Asian Cardiovasc Thorac Ann 2004;12:336-340
© 2004 Asia Publishing EXchange Ltd
Early and Midterm Results with The Ross Procedure: A Study of The First 31 Cases
George Matalanis, FRACS,
Manoj Durairaj, MCh,
Pallav Shah, MCh,
Brian Buxton, FRACS
Department of Cardiac Surgery, Austin Hospital, Melbourne, Australia
For reprint information contact: George Matalanis, FRACS Tel: 61 3 9457 1071 Fax: 61 3 9457 6320 Email: gmatalan{at}bigpond.net.au Department of Cardiac Surgery, Austin Hospital, Studley Road, Heidelberg, Victoria 3084, Australia.
 |
ABSTRACT
|
|---|
Between 1994 and 2002, 31 patients underwent the Ross procedure by a single surgeon. The mean age was 42 years (24–61), 87% were male and 61% were in New York Heart Association (NYHA) class III–IV. Pure aortic stenosis (AS) was present in 32% of patients, pure aortic regurgitation (AR) in 22% and mixed disease in the rest. The aortic valve was bicuspid in 93.5% of the patients. Autograft implantation was by full root replacement in all cases. Concomitant cardiac surgical procedures were carried out in 10/31 (32%). All patients had at least annual clinical and echocardiographic follow-ups. There was one early death (3%). Overall patient survival was 92.7% at 1 year and 86.1% at 5 years. Twenty-eight (96.55%) were in NYHA class I. Echocardiographic follow-up revealed none to trivial AR in 24/29 (82.75%) and mild AR in 4/29 (13.7%). There was no autograft re-operation before 5 years. The mean gradient across the autograft was low (< 4 mm Hg). There were no incidences of endocarditis or thromboembolism. None of the patients required anticoagulation. Our early experience with the Ross procedure has shown good results in relation to early and midterm morbidity, mortality, autograft, and homograft function.
|
INTRODUCTION
|
|---|
The first pulmonary autograft replacement of the aortic and mitral valves was reported by Donald Ross in 1967.1 The procedure was performed on the presumption that living autogenous pulmonary tissue is the best available replacement for diseased valves. Thirty-five years and several publications later, it has been demonstrated that replacement of the diseased aortic valve with a pulmonary autograft is probably the ideal solution in a select group of patients. As the pulmonary autograft is a living structure it offers the unique advantages of unlimited durability,2,3 superior hemodynamics,4 and relative freedom from prosthetic valve and anticoagulant complications.5 Furthermore, the autograft has the potential for growth in children, and for self repair.6,7 These very attractive features are counterbalanced by the increased technical difficulty of the operation and the risk of early and late valve failure requiring re-operation for either the autograft,6 or the right ventricular outflow tract (RVOT) reconstruction valve.8 The pressure on the surgeon to produce consistently good results from the very first operation is great, as more traditional and technically less demanding alternatives are available to most Ross procedure candidates.
A key principle underlying this technique is the maintenance of normal autograft spatial geometry after the translocation to the aortic position to ensure proper leaflet coaptation and avoid autograft insufficiency. Most Ross operations are now performed as a full aortic root replacement with a free-standing autograft, the so-called "modified Ross procedure".9 We present our early and midterm results of the first 31 patients operated on by a single surgeon (GM) using this technique.
|
MATERIALS AND METHODS
|
|---|
Patient population: Between 1994 and 2003, 31 patients underwent aortic valve and root replacement with a pulmonary autograft. There were 27 men and 4 women. Mean age was 42 years (range 24–61 years). The primary diagnosis was pure aortic stenosis (AS) in 10 patients (32%), pure aortic regurgitation (AR) in 7 patients (23%), and mixed disease with predominant stenosis in 14 patients (45%). The valve pathology was bicuspid in 29 (93.5%) patients and tricuspid in 2 patients (6.5%). There were 2 patients with infective endocarditis; one with active and one with healed endocarditis. The functional class was NYHA II in 12 (39%) patients, NYHA III in 16 (52%) and NYHA IV in 3 (9%) patients.
