Impact of porcelain aorta on outcomes in transfemoral aortic valve implantation (TAVI) with balloon-expandable valves

Abstract

Background

Current guidelines recommend transcatheter aortic valve implantation (TAVI) for patients with aortic stenosis and porcelain aorta (PA). Neurological outcomes of patients with PA undergoing TAVI with modern valves require clarification as most trials examined balloon-expandable valves (BEV) and self-expandable valves in intermediate or high-risk patients, but not specifically in patients with PA. Our aim was to compare outcomes, including stroke and mortality, in well-matched patients with and without PA who received BEV during transfemoral TAVI procedures.

Methods

Consecutive patients undergoing TAVI were entered into a registry. For this single-centre (Zentralklinik Bad Berka, Germany), retrospective analysis, we only selected patients who received BEV. PA diagnosis was made when non-contrast axial CT images fulfilled Valve Academic Research Consortium-2 criteria for PA. There was 2:1 nearest neighbour matching of patients without and with PA. The primary outcome measure was 30-day mortality or stroke within 72 hours. Secondary outcome measures were 30-day mortality, stroke within 72 hours, technical success and 30-day device success.

Results

After matching patients with (n=141) and without PA (n=282), the primary outcome of mortality at <30 days or stroke within 72 hours was higher in PA versus non-PA (7.8% vs 2.5%; OR 3.32 (95% CI 1.25 to 8.85); p=0.019). With regard to secondary outcomes, PA was not associated with mortality at 30 days (4.3% vs 2.1%; OR 2.04 (95% CI 0.65 to 6.48); p=0.23); however, stroke within 72 hours was significantly higher in PA versus non-PA (3.5% vs 0.4%; OR 10.33 (95% CI 1.17 to 91.12); p=0.017). Technical and device success were uninfluenced by PA.

Conclusions

Transfemoral TAVI with BEV in patients with PA was associated with a higher risk of the primary combined endpoint of mortality at 30 days or stroke within 72 hours, which was primarily driven by stroke within 72 hours. These findings might influence cerebral embolic protection device use in patients with PA.

Correspondence to Dr Philipp Lauten; [email protected]

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Porcelain aorta (PA) is a risk factor for neurological complications secondary to embolisation during surgical aortic valve replacement; hence, transcatheter aortic valve implantation (TAVI) is recommended; however, some studies have suggested that the risk of stroke is still increased in patients with PA undergoing TAVI. The role of valve type and vascular access route with respect to periprocedural stroke risk in patients with PA undergoing TAVI remains unclear.

WHAT THIS STUDY ADDS

• We found that there was an increased risk of clinically evident stroke in patients with PA who underwent transfemoral TAVI with a balloon-expandable valve as compared with patients without PA who underwent the same procedure.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Consideration could be given to the use of cerebral embolic protection devices when performing TAVI in patients with PA.

• The type of implantation mechanism (ballon-expandable vs self-expandable valves) is particularly important in cases of PA involvement and must be a key point of discussion within the heart valve team.

Introduction

Porcelain aorta (PA) is generally defined as severe circumferential or near circumferential calcification of the ascending aorta, extending to the arch, such that aortic cross-clamping during an open surgical procedure is not considered safe.^1–3 It has been noted that the incidence of PA reported in the literature is highly variable. This may be secondary to the lack of a more precise definition, and radiologically based definitions and classifications are being proposed.⁴ Regardless of its incidence, PA presents the potential for serious neurological complications secondary to embolisation during aortic cross-clamping, and the most recent European Society of Cardiology/European Association for Cardio-Thoracic Surgery Guidelines recommends transcatheter aortic valve implantation (TAVI) over surgical aortic valve replacement (SAVR) when a patient has aortic stenosis (AS) in the setting of PA.⁵ An updated consensus document from Valve Academic Research Consortium-2 (VARC-2) also states that PA is a comorbidity that should be considered as a risk factor when deciding between TAVI and SAVR for a patient with AS, although PA is not part of the European System for Cardiac Operative Risk Evaluation (EuroSCORE) or the Society of Thoracic Surgeons score.¹

For more than 20 years, percutaneous aortic valve (AV) implantation has been used in patients who were both symptomatic and at high surgical risk. Included in these early patients were some with PA.^{6 7} Over time, implantation techniques and available valves have evolved. Recently, there has been interest in the clinical outcomes of patients with PA who undergo TAVI in the current era. It has been reported that TAVI patients with PA have a threefold increased risk of periprocedural stroke in comparison to patients without PA; however, the increased risk was only seen in patients treated by the transapical or trans-subclavian route and not the transfemoral route.⁸ This was surprising as small case series suggested that the transapical approach was superior for patients with PA because the delivery catheter was not manoeuvred by the operator in the highly calcified aorta.^{9 10} Most, but not all, of these patients were treated using balloon-expandable valves (BEV), which naturally raises the question if valve generation and valve type could have had some impact on TAVI outcomes in patients with PA. Most randomised controlled trials and registries looking at the current generations of BEV in comparison to self-expandable valves (SEV) were conducted in high to intermediate-risk patients, but were not specifically addressing patients with PA.^11–14

Recently, clinically relevant outcomes in patients with PA undergoing TAVI with SEV were evaluated and no significant differences were found compared with patients without PA, including stroke and 30-day mortality.¹⁵ Unfortunately, there is still a lack of information on patients with PA receiving BEV via the transfemoral route. In this study, we compared outcomes, including stroke and mortality, in well-matched patients with and without PA who received a BEV as part of their TAVI procedures.

Methods

Patients and database

All patients were treated at the Zentralklinik Bad Berka, which has the largest heart centre in the German Federal State of Thuringia. Consecutive patients were entered into valve registry in place at our Heart Valve Unit and provided written informed consent. This study enrolled patients with symptomatic and/or severe AS who underwent TAVI during the time period from 2017 to 2022. All data were pseudoanonymised prior to undergoing data handling and professional statistical analysis.

Patients were thoroughly evaluated prior to discussion by our institution’s Heart Team. This included current and past medical and surgical histories, preprocedure blood work and EKG, planning multidetector CT, preprocedural transthoracic echocardiography (TTE) and determination of EuroSCORE II. All information was entered into the registry database. Procedural characteristics, valve type used, complications, postprocedural TTE and EKG and clinical findings on follow-up were all entered prospectively into the database.

In order to reduce the risk of thromboembolism, we carefully followed anticoagulation protocols as part of our standard care of TAVI patients. Patients who had no baseline indication for anticoagulation were treated with aspirin (75–100 mg daily) and clopidogrel (75 mg daily) for 6 months, followed by aspirin monotherapy indefinitely. Patients who had an indication for chronic anticoagulation at baselines (eg, atrial fibrillation (AF)) were treated either with a direct oral anticoagulant (DOAC) monotherapy or, in some cases, with combination therapy consisting of a DOAC and a single antiplatelet therapy.

For those patients who exhibited neurological abnormalities immediately after TAVI, we followed, in collaboration with the Neurology Department, institutional standard operating procedures. This included cerebral imaging (cerebral computed tomography) and, if necessary, an interventional procedure such as thrombectomy. Thus, stroke monitoring and therapy were conducted according to dedicated protocols.

PA definition

Patients were considered to have PA when their non-contrast axial CT images fulfilled VARC-2 criteria for PA, namely severe circumferential calcification or severe atheromatous plaques of the entire ascending aorta extending to the arch, as shown in figure 1.¹⁶ All imaging studies were reviewed and ultimately classified by one investigator (PL).

Figure 1. MSCT (Multislice Computed Tomography) classification of the extent of ascending aorta atherosclerotic burden: (A) circular, near confluent calcification of the ascending aorta, (B) absence of relevant atherosclerotic lesions. Axial views of the aorta at the level of pulmonary trunk: (C) circular calcification, (D) no or minor calcification.

Statistical analysis

From the registry database, patients who did not receive a BEV or who had missing data relevant to matching or outcomes analyses were excluded. All patients included underwent TAVI by the femoral access route.

Similar to Eckel and colleagues,¹⁵ the MatchIt package in R programming language (Comprehensive R Archive Network at http://CRAN.R-project.org/package=MatchIt) was used to perform 2:1 nearest neighbour matching of patients without and with PA. The variables used for this propensity matching included sex, hypertension, diabetes, coronary artery disease, peripheral arterial disease, stroke, chronic obstructive pulmonary disease, AF, prior right bundle branch block, prior left bundle branch block, left ventricular outflow tract calcification, eccentric AV calcification, age, EuroSCORE II, cusp calcification and estimated glomerular filtration rate (mL/min/1.73 m²).

