**4. Comment**

178 Front Lines of Thoracic Surgery

Between March 1998 and June 2011, spiral ASO was performed in 57 patients (38.3%), conventional nonspiral ASO with Lecompte maneuver in 92 patients (61.7%) at our hospital. The median age and weight at operation were 9 days and 3.3 kg. Cross-clamp time was significantly lower (p < 0.011) in the spiral than the nonspiral group (128 ± 36 vs 144 ± 37 minutes), because of the common wall technique in spiral ASO, wheras additional time for patch repair on defects in the sinuses of the new MPA was needed in conventional 2-vessel technique. The average follow-up was 6.9 ± 4.2 years (up to 13.5 years). Kaplan Meier survival was 94.1 ± 3.3% at 10 years and the reoperation-free rate 88.3 ± 4.5% for spiral repair. Both ratios were satisfactory and similar to those for the nonspiral group (89.6 ± 3.3% and 89.3 ± 4.1%, respectively). Signicant aortic regurgitatin in the nonspiral group (Chen et al., 2010) was not observed in the spiral group. The supravalvular PS and aortic neocoarctation that occurred in the nonspiral group (7.6% and 2.2 %, respectively) did not occur in the spiral group (0 %) (Chiu et al., 2010); these 2 complications are related to Lecompte maneuver and the unnatural relationship of the great arteries. Tiny aortopulmonary fenestration with a small left to right shunt occurred in 15 cases of spiral group (26.3%), but they closed spontaneously after a median follow up of 6 months by echo. TGA is not merely a reversal of the great arteries; nonexistence of the spiral function in TGA should be appreciated. Recognition of the spiral function and further modification might

In complete TGA with left ventricular outflow obstruction, a segment of the aorta is cut in transverse fashion and donated to the pulmonary circulation to establish the

Fig. 8. Operative techniques for TGA with irreparable left ventricular outflow obstruction.

justify its future application.

**3.2 Modified rastelli operation** 

ventriculoarterial continuity (Figure 8).

The natural spiral relationship of the great arteries and their branches can grow on follow-up (Figure 7 and 9). We have demonstrated that the common wall between the great arteries could grow and become thinner on follow-up (Chiu et al., 2009). Spiral reconstruction in TGA is seldom performed, because the functional implications of spiral relationship of the great arteries remain unknown; thus, the Lecompte maneuver was used either in conventional ASO or Nikaidoh and Lecompte (REV) operation (Emani et al., 2009; Morell & Wearden, 2006;

Restoration of Transposed Great Arteries With or Without Subpulmonary Obstruction to Nature 181

After the Lecompte maneuver, not only flattened anterior MPA (low-pressure), branch PA stenosis, and supravalvular PS at pulmonary bifurcation can occur as a consequence of posterior compression of MPA by the ascending aorta (Williams et al., 1997; Gutberlet et al., 2000); but also the aorta itself (systemic-pressure) may be compromised. Therfore, aortic

Most interestingly, we used computational fluid dynamics on mathematic modeling to compare flow phenomena of the spiral and nonspiral models (Figure 11). The functional superiority of the spiral over the nonspiral model is demonstrated because the blood flow

inside is more streamlined andthere is a smaller power loss ratio (Tang et al., 2001).

Lecomptel Model Spiral Model

**Right**

Fig. 11. The regions of minimal wall shear stress in purple color in the left panel are the site where supravalvular PS is prone to occur. On the contrary, there is no purple color in spiral

Spiral reconstruction would be beneficial to reduce supravalvular PS (Figure 10), supraaortic stenosis (Figure 4), and aortic neocoarctation (Figure 3 & 10). This is to say, the great arteries are normally in a spiral fashion, thus the MPA bifurcation avoids being compressed by the aorta, and the arch window is kept wide open by the adjacent pulmonary bifurcation. Potential limitations of our proposal merit consideration. First, the number of cases studied is small. Second, this is aretrospective study, and longer-term follow-up is mandatory to ascertain the benefit or possible drawbacks of spiral ASO over non-spiral ASO. Subvalvular PS still occurred, but mainly related to infundibular hypertrophy, and not supravalvular as in the non-spiral group. Subvalvular PS was probably related to an unusual coronary artery traversing the infundibulum (Chiu et al., 2010), which may be associated with a thick subaortic muscle bundle that can produce subpulmonary narrowing after ASO (Kurosawa, 1991). It might be questioned whether our technique will reproduce the old dire complicarions that was avoided by Lecompte maneuver such as compression on the transferred coronary arteries (Chiu et al., 2010, discussant). Acually secure coronary redirection like our method is more important than a mere absence of a near-by low-pressure vessel like pulmonary artery. The so-called complications are not seen in our series. Any method that redirect

