Early outcome after Fontan operation for univentricular

Transcription

1 Thrombus Formation After the Fontan Operation Patrick D. Coon, RDCS, RCVT, Jack Rychik, MD, Rita T. Novello, RDCS, Pamela S. Ro, MD, J. William Gaynor, MD, and Thomas L. Spray, MD Cardiac Center at the Children s Hospital of Philadelphia, and Departments of Pediatrics and Surgery, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania Background. Thrombus formation is common after a Fontan operation. We investigated the frequency and location of thrombus in our population of children based on the type of Fontan operation performed. Methods and Results. Between January 1987 and January 1999, 592 patients underwent echocardiography after Fontan operation and 52 (8.8%) had intracardiac thrombus. Median age at Fontan operation was 1.9 years (range 0.8 to 35.1). Freedom from thrombus was 92%, 90%, 84% and 82% at 1, 3, 8, and 10 years after Fontan operation, respectively. There was no difference in freedom from thrombus, based on type of operation (atriopulmonary vs. lateral tunnel) or presence of fenestration. Thrombus was detected in the systemic venous atrium in 26 (48%), in the pulmonary venous atrium in 22 (44%), in both atria in 1 (2%), in the hypoplastic left ventricular cavity in 2 (8%), and in the ligated pulmonary artery stump in 1 (2%). Conclusions. Thrombus formation occurs with equal frequency in all types of modifications and is seen in the pulmonary, as well as the systemic venous atria. Our study suggests that thrombus formation is inherent to the physiology after Fontan operation and is not related to the type of modification performed. (Ann Thorac Surg 2001;71:1990 4) 2001 by The Society of Thoracic Surgeons Early outcome after Fontan operation for univentricular heart continues to improve [1 4]. While many of the modifications to Fontan s original operation have resulted in shorter duration of hospitalization and improved early survival, it is yet unclear if any of these will result in a reduction in the incidence of late complications such as arrhythmia [5], protein-losing enteropathy [6] and thromboembolic disease [7 10]. Although it s etiology is not completely understood, thromboembolism is a significant contributor to late morbidity and mortality after Fontan operation. In part, the high rate of thrombus formation may be related to the presence of venous stasis and low cardiac output, factors that are inherent to the physiology of all patients after Fontan operation, regardless of the type of cavopulmonary connection. Recent data suggest an imbalance in coagulation proteins after Fontan operation that may predispose to the development of thrombus [11 13]. In order to prevent thrombus formation, some have advocated anticoagulation via the administration of warfarin to all patients after Fontan operation [10, 14]. This recommendation is controversial since the efficacy of such a strategy in preventing thromboembolic disease has not yet been demonstrated in this population, and the morbidity related to the risk of bleeding in young children is not inconsequential [15]. We undertook the present study to evaluate the frequency of thrombus formation in our patients after Fontan operation. Specifically, we sought to identify the Accepted for publication Jan 5, Address reprint requests to Dr Rychik, Echocardiography Laboratory, The Children s Hospital of Philadelphia, 34th St and Civic Center Blvd, Philadelphia, PA 19104; rychik@ .chop.edu. prevalence of thrombus formation based on the type of Fontan operation modification performed, as well as the most common anatomical location for the development of thrombus. Material and Methods A retrospective review of the Echocardiography Database at The Children s Hospital of Philadelphia was performed. All patients who had a Fontan operation and an echocardiographic study performed between January 1987 and August 1999 were identified. The presence or absence of a thrombus on transthoracic (surface) echocardiography was noted. Studies were performed using Hewlett-Packard Sonos 1000, 1500, 2000, or 2500 phased array systems. The type of Fontan modification such as atriopulmonary, lateral tunnel (total cavopulmonary connection), right ventricular inclusion, or extracardiac conduit was noted. Patients were also analyzed based on the presence or absence of a fenestration. If thrombus was present on echocardiography, the records were reviewed and data collected concerning the location of the thrombus, as well as the presence or absence of sinus node dysfunction, protein-losing enteropathy, and history of cerebrovascular accident. Patients with a fenestration were routinely placed on low dose aspirin for antiplatelet therapy after surgery. It has been our institutional approach not to routinely anticoagulate patients (ie, warfarin) immediately after Fontan operation. Some patients at high risk, such as those with atrial flutter or prior history of thrombus, were anticoagulated late after Fontan operation at the discretion of the primary cardiologist. Results are expressed as median (range) unless other by The Society of Thoracic Surgeons /01/$20.00 Published by Elsevier Science Inc PII S (01)