The PA and non-PA groups were then tested to confirm that there were no baseline differences between them.

After matching, the outcome analyses were performed. The influence of different variables of the outcomes of interest was tested in the matched population of patients with and without PA. The primary outcome measure was 30-day mortality or stroke (major or minor) within 72 hours. The secondary outcome measures were 30-day mortality, 30-day stroke, technical success and device success at 30 days.

Device success was defined according to VARC-2 criteria and included absence of procedural mortality and correct positioning of a single prosthetic heart valve into the proper anatomical location and no prosthesis-patient mismatch (intended performance of prosthetic valve), and mean AV gradient <20 mm Hg or peak velocity <3 m/s and no moderate/severe prosthetic valve regurgitation.¹⁶

Technical success was defined according to VARC-3 criteria and included freedom from mortality; successful access, delivery of system and retrieval of the delivery system; correct position of single prosthetic valve; and freedom from surgery or intervention related to the device or due to a major vascular or access related or cardiac structural complication.¹⁷

Categorical data were described using frequencies, continuous data were described by mean, SD, median and minimum and maximum (min–max). For the comparison of PA and non-PA, Fisher’s exact test (categorical data) and Mann-Whitney U test (continuous data) were used. ORs with 95% CIs were calculated to quantify associations between and binary or continuous data and outcome variables.

All statistical tests were two sided at a significance level of 0.05.

Results

Comparison of baseline characteristics

From 2017 to 2022, a total of 1779 consecutive patients were recruited into our valve registry. After excluding patients who did not receive a BEV or who had missing data needed for the matching process or analyses, 800 patients without PA and 141 patients with PA were available. Ultimately, 282 patients without PA and 141 patients with PA were used for the 2:1 nearest neighbour matching. These patients all received Edwards SAPIEN BEV (Edwards Lifesciences, Irvine, California, USA). The baseline descriptions and comparisons of non-PA and PA groups are found in table 1. The variables used for matching are labelled with ‘MA’. At baseline, there were no significant differences between the non-PA and PA groups.

Table 1

Description and comparison of groups—matched population (categorical data)

			Non-PA (n=282)	PA (n=141)	P value*
MA	Sex	Male	168 (59.6%)	81 (57.4%)	0.677
		Female	114 (40.4%)	60 (42.6%)
MA	Hypertension	No	48 (17.0%)	25 (17.7%)	0.892
		Yes	234 (83.0%)	116 (82.3%)
MA	Diabetes	No	150 (53.2%)	75 (53.2%)	1.000
		Yes	132 (46.8%)	66 (46.8%)
MA	Coronary artery disease	No	98 (34.8%)	56 (39.7%)	0.336
		Yes	184 (65.2%)	85 (60.3%)
MA	Peripheral arterial disease	No	255 (90.4%)	127 (90.1%)	1.000
		Yes	27 (9.6%)	14 (9.9%)
MA	Stroke	No	275 (97.5%)	137 (97.2%)	1.000
		Yes	7 (2.5%)	4 (2.8%)
MA	COPD	No	249 (88.3%)	125 (88.7%)	1.000
		Yes	33 (11.7%)	16 (11.3%)
MA	AF	No	191 (67.7%)	97 (68.8%)	0.912
		Yes	91 (32.3%)	44 (31.2%)
MA	RBBB (pre-existing)	No	255 (90.4%)	126 (89.4%)	0.732
		Yes	27 (9.6%)	15 (10.6%)
MA	LBBB (pre-existing)	No	254 (90.1%)	127 (90.1%)	1.000
		Yes	28 (9.9%)	14 (9.9%)
	LVEF (%)	55–65	120 (44.4)	58 (42.6)	0.507
		45–55	71 (26.3)	30 (22.1)
		35–45	36 (13.3)	25 (18.4)
		<35	43 (15.9)	23 (16.9)
MA	LVOT calcification	No	266 (94.3%)	135 (95.7%)	0.646
		Yes	16 (5.7%)	6 (4.3%)
MA	Eccentric AV calcification	No	266 (94.3%)	133 (94.3%)	1.000
		Yes	16 (5.7%)	8 (5.7%)
	Predilatation	No	1 (0.4%)	1 (0.7%)	1.000
		Yes	281 (99.6%)	140 (99.3%)
	Postdilatation	No	274 (97.2%)	134 (95.0%)	0.275
		Yes	8 (2.8%)	7 (5.0%)
	Contrast media	No	2 (0.7%)	0 (0%)	0.555
		Yes	280 (99.3%)	141 (100%)
	Protection device	No	268 (95.0%)	137 (97.2%)	0.444
		Yes	14 (5.0%)	4 (2.8%)
	Acute renal failure	No	278 (98.6%)	141 (100%)	0.306
		AKIN I	4 (1.4%)	0 (0%)
	New permanent pacemaker	No	244 (86.5%)	123 (87.2%)	0.880
		Yes	38 (13.5%)	18 (12.8%)
	AV block ()	No	252 (89.4%)	122 (86.5%)	0.422
		Yes	30 (10.6%)	19 (13.5%)
	NYHA	1	9 (3.3%)	1 (0.7%)	0.434
		2	70 (25.6%)	35 (25.2%)
		3	155 (56.8%)	85 (61.2%)
		4	39 (14.3%)	18 (12.9%)
	Porcelain aorta	None	282 (100%)	–	–
		Partial	–	111 (78.7%)
		Circular	–	30 (21.3%)

Description and comparison of groups—matched population (continuous data)
			n	Mean	SD	Median	Min–max	P value†
MA	Age	Non-PA	282	78.0	6.4	79.0	59.0–91.0	0.449
		PA	141	78.2	7.6	80.0	51.0–90.0
	BMI (kg/m2)	Non-PA	282	29.2	5.5	28.4	16.3–50.4	0.958
		PA	141	29.2	5.7	28.4	18.7–49.3
	Length of stay post-TAVR in days	Non-PA	281	8.0	4.8	7.0	1.0–33.0	0.068
		PA	141	9.4	7.0	8.0	0.0–52.0
MA	EuroSCORE II (%)	PA	282	9.3	10.2	5.2	0.8–62.1	0.742
		Non-PA	141	9.4	10.4	5.3	0.8–60.2
	Aortic valve gradient (BV)	PA	256	44.3	15.2	43.0	13.0–106.0	0.203
		Non-PA	130	42.4	14.3	41.0	13.0–100.0
	Aortic valve area (cm2) (BW)	PA	272	0.8	0.2	0.7	0.3–1.8	0.354
		Non-PA	132	0.7	0.2	0.7	0.2–1.5
	Average diameter, mm	PA	277	24.9	2.4	25.0	19.5–31.0	0.224
		Non-PA	141	25.2	2.4	25.0	19.5–31.5
	Area, mm2	PA	194	484.7	94.2	472.0	301.0–726.0	0.082
		Non-PA	87	507.7	103.9	489.0	232.0–780.0
MA	Cusp calcification Agatston	PA	282	3267.4	1705.6	3110.5	365.0–6445.0	0.776
		Non-PA	141	3237.3	1801.0	3046.0	505.0–7890.0
	LVEF, %	PA	277	52.1	13.4	55.0	18.0–80.0	0.098
		Non-PA	138	50.0	12.2	52.0	15.0–76.0
	Maximum velocity, m/s	PA	279	2.3	0.5	2.3	1.2–5.1	0.085
		Non-PA	139	2.3	0.5	2.2	1.1–4.7
	Aortic valve area, cm2	PA	3	1.6	0.9	1.8	0.6–2.4	0.825
		Non-PA	3	1.5	1.0	2.1	0.4–2.1
	Creatinine, mg/dL	PA	282	1.0	0.8	0.9	0.5–11.6	0.961
		Non-PA	141	1.2	1.0	0.9	0.5–6.8
MA	eGFR, mL/min/1.73 m2	PA	282	69.2	21.1	74.4	3.9–114.5	0.726
		Non-PA	141	67.6	24.4	71.7	5.2–116.1

MA indicates a variable used for matching.

*Fisher’s exact test.

†Mann-Whitney U equality-of-populations rank test.

