| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2003;23:719-727
© 2003 Elsevier Science NL
a Department of Pediatric Cardiac Surgery, Bambino Gesù Hospital I.R.C.C.S., Piazza S. Onofrio, 4, 00165 Rome, Italy
b Department of Pediatric Cardiology, Bambino Gesù Hospital I.R.C.C.S., Piazza S. Onofrio, 4, 00165 Rome, Italy
Received 10 September 2002; received in revised form 26 January 2003; accepted 3 February 2003.
* Corresponding author. Tel.: +39-06-6859-2333/5816-904; fax: +39-06-6859-2257
e-mail: carotti{at}opbg.net
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Methods3. Results4. Discussion5. Conclusion6. AddendumAppendix A. Conference...References |
|---|
Key Words: Pulmonary atresia with ventricular septal defect • Major aortopulmonary collateral arteries • Unifocalization
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. Conclusion6. AddendumAppendix A. Conference...References |
|---|
Developing experience has allowed us to further delineate the usefulness of pre- and intra-operative parameters adopted in our approach to surgical treatment of PA, VSD, MAPCAs.
|
2. Methods |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. Conclusion6. AddendumAppendix A. Conference...References |
|---|
Patient selection was mainly based on the preoperative calculation of TNPAI, assuming that a cutoff value of equal to or greater than 150 mm2/m2 was indicative of an overall compliance of the pulmonary vascular tree sufficient to accommodate one-stage unifocalization and repair [9]. In the majority of these cases, the pulmonary blood supply was provided by well developed MAPCAs in the presence of hypoplastic pulmonary arteries with severe branching abnormalities or, even, in their absence. Twenty-five such patients (Fig. 1 : Group 1, 1st stage) underwent unifocalization, irrespective of preoperative PAI and Spa:Sca.
|
2.1. Surgical treatment
2.1.1. Midline one-stage complete unifocalization ± repair
Our approach for one-stage complete unifocalization and intraoperative pulmonary flow study (PFS) was previously described [9]. Briefly, under standard general anaesthesia (continuous intravenous infusion of Fentanyl, 20 µg/kg per min; Midazolam, 200 µg/kg per min; Pancuronium Bromide, 50 µg/kg per min) all collaterals were identified through transverse sinus and posterior mediastinal dissection, controlled and snared just before the cardiopulmonary bypass was begun. This was carried out following 
-stat strategy, using buffered either crystalloid or hematic priming solution in order to maintain the hematocrit of perfusate at 25% throughout the entire procedure. The pump flow was kept at a rate of 120–150 ml/kg per min at a temperature of 30°C during the whole unifocalization time that was performed on the beating decompressed heart. Further cooling to 25°C at a pump flow rate of 100–120 ml/kg per min was achieved at the time of the PFS and maintained throughout the period of cardioplegic cardiac arrest required for either VSD closure (Fig. 1b1) or ventricular conduit anastomosis (Fig. 1b2). The indication of PFS to close the VSD was based on a mean pulmonary arterial pressure value of less than or equal to 30 mmHg when the total neopulmonary arterial district was perfused with an amount of blood flow equivalent to a cardiac index of 2.5 l/min per m2 (Fig. 2a)
.
|
-stat strategy and a short period of cardioplegic cardiac arrest to avoid air embolism. In our early experience, RVOTR was achieved by interposition of a small sized valved homograft in patients with very small (<2.5 mm in size) pulmonary arteries (n=2) and by either prosthetic (polytetrafluoroethylene or dacron) conduit (n=3) or a short transannular right ventricular outflow patch (n=3) in patients with small (>2.5 mm in size, but PAI <100 mm2/m2) pulmonary arteries (n=5) [9]. Recently, however, we moved to extensive use of valved heterografts (Shelhigh Inc., Millburn, NJ, USA: n=4) because of both shortage of small sized homografts and general preference for valved devices.
2.2. Microdeletion of chromosome 22q11.2
All patients were examined at admission from both a phenotypic and a genotypic viewpoint and were screened for the microdeletion of chromosome 22q11.2 (del22q) by fluorescent in situ hybridization.
2.3. Follow-up
Perioperative and follow-up data were collected retrospectively by review of hospital records and reports of referring cardiologists. Mortality was always referred to the first procedure performed and defined as early (within 30 days), hospital (within hospital stay) or late (any time after discharge). Most patients received routine clinical investigations, echocardiograms and, whenever possible, elective cardiac catheterization and 99mTc lung perfusion scan. Two patients were lost to follow-up due to return to foreign country.
