| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 1999;15:735-741
© 1999 Elsevier Science NL
The Heart Unit, Birmingham Children's Hospital, Steelhouse Lane, Birmingham, UK
Received 23 September 1998; received in revised form 22 February 1999; accepted 11 March 1999.
Corresponding author. Tel.: +44-121-3339437; fax: +44-121-3339441
 |
Abstract |
|---|
 Top Abstract 1. Introduction
2. Materials and methods...3. Results4. Discussion
Appendix A. Conference...References |
|---|
Key Words: Total anomalous pulmonary venous connection • Pulmonary vein stenosis
|
1. Introduction |
|---|
|
TopAbstract1. Introduction
2. Materials and methods...3. Results4. Discussion
Appendix A. Conference...References |
|---|
The diagnosis of TAPVC is classically subdivided into four distinct anatomical categories according to the mode of drainage of the pulmonary veins. In addition, each category of TAPVC can be further classified into obstructed or non-obstructed types. Obstruction to PV drainage produces elevation in pulmonary artery pressure (PAP), and usually presents as a neonatal emergency. Prognosis is related to the degree of stenosis and/or obstruction to pulmonary venous return, both at initial presentation, and following interventions [6,7]. These may be considered as primary (at time of diagnosis) or secondary (following intervention) pulmonary venous stenosis (PVS). The treatment of primary PVS is urgent surgical repair of the obstruction to pulmonary venous return. Policies for management of secondary PVS vary according to the aetiology, being different when there is diffuse as opposed to localised PV obstruction.
We report our experience in the management of TAPVC, paying particular attention to episodes of pulmonary venous re-stenosis or recurrent obstruction.
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction
2. Materials and methods...3. Results4. Discussion
Appendix A. Conference...References |
|---|
2.1. Patient demographics
There were 85 patients (60 male and 25 female) who had their primary operation performed at our institution. Three additional patients (two male and one female) had undergone primary repair at another centre, and were referred to us for management of established secondary PVS. Median weight at time of first operation was 3.25 kg (range 1.8–6 kg).
2.2. Preoperative diagnosis
Diagnosis was confirmed in all 88 cases by detailed echocardiological study. Cardiac catheterisation was not performed in any patient prior to their first operation.
Details of the anatomical arrangements for the group are presented in Table 1. Patients with right atrial isomerism were excluded from the study.
|
2.3. Follow-up
All notes of clinic appointments following discharge, and results of any investigations performed, including electrocardiography (ECG) and echocardiography, were recorded. Patients were seen at 4 weeks, 3 months, 6 months and 12 months following discharge from hospital, but some were also seen on other occasions according to clinical requirement. Subsequently, all children were seen in clinic on a yearly basis.
2.4. Surgical procedure
Primary correction was performed shortly after diagnosis in all cases. Age at time of first operation ranged between 1 and 533 days (median 33 days). All operations were performed by one of two consultant surgeons (B.S. or W.J.B.) employing standardised techniques. The heart was dislocated from the pericardium towards the right, and the pulmonary veins repaired from the left side. The procedures were carried out using cardiopulmonary bypass (CPB) and periods of hypothermic circulatory arrest (HCA) in all 80 patients having their initial operation at our institution. Median perfusion times during the procedure were 46 min CPB, 33 min aortic cross-clamping, and 26 min HCA.
2.5. Postoperative care
Intensive care unit (ICU) records and charts were examined for details of the immediate postoperative course. This included inotrope requirement, time to extubation, and episodes of haemodynamic and respiratory compromise. Inhaled nitric oxide and prostacycline infusions were commonly used in the immediate postoperative period to treat pulmonary hypertension, but were only recorded as significant in instances of prolonged requirement. Details of hospital stay after transfer to the ward were also recorded.
2.6. Echocardiography
All pre- and post-operative transthoracic ultrasound examinations were performed using a combination of cross-sectional imaging, pulse and continuous wave doppler studies, and colour flow mapping. Using these methods, it was possible to identify the morphology and the flow patterns of pulmonary venous return, and to estimate pressure gradients across the anastomosis and estimate right ventricular pressure. Studies were performed using the appropriate transducers on either a Hewlett Packard Sonos 1000 system or a Vingmed CFM 750, or system 5.
