Pulmonary Autograft Mitral Valve Replacement (Ross II Procedure): More Than Two Decades of Follow-up From a Single Center in the United States

Abstract

Background

We aim to elucidate the outcomes of patients who underwent a Ross II procedure at our institution.

Methods

Eight patients underwent a Ross II procedure at our institution between 2002 and 2008. Electronic medical records were queried to obtain demographic, procedural, and recent follow-up data.

Results

The mean follow-up time for this cohort is 17 ± 6.2 years, and the pulmonary autograft has stayed in place, on average, for 9.6 ± 8.3 years. At last follow-up, the average pulmonary autograft mean gradient was 6.1 ± 2.4 mm Hg. No new mitral valve reoperations have been required since our last publication. Three early patients developed at least moderate pulmonary homograft (PH) regurgitation, although none of the patients in this cohort had significant PH stenosis at their last follow-up. None of the patients with a pulmonary homograft have required reintervention to date. There have been three deaths in this cohort, two of which were outlined in our previous study. More recently, Patient 2—who underwent a Ross II procedure to surgically correct rheumatic mitral valve stenosis and regurgitation—died 17 years after her operation due to complications of diabetic ketoacidosis.

Conclusion

To our knowledge, this study represents the longest follow-up of a Ross II cohort ever reported, and the largest experience with the procedure in North America. The Ross II procedure offers a durable tissue alternative to conventional prosthetic valves for mitral valve replacement, particularly in younger patients, women of childbearing age, and those who have a contraindication to anticoagulation.

Keywords

autograft congenital heart surgery heart valve mitral valve disease mitral valve replacement outcomes (includes mortality morbidity)Ross operation

Introduction

Finding a suitable option for mitral valve replacement (MVR) can be challenging, particularly in younger patients, women of childbearing age, and in patients for whom anticoagulation is not a viable option. Mechanical valves necessitate lifelong anticoagulation, while bioprosthetic valves are noted to have limited durability. Thus, MVR with a pulmonary autograft (Ross II procedure) offers a durable tissue valve that has the potential to obviate the aforementioned pitfalls of commercially available prosthetic options.¹ We have previously outlined what is to our knowledge, the largest experience with this procedure in North America,² as well as the long-term outcomes from our dataset,³ and described the case with which we have had the longest follow-up.⁴ We now report the outcomes of patients who underwent a Ross II procedure at our institution after nearly two decades of follow-up.

Patients and Methods

This study was approved by Indiana University School of Medicine's Institutional Review Board in 2025 (#25706). Institutional databases were queried to identify patients who had undergone a Ross II procedure at our institution. Ten patients were initially identified, eight of whom underwent MVR using a pulmonary autograft with the technique we have previously described.^2,3 The remaining two patients underwent MVR with aortic homografts using the same technique we described for the Ross II and were thus excluded from the Ross II cohort for this article. Electronic medical records were queried to obtain demographic, procedural, and recent follow-up data.

Given the timing of the operations, echocardiographic parameters to evaluate Ross II and pulmonary homograft (PH) valve function were captured using cardiology reports from the echocardiographic evaluations of our patients. Thus, the degree of valve stenosis and/or insufficiency for both the Ross II and PH valves was made by the cardiologists at the time of the echocardiogram. Pulmonary homografts were considered to be dysfunctional when moderate stenosis and/or regurgitation developed.

Results

Table 1 describes the demographic data for our cohort. Eight patients (seven females, one male) underwent a Ross II procedure at our institution between 2002 and 2008. The mean age at operation was 29 ± 13 years (range: 14-46 years). Five of 8 patients (63%) were noted to have a preoperative supraventricular arrhythmia, and one patient was quite medically complex with coronary artery disease, chronic kidney disease, diabetes mellitus (type 1), and peripheral arterial disease necessitating below knee amputation of the right lower extremity. Indications for the Ross II procedure included rheumatic mitral valve disease (five patients, 63%), outgrown mitral prosthesis (one patient, 13%), and severe isolated mitral regurgitation (MR) (two patients, 25%). Three of the 8 patients had congenital heart defects (Shone's complex with parachute mitral valve [one patient] and partial atrioventricular septal defect (pAVSD) [two patients]). Previous mitral valve procedures included surgical mitral valve repair (two patients, 25%), balloon mitral valvuloplasty (one patient, 13%), mechanical MVR (one patient, 13%), and pAVSD repair (two patients, 25%) (Table 2).

