died from recurrent PE, 3 died from refractory cardiogenic shock, and 4 died from major bleeds. Comparatively, only one patient (7%) died following rescue surgical embolectomy, from refractory cardiogenic shock (p = 0.07). The rate of fatal or nonfatal recurrent PE was significantly higher in patients who underwent repeat thrombolysis compared with those who were operated on (35% vs 0%, respectively; p = 0.015).
Eight patients (20%) experienced bleeding complications, with no significant difference between the two treatment groups. All four major bleeding episodes that occurred in the repeat fibrinolysis group were fatal. Intracranial hemorrhage occurred in one patient (4%) after undergoing repeat thrombolysis. The majority of major bleeding episodes were related to an early invasive procedure (pulmonary angiography, two patients; venous puncture site, one patient; or insertion of intracaval device, one patient). One patient experienced severe GI bleeding. Surgical control of hematoma and blood transfusion were needed in two and four patients, respectively.
Long-term Evolution of Patients Who Survived the Acute Phase
Long-term follow-up data were obtained in all patients who survived the acute phase. The mean follow-up period was 5.7 ± 2.7 years (range, 2.6 to 10.5 years). Long-term condition evolution was uneventful in 62% of patients who underwent rescue embolectomy compared with 69% of those who underwent repeat thrombolysis. The difference was not statistically significant. Of the 29 patients who were alive after the hospital stay, 4 patients died during follow-up from comorbidities. In the repeat-thrombolysis group, one patient died from cancer and one patient died from postoperative complications, while two patients died in the surgical group, one from cancer and one from renal failure. The rate of thromboembolic events (ie, recurrent PE or deep vein thrombosis [DVT]) was 23% in patients who had undergone surgical embolectomy and was 25% in those treated by repeat thrombolysis (p = 0.98), even though approximately half of the study population was permanently treated with oral anticoagulant therapy. The occurrence of CHF or change in NYHA functional class to class III or IV was similar in both groups. All in all, the rate of uneventful evolution was the same in the two groups.
Discussion
To date, no study has ever assessed the management of failed thrombolysis in the setting of acute massive PE. Our study, based on a single-center prospective registry, is the first study to report the immediate clinical course as well as the long-term outcome of acute PE patients who had undergone either rescue embolectomy or repeat thrombolysis after not responding to initial thrombolysis. Patients who were unresponsive to thrombolysis were prospectively defined as patients with both persistent clinical instability and echocardiographic criteria of RV dysfunction.
We found that approximately 8% of patients do not respond to thrombolysis. In this situation, rescue surgical embolectomy should be preferred over repeat thrombolysis. The in-hospital course of patients who had undergone rescue embolectomy was significantly better than that of patients who were treated with a second thrombolysis. This early benefit is the result of a significant reduction in the number of recurrent PEs, which was associated with more recurrent PE. Interestingly, patients who were operated on had a more severe initial profile (shock was present in 36% vs 15% among repeat-thrombolysis patients).
The in-hospital mortality rate of 28% observed in the overall study population is much higher than that usually reported after thrombolysis in patients who have experienced acute massive or submassive PE,1910 but is comparable to the mortality rate reported in the MAPPET registry in patients presenting with cardiogenic shock.11 In the repeat-thrombolysis group, the in-hospital mortality rate reached 38%, and one third of these deaths were caused by recurrent PE, thus confirming that the thromboembolic process is responsible for a higher degree of early mortality than previously believed.19 In contrast, patients who underwent rescue surgical embolectomy had a very low in-hospital mortality rate of 7% and experienced no recurrent PE. Hence, the significant reduction of recurrent PE observed in patients who had been treated surgically may be related to systematic perioperative vena caval filter insertion, as previously suggested.1213
Our results are in accordance with those of a previous nonrandomized trial14 comparing thrombolysis and embolectomy in acute PE patients. The survival rate was 67% in the thrombolysis group compared with 77% in the surgical group. Until more recently, surgical embolectomy was confined to clinically desperate circumstances.7 More recent data,1213 however, reported a high survival rate of 89% at 1 month, thus suggesting that surgical embolectomy might be counted as one of several therapeutic options in acute PE patients. This high survival rate observed in patients with submassive PE was attributed to improved surgical techniques, rapid diagnosis, and triage. The low in-hospital mortality rate observed in our study post-rescue surgical embolectomy confirms that surgery is a valuable therapeutic option, even in hemodynamically unstable patients who have undergone massive PE and are unresponsive to thrombolysis.