AF, atrial fibrillation; AV, aortic valve; BMI, body mass index; COPD, chronic obstructive pulmonary disease ; eGFR, estimated glomerular filtration rate; EuroSCORE, European System for Cardiac Operative Risk Evaluation; LBBB, left bundle branch block; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; NYHA, New York Heart Association; PA, porcelain aorta; RBBB, right bundle branch block ; TAVR, transcatheter aortic valve replacement.

Procedural outcomes and complications in matched population

Using the matched population of patients without and with PA, the influence of the variables in table 1 on the predefined primary and secondary outcomes was determined. These results are available in tables 2–6.

Table 2

Mortality at ≥30 days or stroke within 72 hours in 2:1 matched non-PA and PA groups (categorical data)

		n	Mortality or stroke		P value* OR (95% CI)
			No	Yes
Sex	Male	249	238 (95.6%)	11 (4.4%)	1.000 0.91 (0.34 to 2.39)
	Female	174	167 (96.0%)	7 (4.0%)
Hypertension	No	73	70 (95.9%)	3 (4.1%)	1.000 1.04 (0.29 to 3.71)
	Yes	350	335 (95.7%)	15 (4.3%)
Diabetes	No	225	217 (96.4%)	8 (3.6%)	0.478 1.44 (0.56 to 3.74)
	Yes	198	188 (94.9%)	10 (5.1%)
Coronary artery disease	No	154	149 (96.8%)	5 (3.2%)	0.618 1.51 (0.53 to 4.34)
	Yes	269	256 (95.2%)	13 (4.8%)
Peripheral arterial disease	No	382	365 (95.5%)	17 (4.5%)	1.000 0.54 (0.07 to 4.15)
	Yes	41	40 (97.6%)	1 (2.4%)
Stroke	No	412	395 (95.9%)	17 (4.1%)	0.384 2.32 (0.28 to 19.29)
	Yes	11	10 (90.9%)	1 (9.1%)
COPD	No	374	357 (95.5%)	17 (4.5%)	0.459
	GOLD A	15	15 (100%)	0 (0%)
	GOLD B	22	22 (100%)	0 (0%)
	GOLD C	6	5 (83.3%)	1 (16.7%)
	GOLD D	6	6 (100%)	0 (0%)
COPD (no/yes)	No	374	357 (95.5%)	17 (4.5%)	0.708 0.44 (0.06 to 3.38)
	Yes	49	48 (98.0%)	1 (2.0%)
AF	No	288	278 (96.5%)	10 (3.5%)	0.301 1.75 (0.67 to 4.55)
	Yes	135	127 (94.1%)	8 (5.9%)
RBBB (pre-existing)	No	381	365 (95.8%)	16 (4.2%)	0.697 1.14 (0.25 to 5.15)
	Yes	42	40 (95.2%)	2 (4.8%)
LBBB (pre-existing)	No	381	365 (95.8%)	16 (4.2%)	0.697 1.14 (0.25 to 5.15)
	Yes	42	40 (95.2%)	2 (4.8%)
LVEF (%) (BS)	55–65	178	172 (96.6)	6 (3.4)	0.114
	45–55	101	99 (98.0)	2 (2.0)
	35–45	61	55 (90.2)	6 (9.8)
	<35	66	63 (95.5)	3 (4.5)
LVOT calcification	No	401	384 (95.8%)	17 (4.2%)	1.000 1.08 (0.14 to 8.49)
	Yes	22	21 (95.5%)	1 (4.5%)
Eccentric AV calcification	No	399	382 (95.7%)	17 (4.3%)	1.000 0.98 (0.12 to 7.69)
	Yes	24	23 (95.8%)	1 (4.2%)
Predilatation	No	2	2 (100%)	0 (0%)	1.000
	Yes	421	403 (95.7%)	18 (4.3%)
Postdilatation	No	408	391 (95.8%)	17 (4.2%)	0.485 1.64 (0.20 to 13.27)
	Yes	15	14 (93.3%)	1 (6.7%)
Contrast media	No	2	1 (50.0%)	1 (50.0%)	0.083 0.04 (0.00 to 0.73)
	Yes	421	404 (96.0%)	17 (4.0%)
Protection device	No	405	387 (95.6%)	18 (4.4%)	1.000
	Yes	18	18 (100%)	0 (0%)
Acute renal failure	No	419	404 (96.4%)	15 (3.6%)	<0.001 80.80 (6.81 to 958.51)
	AKIN I	4	1 (25.0%)	3 (75.0%)
New permanent pacemaker	No	367	353 (96.2%)	14 (3.8%)	0.278 1.94 (0.61 to 6.14)
	Yes	56	52 (92.9%)	4 (7.1%)
AV block (BO, BP)	No	374	359 (96.0%)	15 (4.0%)	0.451 1.56 (0.43 to 5.61)
	Yes	49	46 (93.9%)	3 (6.1%)
Porcelain aorta	No	282	275 (97.5%)	7 (2.5%)	0.019 3.32 (1.25 to 8.85)
	Yes	141	130 (92.2%)	11 (7.8%)
NYHA	I	10	10 (100%)	0 (0%)	0.236
	II	105	103 (98.1%)	2 (1.9%)
	III	240	230 (95.8%)	10 (4.2%)
	IV	57	52 (91.2%)	5 (8.8%)

Mortality at ≤30 days or stroke within 72 hours in 2:1 matched non-PA and PA groups (continuous data)
	Mortality or stroke	n	Mean	SD	Median	Min–max	P value† OR (95% CI)
Age	No	405	78.1	6.8	79.0	51.0–91.0	0.616 1.01 (0.94 to 1.08)
	Yes	18	78.5	7.5	80.5	64.0–89.0
BMI (kg/m2)	No	405	29.2	5.5	28.4	16.3–50.4	0.376 0.97 (0.89 to 1.06)
	Yes	18	28.3	6.2	26.7	19.6–39.1
Length of stay post-TAVR (days)	No	405	8.2	4.9	7.0	1.0–38.0	0.001 1.29 (1.18 to 1.40)
	Yes	17	16.4	13.0	12.0	0.0–52.0
EuroSCORE II (%)	No	405	9.2	10.3	5.2	0.8–62.1	0.183 1.02 (0.97 to 1.07)
	Yes	18	11.3	9.6	7.4	1.6–36.8
Aortic valve gradient (BV)	No	369	43.9	15.1	42.0	13.0–106.0	0.262 0.98 (0.95 to 1.01)
	Yes	17	39.2	10.9	40.0	19.0–56.0
Aortic valve area (cm2) (BW)	No	386	0.7	0.2	0.7	0.2–1.8	0.782 1.03 (0.11 to 10.12)
	Yes	18	0.7	0.1	0.8	0.5–1.0
Average diameter, mm	No	400	25.0	2.4	25.0	19.5–31.0	0.727 1.06 (0.87 to 1.29)
	Yes	18	25.3	2.9	25.0	21.1–31.5
Area, mm2	No	271	492.0	96.2	478.0	232.0–780.0	0.576 1.00 (0.99 to 1.01)
	Yes	10	488.0	140.1	461.0	345.0–720.0
Cusp calcification Agatston	No	405	3269.8	1747.3	3066.0	365.0–7890.0	0.521 1.00 (1.00 to 1.00)
	Yes	18	2977.2	1471.7	2944.5	700.0–5536.0
LVEF, %	No	404	51.5	13.0	55.0	15.0–80.0	0.505 0.98 (0.94 to 1.03)
	Yes	11	48.4	15.1	50.0	20.0–70.0
Maximum velocity, m/s	No	405	2.3	0.5	2.3	1.1–5.1	0.021 0.28 (0.09 to 0.90)
	Yes	13	2.0	0.4	2.0	1.5–2.9
Creatinine, mg/dL	No	405	1.1	0.7	0.9	0.5–6.8	0.113 3.06 (1.76 to 5.33)
	Yes	18	1.9	2.5	1.1	0.5–11.6
eGFR, mL/min/1.73 m2	No	405	69.2	21.7	74.3	5.2–116.1	0.128 0.98 (0.96 to 1.00)
	Yes	18	57.8	31.0	60.2	3.9–105.7

*Fisher’s exact test.

†Mann-Whitney U test.

AF, atrial fibrillation; AV, aortic valve; BMI, body mass index; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; EuroSCORE, European System for Cardiac Operative Risk Evaluation; LBBB, left bundle branch block; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; NYHA, New York Heart Association; PA, porcelain aorta; RBBB, right bundle branch block; TAVR, transcatheter aortic valve replacement.