Table 1.

Patient Demographics.

Patient number	Gender	Comorbidities	Age (years)	Age at Ross II operation (years)	Indication for Ross II operation	Congenital heart disease
1	Male	AF	69	46	Rheumatic MS + MR	No
2	Female	PAH, AF, DM (type 2)	59	43	Rheumatic MS + MR	No
3	Female	SVT	36	14	Rheumatic MR	No
4	Female	None	42	19	Mechanical MS	Shone's complex, Parachute MV
5	Female	AF, Trisomy 21	38	16	Severe MR	pAVSD
6	Female	None	59	20	Rheumatic MS + MR	No
7	Female	CAD, CKD, DM (Type 1), PAD with prior right BKA	46	41	Rheumatic MS + MR	No
8	Female	AF, Hypothyroidism	46	29	Severe MR	pAVSD, Left SVC

Abbreviations: AF, atrial fibrillation; BKA, below knee amputation; CAD, coronary artery disease; CKD, chronic kidney disease; DM, diabetes mellitus; MR, mitral valve regurgitation; MS, mitral valve stenosis; MV, mitral valve; PAD, peripheral artery disease; PAH, pulmonary artery hypertension; pAVSD, partial atrioventricular septal defect; SVC, superior vena cava; SVT, supraventricular tachycardia.

Table 2.

Operative Details.

Patient number	Year of Ross II operation	Prior procedures (year)	Native annulus size (mm)	AG size (mm)	RV-PA conduit size (mm)	Total pump time (min)	Ischemic time (min)	Concomitant procedures
1	2002	None	NA	NA	28	242	82	No
2	2003	Balloon mitral valvuloplasty (1991)	NA	24	26	279	95	No
3	2002	MV repair (2002)	35	26	26	263	60	No
4	2002	Coarctation repair (1984), MVR (19 mm, mechanical; 1988), Balloon angioplasty coarctation (1996)	26	23	26	230	82	No
5	2003	pAVSD repair (1992)	32	22	26	264	71	Left-sided maze procedure
6	2003	MV repair (1998)	34	23	26	216	47	No
7	2008	None	NA	24	28	251	94	CABG × 1 (LIMA to LAD)
8	2008	pAVSD repair (NA)	30	20	30	352	127	Maze procedure, Ligation of left SVC

Abbreviations: AG, autograft; CABG, coronary artery bypass grafting; LAD, left anterior descending artery; LIMA, left internal mammary artery; MV, mitral valve; MVR, mitral valve replacement; NA, not available; PA, pulmonary artery; pAVSD, partial atrioventricular septal defect; RV, right ventricle; SVC, superior vena cava.

Mean total cardiopulmonary bypass time was 262 ± 41 min (range: 216-352 min), with a mean ischemic time of 82 ± 24 min (range: 47-127 min). Mean autograft size was 23 ± 1.9 mm (range: 20-26 mm), and the right ventricular outflow tract was most commonly reconstructed with a 26-mm decellularized PH (28 mm PH used in two patients, 30 mm PH used in one patient). Three patients underwent concomitant procedures: one required a left-sided maze procedure, one required a maze procedure and ligation of the left superior vena cava, and the third required a left internal mammary artery to the left anterior descending artery bypass graft (Table 2). Intraoperative transesophageal echocardiographic evaluation of the pulmonary autografts showed excellent function. Five patients had an intraoperative gradient ≤5 mm Hg, and six patients had no/trivial intraoperative pulmonary autograft regurgitation (Table 3). One patient required reoperation in the early postoperative period for mediastinal bleeding, but there were no other major complications in the perioperative or late periods. There were no perioperative or early (within 30 days) deaths (Table 4).

Table 3.

Echocardiographic Data.