model. Wall shear stress is more uniform and the blood flow is more in streamline.

coronary safely can be done in combination with spiral arterial trunk reconstruction.

Our previous publications (Chiu et al., 2000b; Chiu et al., 2002b; Tang et al., 2001) were cited by Dr. Corno at the 86th annual meeting of American Association for Thoracic Surgery, in a symposium about the potential implications of the helical heart to congenital heart disease (Corno, 2006a; Corno, 2006b). In 2001, following we presented our spiral arteial switch at National Cardiovascular Center in Osaka, Japan, Dr. Hideki Uemura named our procedure as "arterial Senning". He is affiliated now with Royal Brompton Hospital in London, UK.

**Inferior**

neocaorctation was reported (Muster et al., 1987; Serraf et al., 1995).

Minimal wall shear stress

Lecompte et al., 1982). Review of surgical treatment for TGA reveals that once the functional implications of two normal ventricles were well known, nobody selected the right ventricle as the systemic pumping chamber. The rationale of spiral reconstruction (Figure 10) at the great arterial and ventricular level includes (1) widening of arch window to avoid aortic neocoarctation (Figure 3), (2) widening of main to branch PA angle to facilitate PA growth (Figure 2, Chen et al., 2007), and (3) avoiding PA compression by the aorta posteriorly.

Fig. 10. Spiral relationship of the great arteries in normal heart (C) was not restored by conventional ASO with Lecompte maneuver (B) on TGA (A). With spiral ASO (D), pulmonary bifurcation is free from posterior compression by high-pressure aorta. Note that acute angulation of arch that is present after Lecompte maneuver (B) is absent in other three illustrations. Acute angulation from the main pulmonary artery to its branches is also present in A and B, but not in C and D, in which main to branch pulmonary artery angle is wider and smoother. Reprinted with permission from Elsevier (Eu J Cardiothorac Surg 2010;37:1239-45).

To facilitate PA growth, mobilise the posterior inclination of the proximal MPA and restore its 'normal' position (Figure 10C) to direct a natural and smooth blood flow into the PA in the neonatal period is the surgical principle. Thus widening this acute angle to both PA in spiral fashion (Figure 10D) is more helpful than the Lecompte maneuver with a huge patch (Figure 10B) for promoting PA growth (Norwood et al., 1988; Paillole et al., 1988; Lupinetti et al., 1992). The systemic high-pressure ascending aorta may compress the neo-MPA from its posterior end towards its anterior end. Insufficient dissection of the distal PAs was suggested to explain this flattened (oval-shaped) MPA in 1988 (Wernovsky et al., 1988), but a later study showed that the cross section of the MPA is still flattened 6 to 22 months after conventional ASO with hilar dissection and the Lecompte maneuver (Massin et al., 1998).

Lecompte et al., 1982). Review of surgical treatment for TGA reveals that once the functional implications of two normal ventricles were well known, nobody selected the right ventricle as the systemic pumping chamber. The rationale of spiral reconstruction (Figure 10) at the great arterial and ventricular level includes (1) widening of arch window to avoid aortic neocoarctation (Figure 3), (2) widening of main to branch PA angle to facilitate PA growth

(Figure 2, Chen et al., 2007), and (3) avoiding PA compression by the aorta posteriorly.