Postoperative pulmonary venous stenosis was diagnosed on documentation of pulmonary venous flow patterns which did not reach baseline throughout the cardiac or respiratory cycles, absence of pulmonary venous return from one lung segment, or the documentation of a tricuspid valve regurgitation jet with a velocity of more than 3 m per second (i.e. an estimated right ventricular pressure in excess of 40 mmHg) [8]. Such patients subsequently underwent cardiac catheterisation, angiography and biplane transoesophageal ultrasound studies. The latter focused on defining the morphology of the pulmonary venous anastomosis of the left atrium, and also the course of the individual pulmonary veins and the definition of the extent of pulmonary venous obstruction. Localised obstructions on transoesophageal imaging were deemed to be amenable to balloon dilatation. At an early stage in our experience, long segment stenoses were treated by either percutaneous or intraoperative stent implantation, but this practice was subsequently abandoned.
|
3. Results |
|---|
|
TopAbstract1. Introduction
2. Materials and methods...3. Results4. Discussion
Appendix A. Conference...References |
|---|
|
|
|
Secondary pulmonary venous stenosis occurred in all four anatomical types, and it also occurred in some cases (5/11) that were unobstructed at initial presentation. In addition, we noted two differing mechanisms of secondary PVS: (1) intrinsic, due to endocardial thickening or sclerosis; (2) extrinsic, representing obstruction at the site of pulmonary venous anastamosis.
The median time period to re-intervention or re-operation was shorter in the presence of intrinsic obstruction (14 days) when compared to extrinsic obstruction (62 days).
Three patients (G, H, I) all developed early secondary PVS which was refractory to all attempts to relieve it, both angiographically and surgically. All three patients died within days of emergency surgery. In one (patient G), the ligature on the ascending vertical vein was found to be compromising the left pulmonary veins (LPV), and was removed. The origins of the LPVs were found to be fibrotic and were incised and dilated. The early postoperative course was stable, but respiratory function started to deteriorate. Angiographic balloon dilatation was unsuccessful, and a further emergency operation was performed to re-open the stenoses. In addition, an atrial septal defect (ASD) was left open. Inotropic support was required to discontinue CPB, and worsening pulmonary oedema eventually led to death on ICU.
Patient H made an excellent recovery from the first re-operation, with documented low right ventricular pressures and no residual obstruction. However, he died 6 months later at home from bronchopneumonia. Post mortem examination did not show any venous obstruction, pulmonary hypertensive changes or right ventricular hypertrophy. One patient (K) had an extremely complicated course, with very small (<3 mm) LPVs initially, which continued to re-stenose despite multiple dilatations and extensive reconstruction and patching. He had a protracted ICU stay, complicated by systemic sepsis and pulmonary haemorrhage. Eventually it was impossible to wean the patient from the ventilator, angiography demonstrated unsalvageable LPVs, and a bronchogram showed widespread bronchomalacia, and hypoplasia of the left mainstem bronchus. It was decided that there was nothing more that could be done and support was withdrawn.
3.3. Morbidity
Non-fatal post-operative complications other than PVS were rare. Although pulmonary hypertension in the immediate postoperative period was not uncommon, it usually responded within 24 h to the administration of inhaled nitric oxide or infused prostacycline. Only three of the surviving patients required longer periods of treatment. One patient developed RSV bronchiolitis and required extracorporeal membrane oxygenation (ECMO) for 8 days, but eventually improved and is currently well. One patient (born prematurely at 32 weeks gestation developed necrotising enterocolitis which settled with conservative management. There were two episodes of cardiac tamponade requiring re-exploration, and four patients developed transient episodes of supraventricular tachycardia (SVT), all of which settled with pharmacological therapy.
3.4. Late outcome
Median total hospital stay for all surviving patients was 8 days (range 5–28 days). Of the entire group of 88 patients, 77 are currently alive (88%). A total of 70 of the 85 patients (82%) having their first operation at our unit had no significant post-operative complications or secondary PVS, and are currently alive and well at a median follow-up interval of 64 months (range 6–119 months). Eleven patients died, four of whom had developed secondary PVS following their first operation at our hospital, and one had been referred in from another centre with secondary PVS. A further five patients initially operated upon at our hospital developed secondary PVS and have been successfully treated to date. Two of the three patients referred in from other hospitals with secondary PVS are currently still alive. None are on any cardiac-related medications or therapy. None of the patients have at any stage during follow-up, had evidence of pulmonary venous obstruction or right ventricular hypertension, either clinically or echocardiographically. All patients are in sinus rhythm with normal atrioventricular conduction as demonstrated electrocardiographically. Five patients were found to have persistent partial right bundle branch block.