Patient number	Intraoperative AG mean gradient (mm Hg)	Intraoperative degree of AG regurgitation	AG mean gradient at last follow-up (mm Hg)	Degree of AG regurgitation at last follow-up	Time to RV-PA conduit moderate regurgitation (years)	Time to RV-PA conduit severe regurgitation (years)
1	7	None	NA	Moderate	NO	NO
2	NA	Trivial	5	None	NO	NO
3	2	Mild	NA	Severe	NO	NO
4	4	None	7	Trivial	12	20
5	5	Trivial	9.6	Mild	10	NO
6	4	Mild	NA	Severe	NO	NO
7	NA	None	6	Mild	5	NO
8	2	None	3	Trivial	NO	NO

Abbreviations: AG, autograft; NA, not available; NO, not occurred; PA, pulmonary artery; RV, right ventricle.

Table 4.

Major Complications and Mortality.

Patient number	ANY major complication (excludes mortality)	Arrhythmia requiring pacemaker	Postoperative bleeding	Pericardial effusion/tamponade	Sternal wound infection	Stroke	VAD placement	Mortality
Patient number	ANY major complication (excludes mortality)	Arrhythmia requiring pacemaker	Postoperative bleeding	Pericardial effusion/tamponade	Sternal wound infection	Stroke	VAD placement	Early	Late
1	Yes	No	Yes (early)	No	No	No	No	No	No
2	No	No	No	No	No	No	No	No	Yes
3	No	No	No	No	No	No	No	No	No
4	No	No	No	No	No	No	No	No	No
5	No	No	No	No	No	No	No	No	Yes
6	No	No	No	No	No	No	No	No	No
7	No	No	No	No	No	No	No	No	Yes
8	No	No	No	No	No	No	No	No	No

Early complications/mortality are defined as those occurring within 30 days of the operation. VAD, ventricular assist device.

The mean follow-up time for this cohort is 17 ± 6.2 years (range: 5.3-23 years), and the pulmonary autograft has stayed in place, on average, for 9.6 ± 8.3 years (range: 1.0-23 years). At last follow-up, the average pulmonary autograft mean gradient was 6.1 ± 2.4 mm Hg (range: 3-9.6 mm Hg). As discussed in our previous publication, three of the patients early in our series required reoperation within 10 years at 1.0, 1.1, and 3.6 years, respectively for progressive autograft regurgitation (moderate in one patient, severe in two patients). Excluding cases where the pulmonary autograft was replaced, all remaining patients at last follow-up have, at maximum, mild autograft regurgitation. No new mitral valve reoperations have been required (Table 5).

Table 5.

Follow-up and Reoperation(s).

Patient Number	Total Follow- up (years)	AG Follow-up (years)	Reoperation(s) for MV	Reoperation(s) for PV	Other Reinterventions (year)
1	22	1	MVR (29 mm, mechanical)	No	No
2	17	17	No	No	RCA PCI (2014)
3	21	3.6	MVR (29 mm, mechanical)	No	No
4	23	23	No	No	No
5	10.5	10.5	No	No	No
6	21	1.1	MVR (25 mm, mechanical)	No	Heart Transplant (2019)
7	5.3	5.3	No	No	No
8	16	16	No	No	No

Abbreviations: AG, autograft gradient; MV, mitral valve; MVR, mitral valve replacement; PCI, percutaneous coronary intervention; PV, pulmonary valve; RCA, right coronary artery.

Three patients developed at least moderate PH regurgitation. Based on echocardiographic surveillance, patient 4 developed moderate PH regurgitation 12 years after her Ross II procedure, which progressed to severe PH regurgitation eight years later. Patients 5 and 7 were noted to have moderate PH regurgitation 10 and 5 years postoperatively, respectively. None of the patients in this cohort had significant PH stenosis at their last follow-up. Additionally, none of the patients in this series have required PH reoperation to date.

Two patients in the cohort have undergone catheter-based procedures after their Ross II procedure: one had a percutaneous coronary intervention and the other underwent an arrhythmia ablation. During the ablation procedure, the patient had a spontaneous left coronary dissection that resulted in left anterior descending artery occlusion, requiring subsequent heart transplantation. Both the catheter-based ablation and the heart transplant procedures were performed at an outside hospital.