2.4. Data analysis
Data are expressed as mean±standard deviation, range or 95% confidence intervals (CI). Statistical analysis was carried out to assess: (a) the influence of del22q on anatomic phenotype and overall mortality; and (b) the postoperative evolution of haemodynamic parameters compared to those detected at PFS. Comparison of mean values of continuous variables was performed by either independent t-test (normal distribution) or Mann–Whitney U-test (non-parametric analysis). Dichotomous variables were analyzed by Fisher's exact test. Survival analysis was performed by the Kaplan–Meier method.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. Conclusion6. AddendumAppendix A. Conference...References |
|---|
Patients of group 2, treated by RVOTR, experienced an uncomplicated postoperative course and showed a significant growth of central pulmonary arteries, within 18±6 months (range: 6–30 months) of the palliation, the mean PAI increasing from 44±28 mm2/m2 (range: 16–89) to 180±65 mm2/m2 (range: 125–337) (P<0.0001) and the acquired TNPAI following the palliation always exceeding 150 mm2/m2 (Fig. 1a). Nine out of 12 patients of group 2 eventually underwent unifocalization and repair (Fig. 1: 2nd stage), while three are still awaiting the second stage.
Overall, one-stage complete unifocalization was carried out in 34 patients (Fig. 1b) and was associated to VSD closure (Fig. 1b1) in 29 (85%; 95% CI: 73–97%). Out of 34 patients, 27 (25 during primary and two during secondary procedures) received intraoperative PFS. Based on this test, the VSD was closed in 23 out of 27 cases.
Two patients of group 1 required reopening of the VSD (Fig. 1c) due to hypersystemic right ventricular pressure at the end of cardiopulmonary bypass, reducing the overall rate of definitive single stage VSD closure to 79% (95% CI: 65–93%). In both cases, the intraoperative PFS proved misleading: more precisely, it was falsely reassuring in one case (with a border-line mean pulmonary arterial pressure of 29 mmHg) and probably misinterpreted in the other case due to pressure transducer dysfunction. These two failures should be included in our learning curve with PFS and represent the only cases of unsuccessful VSD management. As a matter of fact, all remaining 21 patients undergoing VSD closure based on PFS showed postoperative mean pulmonary arterial pressure values (24±3, range 15–28 mmHg) closely approximating the mean pulmonary arterial pressure levels measured during PFS (25±4, range 13–29 mmHg) [difference of means=0.94 (95% CI: -1.53–3.41): P=ns], with a mean right ventricular/left ventricular pressure ratio (pRV/pLV) of 0.5±0.1 (range: 0.3–0.75). Furthermore, in the early postoperative period, all 4 patients whose VSD had been left open based on PFS showed a pulmonary-to-systemic blood flow ratio (Qp:Qs), measured by intracardiac catheters left at operation, persistently less than or equal to 1.
Postoperative major complications included septicemia in three cases, two of which due to fungal agents, and airway bleeding in two cases.
Reoperations in the early postoperative period included extracorporeal life support with a 46-h run of Extracorporeal Membrane Oxygenation (ECMO) in one case and ligation of a residual MAPCA in another one. Both cases occurred in group 1 and were successfully managed. Another patient underwent, twice, replacement of an infected homograft.
There were five hospital deaths after unifocalization procedures (15%; 95% CI: 3–27%). Two patients died early postoperatively: one was patient of group 1, aged 12 years, who required intraoperative VSD reopening and showed diffuse pulmonary vascular obstructive disease at post-mortem examination; the other patient, aged 22 days, also of group 1, had the VSD left open and experienced sudden death in the immediate postoperative period without any apparent cause at autopsy. Other two deaths occurred for infection, after a prolonged (of more than 30 days) hospital stay. The fifth death was due to a major airway bleeding occurred 40 days after total unifocalization with VSD left open.
Infections were the main cause of both early reoperations (n=2 infected homograft replacement in the same patient) and hospital death (n=2).
The surviving patients stayed in hospital a mean of 18±19 days (range: 6–90 days) and generally experienced a smooth postoperative course. Temporary dysfunction of phrenic (n=2) or vagus nerves (n=2) due to surgical dissection recovered always spontaneously within a few days after surgery, whereas postoperative broncospasm (n=6) persisted in all cases for more than hospital stay [11].