Three patients had previous cardiac surgical procedures: two patients had a previous repair of coarctation of the aorta and one patient had previous aortic valve commissurotomy. All patients underwent full preoperative transthoracic echocardiographic (TTE) evaluation including an estimation of pulmonary valve size. Transesophageal echocardiography (TEE) was used routinely intraoperatively. Concomitant procedures were performed in 10 (32%) patients (Table 1
). The two patients who underwent single vessel coronary artery bypass received a left internal thoracic artery graft to the left anterior descending artery.
Operative Technique: Standard cardiopulmonary bypass with cold blood cardioplegia (antegrade and retrograde) and moderate systemic hypothermia (30°C to 32°C) was used. An appropriate size pulmonary homograft (based on preoperative echo pulmonary valve sizing and patient size) was thawed, trimmed and checked for defects or harvesting injuries prior to committing to the procedure. The aortic root was transected at the sinotubular junction and the right and left coronary arteries mobilized with generous aortic buttons. The aortic valve leaflets and all calcium were removed and annulus sized with a Medtronic Freestyle® sizer (Medtronic Inc. Minneapolis, USA). Three equidistant points on the aortic annulus (lining up with native commissures when appropriate) were marked with sutures utilizing the markings on the Medtronic Freestyle® sizer to facilitate subsequent symmetrical autograft insertion. The pulmonary autograft valve was initially inspected for anatomical and functional defects through a transverse arteriotomy just proximal to the pulmonary bifurcation. If acceptable, it was harvested including 2–3 cm of RVOT proximal to the pulmonary valve, preserving as much adventitia as possible onto the autograft to improve the suture holding strength of this rather fragile structure, and exercising care to preserve septal perforators.
The pulmonary autograft was sized without overstretching using the same cylindrical sizer. If the aortic annulus was one size (2 mm) larger than the autograft annulus, an annular reduction procedure was added. For minor discrepancies, sub-commissural plication sutures were placed at either side of the fibrous left ventricular outflow tract (LVOT). For greater than 4 mm differences, a formal annuloplasty of the whole fibrous LVOT was performed with an appropriate length of Dacron Strip (DeBakey® double velour fabric Impa Inc. Tempe, AZ, USA).
The autograft was always inserted as a full root replacement. The proximal suture line was usually completed using the interrupted suture technique for greater accuracy. The autograft was oriented with the left-facing sinus aligned with the aortic noncoronary sinus, so that the thinnest part of the autograft would not need to be used for the coronary anastomoses. The coronary buttons were re-implanted with 6/0 polypropylene into appropriately positioned holes in the autograft, taking care to avoid tension or kinking of the coronaries. Remnants of the native aortic root were included in the autograft to ascending aortic suture line, both to improve hemostasis and to reduce the tendency for future dilatation of the autograft’s sinotubular junction or sinuses. If the ascending aorta and/or arch were dilated, they were replaced with a Dacron graft, and direct anastomosis of the Dacron tube to the autograft was performed. Finally, the right ventricular outflow tract was reconstructed with the previously prepared pulmonary homograft. In all cases, the competence of the autograft aortic and pulmonary homograft valves was confirmed on TEE.
Follow-up: Many patients had pre-discharge TTE, to document normal aortic and pulmonary valve function and to exclude pericardial effusion. Clinical review was performed at 6 weeks, 3–6 months, and annually thereafter. Clinical follow-up spanned 73.9 patient years, with a mean of 2.4 years (range 0–8.2 years). Patients underwent an echo at 3–6 months and annually thereafter.
Statistical Analysis: Kaplan-Meier analysis was used to estimate overall survival, freedom from autograft re-operation, re-operation free survival, freedom from cardiac mortality, freedom from more than mild aortic regurgitation and event free survival (event=death, re-operation, endocarditis or bleeding/thromboembolic episodes).
|
RESULTS
|
|---|
There was one perioperative death (3%). This was the very first patient, a 50-year-old man with a history of Von Willebrand’s disease. He experienced excessive bleeding both intra and postoperatively. Despite multiple explorations and administration of numerous coagulation factors, he continued to experience a refractory coagulopathy and died 14 hours postoperatively. Significant postoperative complications occurred in 6 (20.6%) of the 29 patients (Table 2). One patient required intraoperative conversion to a mechanical conduit due to bleeding from the proximal suture line of a rather fragile autograft.