Optimizing early pulmonary revascularization might be of importance since symptomatic chronic thromboembolic PH has been reported1516 as a relatively common yet serious complication of PE, affecting approximately 4% of patients within 2 years after a first episode of symptomatic PE. In addition, the persistence of PH or RV dysfunction after thrombolytic therapy in patients with acute PE, is associated with increased long-term mortality and adverse outcomes.117 In this context, the better outcome observed in the surgical group could be explained by a more complete pulmonary revascularization, although pulmonary vascular obstruction was not reevaluated after surgery, nor after repeat thrombolysis. In both groups, patients who survived the acute phase could be considered simply as patients who experienced improvement in pulmonary revascularization and right heart function, since long-term outcomes were similar in the two treatment groups. The potential role for catheter embolectomy remains uncertain and has never been assessed in the setting of failed thrombolysis.18 In addition, this percutaneous approach may be associated with an increased bleeding risk in the context of previous recent fibrinolysis.
Our results demonstrate that repeat thrombolysis is successful without adverse outcomes in only 31% of patients. It has previously been reported19 that thrombolysis is more effective in PE patients who have had a relatively short duration of symptoms. There is an inverse association between duration of symptoms and improvement of pulmonary vascular reperfusion after thrombolysis. Although delay attenuates efficacy, thrombolysis was reported in this study to be still useful in patients who had experienced symptoms for 6 to 14 days. In our study, approximately one third of the patients had clinical symptoms dating from 6 to 14 days previously. Furthermore, it is interesting to note that four of the five PE-related deaths in the repeat-thrombolysis group were in patients whose first symptoms had appeared > 6 days prior to treatment. In addition, we cannot exclude the possibility that asymptomatic (silent) PE preceded the onset of clinical symptoms. Hence, failed thrombolysis may be related to prior PEs > 14 days old, for which this treatment is less effective.
It is unlikely that antecedents of streptococcal infections, and thereby, the presence of streptococcal antibodies, were directly responsible for any negative effect on the efficacy of streptokinase. The fact that all patients had a fibrinogen level of < 1 g/L proves that streptokinase was effective.
We did not observe any more episodes of bleeding in patients in the repeat-thrombolysis group than in those in the surgically treated group. Similarly, there was no significant difference in the rate of intracranial hemorrhage between the two treatment groups. Data from the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries 1 study20 showed that repeat thrombolysis for the treatment of early reinfarction was associated with an increased risk of intracranial hemorrhage. On the other hand, the Assessment of the Safety and Efficacy of a New Thrombolytic 2 trial21 did not find any difference in hemorrhagic stroke in the same indication. While our study did not show any significant difference in major bleeding, this result is hard to interpret given the small sample size. Nonetheless, it must be noted that all patients who experienced major bleeding episodes in the repeat-thrombolysis group had a fatal outcome. This important result must be taken into account when evaluating the risk/benefit ratio and in the decision-making process. It underlines the clinical fragility of patients who are hemodynamically unstable, in whom the occurrence of a major bleeding complication is associated with an extremely unfavorable prognosis.
A "watchful waiting" approach after initial thrombolysis in patients who have experienced a massive PE is probably most appropriate. The persistence of echocardiographic signs of RV dysfunction combined with poor clinical tolerance call for a repeat-reperfusion strategy, preferably using surgical embolectomy, based on our experience. In most clinical settings, the readministration of a thrombolytic agent is the only form of treatment available for PE patients who are unresponsive to the initial thrombolysis. In this context, the transfer of these patients to tertiary cardiac surgery centers could be considered as an alternative option.