Table 3

Mortality at ≤30 days in 2:1 matched non-PA and PA groups (categorical data)

		n	30-day mortality		P value* OR (95% CI)
			No	Yes
Sex	Male	249	240 (96.4%)	9 (3.6%)	0.374 0.47 (0.12 to 1.76)
	Female	174	171 (98.3%)	3 (1.7%)
Hypertension	No	73	70 (95.9%)	3 (4.1%)	0.443 0.62 (0.16 to 2.34)
	Yes	350	341 (97.4%)	9 (2.6%)
Diabetes	No	225	219 (97.3%)	6 (2.7%)	1.000 1.14 (0.36 to 3.60)
	Yes	198	192 (97.0%)	6 (3.0%)
Coronary artery disease	No	154	153 (99.4%)	1 (0.6%)	0.063 6.52 (0.82 to 51.68)
	Yes	269	258 (95.9%)	11 (4.1%)
Peripheral arterial disease	No	382	370 (96.9%)	12 (3.1%)	0.616
	Yes	41	41 (100%)	0 (0%)
Stroke	No	412	400 (97.1%)	12 (2.9%)	1.000
	Yes	11	11 (100%)	0 (0%)
COPD	No	374	363 (97.1%)	11 (2.9%)	0.378
	GOLD A	15	15 (100%)	0 (0%)
	GOLD B	22	22 (100%)	0 (0%)
	GOLD C	6	5 (83.3%)	1 (16.7%)
	GOLD D	6	6 (100%)	0 (0%)
COPD (no/yes)	No	374	363 (97.1%)	11 (2.9%)	1.000 0.69 (0.09 to 5.46)
	Yes	49	48 (98.0%)	1 (2.0%)
AF	No	288	281 (97.6%)	7 (2.4%)	0.533 1.54 (0.48 to 4.97)
	Yes	135	130 (96.3%)	5 (3.7%)
RBBB (pre-existing)	No	381	371 (97.4%)	10 (2.6%)	0.338 1.86 (0.39 to 8.79)
	Yes	42	40 (95.2%)	2 (4.8%)
LBBB (pre-existing)	No	381	370 (97.1%)	11 (2.9%)	1.000 0.82 (0.10 to 6.53)
	Yes	42	41 (97.6%)	1 (2.4%)
LVEF (%) (BS)	55–65	178	173 (97.2)	5 (2.8)	0.353
	45–55	101	100 (99.0)	1 (1.0)
	35–45	61	58 (95.1)	3 (4.9)
	<35	66	63 (95.5)	3 (4.5)
LVOT calcification	No	401	390 (97.3%)	11 (2.7%)	0.478 1.69 (0.21 to 13.74)
	Yes	22	21 (95.5%)	1 (4.5%)
Eccentric AV calcification	No	399	388 (97.2%)	11 (2.8%)	0.509 1.53 (0.19 to 12.43)
	Yes	24	23 (95.8%)	1 (4.2%)
Predilatation	No	2	2 (100%)	0 (0%)	1.000
	Yes	421	409 (97.1%)	12 (2.9%)
Postdilatation	No	408	396 (97.1%)	12 (2.9%)	1.000
	Yes	15	15 (100%)	0 (0%)
Contrast media	No	2	1 (50.0%)	1 (50.0%)	0.056 0.03 (0.00 to 0.49)
	Yes	421	410 (97.4%)	11 (2.6%)
Protection device	No	405	393 (97.0%)	12 (3.0%)	1.000
	Yes	18	18 (100%)	0 (0%)
Acute renal failure	No	419	410 (97.9%)	9 (2.1%)	<0.001 136.67 (10.08 to 1853.25)
	AKIN I	4	1 (25.0%)	3 (75.0%)
New permanent pacemaker	No	367	359 (97.8%)	8 (2.2%)	0.060 3.45 (1.00 to 11.96)
	Yes	56	52 (92.9%)	4 (7.1%)
AV block (BO, BP)	No	374	364 (97.3%)	10 (2.7%)	0.638 1.55 (0.33 to 7.30)
	Yes	49	47 (95.9%)	2 (4.1%)
Porcelain aorta	No	282	276 (97.9%)	6 (2.1%)	0.226 2.04 (0.65 to 6.48)
	Yes	141	135 (95.7%)	6 (4.3%)
NYHA	I	10	10 (100%)	0 (0%)	0.353
	II	105	104 (99.0%)	1 (1.0%)
	III	240	232 (96.7%)	8 (3.3%)
	IV	57	54 (94.7%)	3 (5.3%)

Mortality at ≤30 days in 2:1 matched non-PA and PA groups (continuous data)
	Mortality	n	Mean	SD	Median	Min–max	P value† OR (95% CI)
Age	No	411	78.0	6.9	79.0	51.0–91.0	0.215 1.06 (0.98 to 1.15)
	Yes	12	80.8	4.9	81.0	72.0–89.0
BMI (kg/m2)	No	411	29.2	5.6	28.4	16.3–50.4	0.507 0.97 (0.87 to 1.07)
	Yes	12	28.2	5.5	26.7	19.9–37.7
Length of stay post-TAVR (days)	No	411	8.4	5.6	7.0	1.0–52.0	0.163 1.09 (0.98 to 1.21)
	Yes	11	11.3	8.4	10.0	0.0–29.0
EuroSCORE II (%)	No	411	9.2	10.3	5.1	0.8–62.1	0.012 1.04 (0.99–1.10)
	Yes	12	13.6	9.5	12.7	3.4–36.8
Aortic valve gradient (BV)	No	374	43.9	15.0	42.0	13.0–106.0	0.262 0.98 (0.94 to 1.01)
	Yes	12	38.4	12.0	38.5	19.0–56.0
Aortic valve area (cm2) (BW)	No	392	0.7	0.2	0.7	0.2–1.8	0.431 1.88 (0.12 to 30.18)
	Yes	12	0.8	0.2	0.8	0.5–1.0
Average diameter, mm	No	406	25.0	2.4	25.0	19.5–31.5	0.324 1.13 (0.89 to 1.43)
	Yes	12	25.7	2.5	25.5	21.5–29.5
Area, mm2	No	275	491.6	97.3	478.0	232.0–780.0	0.967 1.00 (0.99 to 1.01)
	Yes	6	499.8	128.5	471.0	361.0–718.0
Cusp calcification Agatston	No	411	3266.6	1744.5	3066.0	365.0–7890.0	0.547 1.00 (1.00 to 1.00)
	Yes	12	2940.0	1435.5	2631.0	700.0–5136.0
LVEF, %	No	409	51.5	13.0	55.0	15.0–80.0	p=0.293 0.95 (0.90 to 1.01)
	Yes	6	43.3	17.8	47.5	20.0–60.0
Maximum velocity, m/s	No	410	2.3	0.5	2.3	1.1–5.1	0.002 0.12 (0.03 to 0.51)
	Yes	8	1.8	0.4	1.8	1.5–2.7
Creatinine, mg/dL	No	411	1.0	0.7	0.9	0.5–6.8	0.004 6.44 (3.28 to 12.62)
	Yes	12	2.4	3.0	1.5	0.6–11.6
eGFR, mL/min/1.73 m2	No	411	69.3	21.7	74.4	5.2–116.1	0.005 0.96 (0.93 to 0.98)
	Yes	12	46.6	28.9	44.9	3.9–87.1

*Fisher’s exact test.

†Mann-Whitney U test.

AF, atrial fibrillation; AV, aortic valve; BMI, body mass index; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; EuroSCORE, European System for Cardiac Operative Risk Evaluation; LBBB, left bundle branch block; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; NYHA, New York Heart Association; PA, porcelain aorta; RBBB, right bundle branch block; TAVR, transcatheter aortic valve replacement.