There have been three deaths in this Ross II cohort, two of whom were outlined in our previous study (patients 2, 5, and 7). Briefly, the first mortality occurred due to right heart failure 10.5 years after her Ross II operation (patient 5). The pulmonary autograft was not implicated in her death. The second mortality occurred in a patient with diabetes mellitus (type 1) who died from diabetic ketoacidosis and sepsis 5.3 years after her Ross II procedure (patient 7). More recently, patient 2, who underwent a Ross II procedure to surgically correct rheumatic mitral valve stenosis and regurgitation died 17 years after her operation due to complications of diabetic ketoacidosis. Echocardiographic evaluation in the year of her death showed mild left atrial dilatation with a normally functioning pulmonary autograft in the mitral valve position (mean gradient = 5 mm Hg, no regurgitation).

The two patients who have had the longest reintervention-free follow-up after the Ross II operation were patients 4 and 8, both of whom are presently alive. Both patients are followed with echocardiographic surveillance. Most recently, Patient 4 had an echocardiogram 22 years after the Ross II operation (June 2024), which showed the pulmonary autograft to have a mean gradient of 7 mm Hg with trivial regurgitation and severe PH regurgitation. Patient 8 had an echocardiogram 16 years after the Ross II operation (April 2024), which showed the pulmonary autograft to have a mean gradient of 3 mm Hg with trivial regurgitation and less than moderate PH regurgitation. Neither of these patients have developed PH stenosis nor have they required transcatheter or surgical reintervention to date. Both of these patients are in New York Heart Association Functional Class I and their only medication is daily aspirin (81 mg).

Comment

The appeal of the Ross II procedure in most cases is the durability of the living tissue valve and the absence of a need for anticoagulation. Thus, enthusiasm for the Ross II procedure has largely come from developing countries, where these advantages would be particularly beneficial.^1,5–7 However, most of the research on this procedure is detailed in case reports and a few case series.⁵ The largest study on the Ross II procedure (N = 92) was published by Kabbani and colleagues in 2007, detailing their experience with the operation in Damascus, Syria. At five years after the Ross II procedure, the reported freedom from autograft degeneration, reoperation, and all-cause mortality was 93.4%, 94.2%, and 86.0%, respectively.⁷ The second largest series was reported by Kumar et al and described their experience with the Ross II operation (N = 19) in New Delhi, India. A mean follow-up time of approximately 72 months, three early deaths (15.7%) and one mitral valve reintervention within 30 days were described (5.3%). No late deaths or reoperations were reported.⁶ To our knowledge, this study is the third largest series reported. None of the patients in our cohort died or required mitral valve reintervention within 30 days. There have been three late deaths at 5.3, 10.5, and 17 years post Ross II operation, none of which were due to autograft or homograft dysfunction.

Most of the patients in the two larger series of Ross II cases underwent the operation for surgical management of rheumatic mitral valve disease.^1,5–7 However, our study includes three patients (38%, 3/8) with congenital heart defects rather than a history of rheumatic mitral valve disease. None of the patients with a history of congenital heart disease required mitral valve reintervention, and their pulmonary autografts remained in place for 10.5, 16, and 23 years, respectively. Two of these patients are still alive, and at 16 and 23 years after their Ross II procedures have trivial autograft regurgitation and autograft gradients of 3 and 7 mm Hg, respectively. Given our small dataset, no causal link between the absence of rheumatic disease process and the outcomes of these patients can be determined. However, although most cases of the Ross II procedure are performed to ameliorate rheumatic mitral valve disease, it may be time to more seriously consider its potential benefits in patients with congenital heart disease.

Traditionally, the use of the Ross II procedure has been restricted to adolescents and younger adults because the support structure required to house the new mitral valve was constructed, as we have previously described,^2,3 by securing the pulmonary autograft inside a currently outdated stiff woven Dacron graft that does not permit growth of the pulmonary autograft. Thus, the fixed size of the mitral valve prosthesis has limited the use of the Ross II procedure to address mitral valve pathologies in younger children. However, in 2018, Jeong and colleagues reported a case in which they utilized a variation of the Ross II procedure that involved implanting the pulmonary autograft in a “top hat” configuration without reinforcing it with prosthetic material in a 3.4 kg infant on postnatal day 76. This configuration was chosen to afford the opportunity for the Ross II valve to grow. The patient was reportedly discharged with mild autograft stenosis (mean gradient = 5.5 mm Hg) and mild autograft regurgitation, but no follow-up on this case has been reported.⁸ At present, if utilized to address congenital heart defects or as a reoperative option after initial repair, the Ross II technique used at our center would best serve patients who are adolescents or younger adults who have a left atrium of 5 cm or greater in diameter and an autograft diameter of at least 20 mm by echocardiographic measurement. An autograft of 20 mm would give a potential mitral valve area of 3.14 cm².