There were two late deaths at 6 and 19 months after surgery. The first patient died as a passenger in a motor vehicle accident. This patient was mentally subnormal, undertook risk-taking behaviour and was poorly compliant with medication. It was felt that under those circumstances, a Ross procedure would avoid potential complications with oral anticoagulation. The other was found drowned unattended in a swimming pool. While the post-mortem did not reveal any valvular or cardiac complications, an arrhythmia cannot be excluded. Prior to death, both patients had normally functioning autograft and homograft function on clinical and echocardiographic examination. The Kaplan-Meier overall patient survival was 92.74% (82.98%, 100% 95% CI) at 1 year and 86.12% (70.67 %, 100% 95% CI) at 5 years (Figure 1).
View larger version (10K):
[in this window]
[in a new window]
|
Figure 1. Kaplan-Meier overall survival curve showing 86.12% patient survival at 5 years (70.67%, 100% 95% CI).
|
|
Severe aortic regurgitation (AR) developed in one patient (the second patient in the series) in whom moderate AR was noted on the 1-month echo. The patient was re-operated on 5 years and 3 months post surgery. At surgery, one of the autograft leaflets had a clean laceration in its belly, suggestive of accidental inclusion in one of the implantation sutures. The pulmonary autograft was replaced uneventfully with a mechanical valve. The actuarial freedom from autograft re-operation is summarized in Figure 2. The only other re-operation was a patient who had a Dacron graft replacement of an ascending aortic aneurysm and who developed high gradients in the Dacron graft due to a kink. Although she remained asymptomatic with normal autograft function, she underwent re-operation at two years postoperatively due to the persistence of left ventricular hypertrophy and concern about accelerated wear on the autograft and coronary arteries. She had an uneventful resection of the redundant Dacron and end-to-end anastomosis.
On echocardiography, 24 patients (82.75%) had nil or trace AR, and 4 (13.7%) had mild AR. The only patient with more than mild AR was re-operated on as mentioned above. Apart from the latter, no patient demonstrated progression in the degree of AR during the follow up period. The mean trans-autograft gradient was low (< 4 mm Hg) and also remained stable throughout.
Twenty-eight of 29 patients (96.55%) had minimal gradients across their homografts. Four patients had gradients between 15 and 30 mm Hg. The peak of the gradients occurred between 9 and 60 months postoperatively, and remained stable thereafter. None of these patients were symptomatic nor showed signs of RV dilatation or hypokinesis on echo. None of the homografts had more than mild regurgitation.
There were no incidences of endocarditis or thromboembolism in this series. None of the patients required anticoagulation and 9/29 patients (31%) were on aspirin for reasons other than their valves. One patient with mild RV dysfunction required an implantable defibrillator for 2 episodes of conscious ventricular tachycardia 36 months postoperatively. The actuarial freedom from all events (death, re-operation, thromboembolic and bleeding episodes) was 86.2% (74%, 99% 95% CI) at 1 year and 80.1% (65%, 98% 95% CI) at 5 years (Figure 3). Twenty-five (86.2%) were in NYHA class I, and 4 (13.7 %) were in Class II on last follow-up.
View larger version (11K):
[in this window]
[in a new window]
|
Figure 3. Event-free survival of Ross patients. Event is defined as death, re-operation, bleeding or thromboembolic event.
|
|
|
DISCUSSION
|
|---|
Legitimate concerns about the widespread use of the Ross procedure in aortic valve replacement remain to this day. Powerful arguments include the increased operative complexity and risk, and the potential for problems developing in both the aortic and pulmonary positions in what originated as single valve pathology. The stakes are even higher as the ideal candidates for the operation are the young and relatively well patients who have the most to lose in case of operative complications. On the other hand, a well-performed Ross procedure can represent a near normal restitution of cardiac physiology, and it achieves this result without the need for anticoagulation.