3.2. Follow-up
Follow-up ranged from 1 to 85 months with a mean interval of 43±28 months. Two patients were reoperated on respectively 2 and 28 months postoperatively for right ventricle to pulmonary artery (RV-PA) homograft replacement following sterilized acute endocarditis. A third patient required RV-PA homograft replacement 5 years postoperatively due to conduit degeneration. All these reoperations were successful. The patient with congenital heart block required endocardial pacemaker implantation 6 months postoperatively for dysfunction of previously implanted epicardial leads. Two patients, who had failed intraoperative PFS at the time of total unifocalization, underwent delayed VSD closure respectively 10 and 30 months postoperatively (Fig. 1d). In both cases a RV-PA conduit replacement was associated. The former had a smooth postoperative course with a pRV/pLV=0.5 and a quick discharge home; conversely, the latter experienced a sudden airway bleeding during the weaning off the ventilator and died in spite of an early postoperative pRV/pLV=0.4. No further deaths occurred. Survival up to a 7-year follow-up was 81% (Fig. 3)
.
|
Among 26 completely repaired patients, whose complete follow-up data are available, 22 are asymptomatic and well, whereas four show mild symptoms on effort. Mean peak right ventricular pressure evaluated on two-dimensional echocardiography with Doppler is 45±13 (range: 30–70) mmHg. Sixteen patients underwent routine control cardiac catheterization 14±6 (range: 6–19) months postoperatively, associated to transcatheter procedures in 6 patients: either dilatation (n=4) with or without stenting (n=1) of pulmonary arteries or coil closure of residual collaterals (n=3). Cardiac catheterization (n=16) showed a mean right ventricular pressure value of 45±15 (range: 23–78) mmHg and stable or decreasing mean pulmonary arterial pressure levels (20±3, range 14–25 mmHg versus 24±4, range 15–28 mmHg) and pRV/pLV ratios (0.48±0.2, range 0.22–0.8 versus 0.52±0.1, range 0.35–0.75) compared to early postoperative measurements [mean pulmonary arterial pressure, difference of means=3.6 (95% CI: 1.04–6.16): P=0.007; pRV/pLV, difference of means=0.04 (95% CI: -0.08–0.16): P=ns].
Lastly, six patients also underwent 99mTc lung perfusion scan, which failed to demonstrate major differences of pulmonary blood flow distribution in any of them.
3.3. Microdeletion of chromosome 22q11.2
Microdeletion of chromosome 22q11.2 was present in 15 patients (40%; 95% CI: 24–56%). Its occurrence did not relate to any peculiar anatomic phenotype, neither regarding pulmonary vascular supply [mean number of MAPCAS: 3.7±1.3 (15 pts) versus 3±1.4 (22 pts); P=ns; mean Spa:Sca: 1.5±1.5 (15 pts) versus 2±1.9 (22 pts); P=ns), nor prevalence of right aortic arch [53% (eight pts) versus 36% (eight pts); P=ns] and infundibular atresia [53% (eight pts) versus 50% (11 pts); P=ns].
Early postoperative death occurred in two patients with del22q. Postoperative infections, including two fungal pneumonias, one fungal endocarditis, and 1 Pseudomonas pneumonia occurred in four patients with del22q. Fungal infections were always related to a decreased CD4 lymphocyte count on peripheral blood samples and were responsible for two out of three hospital deaths. The third case of hospital mortality occurred in a del22q patient with VSD left open due to airway bleeding. Late mortality finally occurred in one patient with del22q who experienced sudden airway bleeding early after successful second-stage closure of VSD. In summary, all six deaths occurred in del22q patients either related or unrelated to their immunodeficiency. This association is statistically significant [6/15 versus 0/22, difference -0.4 (95% CI: -0.65–-0.15): P=0.002].
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. Conclusion6. AddendumAppendix A. Conference...References |
|---|
Although deeply inspired by the contributions of the group of the University of California of San Francisco, since 1994 we have adopted a more individualized approach (so called ‘integrated approach’) with patient selection based on preoperative evaluation of pulmonary blood sources [9]. As reported in 1998, we arbitrarily divide patients in those with a TNPAI equal to or greater than 150 mm2/m2, eligible for primary unifocalization and repair, and those with a TNPAI less than 150 mm2/m2, candidate to preliminary palliative RVOTR to promote the growth of the pulmonary arteries and allow for secondary unifocalization and repair [9]. We also heavily rely on the mean pulmonary arterial pressure changes during the intraoperative PFS, to decide whether to close the VSD during unifocalization procedures. Admittedly, our early series, although very satisfactory, was limited in terms of number of cases and length of follow-up and was somewhat biased by the old average age of the patients [9]. Four years later, our experience has grown to 37 cases, with a net decrease of the average age of the patients and an even more extensive application of the mentioned pre- and intraoperative selection criteria. In this study, our management protocols were reviewed in order to assess the usefulness of perioperative parameters.