One potential impediment to performing the Ross II procedure is the reduced availability of the older, stiffer, woven Dacron tube graft used in the construction of the composite mitral valve prosthesis. Unfortunately, this product is no longer commercially available. Kabbani et al elucidated the intraoperative ramifications of utilizing the more contemporary softer Dacron to reinforce the pulmonary autograft, which they coined “autograft stenosis.”⁷ Their solution to this problem was to utilize a “pericardial miniskirt,” which was sewn and glued circumferentially to the Dacron conduit externally with surgical adhesive.⁵ However, Kumar et al proposed a technique where a stent was constructed intraoperatively from thick, nonporous Teflon felt to create a ring and structurally supportive commissural pillars, the dimensions of which could be tailored to the patient's harvested pulmonary autograft.⁶ In 2003, Yamagishi et al published a case report detailing the construction of a composite mitral prosthesis by including a pulmonary autograft in a polytetrafluoroethylene (PTFE) graft and adding a strip of PTFE felt as a sewing ring for the prosthesis. Of note, according to the report, the structural apparatus to house the pulmonary autograft was constructed preoperatively.⁹ At seven years, echocardiographic evaluation of the autograft showed worsening autograft regurgitation, as well as subaortic stenosis that had developed due to fibrous tissue ingrowth. Reportedly, the latter was not related to the Ross II conduit.¹⁰

There are, of course, alternatives to consider when seeking a tissue mitral valve substitute. Our center has also utilized the Ross II technique to implant aortic allografts in the mitral position.³ Briefly, an aortic homograft was placed in the mitral position using the Ross II technique due to thrombosis of a previously placed prosthetic mitral valve (21 mm, mechanical) in an eight-year-old patient with a significant history of multiple infections secondary to a nonspecific T-cell immunodeficiency. Almost ten years later, echocardiographic follow-up showed a mean gradient of 9.5 mm Hg across the aortic homograft with no homograft regurgitation. The patient retained the homograft for 13 years, but subsequently died from sepsis following an eye infection. Additionally, an aortic homograft was placed in the mitral position of a 5-month-old patient with a history of Trisomy 21 and a failed complete atrioventricular septal defect repair. Approximately one year after her Ross II procedure, she died from pneumonia. Prior to her death, echocardiographic follow-up indicated trivial aortic homograft regurgitation with a mean gradient of 7 mm Hg.

Another intriguing option is a living allogenic heart valve transplant—coined the partial heart transplant.¹¹ This technique may allow a living donor mitral valve to be implanted into a recipient with mitral valve disease that is not amenable to repair. Since transplanted living allogenic heart valves have growth potential,^11–14 arguably the most significant application for this procedure would be in children. The transplantation of a living allogenic mitral valve also allows preservation of native valve structure and microarchitecture, which are not retained in other types of valve replacements.¹⁴ However, unlike pulmonary autografts, implanting a living allogenic mitral valve does confer a need for immunosuppression^11,13 although given the novelty of the procedure, a firm protocol on recommended postoperative immunosuppression has not yet been clearly delineated. Additionally, while theoretically sound, no objective data are presently available to verify the projected long-term durability of transplanted living allogenic mitral valves. Should it grow as its aortic and pulmonary allograft cousins¹⁴ and prove durable over time, living allogenic mitral valve transplantation may offer a great solution to the challenging problem of MVR in children. However, in older adolescents and young adults who no longer require a growing mitral valve substitute, the Ross II procedure has proven to be an excellent durable living tissue option, without the need for immunosuppression.

Our adoption of the Ross II procedure was fueled in part by our center's extensive experience with the Ross aortic valve procedure for aortic valve replacement (N > 350 adult and pediatric patients). Thus, some of the technical learning curve associated with the operation was likely mitigated for our Ross II cohort given our familiarity with harvesting the pulmonary autograft. However, in the case of a Ross II procedure, it should be noted that we harvest the pulmonary autograft and reconstruct the right ventricular outflow tract with a decellularized PH while the heart is beating to minimize the cardiac ischemic time. Additionally, we construct the composite mitral valve prosthesis before the aortic cross clamp is applied. Thus, the mean ischemic time for a Ross II procedure at our institution accounts for approximately 31% of the average total cardiopulmonary bypass time.