Our experience with the first 31 patients lends support to the latter view. The single operative mortality (3%) is in line with other series.8,10,11 Like other reports, this was early in our experience, and reflects the "steep learning curve". Along similar lines the only re-operation for autograft failure was our second case, most likely a technical error resulting in laceration of one of the autograft leaflets. Although the postoperative morbidity was significant, it was mainly represented by re-exploration for bleeding. This was largely the consequence of coagulopathy and had a benign course. It is our impression that this is becoming less common, probably due to technical refinements and the routine use of antifibrinolytic agents. Our 2 late deaths at 6 and 19 months postoperatively, were not valve related. The first was a passenger in a high-speed collision. The second was found drowned in a swimming pool. Although both had completely normal valve function on echo prior to death, an arrhythmia cannot be excluded in the case of the drowning.
All patients were in NYHA class I or II at last follow-up. None has greater than mild AR, and all have mean trans-autograft gradients of < 4 ± 2 mm Hg on serial echocardiography. There have been no episodes of thromboembolism, anticoagulant hemorrhage or prosthetic valve endocarditis.
Pulmonary homograft stenosis, defined as a resting gradient of > 30 mm Hg is an infrequent but well recognised problem.12 It generally occurs in the first 1–2 years postoperatively and usually involves the pulmonary trunk in a tubular fashion rather than the pulmonary valve or the anastomosis. Potential immunologically mediated6 mechanisms are evident, although no clear relationship has been found between its frequency and the presence of positive panel reactive antibodies to a wide range of HLA antigens.12,13 Four of our patients developed maximum gradients between 15 and 30 mm Hg at intervals of 9 and 60 months postoperatively and have remained steady since. As none of them have developed symptoms or evidence of right heart strain we have not needed to intervene.
The full root replacement technique of autograft implantation, by minimizing the disturbance to the native 3-dimensional geometry, is associated with the lowest incidence of autograft incompetence.11 We adopted this technique uniformly in our series and did not see any case of significant autograft regurgitation as a result of geometric distortion. There is still a concern that late autograft regurgitation may develop due to annular and/or sinotubular dilatation, especially under the influence of systemic pressures. We have followed the recommendations of others of performing a reduction annuloplasty when there is significant discordance between the autograft and aortic annuli14,15 and of retaining the native aortic root skeleton (that remaining after removal of the coronary artery buttons) around the autograft.16 In addition, when the ascending aorta is significantly dilated, we have no hesitation in replacing it with an appropriately sized Dacron graft. To date we have not observed any major increase in any of the autograft dimensions.
The role of the Ross procedure in active infective endocarditis remains controversial. Criticism levelled against its use in such acutely ill high-risk patients includes increased bleeding and a propensity for myocardial dysfunction. Notwithstanding, excellent results with very low re-infection rates17,18 have been achieved. Our experience is limited to two cases of endocarditis, and only one in the florid acute phase. Both procedures were uneventful with no re-infection. We see no reason to deny the benefits of the Ross procedure to otherwise good candidates just on account of acute endocarditis, and view the living autograft as the best defence against re-infection.
For this operation, patient selection is critical. The ideal patient is young, fit and keen to pursue an active lifestyle, without the encumbrance of Warfarin. We have avoided using the Ross procedure in older patients (more than 60 years), or those with poor LV function, those requiring multiple coronary grafts, other valve procedures or extensive aortic arch replacements. We also do not consider those with inflammatory aortic valve diseases, such as rheumatic fever, to be suitable.
|
CONCLUSION
|
|---|
Our series reflects an eight-year follow-up experience with 31 cases of the Ross procedure. The study has shown very good results with regard to early and midterm morbidity, mortality, autograft and homograft function. Although it is a complex operation with an initial steep learning curve, it is associated with excellent hemodynamics and freedom from anticoagulation. Careful patient selection and close attention to technical details have resulted in a low incidence of technical failure. We believe it is the procedure of choice for young active patients desiring an aortic valve replacement allowing them to lead a productive and unrestricted lifestyle.
|
ACKNOWLEDGMENTS
|
|---|
Dr. Ian Gordon, PhD AStat, Associate Professor Director, Statistical Consulting Centre, University of Melbourne, Melbourne, Victoria, Australia.
Dr. Elizabeth Jones, Cardiologist, Department of Cardiology, Austin Hospital, Melbourne, Victoria, Australia.
|
REFERENCES
|
|---|
- Ross DN. Replacement of aortic and mitral valves with a pulmonary autograft. Lancet
1967;2:956–8.[Medline]
- Chambers JC, Somerville J, Stone S, Ross DN. Pulmonary autograft procedure for aortic valve disease: long-term results of the pioneer series. Circulation
1997;96:2206–14.[Abstract/Free Full Text]
- Ross D, Jackson M, Davies J. The pulmonary autograft – a permanent aortic valve. Eur J Cardiothorac Surg
1992;6:113–7.[Abstract]
- Oury JH, Doty DB, Oswaldt JD, Knapp JF, Mackey SK, Duran CM. Cardiopulmonary response to maximal exercise in young athletes following the Ross procedure. Ann Thorac Surg
1998;66:S153–4.
- Gerosa G, McKay R, Ross DN. Replacement of the aortic valve or root with a pulmonary autograft in children. Ann Thorac Surg
1991;51:424–9.[Abstract]
- Elkins RC, Lane MM, McCue C. Pulmonary autograft reoperation: incidence and management. Ann Thorac Surg
1996;62:450–5.[Abstract/Free Full Text]
- Elkins RC, Knott-Craig CJ, Ward KE, McCue C, Lane MM. Pulmonary autograft in children: realized growth potential. Ann Thorac Surg
1994;57:1387–94.[Abstract]
- Matsuki O, Okita Y, Almeida RS, McGoldrick JP, Hooper TL, RoblesA. Two decades’ experience with aortic valve replacement with pulmonary autograft. J Thorac Cardiovasc Surg
1988;95:705–11.[Abstract]
- Stelzer P, Jones DJ, Elkins RC. Aortic root replacement with pulmonary autograft. Circulation
1989;80:III209–13.
- Stelzer P, Weinrauch S, Tranbaugh RF. Ten years of experience with the modified Ross procedure. J Thorac Cardiovasc Surg
1998;115:1091–100.[Abstract/Free Full Text]
- Oury JH, Hiro SP, Maxwell JM, Lamberti JJ, Duran CM. The Ross Procedure: current registry results. Ann Thorac Surg
1998;66:S162–5.
- Carr-White GS, Kilner PJ, Hon JK, Rutledge T, Edwards S, Burman ED, et al. Incidence, location, pathology, and significance of pulmonary homograft stenosis after the Ross operation. Circulation
2001;104:I16–20.
- Bechtel JF, Bartels C, Schmidtke C, Skibba W, Muller-Steinhardt M, Kluter H. Does histocompatibility affect homograft valve function after the Ross procedure? Circulation
2001;104:I25–8.
- Klena JW, Shweiki E, Mahaffey HW, Woods EL, Benoit CH, Gilbert CL. Annuloplasty and aortoplasty as modifications of the Ross procedure for the correction of geometric mismatch. J Heart Valve Dis
2000;9:195–9.[Medline]
- David TE, Omran A, Webb G, Rakowski H, Armstrong S, Sun Z. Geometric mismatch of the aortic and pulmonary roots causes aortic insufficiency after the Ross procedure. J Thorac Cardiovasc Surg
1996;112:1231–9.[Abstract/Free Full Text]
- Skillington PD, Fuller JA, Grigg LE, Yapanis AG, Porter GF. Ross procedure. Inserting the autograft using a fully supported root replacement method; techniques and results. J Heart Valve Dis
1999;8:593–600.[Medline]
- Oswalt JD, Dewan SJ, Mueller MC, Nelson S. Highlights of a ten-year experience with the Ross procedure. Ann Thorac Surg
2001;71:S332–5.[Abstract/Free Full Text]
- Prat A, Saez de Ibarra JI, Vincentelli A, Decoene C, Fabre OH, Jegou B, et al. Ross operation for active culture-positive aortic valve endocarditis with extensive paravalvular involvement. Ann Thorac Surg
2001;72:1492–6.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
J. J.M. Takkenberg, L. M.A. Klieverik, P. H. Schoof, R.-J. van Suylen, L. A. van Herwerden, P. E. Zondervan, J. W. Roos-Hesselink, M. J.C. Eijkemans, M. H. Yacoub, and A. J.J.C. Bogers
The Ross Procedure: A Systematic Review and Meta-Analysis
Circulation,
January 20, 2009;
119(2):
222 - 228.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