4.1. Management protocol: one-stage approach (Fig. 2a)
In our early experience a TNPAI equal to or greater than 150 mm2/m2 was nearly always related to repairability of the disease (14/15 cases=93%; 95% CI: 80–100%) [9]. Conversely, current results show that, although one-stage unifocalization alone was successfully accomplished in all such patients irrespective of preoperative anatomy of native pulmonary arteries and/or of Spa:Sca, associated VSD closure was possible in 79% (95% CI: 65–93%) of cases only (VSD left open in the first instance: n=5; VSD reopened: n=2), leading to a decrease (although not significant) in the number of one-stage unifocalizations and repairs compared to our earlier experience. Under such circumstances, great importance is concentrated on intraoperative PFS in the decision-making process for VSD closure during one-stage procedures. Accuracy of this test in predicting the post-repair mean pulmonary arterial pressure in patients who underwent VSD closure was 91% (95% CI: 79–100%), whereas Qp:Qs ratio never exceeded 1:1 in any of the patients with VSD left open, accounting for an overall accuracy of PFS up to 93% (95% CI: 83–100%).
4.2. Management protocol: two-stage approach (Fig. 2b)
Current results also corroborate our preliminary experience [9] with the cohort of patients with a TNPAI less than 150 mm2/m2 undergoing palliative RVOTR. Classical palliative procedures providing forward flow to native hypoplastic pulmonary arteries to promote their growth have been proposed by several groups [14–17]. While the recent literature advocates, from one side, very stringent criteria for such a palliative treatment [18], and on the opposite side, the extensive application of first-stage RVOTR for all patients [15,19], our results suggest that patients with hypoplastic though dominant pulmonary arteries are probably those who mostly benefit by RVOTR. All such patients of our series experienced a smooth postoperative course (with no significant degree of pulmonary overcirculation), adequate pulmonary arterial growth within 18±6 months of the procedure, and effective or potential repairability in 100% (12/12) of cases.
We believe that any different approach, such as primary unifocalization, to this cohort of patients would account for unsuitability for associated VSD closure, possible preferential runoff into unifocalized collaterals potentially responsible for early occurrence of significant overcirculation, and related insufficient growth of native pulmonary arteries.
4.3. Management protocol: border-line situations (Fig. 2c)
While results of our experience with patients with a TNPAI less than 150 mm2/m2 suggest clear indications concerning hypoplastic though dominant pulmonary arteries, no informations can be desumed regarding patients with a PAI less than 100 mm2/m2 and a Spa:Sca less than or equal to 1 (hypoplastic not dominant pulmonary arteries) or a PAI between 100 and 150 mm2/m2. In the former, rare situation the surgical approach should be probabably individualized based on patient's preoperative clinical status: preoperative cyanosis should suggest RVOTR, whereas preoperative pulmonary overcirculation should better account for total unifocalization. On the other hand, patients with a PAI between 100 and 150 mm2/m2 should be considered provided of normal sized central pulmonary arteries, although in the presence of MAPCAs their peripheral distribution is usually defective. For this reason, they should undergo primary unifocalization and PFS, although preoperative criteria for such approach are not fulfilled, according to our experience. The same policy should be followed also in cases with a TNPAI less than 150 mm2/m2 and absent pulmonary arteries.
4.4. Optimal age at operation
The advantages of early surgical intervention have already been outlined [7,8]. In particular, early intervention obviates to problems, e.g. progressive stenoses of MAPCAs and segmental pulmonary vasculopathy [20,21].
Such advantages, however, have to be weighed against the perioperative morbidity due to time-related activation of inflammatory response to cardiopulmonary bypass [22] particularly in children [23] together with the conduit-related morbidity peculiarly enhanced early in infancy [24].
Our first report included mainly patients with a somewhat old average age in some way expressing a spontaneous selection of patients with ‘protected’ pulmonary circulation [9]. On the other hand, the significant decrease of mean age at operation in more recent part of our experience, together with substantially confirmed current results compared to previous ones allow us to promote a policy of early surgical approach according to both patient's symptoms and physiology of the disease.
Based on all mentioned considerations, our current protocol prescribes surgery at any age, as far as severe cyanosis is concerned; surgery within 1 year of age, as long as pulmonary overcirculation prevails.
4.5. Microdeletion of chromosome 22q11.2
Microdeletion of chromosome 22q11.2 appears associated with poor overall outcome after surgical treatment of PA, VSD, MAPCAs. Such a peculiarity, not necessarily due to special anatomic features related to the syndrome itself [25], should justify the use of a protocol for preoperative assessment of immunological status of patients with del22q undergoing surgery for PA, VSD, MAPCAs and the administration of a perioperative anti-fungal prophylaxis, when required by a depressed immunological condition.