During our early experience with the Ross II procedure, we did not maintain tight perioperative blood pressure (BP) control. We feel now that some of our early Ross II failures were due to stretching of one or more of the autograft leaflets during perioperative hypertensive episodes. We have learned with both the Ross AVR and MVR that the living autograft leaflets need a few weeks to months to adapt and thicken in response to systemic BP. We feel it is imperative to maximize perioperative BP control to keep systemic BP in the low normal range for 6 to 10 weeks. In addition, as we have discussed in prior reports, the three patients in our cohort who required early reintervention (mechanical MVR) within five years of their Ross II procedure all had rheumatic valve disease. Each of these three pulmonary autografts failed in the same way: prolapse of a single autograft leaflet. We contend that the etiology of the leaflet prolapse was exposure to high systemic BP before the autograft leaflets had time to thicken and adapt. Additionally, as these cases were early in our series, we were quite aggressive about replacing the Ross II valves, even when the valves exhibited moderate regurgitation. In retrospect, it is possible that these three valves may have continued to function adequately had they been left in place and the BP better controlled.

To our knowledge, this study represents the longest follow-up of a Ross II cohort ever reported, and the largest experience with the procedure in North America. In this cohort, the Ross II procedure has offered a durable tissue alternative to conventional prosthetic valves for MVR with the potential to last without reintervention for over two decades.

Footnotes

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) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Aeleia F. Hughes

References

Kabbani

Ross

Jamil

, et al. Mitral valve replacement with a pulmonary autograft: initial experience. J Heart Valve Dis. 1999;8(4):359–366.

Brown

Ruzmetov

Rodefeld

Turrentine

. Mitral valve replacement with Ross II technique: initial experience. Ann Thorac Surg. 2006;81(2):502–508.

Blitzer

Herrmann

Brown

. Pulmonary autograft mitral valve replacement (Ross II): long-term follow-up of a US center. World J Pediatr Congenit Heart Surg. 2018;9(6):645–650.

Hughes

Brown

. Twenty-two-year follow-up after mitral valve replacement with pulmonary autograft. World J Pediatr Congenit Heart Surg. 2025;16(4):526–527.

Kabbani

Jamil

Nabhani

, et al. Analysis of 92 mitral pulmonary autograft replacement (Ross II) operations. J Thorac Cardiovasc Surg. 2007;134(4):902–908.

Kumar

Talwar

Gupta

. Mitral valve replacement with pulmonary autograft: midterm results. J Thorac Cardiovasc Surg. 2009;138(2):359–364.

Kabbani

Jamil

Hammoud

, et al. Use of pulmonary autograft for mitral replacement: short- and medium-term experience. Eur J Cardiothorac Surg. 2001;20(2):257–261.

Jeong

Yun

. Mitral valve replacement with a pulmonary autograft in an infant. J Chest Surg. 2018;51:149–152.

Yamagishi

Shuntoh

Matsushita

, et al. Mitral valve replacement by A Gore-Tex reinforced pulmonary autograft in a child. J Thorac Cardiovasc Surg. 2003;126(4):1218–1219.

10.

Kanzaki

Yamagishi

Yashima

Yaku

. Seven-year outcome of pulmonary valve autograft replacement of the mitral valve in an infant. J Thorac Cardiovasc Surg. 2011;141(5):E33–E35.

11.

Kalfa

Rajab

Cordoves

, et al. Living allogenic heart valve transplantation: relative advantages and unanswered questions. J Thorac Cardiovasc Surg. 2024;167(5):1543–1546.

12.

Rajab

. Partial heart transplantation: growing heart valve implants for children. Artif Organs. 2024;48(4):326–335.

13.

Cordoves

Ferrari

Zorn

Bacha

Vunjak-Novakovic

Kalfa

. Storage, preservation, and rehabilitation of living heart valves to treat congenital heart disease. Med. 2024;9(5):859–862.

14.

Kalfa

Cordoves

LaSala

. Allogenic mitral valve transplant: historical precedent, current considerations, and future implementation. J Thorac Cardiovasc Surg. 2025;170(2):311–315.