Long-term outcomes after the paediatric Ross and Ross-Konno procedures

Abstract

 

OBJECTIVES

The Ross procedure is an attractive option for the management of aortic valve disease in paediatric patients. We reviewed our experience with the paediatric Ross procedure to determine survival and freedom from reoperation in the third decade after surgery.

METHODS

We reviewed the data of 124 paediatric patients [71% male, median age at time of surgery 11.1 years (interquartile range 6–14.8 years); 63.7% bicuspid aortic valve], who underwent the Ross procedure at 2 tertiary centres from April 1991 to April 2020. The Ross-Konno procedures were performed on 14 (11.3%) patients. Deaths were cross-checked with the national health insurance database, and survival status was available for 96.8% of the patients. The median follow-up time was 12.1 years (interquartile range 3–18 years).

RESULTS

There were 3 early and 6 late deaths. All early deaths occurred in patients aged <1 year at the time of surgery. The 25-year survival was 90.3%. Actuarial freedom from reoperation (linearized rates in parentheses) was as follows: Autograft reoperation was 90.8% (0.48%/patient-year) and right ventricular outflow tract (RVOT) reoperation was 67% (2.07%/patient year) at 25 years. The univariable Cox-proportional hazard analysis revealed younger age at time of surgery (P < 0.001), smaller implanted valve size (P < 0.001) and the use of a xenograft rather than a homograft (P < 0.001) as predictors of RVOT reoperation. At multivariable Cox-proportional hazard analysis, only age was an independent risk factor for RVOT reoperation (P = 0.041).

CONCLUSIONS

The Ross and the Ross-Konno procedures are associated with good outcomes in paediatric patients. Reoperation of the RVOT is frequent and associated with younger age.

INTRODUCTION

In paediatric patients the Ross and the Ross-Konno procedures are established options for the surgical management of aortic valve disease. The implantation of the autograft provides a good haemodynamic profile with low transvalvular gradients [1]. The low thrombogenicity makes the Ross procedure a valve replacement option with no necessity for anticoagulation [2, 3]. In patients with multilevel left ventricular outflow tract obstruction, a Ross-Konno procedure can additionally relieve subvalvular outflow tract obstruction. Data on long-term outcomes over 20 years are limited for paediatric patients [4, 5]. Pulmonary autograft and right ventricular outflow tract (RVOT) replacement durability remain of concern regarding reintervention rates. We reviewed our experience with the paediatric Ross and Ross-Konno procedures to report survival and reoperation rates in the third decade after surgery.

PATIENTS AND METHODS

Patients

The multicentre study was carried out at 2 tertiary-care centres with specialized paediatric cardiology and cardiac surgery units. The study was approved by the institutional review boards of both centres (ethics committee submission numbers: 1414/2019 and 1118/2020). Requirement for individual patient consent was waived. A retrospective chart review of all consecutive paediatric patients (<18 years) who underwent a Ross or Ross-Konno procedure at the centres from April 1991 until April 2020 was performed. The follow-up was carried out at both centres. Median follow-up time of patients having been discharged from the hospital after a Ross procedure (96.8%; 120/124) was 12.1 years [interquartile range (IQR) 3–18 years; 1452 patient-years] with a maximum follow-up time of 27.7 years.

Surgical technique

For the Ross procedures, the pulmonary autograft was implanted as a total root below the aortic annulus, and the autograft length was trimmed to keep the root as short as possible. In more recent years, in older children, the autograft was root reinforced with a 3- or 5-mm oversized Valsalva vascular prosthesis (Vascutek, Renfrewshire, Scotland, UK) to prevent dilatation of the autograft. In less recent times, the autograft was wrapped either with Vicryl mesh or the remnant aortic wall. The decision to reinforce the root was based on age rather than on valve anatomy and no additional precautions (e.g. annular reinforcement) were taken in the setting of a bicuspid aortic valve. The Ross-Konno procedure (11.3%; N = 14) was performed in all patients with relevant left ventricular outflow tract stenosis and a hypoplastic aortic annulus. The Konno incision was either closed with the muscular flap of the autograft root or with patch material. Use of homografts and xenografts was determined by availability, which varied throughout the study period. The preference at our centre was to use homografts. Bioprosthetic conduits were used when no suitable homograft was available. RVOT replacement was performed with a homograft in 79% and with a xenograft in 21% of patients. The size of the RVOT replacement was selected accordingly to the body surface area. Patients were discharged with weight-adapted antiplatelet therapy for 6 months postoperatively.