2 Ann Thorac Surg COON ET AL 2001;71: THROMBUS AND FONTAN OPERATIO 1991 Fig 1. Kaplan-Meier curve for % freedom from thrombus after Fontan operation. Broken lines designate the 95% confidence intervals. Numbers indicate patients at risk. wise indicated. Freedom from thrombus curves were generated using Kaplan-Meyer analysis with Log-Rank test to look for differences between groups. Categorical variables were evaluated by 2 analysis. Statistical significance was defined as p less than Results During the study interval, 592 patients underwent echocardiographic imaging after the Fontan operation. Fontan operation was performed at a median age of 1.9 years (0.8 to 35.1 years). Median time interval between Fontan operation and most recent available echocardiogram was 22 months (1 day 20 years). Thrombus was identified in a total of 52 patients; hence, the overall prevalence of thrombus formation in our population was 8.8%. Actuarial freedom from thrombus was 92% at 6 months, 92% at 1 year, 90% at 3 years, 84% at 8 years, and 82% at 10 years after Fontan operation (Fig 1). Thrombus was detected on echocardiography within the first year after Fontan operation in 34 of the 52 patients (65%). Median time interval between Fontan operation and detection of thrombus was 2.3 months (1 day 163 months). Arrhythmia was common at the time of detection of thrombus (Table 1). Protein-losing enteropathy was present in 9 patients (17%) with thrombus. Patients with protein-losing enteropathy had thrombus Table 1. ECG and Thrombus Rhythm No. of Patients Normal sinus 31 (60%) Atrial flutter/fibrillation 8 (15%) Junctional rhythm 3 (6%) No ECG 10 (19%) ECG electrocardiogram. detected at a later date after the Fontan operation than those without protein-losing enteropathy (73 8 months, vs months, p 0.01). A cerebrovascular accident was documented around the time of detection of thrombus on echocardiography in 8 of the 52 patients (15%). Of these 8, 4 were in atrial flutter/fibrillation and 3 had protein-losing enteropathy. Of the 52 patients, 24 (46%) were on low dose aspirin and 6 (12%) were on warfarin at the time of detection of thrombus. One patient developed thrombus while on heparin therapy for protein losing enteropathy. Of the total 592 patients who had the Fontan operation, 480 (82%) had a lateral tunnel type, 73 (12%) had atriopulmonary type, 32 (5%) had extracardiac conduit type, and 7 (1%) had right ventricular inclusion in the Fontan pathway. Thrombus was detected in 36 patients with lateral tunnel, in 12 with atriopulmonary connection, in 1 with extracardiac conduit and in 1 with right ventricular inclusion. The actuarial freedom from thrombus over time for the atriopulmonary and lateral tunnel types of Fontan modifications is shown in Figure 2. Patients with lateral tunnel type Fontan were only slightly more likely to be free of thrombus at 1, 3, and 8 years after surgery (92%, 90%, 83%, respectively) than were patients with atriopulmonary connection (91%, 89%, 81%, respectively). There was no statistical difference between these groups. Of the 592 patients, 212 (36%) had fenestration and 380 (64%) had no fenestration placed at Fontan surgery. Figure 3 compares the actuarial freedom from thrombus for patients with, and without, fenestration. Freedom from thrombus at 1, 3, and 8 years after surgery for patients with fenestration was 91%, 88%, and 82%, respectively, and for patients without fenestration it was 93%, 91%, and 85%, respectively. There was no statistical difference between these groups. Thrombus was located in the systemic venous pathway

3 1992 COON ET AL Ann Thorac Surg THROMBUS AND FONTAN OPERATIO 2001;71: Fig 2. Percent freedom from thrombus varied by type of Fontan modification, atriopulmonary type Fontan (broken line) versus lateral tunnel type Fontan (solid line). There was no statistical difference between the groups. Numbers indicate patients at risk (lateral tunnel/atriopulmonary). in 26 patients (50%), pulmonary venous pathway in 22 (42%), hypoplastic left ventricular cavity in 2 (4%), and the tied off main pulmonary artery stump in 1 (2%). One patient had thrombus in both the systemic and pulmonary venous pathways. Of the 22 patients with thrombus in the pulmonary venous atrium, it was adherent to the wall separating the systemic and pulmonary venous pathways in 18 (78%). Of the 38 patients with lateral tunnel type Fontan and thrombus, it was located in the systemic venous pathway in 17 (45%), pulmonary venous pathway in 18 (47%), hypoplastic left ventricular cavity in 2 (5%) and ligated main pulmonary artery in 1 (3%). Of the 12 patients with atriopulmonary type Fontan and thrombus, it was located in the systemic venous pathway in 7 (58%) and in the pulmonary venous pathway in 5 (42%). There was no statistical difference in the location of thrombus based on type of Fontan modification. Comment Thrombus formation can be a significant cause for morbidity and mortality after Fontan operation [7 10]. Intracardiac thrombus formation can lead to chronic pulmonary embolic disease if formed on the right side, or stroke, if on the left side of the heart. Right-sided embo- Fig 3. Percent freedom from thrombus for patients with fenestrated Fontan (broken line) versus nonfenestrated Fontan (solid line). There was no statistical difference between the groups. Numbers indicate patients at risk (fenestrated/nonfenestrated).

