| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;77:2144-2149
© 2004 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, Providence, Rhode Island, USA
b Department of Anesthesia, Rhode Island Hospital/Hasbro Children's Hospital, Providence, Rhode Island, USA
Accepted for publication October 30, 2003.
* Address reprint requests to Dr Hopkins, Chief Cardiothoracic Surgery, Brown Medical School, 164 Summit Ave, Providence, RI 02906, USA
e-mail: rahopkins{at}lifespan.org
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsCommentReferences |
|---|
METHODS: This retrospective review is based on a single surgeon's experience between July 1, 1988 and December 21, 2002 with 176 consecutive adult (n = 47) and pediatric (n = 129) surgeries, in which ASD was the primary anatomical diagnosis to ascertain current optimal methods and outcomes expected for surgical closure. Patch closure with pericardium was used in all cases. Surgical methods encompassed three phases. The first phase was defined by traditional sternotomy; the second phase involved a series of technical modifications to shorten incisions and reduce surgical trauma; the third phase consisted of standardized less invasive techniques based upon age and gender with "bikini line" incisions for adult females, limited median sternotomy for adult males, and mini-median sternotomy for children. All patients underwent echocardiography to assess ASD closure.
RESULTS: There were no deaths. The most frequent perioperative complications were atrial fibrillation (adult 10%, pediatric 1.2%) and post pericardiotomy syndrome (adult 2%, pediatric 4.7%). All patients had secure and complete closure of ASDs with no residual shunts (trivial or otherwise) documented by echocardiography. No less invasive procedures required conversion.
CONCLUSIONS: Surgical technique evolved from standard sternotomy to limited access incisions using modified cannulation techniques and incision locations determined by age and gender of the patient without deterioration in outcome quality. Both standard and less invasive surgical methods can achieve secure closure of the septum with biological patches, which are incorporated into the tissue structure of the heart and which are free from materials-related failure modes. Patient satisfaction is enhanced by utilizing the least invasive, least traumatic, and most cosmetically appealing techniques for access and cardiopulmonary bypass.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
|
Between July 1, 1996 and December 31, 2002, 67 children and 37 adults (12 males, 25 females) were operated on for primary diagnosis of ASD by the senior author at Rhode Island (adults) and Hasbro Children's Hospitals. In the pediatric series, there were 7 sinus venosus defects, 7 ostium primum (1°) defects and the remainder were secundum (2°) atrial septal defects. Forty-one of these children underwent less invasive surgical closure as described (vida infra). The anatomy of the adults included 2 SVD, 2 1°, and 33 1° ASDs. Twenty-six of the pediatric and 17 of the adult ASD patients had secondary cardiac lesions requiring surgical intervention, such as PDA ligation, anomalous pulmonary venous drainage baffle (SVD), tricuspid valve annuloplasty, mitral cleft suture (all 1°), etc. Two ASDs were reported as common atrium.
Echocardiography
For the Georgetown series, operative notes, discharge summaries, and surgical operative log were reviewed to ascertain patient characteristics and basic surgical outcomes of mortality and echocardiographic evidence for security of ASD closure. Patients during this time period underwent either epicardial echocardiography at the conclusion of the surgery before sternal closure or transthoracic echocardiography during the first week following surgery.
For the Rhode Island Hospital and Hasbro Children's Hospital phase of the review, operative reports, discharge summaries, and the STS National Database (congenital) were used for review. All adult patients and children weighing more than 30 kg had intraoperative TEE routinely performed with an Omniplane 6.2/5-MHz adult probe (Philips Medical Systems, Andover, MA), while children weighing less than 30 kg, but at least 5 kg, had intraoperative TEE utilizing a pediatric biplane 7.5/5.5-MHz probe. Both probes were connected to a Sonos 2000 Ultrasound platform (Hewlett-Packard Co, Andover, MA). Two-dimensional imaging and color flow Doppler examinations of the interatrial septum were performed following repair to confirm closure of the ASD. Following repair, the interatrial septum was examined in both the midesophageal four chamber view (0.20 degree) and the midesophageal bicaval longitudinal plane (90 to 120 degrees), as well as any additional plane as deemed necessary to detect possible residual interatrial shunts. Children less than 5 kg in body weight did not undergo TEE. In these cases, postoperative transthoracic echocardiograms were performed by pediatric cardiologists either in the early hospital phase or after discharge. All intraoperative TEE examinations were videotaped and archived. In those cases where the written reports were not found in the medical records, the original echocardiogram tape was retrieved and reviewed by the author board-certified in perioperative echocardiography (AAB). In only two cases was perioperative echocardiographic confirmation of defect closure not archived and TEE procedure specific notes lost to the record. Assumption that a positive shunt finding would have been recorded combined with documented normalization of auscultatory findings as the only postoperative documentation confirmation of ASD closure occurred in only these two patients.
Results are presented as means ± 1 standard deviation. Populations are compared using Student's t test. A p value of less than 0.05 is significant.
Surgical methods
In all cases, defects were closed with pericardial patches: autologous fresh pericardium, glutaraldehdye treated bovine pericardium (Edwards Life Science, Irvine, CA), or photo-oxidized bovine pericardium (Cardio Fix, Sulzer Medica, Austin, TX). Fenestrated ASDs were resected, converted into single oval defects, and repaired with pericardial patch and running polypropylene suture closure. Primary suture closure of ASDs was never used. A mild hypothermia (34°C) was used.
During the Georgetown era, all patients were operated upon utilizing a relatively short, but classic, median sternotomy (approximately 5 1/2 inches), chest cannulation, and aortic cross clamping with cardioplegia; with two exceptions, two adult females underwent a submammary type right anterior thoracotomy, with femoral arterial and venous cannulation, with separate SVC cannula via chest, cardioplegia, and aortic cross clamping (phase 1).
During the Providence period, an increasing effort was made to limit incisions, avoid the use of cardioplegia unless necessary for complex reconstructions, and ultimately change cannulation techniques to improve access via limited incisions and optimize cosmesis. This evolution (phase 2 or middle phase) included various components of the minimally invasive approaches including 5 port access patients (Heartport, Inc., Redwood City, CA). The final consecutive 41 pediatric and 17 adult patients underwent incisions (phase 3) with uniform techniques based upon age and gender as described below. Induced fibrillation rather than cardioplegia was used in all but one patient during phase 3.
Age and gender determined methods
Adult women
Two "bikini" line incisions; a right submammary (four inches) and an inguinal incision (11/2 to 2 inches transversely at the right inguinal ligament) were used (Fig 2).
Femoral vein cannulation was performed with BioMedicus venous cannulas utilizing the modified Seldinger approach through a 5.0 polypropylene pursestring under direct vision of the exposed femoral vein, usually via the saphenous bulb. In six cases, the femoral artery was also cannulated for arterial perfusion. The submammary incision was performed with the incision beginning just lateral to the right internal mammary artery. Exposure allowed cannulation of the superior vena cava with a right angle metal tip cannula (DLP 673 xx and 693 xx sizes 12 fr, Medtronic Corp., Minneapolis, MN). When femoral artery cannulation was not used, the aorta was cannulated directly with the following: BioMedicus 96820–0 xx series sizes 8 fr-14 fr, Medtronic Corp (pediatric); Sarns 615 mm # 12325 (Terumo Corp., Ann Arbor, MI) (adult). One adult case of SVD-ASD had superior vena caval cannulation using a percutaneous approach (Heartport catheter). Another patient with SVD had superior vena caval drainage established through a short supraclavicular transverse incision for direct cannulation through 5.0 polypropylene pursestring of the jugular vein using a right angle metal tip cannula (# 24 DLP, Medtronic). Two pacemaker wires were secured to the epicardial surface of the right ventricle, exited and secured to the skin (medial and inferior to the chest incision) for use for attachment to the fibrillator and thereafter retained as postoperative short-term external pacemaker wires. The patients were placed on cardiopulmonary bypass and fibrillation induced. Snare was applied only to the superior vena cava (SVC). Cardiopulmonary bypass was maintained with a central venous pressure of 1 to 3 mm Hg. The Medtronic Corporation (Minneapolis, MN) supplied our circuits, which were comprised of medical grade PVC tubing. Affinity hollow fiber oxygenator, MVT-800 closed venous reservoir, arterial filter, filtered intercept cardiotomy reservoir, and two BPX-80 centrifugal pumps. One BPX-80 was utilized as the main blood pump, pumping blood from the venous reservoir into the oxygenator. The other BPX-80 was available to be used as a kinetic venous assist device when necessary. Handling occasional venous air was a simple matter. Affixed to the 2 Luer-lock connectors atop the MVR-800 was a 1/8-degree PVC purge-line, which divided into 2 separate legs, one each for the aforementioned MVR-800 connectors. Its other end was attached to a regular field sucker line via a 1/4 in by 1/4 in Luer-lock connector. Any air entering the reservoir was easily aspirated by the purge-line and deposited into the cardiotomy reservoir. When necessary, placement of a weighted vent (DLP # 12010, Medtronic) in the orifice of the IVC during suturing of the inferior defect rim near the IVC orifice, along the sulcus terminalis, kept the level of blood to just below the orifice of the IVC, which allows full exposure of the interatrial septum without introducing air into the inferior venous line (passive return only). The anatomy is assessed, any accessory fenestrations identified, and the patch (usually oval in shape) is fashioned from the chosen type of pericardium. When prominent, eustachian valve tissue is typically excised. Skin closure with fine monocryl sutures enhances skin cosmesis.
|
Children
Children underwent mini-median sternotomy (21/2 to 23/4 inch incisions) with complete sternotomy (Vernon bone scissors #790315). Arterial and superior vena caval cannulation was accomplished via the median sternotomy, whereas a short (1/2 inch) incision over the right femoral vein allowed placement of venous return cannula, advanced only to the level of the liver (BioMedicus, Medtronic). Fibrillation was performed as described above. Surgical patch closure was accomplished as above.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
Of the 74 patients (1996 to 2002) for whom postoperative ventilator times could be documented, 59 were extubated in the operating room, and 13 within the first few hours following surgery. Only one patient was intubated for greater than 24 but less than 48 hours. One patient had a short episode of ventricular tachycardia (spontaneously resolved). No patient sustained a stroke. No patient had serious bleeding or required a transfusion. Most frequent perioperative complications were atrial fibrillation (adult 10%, pediatric 1.2%) and postpericardiotomy syndrome (adults 2%, pediatric 4.7%).
No patient operated upon with the limited access procedures required conversion to full sternotomy or a larger skin incision. The bikini incision for adult women is truly invisible under a bikini (Fig 2). The limited median sternotomy incision is very acceptable cosmetically and not visible when wearing a tee shirt or a one piece bathing suit (female children). No patients required revision of their incisions. Two pediatric patients underwent laser therapy for incipient keloid formation.
Pediatric length of stay (1996 to 2002) averaged 3.51 ± 1.97 days (admit-to-discharge and procedure-to-discharge were equivalent as all patients were admitted on the day of surgery). Two thirds of these patients (45 of 67) were discharged either on the second or third postoperative day and only two patients remained for more than a week. The 41 pediatric patients undergoing the age-based, less invasive approach had an average length of stay of 3.17 ± 1.30 days (NS p > 0.05, Student's t test compared to standard approach) and only one patient was hospitalized for more than a week.
Length of postoperative stay (LOS) for adults was difficult to interpret as it appeared to be dependent on associated lesions, medical diagnoses, and age rather than the ASD itself; in general there was a trend to shorter stays with the minimally invasive procedures, but the numbers were also insufficient to generate statistical significance (adult less invasive mean procedure-to-discharge LOS = 4.91 ± 2.66 days vs adult standard = 5.62 ± 3.95 days, p > 0.05, Student's t test).
|
Comment |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
Some of the perioperative complications are similar to those seen with transvascular catheter-based techniques (eg, atrial fibrillation) [15]. Postpericardiotomy syndrome appears to be slightly greater with surgery than reported for catheter-based techniques, but conversely, surgery resulted in no residual shunts or leaks (trivial or otherwise), no mortality, and no potential "device" related materials morbidity. Pericardial patches are incorporated into the tissues of the heart with either neointima or pseudointima formation depending on the processing methodology, although glutaraldehyde treatment does seen to ultimately predispose to calcification, despite coverage by host cells.
The anterolateral approach should not be used in prepubescent girls due to the risk of damage to the breast bud. The submammary incisions should follow the lines of Langerhans for optimal wound healing. For ASD closure by itself with these methods, the intracardiac portion of the surgical procedure required, on average, 22 minutes, including atriotomy, patch closure, deairing, and atrial closure (ie, fibrillation time).
This study has a number of limitations. First, it is a retrospective review and is limited to a single surgeon's experience. Second, there was no sharp demarcation between the "old" standard technique and the "new" minimal access techniques. Thus, this review includes three phases. The first phase (chronologically) involved standard sternotomy, while during the second phase an evolution of techniques occurred with various cannulation strategies, sizes of incisions, etc. The final phase included three defined gender-age determined less invasive methods. Other groups have been able to report two different techniques in which the changes were apparently made all at once and at one point in time [6, 16]. However, our evolutionary approach to ASD closure does not invalidate the ultimate results for the optimized technical approaches (phase 3), nor does it invalidate the critical observation that universal use of patch repair obviates residual or recurrent shunts regardless as to whether less invasive or standard incisions are used [14]. And finally, only patient data from the 1996 to 2002 time period were entered into a formal retrievable database and thus the percentages of perioperative complications as quoted are only for the 1996 to 2002 (and thus more recent) cases. Including the 1988 to 1996 complication data likely would have resulted in under-reporting the actual incidence as it would depend upon limited retrospective chart review. Follow-up was more intensive (nurse practitioner driven) and recording of events (including after discharge) was more formal and done in "real-time" (STS Database) in the 1996 to 2002 era.
The consistent use of pericardial patches for surgical closure results in efficient, secure, low profile, and 100% effective defect closure with an interatrial septum that ultimately is relined with native tissue and for which there is no long term bacterial endocarditis (SBE) risk. The ACC/AHA Guidelines suggest SBE prophylaxis does not need to be given after six months following complete surgical septal defect closure [14]. Catheter-based occlusion devices have unknown long-term bacterial endocarditis risks, have potentially catastrophic failure modes (eg, strut fracture, migration, embolization), are not low profile, extending out of the septal plane, and are more liable to residual shunts than surgical closure. Even volumetrically trivial defects, while not an indication for secondary surgery, should be considered methodologic failures (whether device or surgical), which potentially leave a patient at risk for low frequency events (eg, paradoxical emboli, SBE) perhaps for life [17–20]. It is not yet known whether catheter-device patients should have bacterial endocarditis prophylaxis for all indications forever or whether they should continue with anticoagulation (eg, aspirin) for life, with or without a residual shunt (trivial or otherwise). Conversely, secure surgical closure of an ASD in early childhood results in a functionally (and from a natural history standpoint) "normal heart." Virtually all pediatric patients with surgical ASD closure should be treated as such from the standpoint of insurance, employment, and activity. Adult, and especially elderly, patients do not always experience such a benign postoperative course and are at elevated risk for atrial fibrillation, pneumonia, and surgical stress-related complications [21], but nevertheless likely benefit from ASD closure unless truly advanced age, decrepitude, or pulmonary vascular disease are present [22].
In a recent large (n = 459) prospective nonrandomized study using the Amplatzer device (AGA Medical Corporation, Golden Valley, MN), the authors reported a 4.3% device deployment technical failure rate and an additional 3.7% failure to use the device at the time of catheterization despite intent to do so (usually because the ASD was too large), and a 0.9% postprocedure embolism rate; in the 331 (device) patients seen at 12 month follow-up, 1.5% had large or moderately large residual shunts compared to none in the linked surgical arm (n = 154) of their study [15]. The authors reported early 53.7% "trivial" residual shunt and 13.2% "small" residual shunts persisting in the patients treated with the device, but failed to report the trivial-small rates at 12 months for devices [15]. However, they did report, for the surgical arm, these 12 month follow-up data: there were seven patients identified by echocardiography to have trivial or small residual shunts remaining, which is unlike our series in which there were no residual shunts following surgery. However, the surgical protocol reported in their paper allowed for both primary suture and patch closure techniques. There was no reporting of the residual shunt rate for patients closed with a patch. It is our belief that surgical patch closure essentially precludes any residual shunt, trivial or otherwise.
It is important to periodically report methods and results for surgical ASD closure, as current device studies, while prospective in design, often do not randomize to surgery and quote historical surgical literature for comparison. For example, a recent article published in July 2003, from a center that reported 553 percutaneous ASD closures, referenced four surgical journal articles published between 1990 and 1994 and two textbooks (1993 and 1994) as sources for surgical outcomes comparison; such references are documenting results for surgery performed predominantly in the 1970s and 1980s [23]. Our retrospective study does not prove superiority of surgical patch closure over device closure, nor is it intended to do so. However, it is clear that the current generation of devices deployed by experienced interventional cardiologists, and used in carefully selected patients, has at least a 10% failure to achieve the desired result of complete closure of the atrial septum [15]. Failure rate is zero in our experience with patch surgical closure. As results with the devices improve and as the devices themselves improve, this outcome difference may narrow and ultimately the decision to employ either a device or a surgical approach will perhaps depend upon comparisons of cost, preference for the procedure-specific related morbidities, patient satisfaction, and acceptance.
Surgeons should continue to be committed to safe surgical methods that achieve secure closure of the septum preferably with biological patches which are incorporated into the tissue structure of the heart and which are free from materials-related failure modes, while utilizing the least invasive, least traumatic, and most cosmetically appealing techniques for access and cardiopulmonary bypass. Modern less invasive, efficient, and simplified surgical patch closure methods, optimized for individual patient characteristics (eg, age, sex, anatomical location, etc.), seem to offer safe and effective closure for the most common types of atrial septal defect.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Ak, T. Aybek, G. Wimmer-Greinecker, F. Ozaslan, F. Bakhtiary, A. Moritz, and S. Dogan Evolution of surgical techniques for atrial septal defect repair in adults: A 10-year single-institution experience J. Thorac. Cardiovasc. Surg., September 1, 2007; 134(3): 757 - 764. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Misra, A. Sadiq, N. Namboodiri, and J. Karunakaran The 'aortic rim' recount: embolization of interatrial septal occluder into the main pulmonary artery bifurcation after atrial septal defect closure Interactive CardioVascular and Thoracic Surgery, June 1, 2007; 6(3): 384 - 386. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. W. S. Soo, D. G. Healy, K. Walsh, and F. Wood Inferior vena cava and coronary sinus obstruction after percutaneous atrial septal defect device closure requiring surgical revision J. Thorac. Cardiovasc. Surg., June 1, 2006; 131(6): 1405 - 1406. [Full Text] [PDF]
|
|
|
|
 |
|
 M S Spence and S A Qureshi Complications of transcatheter closure of atrial septal defects Heart, December 1, 2005; 91(12): 1512 - 1514. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ando, Y. Takahashi, and T. Kikuchi Short Operation Time: An Important Element to Reduce Operative Invasiveness in Pediatric Cardiac Surgery Ann. Thorac. Surg., August 1, 2005; 80(2): 631 - 635. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Pretre, A. Kadner, H. Dave, A. Dodge-Khatami, D. Bettex, and F. Berger Right axillary incision: A cosmetically superior approach to repair a wide range of congenital cardiac defects J. Thorac. Cardiovasc. Surg., August 1, 2005; 130(2): 277 - 281. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Falk and T. Walther Invited commentary Ann. Thorac. Surg., June 1, 2004; 77(6): 2149 - 2150. [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 |