Definitions

Parameters were obtained and measured as described in the Guidelines for Reporting Mortality and Morbidity after Cardiac Valve Interventions [6]. Mortality was cross-checked with the national health insurance database to provide a complete mortality follow-up until 30 April 2020. Survival status was available for 96.8% of patients because 4 patients were from foreign countries and not systematically followed up. Survival time for these patients was calculated until the last confirmed living follow-up at 1 of the 2 centres. One patient underwent univentricular conversion 56 days after the Ross-Konno procedure, and 1 patient had a cardiac transplant because of dilatative cardiomyopathy 17 years after the Ross procedure. These patients were censored from further Ross-related analysis at the time of the conversion and the transplant, respectively. As reported previously, early failure of the porcine Synergraft accounted for 1 death and 1 reoperation for prophylactic explantation [7]. The prophylactic change was included as the reoperation end-point because structural valve deterioration was seen as early as 2 days after the implant with formation of a dense fibrous sheath around the graft because of a strong inflammatory response.

Statistical analyses

Continuous data were expressed as mean ± standard deviation, while skewed continuous data were expressed as a median with the IQR and minimum and maximum values. Categorical variables were summarized as frequencies and percentages. In order to identify significant differences between 2 subgroups, continuous variables were compared using the independent-samples Mann–Whitney U-test. Time-related end-points were analysed and plotted using Kaplan–Meier actuarial survival curves. Subgroups were compared using the log-rank test. Patients who did not experience outcome events were censored at the time of the last follow-up. Variables with a P < 0.05 or that were deemed clinically relevant were entered into univariable and multivariable Cox proportional hazards models. Linearized event rates per patient-year were calculated. Statistical significance was set at P < 0.05. Data were analysed using the software package SPSS 26 (IBM Corp., Chicago, IL, USA).

RESULTS

Patients

One hundred and twenty-four patients were included during the study period. Demographic characteristics are detailed in Table 1. Seventy-one percent of patients were male (N = 88), and the median age at the time of the operation was 11.1 years (IQR 6–14.8 years). The fundamental aortic valve anatomy was predominantly bicuspid (N = 79; 63.7%). The majority of patients (N = 82; 66.1%) had undergone at least 1 previous percutaneous or surgical aortic valve intervention. The median time from the last aortic valve intervention to the Ross procedure was 5.8 years (IQR 1.5–9.5 years). Patients undergoing the Ross procedure after prior aortic valve intervention were significantly older (P < 0.001) than patients with no previous aortic valve intervention and had a median age at the Ross procedure of 14.2 years (IQR 10.9–16.1 years) and 9.1 years (IQR 5.4–13.6 years), respectively.

Table 1:

Demographic and operative data

Characteristic	Value
Demographic
Patient cohort	124 (100)
Age at time of surgery (years)
<1	13 (10.5)
1–5	18 (14.5)
6–13	54 (43.5)
14–18	39 (31.5)
Weight at time of surgery (kg)	39.1 (21.4–55.8)
Height at time of surgery (cm)	148 (117–165)
Native aortic valve anatomy
Unicuspidalized	4 (3.2)
Bicuspid	79 (63.7)
Tricuspid	23 (18.5)
Unknown	18 (14.5)
Previous interventions
Aortic valve intervention (percutaneous and surgical)
None	42 (33.9)
1	60 (48.4)
2	18 (14.5)
3	4 (3.2)
Indication for surgery
Aortic stenosis	17 (13.7)
Aortic regurgitation	44 (35.5)
Combined aortic valve disease	63 (50.8)
Acute indication for surgery	3 (2.4)
Bacterial endocarditis	2 (1.6)
Rheumatic valve disease	1 (0.8)
Degeneration aortic valve replacement (homograft)	2 (1.6)
Operative
ACCT (min)	123 (96–148)
CPB (min)	174 (143–227)
Circulatory arrest	5 (4.0)
Associated procedures
Ross-Konno	14 (11.3)
Subvalvular myectomy and/or membrane resection	8 (6.5)
Root-reinforced Ross	14 (11.3)
Root reinforced in prosthesis	9 (7.3)
Root reinforced with Vicryl mesh	3 (2.4)
Root reinforced with remnant aortic wall	2 (1.6)
Reduction aortoplasty	32 (25.8)
Aortic arch patch plasty	5 (4.0)
Concomitant mitral valve procedure	5 (4.0)
Mitral valve reconstruction	3 (2.4)
Mitral valve reconstruction with annuloplasty ring	1 (0.8)
Balloon dilatation of Melody valve in mitral position	1 (0.8)
RVOT replacement size (mm)	23 (range 12–30)
Homograft/xenograft
Homograft	98 (79.0)
Pulmonary (homograft bank)	60 (48.4)
Aortic (homograft bank)	9 (7.3)
Cryolife	21 (16.9)
Decellularized Corlife	3 (2.4)
BisLife	5 (4.0)
Xenograft	26 (21.0)
Contegra conduit	19 (15.3)
Hancock conduit	3 (2.4)
Mosaic aortic valve in Dacron prosthesis	1 (0.8)
Porcine pulmonary Synergraft heterograft	3 (2.4)

Values are presented as N, N (%), median (interquartile range) or median (range minimum–maximum) in case of valve size (mm).

ACCT: aortic cross-clamp time; CPB: cardiopulmonary bypass; RVOT: right ventricular outflow tract.

Operative data

Combined aortic valve disease (N = 63; 50.8%) and aortic regurgitation (N = 44; 35.5%) were the most frequent indications for an operation. Operative characteristics are listed in Table 1. The Ross-Konno procedure was performed in 14 patients (11.3%). Patients undergoing the Ross-Konno procedure were younger (P < 0.001) than patients undergoing the Ross procedure, with a median age of 0.5 years (IQR 0.07–2.9 years) and 12.4 years (IQR 7.5–15.2 years), respectively. Root reinforcement of the autograft was performed in 14 patients (11.3%). The root-reinforcement strategies and the RVOT conduits used are summarized in Table 1. Patients receiving a homograft [median age 13.4 years (IQR 8.8–15.4 years)] were older (P < 0.001) compared to patients receiving a xenograft [median age 1.8 years (IQR 0.3–5.7 years)]. Accordingly, the xenografts [median valve size 14 mm (range 12–23 mm)] were smaller (P < 0.001) than the homografts [median valve size 24 mm (range 12–30 mm)].

Early outcomes

Early postoperative outcomes are detailed in Table 2. Early in-hospital mortality was 2.4% (N = 3). All early deaths occurred in patients aged <1 year at the time of surgery, and early mortality was 23.1% (3/13) for patients aged <1 year at the time of surgery. Causes of death are summarized in Table 3. Three patients (2.4%; Ross: N = 2; Ross-Konno: N = 1) required permanent pacemaker implants for complete atrioventricular block in the first 30 postoperative days. Patients aged <1 year had longer times on ventilation, stays in the intensive care (ICU) and overall postoperative hospital stays than patients aged >1 year at the time of surgery [median times on ventilation: 4 days (IQR 3–5 days) and 1 day (IQR 0–1 day), P < 0.001; median ICU stays of 5 days (IQR 2–10 days) and 2 days (IQR 1–3 days), P < 0.001; hospital stays: 22 days (IQR 17–32 days) and 12 days (IQR 10–14 days); P < 0.001].