Fig. 10. Spiral relationship of the great arteries in normal heart (C) was not restored by conventional ASO with Lecompte maneuver (B) on TGA (A). With spiral ASO (D),

2010;37:1239-45).

pulmonary bifurcation is free from posterior compression by high-pressure aorta. Note that acute angulation of arch that is present after Lecompte maneuver (B) is absent in other three illustrations. Acute angulation from the main pulmonary artery to its branches is also present in A and B, but not in C and D, in which main to branch pulmonary artery angle is wider and smoother. Reprinted with permission from Elsevier (Eu J Cardiothorac Surg

To facilitate PA growth, mobilise the posterior inclination of the proximal MPA and restore its 'normal' position (Figure 10C) to direct a natural and smooth blood flow into the PA in the neonatal period is the surgical principle. Thus widening this acute angle to both PA in spiral fashion (Figure 10D) is more helpful than the Lecompte maneuver with a huge patch (Figure 10B) for promoting PA growth (Norwood et al., 1988; Paillole et al., 1988; Lupinetti et al., 1992). The systemic high-pressure ascending aorta may compress the neo-MPA from its posterior end towards its anterior end. Insufficient dissection of the distal PAs was suggested to explain this flattened (oval-shaped) MPA in 1988 (Wernovsky et al., 1988), but a later study showed that the cross section of the MPA is still flattened 6 to 22 months after conventional ASO with hilar dissection and the Lecompte maneuver (Massin et al., 1998). After the Lecompte maneuver, not only flattened anterior MPA (low-pressure), branch PA stenosis, and supravalvular PS at pulmonary bifurcation can occur as a consequence of posterior compression of MPA by the ascending aorta (Williams et al., 1997; Gutberlet et al., 2000); but also the aorta itself (systemic-pressure) may be compromised. Therfore, aortic neocaorctation was reported (Muster et al., 1987; Serraf et al., 1995).

Most interestingly, we used computational fluid dynamics on mathematic modeling to compare flow phenomena of the spiral and nonspiral models (Figure 11). The functional superiority of the spiral over the nonspiral model is demonstrated because the blood flow inside is more streamlined andthere is a smaller power loss ratio (Tang et al., 2001).

Fig. 11. The regions of minimal wall shear stress in purple color in the left panel are the site where supravalvular PS is prone to occur. On the contrary, there is no purple color in spiral model. Wall shear stress is more uniform and the blood flow is more in streamline.

Spiral reconstruction would be beneficial to reduce supravalvular PS (Figure 10), supraaortic stenosis (Figure 4), and aortic neocoarctation (Figure 3 & 10). This is to say, the great arteries are normally in a spiral fashion, thus the MPA bifurcation avoids being compressed by the aorta, and the arch window is kept wide open by the adjacent pulmonary bifurcation. Potential limitations of our proposal merit consideration. First, the number of cases studied is small. Second, this is aretrospective study, and longer-term follow-up is mandatory to ascertain the benefit or possible drawbacks of spiral ASO over non-spiral ASO. Subvalvular PS still occurred, but mainly related to infundibular hypertrophy, and not supravalvular as in the non-spiral group. Subvalvular PS was probably related to an unusual coronary artery traversing the infundibulum (Chiu et al., 2010), which may be associated with a thick subaortic muscle bundle that can produce subpulmonary narrowing after ASO (Kurosawa, 1991).

It might be questioned whether our technique will reproduce the old dire complicarions that was avoided by Lecompte maneuver such as compression on the transferred coronary arteries (Chiu et al., 2010, discussant). Acually secure coronary redirection like our method is more important than a mere absence of a near-by low-pressure vessel like pulmonary artery. The so-called complications are not seen in our series. Any method that redirect coronary safely can be done in combination with spiral arterial trunk reconstruction.

Our previous publications (Chiu et al., 2000b; Chiu et al., 2002b; Tang et al., 2001) were cited by Dr. Corno at the 86th annual meeting of American Association for Thoracic Surgery, in a symposium about the potential implications of the helical heart to congenital heart disease (Corno, 2006a; Corno, 2006b). In 2001, following we presented our spiral arteial switch at National Cardiovascular Center in Osaka, Japan, Dr. Hideki Uemura named our procedure as "arterial Senning". He is affiliated now with Royal Brompton Hospital in London, UK.