4.6. Limitations of the study
The lack of a control group and the somewhat still limited number of patients do not allow meaningful statistical analysis. A longer follow-up interval is also required to draw definitive conclusions on long term results of the reported management protocol.
|
5. Conclusion |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. Conclusion6. AddendumAppendix A. Conference...References |
|---|
|
6. Addendum |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. Conclusion6. AddendumAppendix A. Conference...References |
|---|
|
Footnotes |
|---|
|
Appendix A. Conference discussion |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. Conclusion6. AddendumAppendix A. Conference...References |
|---|
Dr Carotti: We just leave it open.
Dr G. Stellin (Padova, Italy): I would like to know which criteria are you to decide when is the right time to close the VSD. Is it when you achieve a total left-to-right shunting at the ventricular level and normal arterial saturation? Are there other criteria?
Dr Carotti: Usually we wait until the Qp/Qs rises above 1.5/1.
Dr Stellin: And arterial saturation 100%?
Dr Carotti: No, usually we do not take into consideration the arterial saturation.
Dr H. Lindberg (Oslo, Norway): When you did your reperfusion pressure study intraoperatively, did I understand you correctly that you did not ventilate the lungs, you left the lung unventilated while perfusing?
Dr Carotti: Yes, the lungs are unventilated.
Dr Lindberg: You don't think that would influence the pulmonary artery pressure, leaving the lungs unventilated, and that would you have some lower pressure if you ventilated the lungs?
Dr Carotti: This is a setup which we borrowed ‘en block’ from the experience of the University of California in San Francisco. The UCSF group experimented the flow study in neonatal lambs and described a setup with unventilated lungs, at 25° of body temperature and 25% haematocrit. Such protocol was entirely adopted in our experience, so we always maintain lungs unventilated with no positive pressure, by disconnecting the patient from the ventilator during the flow study. Actually, we also perform airway suction just before the flow study is carried out, in order to find out if there is any blood leaking inside, which could theoretically be responsible for mean pulmonary arterial pressure misrecording.
Dr A. Urban (Augustin, Germany): Do you use always median sternotomy only or do you use lateral approach as well, or does it depend, and from what criteria?
Dr Carotti: We do not use lateral thoracotomies anymore. Our approach is always through midline sternotomy. We found a great advantage by performing the one-stage unifocalization through midline sternotomy because we can reach all the collaterals through such an approach. By dissecting the transverse sinus of the pericardium we can usually reach directly the descending aorta, control all the collaterals, and take them anteriorly to the oesophagus or to the airway, when required, in order to anastomose them to each other or to the central pulmonary arteries. In our experience we always found enough tissue to perform direct anastomoses by avoiding interposition of any prosthetic material.
Dr J. Monro (Southampton, UK): Were there borderline cases where you perhaps didn't know whether to close the VSD or not, and as I understood it, you left the VSD open in some cases. A trick that I have used is to have a patch with a hole in it with a flap of pericardium on the back, so that you can have a right-to-left shunt, but not a left-to-right shunt. Have you had experience with that?
Dr Carotti: No, we do not have any experience with such one-way flap valved patch. However, in the future we might follow your suggestion.
|
References |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. Discussion5. Conclusion6. AddendumAppendix A. Conference...References |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. Davies, S. Mussa, P. Davies, J. Stickley, T. J. Jones, D. J. Barron, and W. J. Brawn Unifocalization of major aortopulmonary collateral arteries in pulmonary atresia with ventricular septal defect is essential to achieve excellent outcomes irrespective of native pulmonary artery morphology J. Thorac. Cardiovasc. Surg., December 1, 2009; 138(6): 1269 - 1275. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Zhang, S. Wu, J. Liang, G. Zhang, G. Jiang, M. Zhou, and X. Li Unidirectional Monovalve Homologous Aortic Patch for Repair of Ventricular Septal Defect With Pulmonary Hypertension Ann. Thorac. Surg., June 1, 2007; 83(6): 2176 - 2181. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. M. Amark, T. Karamlou, A. O'Carroll, C. MacDonald, R. M. Freedom, S.-J. Yoo, W. G. Williams, G. S. Van Arsdell, C. A. Caldarone, and B. W. McCrindle Independent Factors Associated With Mortality, Reintervention, and Achievement of Complete Repair in Children With Pulmonary Atresia With Ventricular Septal Defect J. Am. Coll. Cardiol., April 4, 2006; 47(7): 1448 - 1456. [Abstract] [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 |