Table 4

Stroke within 72 hours in 2:1 matched non-PA and PA groups (categorical data)

		n	Stroke within 72 hours		P value* OR (95% CI)
			No	Yes
Sex	Male	249	247 (99.2%)	2 (0.8%)	0.234 2.91 (0.52 to 16.13)
	Female	174	170 (97.7%)	4 (2.3%)
Hypertension	No	73	73 (100%)	0 (0%)	0.596
	Yes	350	344 (98.3%)	6 (1.7%)
Diabetes	No	225	223 (99.1%)	2 (0.9%)	0.425 2.30 (0.41 to 12.74)
	Yes	198	194 (98.0%)	4 (2.0%)
Coronary artery disease	No	154	150 (97.4%)	4 (2.6%)	0.196 0.28 (0.05 to 1.56)
	Yes	269	267 (99.3%)	2 (0.7%)
Peripheral arterial disease	No	382	377 (98.7%)	5 (1.3%)	0.460 1.88 (0.21 to 16.59)
	Yes	41	40 (97.6%)	1 (2.4%)
Stroke	No	412	407 (98.8%)	5 (1.2%)	0.147 8.14 (0.86 to 77.40)
	Yes	11	10 (90.9%)	1 (9.1%)
COPD	No	374	368 (98.4%)	6 (1.6%)	1.000
	GOLD A	15	15 (100%)	0 (0%)
	GOLD B	22	22 (100%)	0 (0%)
	GOLD C	6	6 (100%)	0 (0%)
	GOLD D	6	6 (100%)	0 (0%)
COPD (no/yes)	No	374	368 (98.4%)	6 (1.6%)	1.000
	Yes	49	49 (100%)	0 (0%)
AF	No	288	285 (99.0%)	3 (1.0%)	0.389 2.16 (0.43 to 10.88)
	Yes	135	132 (97.8%)	3 (2.2%)
RBBB (pre-existing)	No	381	375 (98.4%)	6 (1.6%)	1.000
	Yes	42	42 (100%)	0 (0%)
LBBB (pre-existing)	No	381	376 (98.7%)	5 (1.3%)	0.468 1.83 (0.21 to 16.14)
	Yes	42	41 (97.6%)	1 (2.4%)
LVEF (%) (BS)	55–65	178	177 (99.4)	1 (0.6)	0.066
	45–55	101	100 (99.0)	1 (1.0)
	35–45	61	58 (95.1)	3 (4.9)
	<35	66	66 (100)	0 (0)
LVOT calcification	No	401	395 (98.5%)	6 (1.5%)	1.000
	Yes	22	22 (100%)	0 (0%)
Eccentric AV calcification	No	399	393 (98.5%)	6 (1.5%)	1.000
	Yes	24	24 (100%)	0 (0%)
Predilatation	No	2	2 (100%)	0 (0%)	1.000
	Yes	421	415 (98.6%)	6 (1.4%)
Postdilatation	No	408	403 (98.8%)	5 (1.2%)	0.196 5.76 (0.62 to 53.16)
	Yes	15	14 (93.3%)	1 (6.7%)
Contrast media	No	2	2 (100%)	0 (0%)	1.000
	Yes	421	415 (98.6%)	6 (1.4%)
Protection device	No	405	399 (98.5%)	6 (1.5%)	1.000
	Yes	18	18 (100%)	0 (0%)
Acute renal failure	No	419	413 (98.6%)	6 (1.4%)	1.000
	AKIN I	4	4 (100%)	0 (0%)
New permanent pacemaker	No	367	361 (98.4%)	6 (1.6%)	1.000
	Yes	56	56 (100%)	0 (0%)
AV block (BO, BP)	No	374	369 (98.7%)	5 (1.3%)	0.524 1.54 (0.18 to 13.48)
	Yes	49	48 (98.0%)	1 (2.0%)
Porcelain aorta	No	282	281 (99.6%)	1 (0.4%)	0.017 10.33 (1.17 to 91.12)
	Yes	141	136 (96.5%)	5 (3.5%)
NYHA	I	10	10 (100%)	0 (0%)	0.290
	II	105	104 (99.0%)	1 (1.0%)
	III	240	238 (99.2%)	2 (0.8%)
	IV	57	55 (96.5%)	2 (3.5%)

Stroke within 72 hours in 2:1 matched non-PA and PA groups (continuous data)
	Stroke	n	Mean	SD	Median	Min–max	P value† OR (95% CI)
Age	No	417	78.1	6.8	79.0	51.0–91.0	0.376 0.91 (0.81 to 1.03)
	Yes	6	74.0	10.0	73.0	64.0–85.0
BMI (kg/m2)	No	417	29.2	5.5	28.4	16.3–50.4	0.562 0.97 (0.84 to 1.12)
	Yes	6	28.3	7.9	26.0	19.6–39.1
Length of stay post-TAVR (days)	No	416	8.2	5.1	7.0	0.0–38.0	<0.001 1.71 (1.49 to 1.97)
	Yes	6	25.7	15.5	18.5	14.0–52.0
EuroSCORE II (%)	No	417	9.4	10.3	5.2	0.8–62.1	0.216 0.97 (0.90 to 1.05)
	Yes	6	6.6	8.6	2.9	1.6–23.9
Aortic valve gradient (BV)	No	381	43.7	15.0	42.0	13.0–106.0	0.756 0.99 (0.93 to 1.05)
	Yes	5	41.0	8.6	41.0	29.0–53.0
Aortic valve area (cm2) (BW)	No	398	0.7	0.2	0.7	0.2–1.8	0.526 0.32 (0.01 to 15.49)
	Yes	6	0.7	0.1	0.7	0.6–0.8
Average diameter, mm	No	412	25.0	2.4	25.0	19.5–31.0	0.431 0.93 (0.67 to 1.30)
	Yes	6	24.6	3.7	24.2	21.1–31.5
Area, mm2	No	277	492.1	96.7	478.0	232.0–780.0	0.410 1.00 (0.99 to 1.01)
	Yes	4	470.2	175.2	408.0	345.0–720.0
Cusp calcification Agatston	No	417	3260.3	1738.5	3066.0	365.0–7890.0	0.803 1.00 (1.00 to 1.00)
	Yes	6	3051.7	1679.3	3190.0	1034.0–5536.0
LVEF, %	No	410	51.4	13.1	55.0	15.0–80.0	0.867 1.02 (0.95 to 1.09)
	Yes	5	54.4	9.7	50.0	45.0–70.0
Maximum velocity, m/s	No	413	2.3	0.5	2.3	1.1–5.1	0.791 1.20 (0.19 to 7.72)
	Yes	5	2.4	0.3	2.2	2.1–2.9
Creatinine, mg/dL	No	417	1.1	0.9	0.9	0.5–11.6	0.193 0.66 (0.26 to 1.71)
	Yes	6	0.8	0.3	0.8	0.5–1.2
eGFR, mL/min/1.73 m2	No	417	68.5	22.2	73.0	3.9–116.1	0.177 1.02 (0.99 to 1.06)
	Yes	6	80.3	22.9	85.2	42.0–105.7

*Fisher’s exact test.

†Mann-Whitney U test.

AF, atrial fibrillation; AV, aortic valve; BMI, body mass index; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; EuroSCORE, European System for Cardiac Operative Risk Evaluation; LBBB, left bundle branch block; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; NYHA, New York Heart Association; PA, porcelain aorta; RBBB, right bundle branch block; TAVR, transcatheter aortic valve replacement.

Table 5

Technical success in 2:1 matched non-PA and PA groups (categorical data)