Restoration of Transposed Great Arteries With or Without Subpulmonary Obstruction to Nature 183

Chen, YS., Chiu, IS., Huang, SC. & Chen, SJ. (2010). Valve-sparing procedure and

Chiu, IS., Anderson, RH., Macartney, FJ. & et al. (1984). Morphologic features of an intact

Chiu, IS., Chu, SH., Wang, JK. & et al. (1995). Evolution of coronary artery pattern according

transposition of the great arteries, *J Am Coll Cardiol* 26:250-8, ISSN 0735-1097 Chiu, IS., Chou, TF., Lin, SF. & et al. (1996). Utilization of the aortic flap above the facing commissure in arterial switch operations, *J Card Surg* 11:187-91, ISSN 0886-0440 Chiu, IS., Wang, JK., Wu, MH. & et al. (1996). Angiographic evidence of long-axis rotation in

Chiu IS, Wu MH, Chang CI. & et al. (1997). Clinical implications of short-axis

Chiu, IS., Wu, CS., Wang, JK. & et al. (2000). Influence of aortopulmonary rotation on the

Chiu, IS., Wu, SJ., Chen, MR. & et al. (2000). Modified arterial switch operation by spiral

Chiu, IS., Chen, SJ., Wu, SJ. & et al. (2001). Modified arterial switch operation by sharing the

Chiu, IS., Wu, SJ., Chen, MR. & et al. (2002). Anatomic relationship of the coronary orifice

Chiu, IS., Wang, JK. & Wu, MH. (2002). Spiral arterial switch operation in transposition of the great arteries. *J Thorac Cardiovasc Surg* 124:1050-2, ISSN 0022-5223 Chiu, IS., Wu, SJ., Chen, SJ., Wang, JK., Wu, MH. & Lue, HC. (2003). Sequential diagnosis of

Chiu, IS., Huang, SC., Wu, SJ. & et al. (2009). Surgical treatment of anomalies with

Chiu, IS., Huang, SC., Chen, YS. & et al. (2010). Restoring the spiral flow of nature in transposed great arteries. *Eu J Cardiothorac Surg* 37:1239-45, ISSN 1010-7940 Chiu, IS., Chen, MR., Huang, SC. & et al. (2011). Restoration of Transposed Great Arteries to

an extracardiac conduit. *Ann Thorac Surg* 87:1220-6, ISSN 0003-4975

the great arteries, *Cathet Cardiovasc Diagn* 39:21-30, ISSN 0098-6569

transposition, *Am J Cardiol* 53:1633-8, ISSN 0002-9149

0022-5223

0167-5273

ISSN 0002-9149

ISSN 0003-4975

*Cardiovasc Surg* 123:350-2, ISSN 0022-5223

arteries. *Ann Thorac Surg* 75:422-9, ISSN 0003-4975

6646

2150-1351

de-Lecompte maneuver in patients with late aortic regurgitation and aortic aneurysm after arterial switch operation, *J Thorac Cardiovasc Surg* 140:1191-2, ISSN

ventricular septum susceptible to subpulmonary obstruction in complete

to short-axis aortopulmonary rotation: A new categorization for complete

addition to short-axis aortopulmonary rotation: Its implication in transposition of

aortopulmonary rotation on juxtacommissural origin of the coronary artery in transposition of the great arteries and surgical strategy, *J Card Surg* 12:23-31, ISSN

anomalous coronary artery pattern in tetralogy of Fallot. Am J Cardiol 85:780-4,

reconstruction of the great arteries in transposition, *Ann Thorac Surg* 69:1887-92,

common wall between the great arteries.; *J Formos Med Assoc* 100:77-82, ISSN 0929-

and truncal valve in truncus arteriosus and their surgical implication. *J Thorac* 

coronary arterial anatomy in congenitally corrected transposition of the great

discontinuity between the right ventricle and the pulmonary arteries without use of

Nature, *World Journal for Pediatric and Congenital Heart Surgery*, 2:287-295, ISSN

We are grateful to his thoughtful nomenclature for our procedure. We dare not call the spiral arterial switch this name by ourselves before his proposal.