4 Ann Thorac Surg COON ET AL 2001;71: THROMBUS AND FONTAN OPERATIO 1993 lism may result in ventilation/perfusion mismatch or elevation of pulmonary vascular resistance, both of which may seriously hamper cavo-pulmonary physiology. Left-sided embolism can lead to neurologic deficit or death [16]. Thrombus was detected on echocardiography in approximately 9% of our patients after Fontan operation. Of note, thrombus occurred in our patients on aspirin therapy as well as in 6 patients on warfarin therapy. Although most commonly detected within the first year after surgery, there was a continuing risk of thrombus formation throughout the study period. Our curve for thrombus formation mimics very closely the previously published curves for development of arrhythmia [5] and protein-losing enteropathy [6] after Fontan operation. This suggests a life-long risk of thrombus formation in these patients and the possibility of a codependent relationship between these late complications. Although we initially suspected that patients with atriopulmonary type Fontan would likely have a higher frequency of thrombi relative to other types of surgical modifications, this was not the case. There was no difference in freedom from thrombus up to 10 years after Fontan operation between patients with atriopulmonary and lateral tunnel type modification. In addition, thrombus formation was not overwhelmingly isolated to the systemic venous pathway, as might be expected if low velocity flow were the primary factor influencing the development of thrombus. Instead, thrombus was equally common in the pulmonary venous pathway as in the systemic venous pathway, with no difference in location, based on type of Fontan modification. Our study findings support the hypothesis that factors other than venous stasis and low velocity flow contribute to thrombus formation after Fontan operation. Deficiencies in protein S, protein C, factor VII, and antithrombin III have been described, suggesting a generalized hypercoagulable state [11 13]. These coagulation factor abnormalities may be related to poor hepatic function. Abnormalities of liver enzymes as well as hepatic fibrosis have been described after Fontan operation [17]. Alternatively, deficiencies in coagulation factors may be related to enteric losses. Elevated fecal 1 -antitrypsin clearance has been demonstrated in subjects without overt signs or symptoms of protein-losing enteropathy [18]. Subclinical protein loss within the gut may result in an imbalance of coagulation factors and a predilection towards thrombus formation. A lowered threshold towards thrombosis due to a chemical coagulopathy will result in clot formation equally in the systemic and pulmonary venous atria. The placement of foreign/prosthetic material within the heart can then act as a nidus for clot formation. This may explain why the vast majority of clots formed on the pulmonary venous side in our study were adherent to the baffle/patch material separating the systemic from the pulmonary venous pathways. If in fact there is a fundamental predisposition to clot formation in these patients, then the use of the extracardiac conduit for the systemic venous pathway, a popular modification at present, may theoretically result in an increase in the frequency of right-sided thrombi. To date, the follow-up of patients with the extracardiac conduit type Fontan is limited, although preliminary data would suggest no difference in the frequency of thrombi from other modifications [19]. Of note, only 1 patient out of 32 in our study with extracardiac conduit type Fontan modification had thrombus identified, however, the follow-up in this group of patients has been too short to draw any conclusions. Careful follow-up of all patients with extracardiac Fontan modifications to assess for thrombus formation is indicated. Our study is limited due to its retrospective nature. Echocardiography was performed at various time intervals for different patients. Our prevalence for thrombus formation may underestimate the true frequency. There is no specific schedule for timing of echocardiographic evaluation after Fontan operation at our institution. Echocardiograms are performed at the discretion of the primary cardiologist, hence, it is possible that patients may exist with undetected thrombi during periods in which imaging had not taken place. This may also explain why thrombus detection was common early after surgery, since echocardiography is likely performed more frequently within the first year after surgery than thereafter. In addition, transesophageal echocardiography is superior to surface transthoracic study in identification of small thrombi [20 21]. All of our patients had either complete identification, or at least a suspicion of thrombus, on transthoracic echocardiography that then prompted a transesophageal study in some. This suggests that most of the thrombi detected in our review were of a relatively large nature. In addition, complete imaging of the systemic venous pathway (in particular extracardiac conduits) can be difficult from the surface. Prospective evaluation using transesophageal echocardiography in all patients after Fontan operation may in fact yield a higher rate of detection of small thrombi, or even large thrombi, within the systemic venous baffle [10]. In conclusion, thrombus formation occurs with equal frequency after atriopulmonary or lateral tunnel type Fontan modification, as well as in patients with or without fenestration. Thrombi are as commonly seen on the pulmonary venous side as they are on the systemic venous side, and are usually adherent to the baffle/patch separating the venous circulations. Our study lends support to the suggestion that thrombus formation after Fontan operation may be inherent to the physiology of cavopulmonary flow and not specifically related to the type of Fontan connection created. How to protect children after the Fontan operation and prevent the occurrence of thrombosis remains unclear, particularly in light of our findings of thrombus development even in patients on aspirin or warfarin. Prospective trials detailing the role of transesophageal echocardiography in the detection of thrombus [10] and the potential benefits of routine anticoagulation of all children after Fontan operation [22] are warranted.