Table 2:

Postoperative outcomes

Characteristic	Value
Patient cohort	124 (100)
Permanent pacemaker insertion	4 (3.2)
AV block III	3 (2.4)
Bradycardia	1 (0.8)
Reoperation for bleeding	3 (2.4)
Subxiphoidal drainage for pericardial effusion	6 (4.8)
Delayed sternal closure	6 (4.8)
Prolonged ventilation (> 5 days)a	2 (1.6)
Ventilation (days)a	1 (0–1)
ICU stay (days)a	2 (2–4)
Hospital stay (days)a,b	12 (10–15)
Peritoneal dialysis	4 (3.2)
ECMO	3 (2.4)
Early in-hospital deaths	3 (2.4)

Values are presented as N, N (%), median (interquartile range).

^aTimes of the 3 patients who died were excluded.

^bTime of patient who underwent univentricular conversion is excluded because the patient was censored at the time of surgery during the same hospital stay after transfer from the ICU to the normal ward.

ECMO: extracorporeal membrane oxygenation; ICU: intensive care unit.

Table 3:

Early and late deaths

Patient no. (sex)	Surgery year	Age	Diagnosis	Procedure	Time of death after surgery	Cause of death
Early in-hospital deaths
No. 1 (f)	2004	8 days	Critical AS	Ross	8 days	Multiorgan failure after reoperation for valve thrombosis of Hancock conduit on 7th postoperative day
No. 2 (m)	2000	23 days	Critical AS Subvalvular AS	Ross-Konno	12 days	Early onset sepsis (mother Streptococcus B positive)
No. 3 (m)	2015	4 months	Critical AS Preoperative SIMDAX cycles in the setting of highly reduced LVF	Ross-Konno	53 days	Multiorgan failure
Late deaths
No. 4 (f)	2006	13 years	AR VSD	Ross	4 months	Brain death due to status epilepticus with cardiac arrest and out-of-hospital CPR leading to ROSC
No. 5 (m)	2001	2 years	Critical AS Subvalvular AS	Ross-Konno	1 year	Sudden cardiac death (severely degenerated Synergraft valve)
No 6. (m)	1996	16 years	AS	Ross	2 years	Accidental traumatic injury, CCI (car accident)
No. 7 (m)	1999	13 years	Critical AS	Ross	8 years	Accidental traumatic injury (car accident)
No 8. (f)	1999	17 years	AS CoA	Ross	16 years	Brain death due to hypoxic brain damage after out-of-hospital CPR leading to ROSC
No. 9 (m)	2001	3 years	Critical AS	Ross	17 years	Unknown non-cardiac death

AR: aortic regurgitation; AS: aortic stenosis; CCI: craniocerebral injury; CoA: coarctation of the aorta; CPR: cardiopulmonary resuscitation; f: female; LVF: left ventricular function; m: male; MR: mitral regurgitation; MV: mitral valve; ROSC: return of spontaneous circulation; VSD: ventricular septal defect.

Follow-up

There were 6 late deaths. The linearized rate of late deaths after discharge was 0.41%/patient-year. Causes of death are detailed in Table 3. Overall survival was 94.9% ± 2 at 5 years, 93.8% ± 2.3 at 15 years and 90.3% ± 3.3 at 25 years (Fig. 1). Overall survival for patients aged <1 year at the time of the operation (N = 13) was 76.9% ± 11.7 at 15 years compared to overall survival for patients aged >1 year at the time of the surgery (N = 111) with 95.9% ± 2 at 15 years (P = 0.001). Freedom from any Ross-related reoperation was 86.5% ± 3.4 at 5 years, 66.2% ± 5.3 at 15 years and 58.3% ± 6.7 at 25 years (Supplementary Material). There were no deaths associated with reoperation during the follow-up period. Additional Kaplan–Meier estimated rates can be seen in the Supplementary Material.

Figure 1:

Survival probability following the Ross procedure. Kaplan–Meier estimated survival curve with 95% CI. CI: confidence interval.

Autograft reoperation

Freedom from autograft reoperation was 96.9% ± 1.8 at 5 years and 90.8% ± 3.4 at 15 and 25 years (Fig. 2). Seven patients (5.6%; 7/124) underwent reoperation of the autograft with linearized event rates of 0.48%/patient-year for any autograft reoperation and 0.07%/patient-year for aortic valve replacement. No reoperation occurred in patients who had received root reinforcement. Autograft-sparing root replacement (Tirone David operation) for neoaortic root dilatation was performed in 4 patients (57.1%; 4/7), a mechanical Bentall procedure was performed in 1 patient (14.3%; 1/7) and 2 patients (28.6%; 2/7) underwent annuloplasty. There has been no repeat autograft reintervention, with a mean time of 7.9 ± 7.8 years (range 0.05–19.4 years) postoperatively.

Figure 2:

Freedom from autograft reoperation. Kaplan–Meier estimated freedom from autograft reoperation. Curve with 95% CI. CI: confidence interval.

Right ventricular outflow tract reoperation

Freedom from RVOT reoperation was 90.9% ± 2.9, 75.7% ± 4.9 and 67.0% ± 7 at 5, 15 and 25 years (Fig. 3A). Freedom from RVOT reintervention (percutaneous and surgical valve replacement) was 88.7% ± 3.2, 64.5% ± 5.5 and 58.5% ± 6 at 5, 15 and 25 years (Fig. 3B). Thirty-two patients (25.8%; 32/124) underwent 48 RVOT reinterventions (percutaneous valve replacements: 37.5%; 18/48, surgical valve replacements: 62.5%; 30/48). The linearized event rate per patient-year for any RVOT reintervention was 3.31%.

Figure 3:

Freedom from RVOT reintervention. (A) Kaplan–Meier estimated freedom from RVOT reoperation. Curve with 95% CI. (B) Kaplan–Meier estimated freedom from RVOT reintervention (surgical and percutaneous pulmonary valve replacement). Curve with 95% CI. CI: confidence interval; RVOT: right ventricular outflow tract.

Fifteen patients (12.1%, 15/124) underwent a percutaneous pulmonary valve implant with a Melody transcatheter pulmonary valve (Medtronic plc, Dublin, Ireland) (stenosis: 53.3%, 8/15; combined valve disease: 46.7%, 7/15). One patient had to be converted to a surgical valve replacement for perforation during the implant procedure. Five patients (35.7%, 5/14) underwent reoperation after the percutaneous valve was implanted (stenosis: N = 2; combined valve disease: N = 2; endocarditis: N = 1) with a mean time of 7.1 ± 3.9 years (range 0.2–9.6 years) from initial percutaneous implant. Two patients had a Melody valve-in-valve implant after 3 years and 6.6 years from the initial percutaneous valve implant, with 1 being reoperated 5.6 years after the valve-in-valve implant. One patient received an Edwards Sapien 3 valve (Edwards Lifesciences Corp., Irvine, CA USA) as a second percutaneous pulmonary valve replacement after a previous Melody implant and an RVOT reoperation.

Twenty-three patients underwent 30 RVOT reoperations with a linearized event rate of 2.07%/per patient-year. The mean time-to-first RVOT reoperation was 7.9 ± 5.6 years (range 0.005–20.7 years). Indications for an RVOT reoperation were stenosis (43.5%; 10/23), regurgitation (13%; 3/23) and combined valve disease (17.4%; 4/23). Three patients underwent reoperation for homograft endocarditis (13%; 3/23). One patient (4.3%; 1/23) underwent acute RVOT reoperation for perforation during a percutaneous valve implant procedure. Five patients (21.7%, 5/23; stenosis: N = 1, combined valve disease: N = 2, Melody endocarditis: N = 1, stenosis and subvalvular stenosis: N = 1) underwent a second reoperation after a mean time of 8 ± 4 years (range 1.7–12.5 years) and 2 patients (8.6%, 2/23) had a third reoperation for pulmonary stenosis after 2 and 5.3 years. The linearized event rate per patient-year for pulmonary valve endocarditis was 0.28% with 4 with endocarditis (homograft: N = 3; Melody valve: N = 1).

