Endoscopic aortic valve surgery in isolated and concomitant procedures

Abstract

OBJECTIVES

To evaluate early outcomes of endoscopic aortic valve replacement (AVR) and risks of concomitant procedures done through the same working port.

METHODS

At our institution, we performed a data analysis of 342 consecutive patients (from July 2013 to May 2021) who underwent endoscopic AVR with or without associated major procedure. Preoperative, intraoperative, postoperative data were evaluated. Subsequently, we perform a comparative analysis between the isolated and concomitant surgery group. The surgical access was a 3- to 4-cm working port in the second right intercostal space and 3 additional 5-mm mini-ports for the introduction of the thoracoscope, the transthoracic clamp and the vent line. Cardiopulmonary by-pass was achieved through peripheral cannulation.

RESULTS

105 patients (30.7%) underwent combined procedure: 2 coronary artery bypass (1.9%), 21 ascending aorta replacement (19.6%), 41 mitral surgery (38.3%), 16 mitral and tricuspid surgery (15%) and 25 other procedure (27%). Death occurred in 1 patient (0.4%) in the isolated group versus 2 patients (1.9%) in the combined group (P = 0.175). Seven strokes were observed, 4 in isolated procedures (1.7%) and 3 in the concomitant ones (2.85%) (P = 0.481). Surgical revision for bleeding was performed always through the same access in 13 patients (5.4%) versus 11 patients (10.4%) (P = 0.096). Pacemaker implantation was necessary in 5 patients (2.1%) versus 8 patients (7.6%) (P = 0.014). Median intubation time was 5 (2) h vs 6 (8) (P < 0.080).

CONCLUSIONS

Through a single working port made for endoscopic AVR, a concomitant procedure may be done without affecting in-hospital mortality and postoperative stroke rate.

The concept of minimally invasive access for the treatment of the aortic valve disease is not new, since Cosgrove and Sabik [1] first described it in 1996.

INTRODUCTION

The concept of minimally invasive access for the treatment of the aortic valve disease is not new, since Cosgrove and Sabik [1] first described it in 1996. Since then, it has increased steadily over time, although the most used option around the world continues to be the traditional full sternotomy [2].

The introduction of the transcatheter aortic valve replacement (TAVR) has further revolutionized the treatment of aortic valve stenosis and it is considered a valid alternative to conventional surgery, providing good outcomes also in low-risk patients [3]. The evolution of the prostheses and catheters has reduced the incidence of vascular complications and improved the valve performance hence the indications for TAVR have greatly increasing [4]. Nevertheless, this approach still has some limitations in terms of uncertain long-term valve’s durability and the treatment of combined pathologic processes is still at the beginning. Moreover, a recent metanalysis showed a significantly worse survival compared to surgery after 40 months, and therefore, it should be carefully evaluated in patients with long life expectancy [5].

In consideration of the widespread success of TAVR technique and together with its intrinsic limits, the interest in minimally invasive aortic valve replacement (AVR) has further increased and inspired the cardiac surgery community to find new minimally invasive techniques such as video-assisted surgery [6], robotic-assisted surgery [7] or endoscopic surgery [8]. Recently, we published our new minimally invasive endoscopic technique and documented its technical feasibility, safety and effectiveness evaluating its operative results in 125 isolated AVR, regardless of the type of implanted prosthetic valve and the preoperative patients’ characteristics [9]. With this study, we want to analyze early outcomes of 342 consecutive patients who underwent endoscopic aortic valve replacement (E-AVR). In this cohort, we have also included concomitant procedures in order to evaluate the feasibility and risks of concomitant procedures performed through the same single working port, such as mitral, tricuspid or ascending aorta replacement.

PATIENTS AND METHODS

Study design

A retrospective, observational, single-center study was conducted in all patients who underwent totally endoscopic AVR with or without concomitant major procedure. Inclusion criteria was the use of a thoracoscope for a totally endoscopic approach and a minithoracotomic access. Exclusion criteria were the impossibility to obtain a peripheral cannulation, whereas a relative contraindication was the presence of pleural and pericardial adhesion that could make the access very difficult and dangerous. A preoperative computed tomography (CT) scan screening for intrathoracic aortic position, pathology or peripheral vascular disease was not made routinely in isolated AVR cases. We did it in all cases of concomitant aorta surgery or in cases where we aimed to further investigate on patients with some dubious medical history or preoperative angiography images.