		n	Technical success		P value* OR (95% CI)
			No	Yes
Sex	Male	249	15 (6.0%)	234 (94.0%)	0.140 0.56 (0.27 to 1.14)
	Female	174	18 (10.3%)	156 (89.7%)
Hypertension	No	73	1 (1.4%)	72 (98.6%)	0.027 0.14 (0.02 to 1.04)
	Yes	350	32 (9.1%)	318 (90.9%)
Diabetes	No	225	14 (6.2%)	211 (93.8%)	0.209 0.63 (0.30 to 1.29)
	Yes	198	19 (9.6%)	179 (90.4%)
Coronary artery disease	No	154	11 (7.1%)	143 (92.9%)	0.851 0.86 (0.41 to 1.83)
	Yes	269	22 (8.2%)	247 (91.8%)
Peripheral arterial disease	No	382	28 (7.3%)	354 (92.7%)	0.351 0.57 (0.21 to 1.57)
	Yes	41	5 (12.2%)	36 (87.8%)
Stroke	No	412	33 (8.0%)	379 (92.0%)	1.000
	Yes	11	0 (0%)	11 (100%)
COPD	No	374	31 (8.3%)	343 (91.7%)	0.454
	GOLD A	15	1 (6.7%)	14 (93.3%)
	GOLD B	22	0 (0%)	22 (100%)
	GOLD C	6	1 (16.7%)	5 (83.3%)
	GOLD D	6	0 (0%)	6 (100%)
COPD (no/yes)	No	374	31 (8.3%)	343 (91.7%)	0.404 2.12 (0.49 to 9.20)
	Yes	49	2 (4.1%)	47 (95.9%)
AF	No	288	23 (8.0%)	265 (92.0%)	1.000 1.08 (0.50 to 2.35)
	Yes	135	10 (7.4%)	125 (92.6%)
RBBB (pre-existing)	No	381	28 (7.3%)	353 (92.7%)	0.356 0.59 (0.21 to 1.62)
	Yes	42	5 (11.9%)	37 (88.1%)
LBBB (pre-existing)	No	381	29 (7.6%)	352 (92.4%)	0.556 0.78 (0.26 to 2.35)
	Yes	42	4 (9.5%)	38 (90.5%)
LVEF (%) (BS)	55–65	178	17 (9.6)	161 (90.4)	0.234
	45–55	101	4 (4.0)	97 (96.0)
	35–45	61	7 (11.5)	54 (88.5)
	<35	66	4 (6.1)	62 (93.9)
LVOT calcification	No	401	30 (7.5%)	371 (92.5%)	0.241 0.51 (0.14 to 1.84)
	Yes	22	3 (13.6%)	19 (86.4%)
Eccentric AV calcification	No	399	31 (7.8%)	368 (92.2%)	1.000 0.93 (0.21 to 4.13)
	Yes	24	2 (8.3%)	22 (91.7%)
Predilatation	No	2	0 (0%)	2 (100%)	1.000
	Yes	421	33 (7.8%)	388 (92.2%)
Postdilatation	No	408	33 (8.1%)	375 (91.9%)	0.619
	Yes	15	0 (0%)	15 (100%)
Contrast media	No	2	0 (0%)	2 (100%)	1.000
	Yes	421	33 (7.8%)	388 (92.2%)
Protection device	No	405	33 (8.1%)	372 (91.9%)	0.382
	Yes	18	0 (0%)	18 (100%)
Acute renal failure	No	419	32 (7.6%)	387 (92.4%)	0.278 0.25 (0.02 to 2.47)
	AKIN I	4	1 (25.0%)	3 (75.0%)
New permanent pacemaker	No	367	27 (7.4%)	340 (92.6%)	0.419 0.66 (0.26 to 1.69)
	Yes	56	6 (10.7%)	50 (89.3%)
AV block (BO, BP)	No	374	27 (7.2%)	347 (92.8%)	0.251 0.56 (0.22 to 1.43)
	Yes	49	6 (12.2%)	43 (87.8%)
Porcelain aorta	No	282	19 (6.7%)	263 (93.3%)	0.254 0.66 (0.32 to 1.35)
	Yes	141	14 (9.9%)	127 (90.1%)
NYHA	I	10	0 (0%)	10 (100%)	0.322
	II	105	5 (4.8%)	100 (95.2%)
	III	240	20 (8.3%)	220 (91.7%)
	IV	57	7 (12.3%)	50 (87.7%)

Technical success in 2:1 matched non-PA and PA groups (continuous data)
	Technical success	n	Mean	SD	Median	Min–max	P value† OR (95% CI)
Age	No	33	77.1	6.6	79.0	59.0–89.0	0.298 1.02 (0.97 to 1.08)
	Yes	390	78.2	6.8	79.0	51.0–91.0
BMI (kg/m2)	No	33	29.2	6.9	28.7	19.6–50.4	0.675 1.00 (0.94 to 1.06)
	Yes	390	29.2	5.5	28.4	16.3–49.3
Length of stay post-TAVR (days)	No	33	12.6	10.1	10.0	0.0–52.0	0.001 0.87 (0.82 to 0.93)
	Yes	389	8.1	5.0	7.0	1.0–38.0
EuroSCORE II (%)	No	33	10.5	10.4	6.0	1.1–36.8	0.584 0.99 (0.95 to 1.02)
	Yes	390	9.2	10.3	5.2	0.8–62.1
Aortic valve gradient (BV)	No	31	41.4	13.4	41.0	18.0–68.0	0.509 1.01 (0.99 to 1.04)
	Yes	355	43.9	15.1	42.0	13.0–106.0
Aortic valve area (cm2) (BW)	No	33	0.8	0.2	0.8	0.4–1.4	0.537 0.69 (0.12 to 3.85)
	Yes	371	0.7	0.2	0.7	0.2–1.8
Average diameter, mm	No	32	24.6	2.8	24.6	20.0–29.5	0.371 1.09 (0.94 to 1.26)
	Yes	386	25.0	2.4	25.0	19.5–31.5
Area, mm2	No	18	485.2	118.5	473.5	317.9–780.0	0.643 1.00 (1.00 to 1.01)
	Yes	263	492.3	96.4	480.0	232.0–726.0
Cusp calcification Agatston	No	33	3368.7	1676.9	3100.0	923.0–6388.0	0.686 1.00 (1.00 to 1.00)
	Yes	390	3247.9	1742.6	3064.0	365.0–7890.0
LVEF, %	No	28	50.8	11.7	51.0	20.0–70.0	0.658 1.00 (0.97 to 1.03)
	Yes	387	51.4	13.2	55.0	15.0–80.0
Maximum velocity, m/s	No	29	2.2	0.4	2.2	1.3–3.0	0.214 1.76 (0.79 to 3.90)
	Yes	389	2.3	0.5	2.3	1.1–5.1
Creatinine, mg/dL	No	33	1.1	0.6	1.0	0.5–3.1	0.646 0.97 (0.64 to 1.48)
	Yes	390	1.1	0.9	0.9	0.5–11.6
eGFR, mL/min/1.73 m2	No	33	65.3	24.1	71.0	17.3–105.7	0.409 1.01 (0.99 to 1.02)
	Yes	390	69.0	22.1	74.4	3.9–116.1

*Fisher’s exact test.

†Mann-Whitney U test.

AF, atrial fibrillation; AV, aortic valve; BMI, body mass index; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; EuroSCORE, European System for Cardiac Operative Risk Evaluation; LBBB, left bundle branch block; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; NYHA, New York Heart Association; PA, porcelain aorta; RBBB, right bundle branch block; TAVR, transcatheter aortic valve replacement.

Table 6

Device success in 2:1 matched non-PA and PA groups (categorical data)

		n	Device success		P value* OR (95% CI)
			No	Yes
Sex	Male	248	14 (5.6%)	234 (94.4%)	<0.001 0.29 (0.15 to 0.57)
	Female	174	30 (17.2%)	144 (82.8%)
Hypertension	No	72	8 (11.1%)	64 (88.9%)	0.833 1.09 (0.48 to 2.46)
	Yes	350	36 (10.3%)	314 (89.7%)
Diabetes	No	225	29 (12.9%)	196 (87.1%)	0.081 1.80 (0.93 to 3.47)
	Yes	197	15 (7.6%)	182 (92.4%)
Coronary artery disease	No	154	20 (13.0%)	134 (87.0%)	0.246 1.52 (0.81 to 2.85)
	Yes	268	24 (9.0%)	244 (91.0%)
Peripheral arterial disease	No	381	39 (10.2%)	342 (89.8%)	0.600 0.82 (0.30 to 2.22)
	Yes	41	5 (12.2%)	36 (87.8%)
Stroke	No	411	43 (10.5%)	368 (89.5%)	1.000 1.17 (0.15 to 9.37)
	Yes	11	1 (9.1%)	10 (90.9%)
COPD	No	373	42 (11.3%)	331 (88.7%)	0.399
	GOLD A	15	1 (6.7%)	14 (93.3%)
	GOLD B	22	0 (0%)	22 (100%)
	GOLD C	6	1 (16.7%)	5 (83.3%)
	GOLD D	6	0 (0%)	6 (100%)
COPD (no/yes)	No	373	42 (11.3%)	331 (88.7%)	0.142 2.98 (0.69 to 12.80)
	Yes	49	2 (4.1%)	47 (95.9%)
AF	No	287	23 (8.0%)	264 (92.0%)	0.025 0.47 (0.25 to 0.89)
	Yes	135	21 (15.6%)	114 (84.4%)
RBBB (pre-existing)	No	380	40 (10.5%)	340 (89.5%)	1.000 1.12 (0.38 to 3.30)
	Yes	42	4 (9.5%)	38 (90.5%)
LBBB (pre-existing)	No	380	37 (9.7%)	343 (90.3%)	0.180 0.54 (0.22 to 1.30)
	Yes	42	7 (16.7%)	35 (83.3%)
LVEF (%) (BS)	55–65	178	20 (11.2)	158 (88.8)	0.812
	45–55	101	8 (7.9)	93 (92.1)
	35–45	60	7 (11.7)	53 (88.3)
	<35	66	6 (9.1)	60 (90.9)
LVOT calcification	No	400	44 (11.0%)	356 (89.0%)	0.150
	Yes	22	0 (0%)	22 (100%)
Eccentric AV calcification	No	398	44 (11.1%)	354 (88.9%)	0.157
	Yes	24	0 (0%)	24 (100%)
Predilatation	No	2	0 (0%)	2 (100%)	1.000
	Yes	420	44 (10.5%)	376 (89.5%)
Postdilatation	No	407	43 (10.6%)	364 (89.4%)	1.000 1.65 (0.21 to 12.93)
	Yes	15	1 (6.7%)	14 (93.3%)
Contrast media	No	2	0 (0%)	2 (100%)	1.000
	Yes	420	44 (10.5%)	376 (89.5%)
Protection device	No	404	43 (10.6%)	361 (89.4%)	0.708 2.02 (0.26 to 15.65)
	Yes	18	1 (5.6%)	17 (94.4%)
Acute renal failure	No	418	43 (10.3%)	375 (89.7%)	0.357 0.34 (0.03 to 3.40)
	AKIN I	4	1 (25.0%)	3 (75.0%)
New permanent pacemaker	No	366	40 (10.9%)	326 (89.1%)	0.487 1.60 (0.55 to 4.65)
	Yes	56	4 (7.1%)	52 (92.9%)
AV block (BO, BP)	No	373	41 (11.0%)	332 (89.0%)	0.454 1.89 (0.56 to 6.38)
	Yes	49	3 (6.1%)	46 (93.9%)
PA	No	281	35 (12.5%)	246 (87.5%)	0.063 2.09 (0.97 to 4.49)
	Yes	141	9 (6.4%)	132 (93.6%)
NYHA	I	10	0 (0%)	10 (100%)	0.752
	II	105	12 (11.4%)	93 (88.6%)
	III	239	25 (10.5%)	214 (89.5%)
	IV	57	4 (7.0%)	53 (93.0%)

Device success in 2:1 matched non-PA and PA groups (continuous data)
	Device success	n	Mean	SD	Median	Min–max	P value† OR (95% CI)
Age	No	44	79.5	6.1	79.5	64.0–90.0	0.285 0.97 (0.92 to 1.01)
	Yes	378	77.9	6.9	79.0	51.0–91.0
BMI (kg/m2)	No	44	28.3	4.4	28.6	19.6–36.2	0.588 1.03 (0.98 to 1.09)
	Yes	378	29.3	5.7	28.4	16.3–50.4
Length of stay post-TAVR (days)	No	44	10.1	7.7	9.5	0.0–37.0	0.168 0.95 (0.90 to 1.00)
	Yes	378	8.3	5.4	7.0	1.0–52.0
EuroSCORE II (%)	No	44	10.2	10.8	6.0	1.8–45.6	0.405 0.99 (0.96 to 1.02)
	Yes	378	9.2	10.2	5.2	0.8–62.1
Aortic valve gradient (BV)	No	42	47.4	17.6	46.5	14.0–91.0	0.101 0.98 (0.96 to 1.00)
	Yes	343	43.3	14.6	41.0	13.0–106.0
Aortic valve area (cm2) (BW)	No	39	0.7	0.2	0.7	0.4–1.1	0.278 2.67 (0.54 to 13.15)
	Yes	364	0.8	0.2	0.7	0.2–1.8
Average diameter, mm	No	44	24.0	2.2	23.5	19.5–28.5	0.003 1.22 (1.07 to 1.39)
	Yes	373	25.1	2.4	25.0	19.5–31.5
Area, mm2	No	27	442.2	86.0	419.0	232.0–575.0	0.013 1.01 (1.00 to 1.01)
	Yes	253	496.8	97.7	481.0	301.0–780.0
Cusp calcification Agatston	No	44	2868.5	1497.2	3000.0	551.0–5317.0	0.139 1.00 (1.00 to 1.00)
	Yes	378	3301.9	1760.0	3064.0	365.0–7890.0
LVEF, %	No	40	51.4	12.9	50.0	20.0–75.0	0.779 1.00 (0.98 to 1.03)
	Yes	375	51.4	13.1	55.0	15.0–80.0
Maximum velocity, m/s	No	40	2.9	0.8	3.0	1.2–5.1	<0.001 0.06 (0.03 to 0.13)
	Yes	378	2.3	0.4	2.3	1.1–4.2
Creatinine, mg/dL	No	44	0.9	0.4	0.8	0.5–2.3	0.187 1.32 (0.84 to 2.09)
	Yes	378	1.1	0.7	0.9	0.5–6.8
eGFR, mL/min/1.73 m2	No	44	68.0	21.8	71.0	26.1–105.7	0.760 1.00 (0.99 to 1.02)
	Yes	378	68.9	22.1	74.8	5.2–116.1

*Fisher’s exact test.

†Mann-Whitney U test.

AF, atrial fibrillation; AV, aortic valve; BMI, body mass index; COPD, chronic obstructive pulmonary disease; eGFR, estimated glomerular filtration rate; EuroSCORE, European System for Cardiac Operative Risk Evaluation; LBBB, left bundle branch block; LVEF, left ventricular ejection fraction; LVOT, left ventricular outflow tract; NYHA, New York Heart Association; PA, porcelain aorta; RBBB, right bundle branch block; TAVR, transcatheter aortic valve replacement.

The primary outcome of mortality at <30 days or stroke within 72 hours was higher in PA versus non-PA (7.8% vs 2.5%; OR 3.32 (95% CI 1.25 to 8.85); p=0.019), as detailed in table 2.

With regard to secondary outcomes, in table 3 it can be seen that PA was not significantly associated with mortality at 30 days (4.3% vs 2.1%; OR 2.04 (95% CI 0.65 to 6.48); p=0.23). In contrast, stroke within 72 hours was significantly higher in PA versus non-PA (3.5% vs 0.4%; OR 10.33 (95% CI 1.17 to 91.12); p=0.017), as found in table 4.

Technical success (table 5) was not affected by PA (90.1% vs 93.3%; OR 0.66 (95% CI 0.32 to 1.35); p=0.25, PA vs non-PA, respectively), whereas there was a trend for PA to have higher device success (93.6% vs 87.5%; OR 2.09 (95% CI 0.97 to 4.49); p=0.063, PA vs non-PA, respectively), as found in table 6.

Discussion

We report here that stroke occurred more often in patients with PA who underwent TAVI with a BEV than in patients without PA who underwent the same procedure. Specifically, PA negatively affected the primary outcome of mortality within 30 days or stroke within 72 hours; however, given that it did not affect the secondary outcome of mortality within 30 days, but did affect the secondary outcome of stroke within 72 hours, the influence of PA on the primary outcome may be largely due to its association with more strokes. Neither device success nor technical success of the procedure appears to be influenced by the presence of PA.

These results differed from those of a recent study by Eckel et al that looked at the risk of neurological complications in patients with PA versus patients without PA when SEVs were used during TAVI performed via the transfemoral route. In their population, the presence of PA did not result in more peri-TAVI or post-TAVI neurological dysfunction when compared with patients without PA. This study also performed nearest neighbour 2:1 matching for the analysis.¹⁵ Conversely, a Swiss registry study, which did not do nearest neighbour matching, found that patients with PA had a threefold higher risk of stroke than patients without PA undergoing TAVI, and that when broken down by device type, this was specifically seen with BEV. The authors noted that this increased stroke risk with BEV was probably driven by the use of alternative access routes and that the BEV was primarily used with transapical access in their registry.⁸ However, the role of alternative access routes, such as the transapical route, in patients with PA and other aortic pathologies is not unambiguous. It has also been reported that transapical access for patients with PA during TAVI is not associated with adverse outcomes and may actually be associated with a lower stroke rate.¹⁸

TheSOLVE-TAVI trial (Comparison of newer generation self-expandable vs. balloon-expandable valves in transcatheter aortic valve implantation: the randomized SOLVE-TAVI trial)did not compare PA and non-PA; rather, it examined BEV and SEV in a population that presumably contained a small percentage of patients with PA. Although the BEV and newer generation SEV studied at that time were equivalent with respect to the composite of all-cause mortality, stroke, permanent pacemaker implantation and paravalvular leak, the stroke rate was higher in the BEV group than in the SEV group but without reaching statistical significance after superiority testing.¹³ On the other hand, the CENTER–collaboration, which likewise did not compare patients with and without PA, found lower stroke rates with BEV versus SEV. This propensity-matched study pooled data from 10 registries and looked at mortality and stroke rates up to 30 days post-TAVI and saw fewer strokes with BEV.¹² The comparability to our findings again is limited as the specific stroke rate of patients with PA was not taken into consideration.