Of the 12 patients treated for secondary PVS (nine local and three referred in), seven have survived (five local and two referred in) and are regularly followed up in clinic. One patient (L) developed obstruction of all four veins, so the right pulmonary vein (RPV) anastomosis was reconstructed, and the LPVs were reimplanted into the left atrial appendage. A further reconstruction of the RPVs was required 19 months later, whereby a reversed azygos vein graft was used as an interposition graft. Two subsequent balloon dilatations achieved a reduction in gradient from 16 to 8 mm Hg, and resulted in continuing clinical improvement. Angiography 5 months later demonstrated complete occlusion of the graft at the atrial anastomosis, however the patient remains well.
Two patients (M, O) each underwent two stenting procedures during the course of their secondary PVS management. Both cases required further balloon dilatation procedures and ultimately surgery following re-occlusion, but are currently alive and well.
Two patients (N, R) are well with normal haemodynamic and ECG appearances, and with right ventricular pressures are less than one third of systemic in both cases. Patient R was found to have severe endocardial thickening, and underwent resection of the left atrial endocardium, and a partial endarterectomy of all four orifices, with excellent result and no sign of recurrent PVS to date.
Patient Q was referred from another centre for relief of secondary PVS, with supra-systemic right ventricular pressures on arrival. After limited success with balloon dilatation on repeated occasions, three further surgical procedures were performed. At the first re-operation, all four ostia were dilated and an ASD formed, and at the second re-operation the dilatations were repeated. Three years later, the ASD was closed uneventfully, and the child continues to thrive.
|
4. Discussion |
|---|
|
TopAbstract1. Introduction
2. Materials and methods...3. Results4. Discussion
Appendix A. Conference...References |
|---|
There have been a number of reports in the literature describing experience with surgical management of TAPVC, and analysing risk factors for outcome. Until recently, it has been widely accepted that venous obstruction on presentation, and infradiaphragmatic anatomical type were both important independent predictors of adverse outcome [6,7]. In a univariate analysis of risk factors for death and reoperation for PVS in 1996, Bando et al. [10] made the controversial statement that primary obstruction at presentation ceased to be correlated with early mortality after 1991, whereas the two had been correlated in the 25 years prior to that. They also reported that anatomical category was not found to be correlated with outcome at any stage. This refuted all previous published work on risk factors for morbidity and mortality in this condition. Several groups had previously stated that anatomical category (infradiaphragmatic) and obstruction on presentation were both adverse risk factors for 30 day survival [11–13]. Bando's findings also contrasted with those of the recent large series described by Sinzobahamvya et al. [14], who found obstructed drainage preoperatively to be a main incremental risk factor for adverse outcome, along with operation before 1987, and poor preoperative condition. The latter two risk factors, as well as failure to monitor postoperative pulmonary artery pressure were also noted by Serraf et al. [15]. Multiple logistic regression analysis by Bando et al. showed that a small pulmonary confluence associated with diffuse stenotic disease was the only independent risk factor for both early (P<0.02) and late (P<0.01) mortality. Our results concur with those of Bando's group, in the respect that neither site nor obstruction appear to correlate with outcome.
There does not seem to be any single operative method of preventing the development of secondary PVS. In 1983, Hawkins et al. [16] examined surgical techniques for TAPVD repair. They found that the technique used for repair was one of the most important factors affecting mortality. Prior to 1976, the techniques used for the anastomosis of the left atrium to the common PV involved displacement of the heart from its normal anatomical position. Subsequently, a right atriotomy approach was adopted, with less likelihood of kinking the anastomosis, and better results. In addition, cardiac TAPVD was repaired using a pericardial patch to direct flow through the ASD to the left atrium. Another study by the same author has investigated the theoretical advantages of absorbable sutures for the anastomosis to allow ‘growth’ [17]. They found there to be a significantly lower incidence of secondary PVS and death with polydioxanone when compared with polypropylene (P<0.05). Although various aspects have been examined, experience and scrupulous technique still remain the best option for success [18].