Ethical statement

This study was approved by our local hospital institutional ethics committee (IRB Number: 4009, date 14 January 2021).

Operative technique

Our minimally invasive E-AVR technique [9] is an endoscopic completely video-guided operation that requires the use of a video column and specific long shaft instruments. Cardiopulmonary bypass (CPB) is usually established through a peripheral femoral arterial and venous cannulation. Then, a 3- to 4-cm right mini-thoracotomy in the 2nd or 3rd intercostal space at the midclavicular line is done as principal working port. Soft tissue retractor is used with no rib-spreading and without ligation of the mammary vessels. A 5-mm 30° thoracoscope is introduced into the chest through a 5-mm mini-port in the 2nd intercostal space. The aortic cross clamping is achieved using a transthoracic Chitwood^® clamp (Scanlan International, Inc, St Paul, MN, USA) introduced through a 5-mm stab-wound incision in the 1st intercostal space. The cardioplegia is delivered into the ascending aorta and then, after the aortotomy, directly into the coronary ostia if needed. The vent line is placed through the superior right pulmonary vein and passed through a 5-mm stab-wound incision into the 4th intercostal space. In case of isolated AVR, after the aortotomy, the aortic valve exposure is obtained by using 3 retraction stitches placed at the aortic valve commissures and, in addition, by 2 or 3 stitches keeping the aortic wall away from the thoracoscope field. The native valve is removed in standard fashion and the prosthesis is implanted in the usual manner and under a completely endoscopic vision. Finally, the aorta is normally closed using a double 4/0 propylene running suture.

In case of concomitant double (aortic+mitral) or triple (aortic + mitral + tricuspid) valve surgery, the operation is performed through the same working port (Fig. 1). In this case, a slightly more lateral mini-thoracotomy can help in obtaining a more favourable angle of visualization. In more complex anatomy, the working port can be moved through the adjacent intercostal space maintaining the same skin incision. Pericardial stay sutures are placed at the level of the inferior vena cava, pulmonary veins and/or superior vena cava in order to better expose the interatrial sulcus. An additional pericardial stay suture is placed lateral to the aorta to obtain further retraction and the exposition of the right atrium wall as well. The aortic cross clamping is achieved in the same way. Usually, after removing the aortic valve and debridement of the annulus, the mitral valve is addressed as needed. A standard left atriotomy is done at the Sondergaard’s interatrial groove. An appropriately-sized retractor blade is used to expose the mitral valve. The 3-mm shaft of the retractor is inserted into the thorax through the 3rd or 4th intercostal space, paying attention to avoid any injury to the mammary vessels.

Figure 1:

(A) Schematic view of the operative field and the working port. (B) Rapid deployment valve. (C) Mitral valve repair. (D) Tricuspid valve repair. (E) Wound dimension at the end of surgery.

In case of concomitant tricuspid valve, the right atrium is opened under cardiac arrest or, in some cases, after removing the aortic clamp, leaving the heart beating (Video 1). In case of concomitant aorta surgery, the technique does not differ (Fig. 2).

Figure 2:

Concomitant endoscopic aortic valve and ascending aorta replacement: operative view and thoracoscopic vision (box).

In case of concomitant atrial fibrillation (AF) surgery, a completely video-guided maze procedure is done using the Atricure Cryo Probe (Atricure Inc, Westchester, OH, USA) and the left appendage directly sutured or closed using the AtriClip^® Exclusion System.

Data collection

We enrolled patients who consecutively underwent EAVR at our institution. We collected baseline characteristic before surgery, intraoperative variables, immediate postoperative and 30-day results.

Outcomes

The primary outcome was mortality at 30 days, secondary outcomes were incidence of major adverse cardiovascular events as stroke, acute myocardial infarction (AMI), defined as type 5 myocardial infarction from the Fourth Universal Definition of Myocardial Infarction [10], surgical re-exploration and acute kidney injury as defined by the AKIN classification. Then, we evaluated surgical outcomes as CPB time, cross-clamp time, sternotomic conversion, pacemaker (PM) implantation, hospitalization days and anesthesiologic outcomes as intensive care unit (ICU) hours and oro-tracheal (OT) intubation hours. Preoperative variables were defined according to EuroSCORE definitions [11]. Renal function was evaluated as peak of creatinine at 72 h, stroke was defined as neurological deficit with a documented CT scan cerebral lesion.