5 1994 COON ET AL Ann Thorac Surg THROMBUS AND FONTAN OPERATIO 2001;71: References 1. Jacobs ML, Norwood WI. Fontan operation: influence of modifications on morbidity and mortality. Ann Thorac Surg 1994;58: Cetta F, Feldt RH, O Leary PW, et al. Improved early morbidity and mortality after Fontan operation: the Mayo experience. J Am Coll Cardiol 1996;28: Hsu DT, Quaegebur JM, Ing FF, et al. Outcome after single stage, non-fenestrated Fontan procedure. Circulation 1997; 96(Suppl II): Thompson LD, Petrossian E, McElhinney DB, et al. Is it necessary to routinely fenestrate an extracardiac Fontan? J Am Coll Cardiol 1999;34: Cohen MI, Wernovsky G, Vetter VL, et al. Sinus node function after a systematically staged Fontan procedure. Circulation 1998;(Suppl II): Feldt RH, Driscoll DJ, Offord KP, et al. Protein-losing enteropathy after the Fontan operation. J Thorac Cardiovasc Surg 1996;112: Dobell AR, Trusler GA, Smallhorn JF, et al. Atrial thrombi after the Fontan operation. Ann Thorac Surg 1986;42: Rosenthal DN, Friedman AH, Kleinman CS, et al. Thromboembolic complications after Fontan operations. Circulation 1995;92(Suppl 9): Wilson DG, Wisheart JD, Stuart AG. Systemic thromboembolism leading to myocardial infarction and stroke after fenestrated total cavopulmonary connection. Br Heart J 1995; 73: Balling G, Vogt M, Kaemmerer H, et al. Intracardiac thrombus formation after the Fontan operation. J Thorac Cardiovasc Surg 2000;119: Cromme-Dijkhuis AH, Henkens CMA, Bijleveld CMA, et al. Coagulation factor abnormalities as possible thrombotic risk factors after Fontan operation. Lancet 1990;336: Jahingiri M, Shore D, Kakkar V, et al. Coagulation factor abnormalities after the Fontan procedure and it s modifications. J Thorac Cardiovasc Surg 1997;113: Van Nieuwenhuizen RC, Peters M, Lubbers LJ, et al. Abnormalities in liver function and coagulation profile following the Fontan procedure. Heart 1999;82: Monagle P, Cochrane A, McCrindle B, et al. Thromboembolic complications after Fontan procedures: the role of prophylactic anticoagulation. J Thorac Cardiovasc Surg 1998; 115: Streif W, Andrew M, Marzinotto V, et al. Analysis of warfarin therapy in pediatric patients: a prospective cohort study of 319 patients. Blood 1999;94: Du Plessis AJ, Chang AC, Wessel DL, et al. Cerebrovascular accidents following the Fontan operation. Pediatr Neurol 1995;12: Lemmer JH, Coran A, Behrendt DM, et al. Liver fibrosis (cardiac cirrhosis) five years after modified Fontan operation for tricuspid atresia. J Thorac Cardiovasc Surg 1983;86: Thorne SA, Hooper J, Kemp M, Somerville J. Gastrointestinal protein loss in late survivors of Fontan surgery and other congenital heart disease. Eu Heart J 1998;19: Shirai LK, Rosenthal DN, Reitz BA, et al. Arrhythmias and thromboembolic complication after the extracardiac Fontan operation. J Thorac Cardiovasc Surg 1998;115: Fyfe DA, Kline CH, Sade RM, Gillette PC. Transesophageal echocardiography detects thrombus formation not identified by transthoracic echocardiography after the Fontan operation. J Am Coll Cardiol 1991;18: Stumper O, Sutherland GR, Geuskens R, et al. Transesophageal echocardiography in evaluation and management after a Fontan procedure. J Am Coll Cardiol 1991;17: Jonas RA. Intracardiac thrombus after Fontan procedure. J Thorac Cardiovasc Surg 1995;110: