Original Article
Repair of Anomalous Left Coronary Artery From the Right Pulmonary Artery: A Series of Nine Cases
World Journal for Pediatric and Congenital Heart Surgery 2015, Vol. 6(3) 382-386 ª The Author(s) 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/2150135115579918 pch.sagepub.com
Clement D. Marshall, MD1, Justin Weigand, MD2, Peter Sambatakos, MD2, Denise A. Hayes, MD2, Jonathan M. Chen, MD1, Jan M. Quaegebeur, MD, PhD1, Emile Bacha, MD1, and Marc E. Richmond, MD, MS2
Abstract Background: Repair of anomalous left coronary artery from the right pulmonary artery presents a particular technical challenge to the congenital cardiac surgeon. There is disagreement in the literature over the optimal technique for this defect, with some authors advocating for unroofing of the periaortic segment of coronary artery, while others prefer direct aortic reimplantation of the artery. Methods: We performed a retrospective study examining outcomes of patients who were repaired for this anomaly at our institution. In-hospital and outpatient follow-up data were analyzed. Results: Nine patients were identified. Most patients had poor left ventricular function at the time of surgery. All patients in our series were repaired using the direct coronary transfer technique. To date there were no mortalities among the study participants. At last follow-up, all patients with available echocardiograms had normal ventricular function. One patient required reoperation for anastomotic stenosis. Conclusions: We demonstrate that using the technique of direct coronary transfer to the aorta, we have achieved excellent results with repair of this defect. Keywords coronary artery anomaly, congenital heart disease, heart failure, myocardial infarction Submitted February 9, 2015; Accepted March 10, 2015.
Introduction Anomalous left coronary artery from the pulmonary artery (ALCAPA) is a congenital cardiac anomaly affecting about 1 in 300 000 live births.1 This rare lesion is characterized by impaired perfusion and eventual infarction of the myocardium supplied by the left coronary artery (LCA). Although extensive coronary collateralization may permit survival into childhood or even adulthood, ALCAPA is usually fatal in infancy without surgical intervention. The currently preferred reparative technique for ALCAPA is direct reimplantation of the LCA into the aorta.2 In the most common anatomical configuration of ALCAPA, the anomalous LCA arises from the posterior sinus of the pulmonary artery (PA).2 In a less common variant, the anomalous LCA arises from the right pulmonary artery (RPA, anomalous left coronary artery from the right pulmonary artery [ALCARPA]; Figures 1 and 2). When this variation is present, the LCA often follows an intramural course, running vertically within the wall of the ascending aorta. In such cases, the LCA cannot always be easily reimplanted by usual techniques.
From 1994 to 2012, there were 43 cases of ALCAPA repaired at Columbia University’s NewYork-Presbyterian/ Morgan Stanley Children’s Hospital. In nine of these cases, the LCA originated from the RPA. The direct coronary transfer technique was used in all nine patients, with excellent results. We wish to report our experience with the repair of this defect.
1 Division of Cardiothoracic and Vascular Surgery, Columbia University College of Physicians and Surgeons, Morgan Stanley Children’s Hospital of NewYork-Presbyterian, New York, NY, USA 2 Division of Pediatric Cardiology, Columbia University College of Physicians and Surgeons, New York, NY, USA
Corresponding Author: Emile Bacha, Division of Cardiothoracic and Vascular Surgery, Columbia University College of Physicians and Surgeons, Morgan Stanley Children’s Hospital of NewYork-Presbyterian, 3959 Broadway, New York, NY 10032, USA. Email: [email protected]
Downloaded from pch.sagepub.com at CALIFORNIA STATE UNIV FRESNO on July 22, 2015
Marshall et al
383
Patients and Methods
Abbreviations and Acronyms ALCAPA ALCARPA EF LAD LCA LV LVAD PA RPA SD SF TEE
anomalous left coronary artery from the pulmonary artery anomalous left coronary artery from the right pulmonary artery ejection fraction left anterior descending left coronary artery left ventricle left ventricular assist device pulmonary artery right pulmonary artery standard deviation shortening fraction transthoracic echocardiogram
We used a hospital database to identify all cases of ALCAPA repaired at our institution since 1994. Patients with ALCARPA were identified from the operative reports, and data were collected from electronic and paper medical records. Data included electrocardiogram, echocardiogram, and laboratory results as well as clinical information from the pre- and postoperative periods. Outside cardiologists were contacted to determine the status of the patients (living, transplanted, or dead) and to provide any postoperative echocardiograms. This study was approved by the Columbia University Medical Center Institutional Review Board.
Results
Figure 1. Aortogram in a patient with anomalous left coronary artery from the right pulmonary artery (ALCARPA) shows opacification of the normal right coronary artery (RCA) and absent left coronary artery (LCA; A). A still frame from selective right coronary angiogram in the same patient taken just after injection (B). The RCA is still seen (arrow) and the anomalous LCA that fills via collaterals can be seen emptying into the pulmonary artery (PA; arrowheads). It is not obvious from this angiogram that the origin of the LCA is from the right pulmonary artery (RPA), highlighting the difficulty of diagnosing the ALCARPA subtype preoperatively.
Figure 2. Parasternal short axis view demonstrating abnormal retrograde flow in the left coronary artery (LCA; arrow) at its origin on the right pulmonary artery (RPA).
We identified nine patients who underwent repair of ALCARPA. Preoperative and anatomic data are summarized in Table 1. The median age at repair was 4.5 months (range 2 months to 5 years) and median weight at repair was 6.1 kg (range 4.4-17.5 kg). In our cohort, seven (78%) of the nine patients presented at less than one year of life, with the majority presenting at less than six months (six of nine or 67%). We were only able to obtain the clinical presentation on seven of the nine patients. Of the seven patients, three presented with respiratory symptoms, three presented with abnormal cardiovascular examinations, and one was found on screening echo after being diagnosed with cri du chat syndrome. Of the nine patients, five were diagnosed with an anomalous left coronary via echocardiogram, while three required a cardiac catheterization to further evaluate the coronary arteries in order to establish a diagnosis. The mode of diagnosis was not clear in the last patient. Eight of the nine patients had the LCA arising from the RPA. In one patient (case #7), the circumflex artery arose from the RPA and the left anterior descending (LAD) artery arose normally from the left sinus of Valsalva. In six of the nine patients, the anomalous LCA had an intramural course within the aortic media after arising from the PA. In one patient (case #4), the LCA was fused with the aortic adventitia but was not intramural per se. In case #7, whose circumflex artery arose from the PA with a normal LAD origin, there was no fused or intramural segment. In the earliest patient (case #9), it is not clear from the record whether there was a fused or intramural segment. Intraoperative surgical data are summarized in Table 2. In all cases, direct reimplantation of the LCA into the aorta was performed (Figure 2). The mean cardiopulmonary bypass time was 108 minutes (standard deviation [SD] 25 minutes) and the mean aortic cross-clamp time was 58 minutes (SD 13 minutes). One patient (case #8) required a brief period of hypothermic circulatory arrest due to persistent back bleeding from the LCA coronary ostium (Figure 3). Surgical steps were generally as follows. The PA and branches are mobilized and the origin of the LCA is identified. The patient is placed on cardiopulmonary bypass via the cannulae in the ascending aorta and the right atrium and both PA
Downloaded from pch.sagepub.com at CALIFORNIA STATE UNIV FRESNO on July 22, 2015
384
World Journal for Pediatric and Congenital Heart Surgery 6(3)
Table 1. Summary of Preoperative Patient Data. Patient 1 2 3 4 5 6 7 8 9
Year of Surgery
Age, mo
2012 2011 2011 2010 2010 2009 2008 2007 1995
4.6 2.5 2 4.5 7.5 22 66 4.5 3
Sex M F M M F M M M M
Wt, kg 5.7 4.4 4.5 6.1 6.4 11.6 17.5 6.8 4.8
LCA Origin
Intramural Course?
EF
SF
LVEDD Z Score
RPA RPA RPA RPA RPA RPA Circ ¼ RPA, LAD ¼ aorta RPA RPA
Yes Yes Yes No Yes Yes No No Unknown
31 7 11 24 32 62 62 16
18 6 10 11 7 37 34 8 13
9 10 9 17 10 2 1 13
Abbreviations: EF, ejection fraction; SF, shortening fraction; LVEDD, left ventricular end-diastolic dimension; Circ, circumflex artery; LCA, left coronary artery; RPA, right pulmonary artery; LAD, left anterior descending; F, female; M, male.
Table 2. Perioperative Data. Postop Mechanical Patient Support
Delayed Sternal IV Inotropic ICU Stay, Closure Support, Days Days
1 2 3 4 5 6 7 8 9
No No Yes No No No No No Yes
No No No No No No No No LVAD (3 days)
8 17 6 22 3 0 0 7 13
8 18 7 21 4 2 2 10 15
Abbreviations: LVAD, left ventricular assist device; ICU, intensive care unit; IV, intravenous.
Figure 3. Direct reimplantation without coronary unroofing for anomalous left coronary artery from the right pulmonary artery (ALCARPA). The left coronary artery often runs vertically, in close approximation to the normal aortic origin, often with aortic fusion (A). Care must be taken to avoid kinking or axial twist upon direct reimplantation (B).
branches are snared. Antegrade cardioplegia is administered into the aortic root and reaches the LCA circulation via intracoronary collaterals. The branch PAs are kept snared to prevent cardioplegia from running off into the PA. The RPA is typically
opened and a button of tissue around the origin of the LCA is excised. The aorta is transected in most cases to facilitate exposure. If necessary (and possible), the intramural segment of the coronary artery is dissected free from the aortic wall over several millimeters to obtain enough of a proximal coronary conduit to perform the 90 rotation of the button that is necessary as one moves from the RPA (axial) plane to the aortic (sagittal) plane. Mobilization proceeds from the most cranial aspect of the intramural segment toward distal. The coronary artery may not be intramural over its entire length and may simply be embedded in periaortic adventitia. This is sometimes difficult to ascertain during surgery. The important point is that just enough of the proximal coronary needs to be dissected out to allow it to be turned rightward toward the ascending aorta. Typically the coronary button is anastomosed end-to-side to the aorta. The implantation site is quite a bit higher than the sinotubular junction, at the level of the RPA crossing. The RPA is reconstructed with a patch and the patient is weaned off cardiopulmonary bypass per protocol. Delayed sternal closure occurred in two patients (cases #3 and #9). One patient (case #9) required three days of postoperative mechanical support in the form of a left ventricular assist device (LVAD). This patient had poor LV function (shortening fraction [SF] 13%) and marked LV dilation and mild mitral regurgitation preoperatively and the LVAD was implanted at the end of the case in anticipation of continued poor function. At the time of hospital discharge (postoperative day 18), the function was normalized (SF 35%). Long-term outcomes are summarized in Table 3. All patients were known to be alive at the time of data gathering, and no patient underwent cardiac transplantation. Long-term echocardiographic data are available for five of the nine patients. Median duration from surgery to last echo was 2.3 years (range 0.8-3.9 years). All patients with available data had normal function at last echo defined as SF > 28% or ejection fraction (EF) > 55%. Median time from surgery to normalization of function was 82 days (range 18-126 days). This excludes one patient (case #6), in whom the function was normal before and immediately after surgery. These values may overestimate the time to normalization because the function
Downloaded from pch.sagepub.com at CALIFORNIA STATE UNIV FRESNO on July 22, 2015
Marshall et al
385
Table 3. Long-Term Data.a
Patient 1 2 3 4 5 6 7 8 9
Time to Last Follow-Up, year 1 2.4 2.5 3.7 3.9 4.2 5.5 6.7 18.3
Time to Time to Last Function at Normalization, day Echo, year Last Echo 0.8
EF 63%
126
2.1 2.3 3.9
EF 69% SF 31% EF 64%
89 103 N/A
2.3
SF 35%
18
Abbreviations: EF, ejection fraction; SF, shortening fraction; N/A, not applicable. a Blank spaces represent unavailable or missing data. In patient 6, the function was normal.
could have become normal prior to the echo being acquired. Reoperation was required in one patient (case #6) who underwent revision of a stenotic LCA-aorta anastomosis two years after initial repair. The stenosis was discovered on routine echo and the EF at the time was 70%.
Comment Surgical repair of ALCAPA has dramatically improved the outcomes of patients with this rare and potentially fatal congenital anomaly. Techniques to reimplant the anomalous LCA, which typically arises from the posterolateral PA sinus, have been well described.2 The subtype of ALCARPA is a much more challenging lesion. Echocardiographic or even catheter diagnosis can be difficult, as the LCA typically courses in close approximation to its usual aortic origin and ultrasound ‘‘dropout’’ in that region may be misleading. It is easy to injure the LCA during dissection of the aortopulmonary fold as it rises in that area to reach the RPA, where it typically originates at the lesser curvature of the angle between the main PA and RPA, making PA reconstruction more challenging. The ALCARPA with aortic fusion presents a technical challenge to the surgeon because the ostium of the coronary artery frequently faces away from the aortic wall in an axial plane, and there is typically a very short length of artery between the RPA and the area of aortic fusion. This makes it difficult to achieve direct reimplantation of the LCA without kinking or twisting. Thus, the surgical options include (1) dissecting the entire LCA from the aortic wall and reimplanting it, (2) dissecting only the most proximal LCA and reimplanting it, leaving the intramural portion untouched, and (3) ‘‘unroofing’’ the intramural portion of the LCA (creating a new ostium) and ligating the proximal LPA. In the former two options, the LCA will be reimplanted quite high on the ascending aorta, and care must be taken to cannulate the aorta distally. Kumar et al performed a thorough analysis of 12 reported cases of ALCARPA with aortic fusion in which the following two predominant surgical techniques were used: coronary
transfer with direct reimplantation and unroofing of the intramural segment.3 Their outcomes suggest that unroofing of the intramural segment is the superior surgical technique, as it avoids kinking of the artery with resulting obstruction of flow. Adachi et al further argue that in cases of ALCARPA with an intramural segment of the LCA, direct reimplantation alone leaves an intramural segment that may predispose to ischemia and sudden death.4 In our series, no patient developed ischemia due to intramural narrowing and the only coronary stenosis (case #6) involved the anastomosis. We believe that the risk of ischemia arising from the intramural segment is a theoretical concern. In our experience, the small size of the LCA makes unroofing its intramural segment technically challenging. Additionally, in some cases the LCA is not truly intramural but rather tightly embedded in the aortic adventitia, making an unroofing procedure hazardous if not impossible. Therefore, the preferred technique at our center has been coronary dissection and direct reimplantation. If it is necessary we prefer to dissect a few millimeters of the LCA from the aorta, to avoid causing an axial twist upon transfer of the coronary button to the midascending aorta. One must take care during this dissection to respect the integrity of both the LCA and the aorta, and when in doubt the easily reconstructed aortic wall should be the structure that is breached. The aortic wall was not intentionally breached during LCA dissection in any of the above-mentioned cases. This delicate surgical strategy, which has been utilized on all patients since the beginning of our series, has yielded excellent results.
Limitations This study assumes the typical limitations of a retrospective, chart-based review. In particular, we likely failed to identify all patients who underwent surgery for this condition during this study time period, which is evidenced by the fact that we only identified one patient between 1995 and 2007. Although we have demonstrated excellent perioperative outcomes, more complete long-term follow-up would be valuable to this report. Unfortunately, we were not able to obtain follow-up echocardiography on all of our patients, which limits our ability to make conclusions about long-term functional outcomes.
Conclusion In conclusion, we describe a series of patients with ALCARPA who underwent LCA dissection and direct coronary reimplantation with excellent results. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study
Downloaded from pch.sagepub.com at CALIFORNIA STATE UNIV FRESNO on July 22, 2015
386
World Journal for Pediatric and Congenital Heart Surgery 6(3)
was supported by the Columbia University College of Physicians and Surgeons.
References 1. Cowles RA, Berdon WE. Bland-White-Garland syndrome of anomalous left coronary artery arising from the pulmonary artery (ALCAPA): a historical review. Pediatr Radiol. 2007; 37(9): 890-895.
2. Dodge-Khatami A, Mavroudis C, Backer CL. Anomalous origin of the left coronary artery from the pulmonary artery: collective review of surgical therapy. Ann Thorac Surg. 2002;74(4): 946-955. 3. Kumar TKS, Sinha P, Donofrio MT, Jonas RA. Anomalous left coronary artery from the right pulmonary artery with aortic fusion. J Thorac Cardiovasc Surg. 2012;143(2): 505-507. 4. Adachi I, Kagisaki K, Yagihara T, et al. Unroofing aortic intramural left coronary artery arising from right pulmonary artery. Ann Thorac Surg. 2008;85(2): 675-677.
Downloaded from pch.sagepub.com at CALIFORNIA STATE UNIV FRESNO on July 22, 2015
Repair of Anomalous Left Coronary Artery From the Right Pulmonary Artery: A Series of Nine Cases
World Journal for Pediatric and Congenital Heart Surgery 2015, Vol. 6(3) 382-386 ª The Author(s) 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/2150135115579918 pch.sagepub.com
Clement D. Marshall, MD1, Justin Weigand, MD2, Peter Sambatakos, MD2, Denise A. Hayes, MD2, Jonathan M. Chen, MD1, Jan M. Quaegebeur, MD, PhD1, Emile Bacha, MD1, and Marc E. Richmond, MD, MS2
Abstract Background: Repair of anomalous left coronary artery from the right pulmonary artery presents a particular technical challenge to the congenital cardiac surgeon. There is disagreement in the literature over the optimal technique for this defect, with some authors advocating for unroofing of the periaortic segment of coronary artery, while others prefer direct aortic reimplantation of the artery. Methods: We performed a retrospective study examining outcomes of patients who were repaired for this anomaly at our institution. In-hospital and outpatient follow-up data were analyzed. Results: Nine patients were identified. Most patients had poor left ventricular function at the time of surgery. All patients in our series were repaired using the direct coronary transfer technique. To date there were no mortalities among the study participants. At last follow-up, all patients with available echocardiograms had normal ventricular function. One patient required reoperation for anastomotic stenosis. Conclusions: We demonstrate that using the technique of direct coronary transfer to the aorta, we have achieved excellent results with repair of this defect. Keywords coronary artery anomaly, congenital heart disease, heart failure, myocardial infarction Submitted February 9, 2015; Accepted March 10, 2015.
Introduction Anomalous left coronary artery from the pulmonary artery (ALCAPA) is a congenital cardiac anomaly affecting about 1 in 300 000 live births.1 This rare lesion is characterized by impaired perfusion and eventual infarction of the myocardium supplied by the left coronary artery (LCA). Although extensive coronary collateralization may permit survival into childhood or even adulthood, ALCAPA is usually fatal in infancy without surgical intervention. The currently preferred reparative technique for ALCAPA is direct reimplantation of the LCA into the aorta.2 In the most common anatomical configuration of ALCAPA, the anomalous LCA arises from the posterior sinus of the pulmonary artery (PA).2 In a less common variant, the anomalous LCA arises from the right pulmonary artery (RPA, anomalous left coronary artery from the right pulmonary artery [ALCARPA]; Figures 1 and 2). When this variation is present, the LCA often follows an intramural course, running vertically within the wall of the ascending aorta. In such cases, the LCA cannot always be easily reimplanted by usual techniques.
From 1994 to 2012, there were 43 cases of ALCAPA repaired at Columbia University’s NewYork-Presbyterian/ Morgan Stanley Children’s Hospital. In nine of these cases, the LCA originated from the RPA. The direct coronary transfer technique was used in all nine patients, with excellent results. We wish to report our experience with the repair of this defect.
1 Division of Cardiothoracic and Vascular Surgery, Columbia University College of Physicians and Surgeons, Morgan Stanley Children’s Hospital of NewYork-Presbyterian, New York, NY, USA 2 Division of Pediatric Cardiology, Columbia University College of Physicians and Surgeons, New York, NY, USA
Corresponding Author: Emile Bacha, Division of Cardiothoracic and Vascular Surgery, Columbia University College of Physicians and Surgeons, Morgan Stanley Children’s Hospital of NewYork-Presbyterian, 3959 Broadway, New York, NY 10032, USA. Email: [email protected]
Downloaded from pch.sagepub.com at CALIFORNIA STATE UNIV FRESNO on July 22, 2015
Marshall et al
383
Patients and Methods
Abbreviations and Acronyms ALCAPA ALCARPA EF LAD LCA LV LVAD PA RPA SD SF TEE
anomalous left coronary artery from the pulmonary artery anomalous left coronary artery from the right pulmonary artery ejection fraction left anterior descending left coronary artery left ventricle left ventricular assist device pulmonary artery right pulmonary artery standard deviation shortening fraction transthoracic echocardiogram
We used a hospital database to identify all cases of ALCAPA repaired at our institution since 1994. Patients with ALCARPA were identified from the operative reports, and data were collected from electronic and paper medical records. Data included electrocardiogram, echocardiogram, and laboratory results as well as clinical information from the pre- and postoperative periods. Outside cardiologists were contacted to determine the status of the patients (living, transplanted, or dead) and to provide any postoperative echocardiograms. This study was approved by the Columbia University Medical Center Institutional Review Board.
Results
Figure 1. Aortogram in a patient with anomalous left coronary artery from the right pulmonary artery (ALCARPA) shows opacification of the normal right coronary artery (RCA) and absent left coronary artery (LCA; A). A still frame from selective right coronary angiogram in the same patient taken just after injection (B). The RCA is still seen (arrow) and the anomalous LCA that fills via collaterals can be seen emptying into the pulmonary artery (PA; arrowheads). It is not obvious from this angiogram that the origin of the LCA is from the right pulmonary artery (RPA), highlighting the difficulty of diagnosing the ALCARPA subtype preoperatively.
Figure 2. Parasternal short axis view demonstrating abnormal retrograde flow in the left coronary artery (LCA; arrow) at its origin on the right pulmonary artery (RPA).
We identified nine patients who underwent repair of ALCARPA. Preoperative and anatomic data are summarized in Table 1. The median age at repair was 4.5 months (range 2 months to 5 years) and median weight at repair was 6.1 kg (range 4.4-17.5 kg). In our cohort, seven (78%) of the nine patients presented at less than one year of life, with the majority presenting at less than six months (six of nine or 67%). We were only able to obtain the clinical presentation on seven of the nine patients. Of the seven patients, three presented with respiratory symptoms, three presented with abnormal cardiovascular examinations, and one was found on screening echo after being diagnosed with cri du chat syndrome. Of the nine patients, five were diagnosed with an anomalous left coronary via echocardiogram, while three required a cardiac catheterization to further evaluate the coronary arteries in order to establish a diagnosis. The mode of diagnosis was not clear in the last patient. Eight of the nine patients had the LCA arising from the RPA. In one patient (case #7), the circumflex artery arose from the RPA and the left anterior descending (LAD) artery arose normally from the left sinus of Valsalva. In six of the nine patients, the anomalous LCA had an intramural course within the aortic media after arising from the PA. In one patient (case #4), the LCA was fused with the aortic adventitia but was not intramural per se. In case #7, whose circumflex artery arose from the PA with a normal LAD origin, there was no fused or intramural segment. In the earliest patient (case #9), it is not clear from the record whether there was a fused or intramural segment. Intraoperative surgical data are summarized in Table 2. In all cases, direct reimplantation of the LCA into the aorta was performed (Figure 2). The mean cardiopulmonary bypass time was 108 minutes (standard deviation [SD] 25 minutes) and the mean aortic cross-clamp time was 58 minutes (SD 13 minutes). One patient (case #8) required a brief period of hypothermic circulatory arrest due to persistent back bleeding from the LCA coronary ostium (Figure 3). Surgical steps were generally as follows. The PA and branches are mobilized and the origin of the LCA is identified. The patient is placed on cardiopulmonary bypass via the cannulae in the ascending aorta and the right atrium and both PA
Downloaded from pch.sagepub.com at CALIFORNIA STATE UNIV FRESNO on July 22, 2015
384
World Journal for Pediatric and Congenital Heart Surgery 6(3)
Table 1. Summary of Preoperative Patient Data. Patient 1 2 3 4 5 6 7 8 9
Year of Surgery
Age, mo
2012 2011 2011 2010 2010 2009 2008 2007 1995
4.6 2.5 2 4.5 7.5 22 66 4.5 3
Sex M F M M F M M M M
Wt, kg 5.7 4.4 4.5 6.1 6.4 11.6 17.5 6.8 4.8
LCA Origin
Intramural Course?
EF
SF
LVEDD Z Score
RPA RPA RPA RPA RPA RPA Circ ¼ RPA, LAD ¼ aorta RPA RPA
Yes Yes Yes No Yes Yes No No Unknown
31 7 11 24 32 62 62 16
18 6 10 11 7 37 34 8 13
9 10 9 17 10 2 1 13
Abbreviations: EF, ejection fraction; SF, shortening fraction; LVEDD, left ventricular end-diastolic dimension; Circ, circumflex artery; LCA, left coronary artery; RPA, right pulmonary artery; LAD, left anterior descending; F, female; M, male.
Table 2. Perioperative Data. Postop Mechanical Patient Support
Delayed Sternal IV Inotropic ICU Stay, Closure Support, Days Days
1 2 3 4 5 6 7 8 9
No No Yes No No No No No Yes
No No No No No No No No LVAD (3 days)
8 17 6 22 3 0 0 7 13
8 18 7 21 4 2 2 10 15
Abbreviations: LVAD, left ventricular assist device; ICU, intensive care unit; IV, intravenous.
Figure 3. Direct reimplantation without coronary unroofing for anomalous left coronary artery from the right pulmonary artery (ALCARPA). The left coronary artery often runs vertically, in close approximation to the normal aortic origin, often with aortic fusion (A). Care must be taken to avoid kinking or axial twist upon direct reimplantation (B).
branches are snared. Antegrade cardioplegia is administered into the aortic root and reaches the LCA circulation via intracoronary collaterals. The branch PAs are kept snared to prevent cardioplegia from running off into the PA. The RPA is typically
opened and a button of tissue around the origin of the LCA is excised. The aorta is transected in most cases to facilitate exposure. If necessary (and possible), the intramural segment of the coronary artery is dissected free from the aortic wall over several millimeters to obtain enough of a proximal coronary conduit to perform the 90 rotation of the button that is necessary as one moves from the RPA (axial) plane to the aortic (sagittal) plane. Mobilization proceeds from the most cranial aspect of the intramural segment toward distal. The coronary artery may not be intramural over its entire length and may simply be embedded in periaortic adventitia. This is sometimes difficult to ascertain during surgery. The important point is that just enough of the proximal coronary needs to be dissected out to allow it to be turned rightward toward the ascending aorta. Typically the coronary button is anastomosed end-to-side to the aorta. The implantation site is quite a bit higher than the sinotubular junction, at the level of the RPA crossing. The RPA is reconstructed with a patch and the patient is weaned off cardiopulmonary bypass per protocol. Delayed sternal closure occurred in two patients (cases #3 and #9). One patient (case #9) required three days of postoperative mechanical support in the form of a left ventricular assist device (LVAD). This patient had poor LV function (shortening fraction [SF] 13%) and marked LV dilation and mild mitral regurgitation preoperatively and the LVAD was implanted at the end of the case in anticipation of continued poor function. At the time of hospital discharge (postoperative day 18), the function was normalized (SF 35%). Long-term outcomes are summarized in Table 3. All patients were known to be alive at the time of data gathering, and no patient underwent cardiac transplantation. Long-term echocardiographic data are available for five of the nine patients. Median duration from surgery to last echo was 2.3 years (range 0.8-3.9 years). All patients with available data had normal function at last echo defined as SF > 28% or ejection fraction (EF) > 55%. Median time from surgery to normalization of function was 82 days (range 18-126 days). This excludes one patient (case #6), in whom the function was normal before and immediately after surgery. These values may overestimate the time to normalization because the function
Downloaded from pch.sagepub.com at CALIFORNIA STATE UNIV FRESNO on July 22, 2015
Marshall et al
385
Table 3. Long-Term Data.a
Patient 1 2 3 4 5 6 7 8 9
Time to Last Follow-Up, year 1 2.4 2.5 3.7 3.9 4.2 5.5 6.7 18.3
Time to Time to Last Function at Normalization, day Echo, year Last Echo 0.8
EF 63%
126
2.1 2.3 3.9
EF 69% SF 31% EF 64%
89 103 N/A
2.3
SF 35%
18
Abbreviations: EF, ejection fraction; SF, shortening fraction; N/A, not applicable. a Blank spaces represent unavailable or missing data. In patient 6, the function was normal.
could have become normal prior to the echo being acquired. Reoperation was required in one patient (case #6) who underwent revision of a stenotic LCA-aorta anastomosis two years after initial repair. The stenosis was discovered on routine echo and the EF at the time was 70%.
Comment Surgical repair of ALCAPA has dramatically improved the outcomes of patients with this rare and potentially fatal congenital anomaly. Techniques to reimplant the anomalous LCA, which typically arises from the posterolateral PA sinus, have been well described.2 The subtype of ALCARPA is a much more challenging lesion. Echocardiographic or even catheter diagnosis can be difficult, as the LCA typically courses in close approximation to its usual aortic origin and ultrasound ‘‘dropout’’ in that region may be misleading. It is easy to injure the LCA during dissection of the aortopulmonary fold as it rises in that area to reach the RPA, where it typically originates at the lesser curvature of the angle between the main PA and RPA, making PA reconstruction more challenging. The ALCARPA with aortic fusion presents a technical challenge to the surgeon because the ostium of the coronary artery frequently faces away from the aortic wall in an axial plane, and there is typically a very short length of artery between the RPA and the area of aortic fusion. This makes it difficult to achieve direct reimplantation of the LCA without kinking or twisting. Thus, the surgical options include (1) dissecting the entire LCA from the aortic wall and reimplanting it, (2) dissecting only the most proximal LCA and reimplanting it, leaving the intramural portion untouched, and (3) ‘‘unroofing’’ the intramural portion of the LCA (creating a new ostium) and ligating the proximal LPA. In the former two options, the LCA will be reimplanted quite high on the ascending aorta, and care must be taken to cannulate the aorta distally. Kumar et al performed a thorough analysis of 12 reported cases of ALCARPA with aortic fusion in which the following two predominant surgical techniques were used: coronary
transfer with direct reimplantation and unroofing of the intramural segment.3 Their outcomes suggest that unroofing of the intramural segment is the superior surgical technique, as it avoids kinking of the artery with resulting obstruction of flow. Adachi et al further argue that in cases of ALCARPA with an intramural segment of the LCA, direct reimplantation alone leaves an intramural segment that may predispose to ischemia and sudden death.4 In our series, no patient developed ischemia due to intramural narrowing and the only coronary stenosis (case #6) involved the anastomosis. We believe that the risk of ischemia arising from the intramural segment is a theoretical concern. In our experience, the small size of the LCA makes unroofing its intramural segment technically challenging. Additionally, in some cases the LCA is not truly intramural but rather tightly embedded in the aortic adventitia, making an unroofing procedure hazardous if not impossible. Therefore, the preferred technique at our center has been coronary dissection and direct reimplantation. If it is necessary we prefer to dissect a few millimeters of the LCA from the aorta, to avoid causing an axial twist upon transfer of the coronary button to the midascending aorta. One must take care during this dissection to respect the integrity of both the LCA and the aorta, and when in doubt the easily reconstructed aortic wall should be the structure that is breached. The aortic wall was not intentionally breached during LCA dissection in any of the above-mentioned cases. This delicate surgical strategy, which has been utilized on all patients since the beginning of our series, has yielded excellent results.
Limitations This study assumes the typical limitations of a retrospective, chart-based review. In particular, we likely failed to identify all patients who underwent surgery for this condition during this study time period, which is evidenced by the fact that we only identified one patient between 1995 and 2007. Although we have demonstrated excellent perioperative outcomes, more complete long-term follow-up would be valuable to this report. Unfortunately, we were not able to obtain follow-up echocardiography on all of our patients, which limits our ability to make conclusions about long-term functional outcomes.
Conclusion In conclusion, we describe a series of patients with ALCARPA who underwent LCA dissection and direct coronary reimplantation with excellent results. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study
Downloaded from pch.sagepub.com at CALIFORNIA STATE UNIV FRESNO on July 22, 2015
386
World Journal for Pediatric and Congenital Heart Surgery 6(3)
was supported by the Columbia University College of Physicians and Surgeons.
References 1. Cowles RA, Berdon WE. Bland-White-Garland syndrome of anomalous left coronary artery arising from the pulmonary artery (ALCAPA): a historical review. Pediatr Radiol. 2007; 37(9): 890-895.
2. Dodge-Khatami A, Mavroudis C, Backer CL. Anomalous origin of the left coronary artery from the pulmonary artery: collective review of surgical therapy. Ann Thorac Surg. 2002;74(4): 946-955. 3. Kumar TKS, Sinha P, Donofrio MT, Jonas RA. Anomalous left coronary artery from the right pulmonary artery with aortic fusion. J Thorac Cardiovasc Surg. 2012;143(2): 505-507. 4. Adachi I, Kagisaki K, Yagihara T, et al. Unroofing aortic intramural left coronary artery arising from right pulmonary artery. Ann Thorac Surg. 2008;85(2): 675-677.
Downloaded from pch.sagepub.com at CALIFORNIA STATE UNIV FRESNO on July 22, 2015