Limitations
The treatment assignment, after the patient has not responded to thrombolysis, to either the readministration of a thrombolytic agent or to rescue surgical embolectomy, was not randomized. There could have been several potential selection biases. As the management strategy of patients was left entirely to the discretion of the attending physician, it is quite possible that older age and the presence of right heart thrombi led to a decision in favor of surgical therapy rather than for repeat thrombolysis.22 There were no significant differences between treatment groups in terms of clinical presentation, echocardiographic findings, and the location of pulmonary emboli, before and after thrombolytic therapy, except with regard to intracardiac thrombus. However, it should be noted that there were nearly twice as many patients in the repeat-thrombolysis group. Despite the fact that this study was conducted over a long period of time, the thrombolytic agents and surgical embolectomy techniques used remained unchanged throughout the whole study period.
In addition, the bleeding rate observed in the repeat-thrombolysis group may have been underestimated, since thrombolytic therapy is usually not readministered to patients who experience a bleeding complication after receiving the first thrombolytic dose. Hence, selection bias may have caused less bleeding in the patients who underwent repeat thrombolysis.
Furthermore, the extrapolation of our results to other centers can be envisaged only if an experienced surgical team similar to ours is available, since few centers have surgeons who are experienced in performing pulmonary embolectomy. Similarly, it should be noted that most of the patients who underwent thrombolysis (73%) were initially treated with streptokinase. Therefore, some caution should be exercised in the extrapolation of these results to other groups of patients undergoing thrombolysis. Despite these limitations, a more aggressive approach toward patients presenting after not responding to thrombolysis is associated with improved prognosis.
Conclusions
Lack of response to thrombolysis, which was defined as persistent RV dysfunction and hemodynamic instability, can be expected in about 8% of patients who experience massive PE. Rescue surgical embolectomy led to a better in-hospital course when compared with repeat thrombolysis in such patients. An uneventful in-hospital evolution was significantly higher in the surgical group, due to a lower in-hospital mortality rate, less frequent episodes of major bleeding, and recurrent PE. Based on our results, and pending confirmation in a randomized trial, repeat thrombolysis is not to be recommended in this setting. The transfer of patients who have not responded to thrombolysis to tertiary cardiac surgery centers could be considered as an option in this context.
|
|
Abbreviations: CHF = congestive heart failure; CI = confidence interval; DVT = deep vein thrombosis; NYHA = New York Heart Association; OR = odds ratio; PE = pulmonary embolism; PH = pulmonary hypertension; RV = right ventricular; t-PA = tissue plasminogen activator
Received for publication August 5, 2005. Accepted for publication September 29, 2005.
References
This article has been cited by other articles:
|
|
 |
|
  G. Eid-Lidt, J. Gaspar, J. Sandoval, F. D. de los Santos, T. Pulido, H. Gonzalez Pacheco, and C. Martinez-Sanchez Combined Clot Fragmentation and Aspiration in Patients With Acute Pulmonary Embolism Chest, July 1, 2008; 134(1): 54 - 60. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Kumar, Y. Janjigian, and D. R. Schwartz Paradoxical Worsening of Shock After the Use of a Percutaneous Mechanical Thrombectomy Device in a Postpartum Patient With a Massive Pulmonary Embolism Chest, August 1, 2007; 132(2): 677 - 679. [Full Text] [PDF]
|
|
|
|
|
|
D. J. Carlbom and B. L. Davidson Pulmonary Embolism in the Critically Ill Chest, July 1, 2007; 132(1): 313 - 324. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Davidson and R. Karmy-Jones When Pulmonary Embolism Treatment Isn't Working. Chest, April 1, 2006; 129(4): 839 - 840. [Full Text] [PDF]
|
|
Read all eLetters
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