Freedom from reoperation was higher (P < 0.001) in the homograft cohort (N = 98) than in the xenograft cohort (N = 26), with 83.5% ± 4.7 and 17% ± 14.2 at 15 years after the Ross procedure, respectively. At univariable analysis, younger age at the time of the operation [hazard ratio (HR) 0.8 for each increase in year; P < 0.001) and, in particular, age <6 years (HR 3.1, 95% confidence interval (CI) 1.3–7.1; P = 0.01] were risk factors associated with an RVOT reoperation. Also, at univariable analysis, the use of a xenograft rather than a homograft (HR 10, 95% CI 4–25.1; P < 0.001) and a smaller implanted valve size (HR 0.8 for each increase in valve size in mm; P < 0.001) were associated risk factors for reoperation. At multivariable analysis, only age was an independent risk factor for an RVOT reoperation (HR 0.84, 95% CI 0.7–1; P = 0.041).

DISCUSSION

The Ross and the Ross-Konno procedures provide good long-term outcomes in paediatric patients with the advantage of longevity of a good haemodynamic profile in the absence of anticoagulation [1–3, 8]. Only a few publications reported on outcomes longer than 20 years [4, 5]. The Ross procedure is offered to paediatric patients when aortic valve replacement becomes inevitable because the durability of a valve repair is limited [9–11], though it is preferable to postpone valve replacement until the patient is older and therefore has grown somatically [12, 13]. In our cohort, 66.1% of patients had undergone at least 1 percutaneous or surgical aortic valve intervention prior to the Ross procedure. Patients undergoing the Ross procedure after prior aortic valve intervention were significantly older (P < 0.001) than patients who had not undergone previous aortic valve intervention, thus allowing them to delay valve replacement. Previous aortic balloon valvuloplasty, surgical valvulotomy or valve repair allowed them to defer the Ross procedure by a median time of 5.8 years and likely decreased periprocedural risk. The median aortic cross-clamp time (P = 0.002) and the median cardiopulmonary bypass time (P = 0.004) were longer in patients aged <6 years, demonstrating the more complex surgical course in younger patients, who more often have complex left heart disease. Also, we observed longer median stays in the ICU (P < 0.001) and in the hospital (P = 0.045) in patients aged <6 years.

Early in-hospital deaths (2.4%) were comparable to those reported in other recent studies with larger cohorts ranging from 1.6% to 5% [4, 5, 14–16]. All early deaths occurred in patients aged <1 year at the time of surgery. For patients aged <1 year at surgery, early deaths were 23.1% (3/13), which is similar to data reported for other cohorts that showed higher periprocedural deaths in infants and neonates, with early mortality rates for patients aged <1 year at the time of surgery ranging from 18.2% to 46.2%. [14, 16–18]. We had 6 late deaths and an estimated 25-year survival of 90.3%, which is similar to an estimated 25-year survival of 94.4% (1 early in-hospital death: 1.6%, 1/63) reported by Martin et al. [5], who to our knowledge is the only group reporting 25-year outcomes. Among other recent long-term paediatric studies, the estimated 15-year survival ranges from 86.6% to 94.4% [4, 5, 14, 15, 19], which is consistent with our estimated 15-year survival of 93.8%. Etnel et al. [3] reported a pooled actuarial survival of 93.5% at 15 years.

Reoperation of the autograft was rare, with 7 autograft reoperations (5.6%, 7/124) and an estimated 25-year freedom from autograft reoperation of 90.8%, compared to Martin et al. [5], who reported 15 reoperations (24%, 15/63) and an estimated 25-year freedom from autograft reoperation of 61.2%. Studies covering 15-year outcomes reported freedom from reoperation rates from 59% to 81% [4, 5, 14–16, 18], and Etnel et al. [3] calculated a pooled freedom from autograft reoperation of 77.3% at 15 years. Mean time to autograft reoperation was 7.1 years with 2 early annuloplasties (1.2 and 3.7 years after the Ross procedure) in the first paediatric Ross patient of each centre, respectively. Valve-sparing root replacement is offered when leaflet function is sufficient in the setting of neoaortic root dilatation to allow for delay of replacement and the often-accompanying need for anticoagulation. Only 1 patient required aortic valve replacement (mechanical Bentall).