Statistics

All statistical analyses were performed using SPSS 21.0 (SPSS, Inc, Chicago, IL, USA) at 95% confidence interval level. Alpha was 0.05. Assumption of normality was verified with Kolmogorov–Smirnoff and Shapiro–Wilk test for continuous variables. Continuous variables with normal distribution are presented as mean (standard deviation), continuous variables with non-normal distribution are presented as median (interquartile range), while categorical or ordinal variables are presented as frequencies and percentages. Statistical analysis comparing the isolated EAVR group versus the Concomitant-EAVR group was performed using the unpaired Student’s t-test for normal continuous variables, Mann–Whitney test for not normal continuous variables, Pearson’s chi-squared test for categorical variables and Fisher’s exact test when categorical variables had expected and observed frequencies <5 in more of 20% of the cross-tabs cells.

RESULTS

Demographic and baseline data

From January 2014 to May 2021, we enrolled 342 patients who consecutively underwent EAVR at our institution. Among them, 237 underwent isolated AVR and 105 had concomitant procedures. All preoperative variables are reported in Table 1. The mean age between the isolated EAVR and concomitant EAVR was similar with non-statistical significative difference {67.3 [standard deviation (SD): 11] vs 66.8 (SD: 12), P = 0.903}. As expected, EuroSCORE II percentage was higher in the concomitant group [1.79 (SD: 2.09) vs 3.31 (SD: 4.29), P < 0.001]. In the isolated group, the pathology of presentation was aortic stenosis in 127 patients (53.6%), aortic regurgitation in 68 patients (28.8%), aortic steno-regurgitation in 42 patients (17.6%); in the concomitant group, artic stenosis was present in 44 patients (41.5%), aortic regurgitation in 42 patients (40.6%), aortic steno-regurgitation in 19 patients (17.9%) with P = 0.071. Isolated EAVR patients had higher BMI [27.2 (SD: 4.4) vs 25.4 (SD: 3.78), P < 0.001] and higher incidence of obesity (27.2% vs 14.3%, P = 0.009). CT scan exam was performed more frequently in concomitant procedure (34.6% vs 10.2%, P < 0.001). New York Heart Association class divided in the subgroups was close to statistical significance (P = 0.057). Cardiac rhythm analysis showed a higher prevalence of paroxysmal AF and permanent AF in the concomitant group (27.1% and 13.1% vs 8.5%, P < 0.001).

Table 1:

Preoperative variables

	Isolated	Concomitant	P-Value
Age (years), mean (SD)	67.3 (11)	66.8 (12)	0.903
EuroSCORE II (%), mean (SD)	1.79 (2.09)	3.31 (4.29)	<0.001
BSA (m2), mean (SD)	1.8 (0.29)	1.74 (0.3)	<0.001
BMI (kg/m2), mean (SD)	27.2 (4.4)	25.4 (3.78)	<0.001
GFR (ml/min), mean (SD)	90 (34)	83 (26)	0.053
Peak gradient (mmHg), mean (SD)	71 (27.1)	58 (32.1)	0.001
EF (%), mean (SD)	59 (10)	58 (9)	0.538
Sex (M = 0, F = 1)			<0.001
0
n	150	45
%	63	42
1
n	87	62
%	37	58
Total	237	105
Arterial hypertension (0 = no, 1 = yes)			0.687
0
n	49	24
%	20.5	22.4
1
n	188	81
%	79.5	77.6
Total	237	105
Dyslipidaemia (0 = no, 1 = yes)			0.078
0
n	118	62
%	49.1	59.4
1
n	119	43
%	50.9	40.6
Total	237	105
Diabetes mellitus (0 = no, 1 = yes)			0.076
0
n	203	98
%	86.8	93.3
1
n	34	7
%	13.2	6.7
Total	237	105
Obesity (0 = no, 1 = yes)			0.009
0
n	173	90
%	72.8	85.7
1
n	64	15
%	27.2	14.3
Total	237	105
Smoke (0 = no, 1 = ex, 2 = active)			0.007
0
n	179	93
%	75.3	88.8
1
n	50	8
%	21.3	7.5
2
n	8	4
%	3.4	3.7
Total	237	105
Extracardiac arteriopathy (0 = no, 1 = yes)			0.796
0
n	215	94
%	90.6	91.5
1
n	22	9
%	9.4	8.5
Total	237	105
COPD (0 = no, 1 = yes)			0.426
0
n	209	89
%	88	84.9
1
n	28	16
%	12	15.1
Total	237	105
Preoperative CT (0 = no, 1 = yes)			<0.001
0
n	213	70
%	89.8	65.4
1
n	24	35
%	10.2	34.6
Total	237	105
NYHA (I, II, III, IV)			0.0576
I
n	18	8
%	6.9	5.8
II
n	140	52
%	59.5	51
III
n	67	38
%	27.2	34.6
IV
n	12	7
%	4.3	5.8
Total	237	105
Chronic kidney disease			0.589
0
n	221	96
%	93.2	91.5
1
n	16	9
%	6.8	8.5
Total	237	105
Preop cardiac rythm (0 = SR, 1 = permanent AF, 2 = paroxysmal AF, 3 = PM)			<0.001
0
n	193	61
%	81.6	58.9
1
n	21	29
%	8.5	27.1
2
n	21	14
%	8.5	13.1
3
n	2	1
%	0.9	0.9
Total	237	105
Previous cardiac surgery (0 = no, 1 = yes)			0.462
0
n	226	98
%	95.3	93.4
1
n	11	7
%	4.7	6.6
Total	237	105
Aortic pathology 0 = AS, 1 = AR, 2 = ASI			0.071
0
n	127	44
%	53.6	41.5
1
n	68	42
%	28.8	40.6
2
n	42	19
%	17.6	17.9
Total	237	105

AF: atrial fibrillation; AR: aortic regurgitation; AS: aortic stenosis; ASI: aortic steno-insufficiency; BMI: body mass index; BSA: body surface area; COPD: chronic obstructive pulmonary disease; CT: computerized tomography; EF: ejection fraction; GFR: glomerular filtration rate; NYHA: New York Heart Association; PM: pacemaker; SD: standard deviation; SR: sinus rhythm.

Intraoperative data

All preoperative variables are reported in Table 2. The mean CPB time and Cross-Clamp time was lower in the isolated group [130 (SD: 34) vs 184 (SD: 48) min, P < 0.001 and 88 (SD: 23) vs 130 (SD: 33) min, P < 0.001, respectively]. Prosthesis type in the isolated group were 89 stented (37%), 74 rapid deployment (31%) Intuity Elite (Edwards Lifesciences, Irvine CA), 74 sutureless (31%) Perceval Sutureless Valves (Corcym, London, UK); in the concomitant group were 51 stented (48%), 29 rapid deployment (27%), 26 sutureless (25%; P < 0.001). Concomitant procedures were 2 coronary artery bypass grafting on right coronary artery (1.9%), 21 ascending aorta replacement (19.6%), 41 mitral surgery (38.3%), 16 mitral and tricuspid (15%) and 25 other procedure (23.4%) (myectomy, AF Cryoablation, neoformation removal). Significant regurgitation requiring repositioning occurred only in 2 patients 0.80% of isolated group (P = 0.34), meanwhile conversion to full sternotomy occurred in 3 patients (1.30%) in the isolated group vs 1 patients (0.9%) in the concomitant group (P = 0.788). The reasons of conversions were: 1 case for impossibility to continue CPB due to elevated circuit resistance and no other peripheral vessel accessible because of severe vasculopathy, 1 case for left main coronary dissection treated with coronary artery bypass grafting and 2 cases of aortic haemorrhage non manageable trough mini-thoracotomy.

Table 2:

Intraoperative patients variables

	Isolated	Concomitant	P-Value
CPB time (min), mean (SD)	130 (34)	184(48)	<0.001
Cross-clamp time (min), mean (SD)	88 (23)	130 (33)	<0.001
Prosthesis type (0 = stented, 1 = rapid deployment, 2 = sutureless)			0.01
0
n	89	51
%	37	48
1
n	74	29
%	31	27
2
n	74	26
%	31	24.3
Total	237	105
Other surgical procedure (0 = no, 1 = CABG, 2 = ascending aorta replacement, 3 = mitral surgery, 4 = mitral + tricuspid, 5 = other)			<0.001
0
n	237	0
%	100	0
1
n	0	2
%	0	1.9
2
n	0	21
%	0	19.6
3
n	0	41
%	0	38.3
4
n	0	16
%	0	15
5
n	0	25
%	0	23.4
Total	237	105
Significant regurgitation requiring repositioning/new prosthesis (0 = no, 1 = yes)			0.34
0
n	235	105
%	99.2	100
1
n	2	0
%	0.8	0
Total	237	105
Conversion to full sternotomy (0 = no, 1 = yes)			0.788
0
n	234	104
%	98.7	99.1
1
n	3	1
%	1.3	0.9
Total	236	105

CABG: coronary artery bypass grafting; CPB: cardiopulmonary bypass; SD: standard deviation.

Postoperative outcome

All postoperative outcomes are reported in Table 3. Median ICU (hours) differ between the 2 groups [24 (28) vs 40 (52), P = 0.007], meanwhile OT Intubation time was not different [5 (2) vs 6 (8), P = 0.080], as well as median hospitalization days [5 (3) vs 6 (4), P = 0.083]. Death occurred 1 isolated group patient (0.4%) vs 2 concomitant group patients (1.9%; P = 0.17). AMI occurred only in 2 patients of concomitant group (1.9%, P = 0.034), stroke happened in 4 isolated group patients (1.7%) vs 3 concomitant group patients (2.85%; P = 0.48). Isolated group EAVR had lower permanent PM implant [5 (2.2%) vs 8 (7.6%), P = 0.01] and lower incidence of transfusions [63 (28.9%) vs 69 (64.4%) P < 0.001]. There was no difference in the new onset of AF [50 (21.6%) vs 20 (19.4%) patients, P = 0.658], continuous veno-venous hemofiltration need [1 (0.4%) vs 1 (0.9%) patients, P = 0.549] and re-exploration for bleeding [13 (5.4%) vs 11 (10.4%), P = 0.096] always performed through the same access.

Table 3:

Postoperative results

	Isolated	Concomitant	P-Value
ICU (h), median [interquartile range]	24 [28]	40 [52]	0.007
OT intubation time in ICU (h), median [interquartile range]	5 [2]	6 [8]	0.080
Postoperative Hospitalization (days), median [interquartile range]	5 [3]	6 [4]	0.083
Creatinine (peak within 72 h) (mg/dl), mean (SD)	0.88 (0.78)	0.90 (0.44)	0.805
All-cause mortality (0 = no, 1 = yes)			0.17
0
n	236	103
%	99.6	97.2
1
n	1	2
%	0.4	1.9
Total	237	105
AMI (0 = no, 1 ≤ 72 h)			0.034
0
n	237	103
%	100	98.1
1
n	0	2
%	0	1.9
Total	237	105
Stroke (0 = no, 1 = yes)			0.48
0
n	233	102
%	98.3	97.1
1
n	4	3
%	1.7	2.8
Total	237	105
Re-exploration for bleeding (0 = no, 1 = same access, 2 = sternotomy)			0.096
0
n	224	94
%	94.6	89.6
1
n	13	11
%	5.4	10.4
2
n	0	0
%	100	100
Total	237	105
Vascular complication (0 = no, 1 = minor, 2 = major)			0.569
0
n	235	103
%	98.7	98.1
1
n	1	0
%	0.4	0
2
n	2	2
%	0.9	1.9
Total	237	105
New permanent PM implantation (0 = no, 1 = yes)			0.01
0
n	232	97
%	97.8	92.4
1
n	5	8
%	2.2	7.6
Total	237	105
New-onset AF during hospitalization (0 = no, 1 = yes)			0.658
0
n	187	85
%	78.4	80.6
1
n	50	20
%	21.6	19.4
Total	237	105
CVVH (0 = no, 1 = yes)			0.549
0
n	236	104
%	99.6	99
1
n	1	1
%	0.4	1
Total	237	105
Trasfusions (0 = no, 1 = yes)			<0.001
0
n	174	36
%	71.1	38.8
1
n	63	69
%	28.9	64.4
Total	237	105
All-cause mortality (0 = no, 1 = yes)			0.175
0
n	236	103
%	99.6	98.1
1
n	1	2
%	0.4	1.9
Total	237	105

AF: atrial fibrillation; AMI: acute myocardial infarction; CVVH: continous veno-venous haemodialysis; ICU: intensive care unit; OT: oro-tracheal; PM: pacemaker; SD: standard deviation.

DISCUSSION

In this study, we aimed to report early outcomes of 342 consecutive patients who underwent minimally invasive E-AVR at our Institution, including 105 cases were a concomitant procedure was needed. We also sought to evaluate the feasibility and any additional risks of a concomitant procedure performed through the same working port.

In recent years, the interest in minimally invasive techniques for aortic valve surgery has become increasingly popular, encouraged by the widespread success of the TAVR and by the patients request for the least invasive approach. In the last years, it has been showed that MICS compared to full sternotomy is associated with reduced death, ICU stay, hospitalization, renal failure, transfusions and postoperative pain, with slightly longer durations of cross-clamping and CPB time [12]. Moreover, it has been showed that the mini-right thoracotomy approach had lower incidence of postoperative AFib, blood transfusion, shorter ventilation time and hospital length of stay, when compared to the hemi-sternotomy approach [13].

Despite these findings, the incidence of minimal access for AVR has not increased enough across Europe over the last years and nowadays, it is reported that it is ∼25% in Germany but only 12% in the UK [2], while in the USA, it is 12% for hemi-sternotomy and only 3.6% for a thoracotomy approach [14].

Moreover, to the best of our knowledge, there are few studies that have analyzed feasibility and outcomes of minimally invasive approach for aortic pathology and concomitant major procedures [15–17].

As previously described by our group [9], all types of aortic prosthesis can be used with this approach and the choice of the prosthetic valve was not influenced by the surgical technique. Aortic stented valve were preferred in elliptical annulus (e.g. bicuspid valve), which have higher chances of perivalvular leaks; in aortic regurgitation with big annulus to avoid an overdistension effect expected in rapid deployment/sutureless prosthesis; in endocarditis to reduce at minimum the foreign prosthetic material; in patients with complete mitral ring or mitral prosthesis to reduce the risk of movement interference of mitro-aortic prosthesis; in young patients because there is a well-known follow-up and durability. Sutureless prosthesis were preferably implanted in patients with tricuspid valve; small circular annulus with adequate amount of calcium to let the prosthesis anchor; in case of calcific aortic root to reduce manipulation and plaque dislodgement risk; in leftward aortic position which renders more difficult hand-tying. Rapid deployment prosthesis were preferred in tricuspid valve, circular annulus, in case of higher risk of patient prosthesis mismatch and leftward aortic position which renders more difficult hand-tying.

In our study, the overall 30-day mortality rate was 0.9% and it was higher in the subgroup of concomitant cases than isolated AVR (1.9% vs 0.4%) although the difference was not statistically significant (P = 0.17). Our finding is similar to the 1.3% in isolated and 3.2% in concomitant mini-AVR 30-day mortality reported by Lamelas et al. [15]. Moreover, these data are very encouraging even if compared to data from the Society of Thoracic Surgeon database, showing a 30-day and operative mortality of 1.6% and 2.2% in a study population of 28 037 patients who underwent AVR [18].

It is well known that stroke is one of the worst complications following aortic valve surgery that adversely affects survival and quality of life [19]. In a recent study, Kapadia et al. aimed to assess stroke risk after surgical and transcatheter AVR performing a propensity-matched study of 1204 pairs of patients with severe aortic stenosis in the PARTNER trial and they showed that 30-day major stroke rate was 3.9% in the surgical group vs 2.2% in TAVR patients [20]. In our series, the overall stroke rate was 2% and an additional procedure did not increase the risk of stroke in a statistically significative way, although the stroke rate was higher in concomitant cases (1.7% vs 2.85%, P = 0.48).

It is also demonstrated that aortic valve surgery is associated to the risk of periprocedural conduction disturbances requiring new PM implantation. Data from the ‘German Aortic Valve Registry’ showed that the incidence of new PM implantation was 16.6% after TAVR and 3.6% after surgical AVR [21]. In our study, the new permanent PM implantation rate was 2.1% in isolated AVR and 7.6% in concomitant cases (P < 0.01). Our data confirm the superiority of surgical AVR versus TAVR in terms of periprocedural new PM implantation rate. We can also speculate that the low incidence of the new PM implantation in our isolated AVR cases is due to the use of the thoracoscope, which allows a greater surgical accuracy. Instead, the higher PM implantation rate in concomitant group can be explained by the percentage of tricuspid surgery (19 cases, 18%), AF cryo-ablation (5 cases, 4.7%) and myectomy (9 cases, 8.5%). Procedures that in literature have incidence of PM implantation up to 10% [22, 23].

From our casuistic, it emerged that adding a procedure to the isolated AVR resulted in an increase in AMI rate, which occurred only in 2 patients (1.9%) who underwent concomitant EAVR P = 0.034, increase in blood transfusion that occurred in 63 (28.9%) insolated EAVR patients vs 69 (64.4%) concomitant EAVR (P < 0.001), and prolonged ICU hours [median 24 (28) in isolated EAVR vs 40 (52) in concomitant EAVR, P = 0.007].

On the other hand, the incidence of postoperative AF was not influenced by the presence of a concomitant procedure, since it has been observed in 50 cases in isolated AVR and 20 in concomitant cases (21.6% vs 19.4%, P = 0.65), as well as the re-exploration for bleeding, always performed through the same access, which show a higher but non significative rate [13 (5.4%) vs 11 (10.4%), P = 0.096). Furthermore, the median OT intubation time [5 (2) vs 6 (8), P = 0.080) and median hospitalization days [5 (3) vs 6 (4), P = 0.083] did not different between the 2 groups.

Not surprisingly CPB and cross-clamp time were higher in the concomitant group due to the associated procedure [130.1 (SD: 33.1) and 184.2 (SD: 48.2) vs 88.1 (SD: 23.1) and 130.5 (SD: 34.4) min, P < 0.001].

Depending of the center’s experience, the need to convert from mini-right thoracotomy to sternotomy occurs in 0–8.0% in all series [24]. In our experience, conversion to full sternotomy occurred in 3 isolated EAVR patients (1.3%) vs 1 concomitant EAVR patient (0.9%; P = 0.78). We believe that the video-thoracoscopic vision allows a superior control of any minor complications and it can avoid some conversion to sternotomy, but from our experience, it is advisable to start with isolated AVR, then adding concomitant procedures due to the longer learning curve.

Limitations

This is a retrospective, observational, single-center study with relatively small casuistic therefore small sample size that could be underpowered to detect statistically significant difference in some comparisons. Furthermore, imputation was not adjusted for confounders so there could be some undetected bias. Moreover, we reported only the operative and 30-day outcomes and we did not determine the long-term outcomes. Our database contains only EuroSCORE II records and the Society of Thoracic Surgeon score was not assessed. Furthermore, this study is a single-surgeon (Loris Salvador) experience with this approach and it includes an unavoidable learning curve.

CONCLUSIONS

This study shows that through a single working port made for E-AVR a concomitant procedure is feasible and in our casuistic and experienced center, it can be done safely and effectively maintaining good in-hospital mortality, postoperative stroke rate, hospitalization days and mechanical ventilation hours at the expense of higher incidence of new PM implantation, of AMI and longer ICU hours stay.

In our experience, the use of the thoracoscope was helpful to conduct this kind of operation. It allowed a larger visualization field, was very useful to reach the aorta regardless of its position and to manage concomitant surgical procedures through the same single working port. This technique requires a high experience in minimally invasive surgery, especially in endoscopic surgery. Before approaching this technique, a proper training is mandatory.

Glossary

ABBREVIATIONS

AF

Atrial fibrillation

AMI

Acute myocardial infarction

AVR

Aortic valve replacement

CPB

Cardiopulmonary bypass

E-AVR

Endoscopic aortic valve replacement

ICU

Intensive care unit

OT

Oro-tracheal

PM

Pacemaker

SD

Standard deviation

TAVR

Transcatheter aortic valve replacement

 

Presented at the EACTS Annual Meeting, Milan, Italy, 7 October 2022.

Funding

All authors have nothing to disclose with regard to commercial support.

Conflict of interest: none declared.

DATA AVAILABILITY

All relevant data are available on request to the authors.

Author contributions

Daniele Zoni: Conceptualization. Giovanni Domenico Cresce: Supervision. Tommaso Hinna-Danesi: Investigation. Luciana Benvegnù: Data curation. Salvatore Poddi: Data curation. Michele Gallo: Software. Massimo Sella: Data curation. Loris Salvador: Conceptualization.

Reviewer information

Interdisciplinary CardioVascular and Thoracic Surgery thanks Antonios Pitsis, Tomislav Kopjar and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.