Our experience in the management of secondary PVS emphasises the importance of an aggressive approach when treating this major complication. Our results suggest that surgical repair of extrinsic obstruction is more easily achieved than relief of PVS secondary to endovascular sclerosis. The exact role of balloon dilatation in the management of secondary PVS has yet to be resolved, although its beneficial use has previously been reported [19]. Although balloon dilatation may only provide short term relief, it may, on occasion, usefully defer the timing of surgical re-intervention.
In conclusion, we have adapted our management policies according to both published reports, and results attained during this ten-year study period. We favour an approach whereby a small ASD is left electively at the time of any re-operation for secondary PVS. This will facilitate catheter access and balloon dilatation so that this technique can be employed as required. Subsequent episodes of PVS may then treated by repeated balloon dilatation or further surgery. We have abandoned the use of stents in the management of secondary PVS, due to the uniform lack of success with this approach. Finally, the ASD may be closed utilising surgical or trans-catheter techniques, once there is no further evidence of residual or significant pulmonary venous obstruction.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
Appendix A. Conference discussion |
|---|
|
TopAbstract1. Introduction
2. Materials and methods...3. Results4. Discussion
Appendix A. Conference...References |
|---|
Dr Hyde: Through the atrium. The actual surgical technique – again, it depends whether it is an iatrogenic restenosis, or whether it is due to the presence of this, what we call, intrinsic or endocardial thickening. And clearly the procedure will vary according to which type.
Dr Ebels: And do you then patch the stenosis?
Dr Hyde: We have used patches, yes. We have used autologous and bovine pericardium. They're the commonest methods we have used.
Dr Ebels: There has been a report from Paris where they open the pulmonary veins very widely in these situations and let it drain through the pericardium. Have you used that technique?
Dr Hyde: That technique was used in only one patient, so we don't feel we're able to comment on its use. We don't feel that we are able to really give a valid opinion on that.
Dr M. de Leval (London, UK): Are the pulmonary veins normal when you operate on a patient with primary stenosis? Do those veins look normal at the time of the operation?
Dr Hyde: There were some, very few cases, with the intrinsic stenosis where they were very, very narrowed down and thickened; but in the majority of cases they would look normal.
Dr De Leval: So it is an acquired problem as well?
Dr Hyde: Yes.
Dr H. Sairanen (Helsinki, Finland): We have had three cases with a pulmonary venous stenosis after TAPVD. In all cases we have done endocardectomy as we call it. It's endarterectomy as you call it. We have been very successful in all of these cases with no deaths. One of these patients underwent another endocardectomy was performed at that time as well and the patient recovered and is doing well. So in our hands this kind of technique has been very successful.
Dr Hyde: We found it the biggest problem to deal with, the presence of this endocardial sclerosis, and it's impressive that you are having survivors regularly with that.
Dr J. Waldhausen (Hershey, PA, USA): How many of the intrinsic types were you able to fix?
Dr Hyde: I can't tell you the exact number right now.
Dr Waldhausen: Recently, in discussion, Tom Spray from Philadelphia says that you can't fix them and that they do lung transplants on these patients because they're literally not repairable.
Dr Hyde: Well, I think I'd disagree with that. We had successful results in a number of patients. The worst case that I can remember is one patient where the pulmonary veins were sclerosed away almost to nothing, that we weren't even able to show the drainage of the right lung, and a right pneumonectomy was performed at the same time, but the patient did badly and was an early death postoperatively, almost inevitably.
Dr B. Maruszewski (Warsaw, Poland): As far as iatrogenic complications are concerned, I'd like to know a bit more about your original technique. Do you use the patch to close the ASD and enlarge the left atrium? And do you ever use the patch to enlarge the right atrium around the ostia of the right pulmonary veins to avoid the compression when closing direct the right atrium?
Dr Hyde: The original technique for the first operation, as I mentioned, we lift the apex up towards the superior vena cava, approach it from that side. The pericardium is used to patch both the ASD and, if required, around the site of entry of the right pulmonary veins.
|
References |
|---|
|
TopAbstract1. Introduction
2. Materials and methods...3. Results4. Discussion
Appendix A. Conference...References |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