The longevity of an RVOT replacement remains a concern across the paediatric age spectrum, especially in infants, who will inevitably outgrow their RVOT conduit and will require a reoperation [20]. In our cohort, freedom from RVOT reoperation was 67.0% at 25 years and freedom from RVOT reintervention (percutaneous and surgical valve replacement) was 58.5% at 25 years. Martin et al. [5] reported freedom from RVOT reintervention (97% homografts) of 28.3% at 25 years. In the Italian paediatric Ross registry cohort of Luciani et al. [14], who reported more autograft than RVOT reoperations, 15-year freedom from RVOT reintervention (percutaneous and surgical valve replacement) was 89%. In recent studies of larger cohorts, the 15-year freedom from RVOT reoperation ranged from 53% to 59% [4, 16, 18], and Etnel et al. [3] calculated a pooled freedom from RVOT reoperation of 67.4% at 15 years. In our cohort, freedom from reoperation at 15 years was 83.5% in the homograft group compared to 17% in the xenograft cohort, with an overall 15-year freedom from RVOT reoperation of 75.7%. Consistently with other publications, younger age at the time of the Ross procedure, the use of a xenograft rather than a homograft and a smaller implanted valve size were identified as risk factors associated with RVOT reoperation [4, 14, 16, 17, 21]. Postponing an RVOT reoperation with a transcatheter valve is an attractive option to reduce the reoperation rate over a lifetime. In our cohort, the mean time from the Ross procedure to a percutaneous pulmonary valve implant was 9.4 years. Nine out of 14 patients who underwent percutaneous pulmonary valve replacements were free from reoperation after a mean time of 8.9 ± 3.1 years (range 2–12.7 years) post-implant. In the 5 patients who required reoperation, the timing of reoperation could be postponed by a mean time of 7.1 ± 3.9 years (range 0.2–9.6 years) from the percutaneous valve implant to surgery. One patient had received a Melody valve-in-valve implant earlier, which allowed postponement of the reoperation for 5.6 years after the valve-in-valve was implanted.

The typical limitations of a retrospective study are evident in this study. Variability regarding indication, choice and the evolution of surgical and catheter-based techniques changed over the decades. The small number of events limited our ability to identify predictors of adverse events and perform multivariable analyses. Nevertheless, this study offers the strengths of presenting one of the larger cohorts studied within this disease spectrum and a long follow-up period of 27.7 years.

CONCLUSION

The paediatric Ross and Ross-Konno procedures are associated with good long-term outcomes. RVOT reoperation is frequent and associated with younger age at the time of the operation and the use of xenografts. Transcatheter valve replacement can result in a postponed RVOT reoperation and therefore in reduced reoperation rates.

SUPPLEMENTARY MATERIAL

Supplementary material is available at ICVTS online.

Funding

This research received no external funding.

Conflict of interest: All authors have reported that they have no conflict of interest relevant to the content of the manuscript, the design of the study, the collection, analyses or interpretation of the data, or the decision to publish the results. Prof. Zimpfer has received research grants from Edwards Lifesciences, Medtronic, Abbott and Berlin Heart.

Author contributions

Johanna Schlein: Conceptualization; Data curation; Formal analysis; Investigation; Visualization; Writing—original draft. Barbara Elisabeth Ebner: Data curation; Investigation; Writing—review & editing. Ralf Geiger: Conceptualization; Resources; Validation; Writing—review & editing. Paul Simon: Writing—review & editing. Gregor Wollenek: Validation; Writing—review & editing. Anton Moritz: Writing—review & editing. Andreas Gamillscheg: Validation; Writing—review & editing. Eva Base: Validation; Writing—review & editing. Günther Laufer: Resources; Validation; Writing—review & editing. Daniel Zimpfer: Conceptualization; Resources; Supervision; Validation; Writing –review & editing.

Reviewer information

Interactive CardioVascular and Thoracic Surgery thanks Hani Najm, Marco Pozzi and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.

Glossary

ABBREVATIONS

CI

Confidence interval

HR

Hazard ratio

ICU

Intensive care unit

IQR

Interquartile range

RVOT

Right ventricular outflow tract

 

Presented at the 34th Annual Meeting of the European Association for Cardio-Thoracic Surgery, Barcelona, Spain, 8–10 October 2020.