A cerebral embolic protection device (SENTINEL Cerebral Protection System, Boston Scientific, Massachusetts, USA) was rarely used in our population, with its being deployed in only 14 patients without PA (5.0%) and four patients with PA (2.8%, p=0.444). The use of a cerebral embolic protection device was so seldom that it was not a variable used in matching. Given the possible morbidity, increased length of hospital stay and potential mortality associated with a procedure-related stroke, it is tempting to think that our findings may be a signal that cerebral protection devices should be used more often with BEV in patients with PA. The results of the PROTECTED-TAVR (Cerebral Embolic Protection during Transcatheter Aortic-Valve Replacement) trial do not support this line of thought entirely. In PROTECTED-TAVR, the use of cerebral embolic protection did not change the incidence of preprocedural stroke, but a benefit from cardioembolic protection could not be excluded given the 95% CIs. Interestingly, on multivariate analysis, BEVs were associated with a lower risk of stroke.¹⁹ Furthermore, the rate of stroke overall in the PROTECTED-TAVR trial was low at 2.6% and did not differ significantly between those who received cerebral embolic protection and those who did not.¹⁹ This stroke rate of 2.6% is the same as the overall stroke rate in our study. (11 strokes occurred in the 423 patients.) Although we had an overall low stroke rate within 72 hours, it was still significantly higher in the patients with PA as compared with the patients without PA (3.5% vs 0.4%; OR 10.33 (95% CI 1.17 to 91.12); p=0.017).

Postdilatation of the prosthesis during TAVI is well recognised to be associated with more stroke, especially in the setting of highly calcified valves.²⁰ However, we doubt that it played a role in our findings as postdilatation was rarely performed in our population and there was no difference between groups in terms of rate of postdilatation (p=0.275). Given its rarity, postdilatation was also not a matching variable in our study.

In a PA population, it would be interesting to compare the stroke rate found with BEV to that seen with SAVR. Our study was not designed to address that question; however, a recent Cleveland Clinic study did look at stroke as a complication in matched patients with PA who underwent SAVR or TAVI. The investigators found that 14% of the SAVR patients and 3.4% of the TAVI patients (p=0.4) had a stroke.²¹ It is possible that now the TAVI stroke rate would be considerably lower, as this study included patients going back to 2006, and techniques and devices have since improved.²² Furthermore, the study did not distinguish between TAVI patients who received BEV versus other styles of implantable prostheses.

In summary, we have found transfemoral TAVI with BEV in patients with PA seems to be associated with more strokes within 72 hours than the same procedure in patients without PA. Potential measures to avoid stroke in this high-risk population remain debatable and need to be further studied as we still lack unequivocal answers in terms of preferred valve type, vascular access choice or use of cerebral embolic protection. Nevertheless, debris that is mobilised along the retrograde path via the aortic arch into the ascending aorta certainly plays a role in the increased stroke rate. Careful passage and manipulation of the device would help reduce the risk of stroke, and consideration could be given to the use of cerebral embolic protection systems when performing TAVI in patients with PA. Furthermore, it might be useful to incorporate the use of cerebral embolic protection devices into preprocedure checklists when performing TAVI procedures in these high-risk patients.

Limitations

By design this is a retrospective, observational registry study and so had the usual limitations associated with that, especially incomplete data. Also, the diagnosis of stroke was based on clinical findings. The results might be different if all patients had postprocedural cerebral imaging, which would detect cases of microinfarcts.

We thank Frau Hiltrud Niggemann for the excellent statistical support.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study involves human participants and was approved by the Ethics Committee of the State Medical Association of the Federal State of Thuringia (Landesärztekammer Thüringen; ID DE/EKTH47). Participants gave informed consent to participate in the study before taking part.

Footnote

X @PhilippLauten

Contributors Design, planning, data collection, manuscript writing: PL, LCC-B. Data collection and proofing, manuscript review: GB. Planning, manuscript review: CA. Design and manuscript review: BS, HL. Manuscript editing and completion: all authors. Corresponding author: PL. Guarantor: PL.

Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

Competing interests PL received honoraria for lectures by Edwards Lifesciences and Medtronic.

Provenance and peer review Not commissioned; externally peer reviewed.

References

¹ Kappetein AP, Head SJ, Généreux P, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. Eur Heart J 2012; 33: 2403–18. doi:10.1093/eurheartj/ehs255

² Abramowitz Y, Jilaihawi H, Chakravarty T, et al. Porcelain Aorta. Circulation 2015; 131: 827–36. doi:10.1161/CIRCULATIONAHA.114.011867

³ DeAnda A, Balsam LB. Another crack at the porcelain aorta. J Thorac Cardiovasc Surg 2018; 156: 1035–6. doi:10.1016/j.jtcvs.2018.04.031

⁴ Snow T, Semple T, Duncan A, et al. ‘Porcelain aorta’: a proposed definition and classification of ascending aortic calcification. Open Heart 2018; 5: e000703. doi:10.1136/openhrt-2017-000703

⁵ Vahanian A, Beyersdorf F, Praz F, et al. 2021 ESC/EACTS Guidelines for the management of valvular heart disease. Eur Heart J 2022; 43: 561–632. doi:10.1093/eurheartj/ehab395

⁶ Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description. Circulation 2002; 106: 3006–8. doi:10.1161/01.cir.0000047200.36165.b8

⁷ Webb JG, Pasupati S, Humphries K, et al. Percutaneous transarterial aortic valve replacement in selected high-risk patients with aortic stenosis. Circulation 2007; 116: 755–63. doi:10.1161/CIRCULATIONAHA.107.698258

⁸ Asami M, Bernhard B, Demirel C, et al. Clinical outcomes following transcatheter aortic valve implantation in patients with porcelain aorta. J Cardiovasc Comput Tomogr 2022; 16: 215–21. doi:10.1016/j.jcct.2021.10.006

⁹ Useini D, Beluli B, Christ H, et al. Transapical transcatheter aortic valve implantation in patients with aortic diseases. Eur J Cardiothorac Surg 2021; 59: 1174–81. doi:10.1093/ejcts/ezab050

¹⁰ Francica A, Tonelli F, Saran A, et al. Transapical Transcatheter Aortic Valve Replacement: A Real-World Early and Mid-Term Outcome of a Third-Level Centre. J Clin Med 2022; 11: 4158. doi:10.3390/jcm11144158

¹¹ Chieffo A, Buchanan GL, Van Mieghem NM, et al. Transcatheter aortic valve implantation with the Edwards SAPIEN versus the Medtronic CoreValve Revalving system devices: a multicenter collaborative study: the PRAGMATIC Plus Initiative (Pooled-RotterdAm-Milano-Toulouse In Collaboration). J Am Coll Cardiol 2013; 61: 830–6. doi:10.1016/j.jacc.2012.11.050

¹² Vlastra W, Chandrasekhar J, Muñoz-Garcia AJ, et al. Comparison of balloon-expandable vs. self-expandable valves in patients undergoing transfemoral transcatheter aortic valve implantation: from the CENTER-collaboration. Eur Heart J 2019; 40: 456–65. doi:10.1093/eurheartj/ehy805

¹³ Thiele H, Kurz T, Feistritzer H-J, et al. Comparison of newer generation self-expandable vs. balloon-expandable valves in transcatheter aortic valve implantation: the randomized SOLVE-TAVI trial. Eur Heart J 2020; 41: 1890–9. doi:10.1093/eurheartj/ehaa036

¹⁴ van Nieuwkerk AC, Santos RB, Andraka L, et al. Balloon-Expandable versus Self-Expandable Valves in Transcatheter Aortic Valve Implantation: Complications and Outcomes from a Large International Patient Cohort. J Clin Med 2021; 10: 4005. doi:10.3390/jcm10174005

¹⁵ Eckel C, Blumenstein J, Grothusen C, et al. Procedural Outcomes of a Self-Expanding Transcatheter Heart Valve in Patients with Porcelain Aorta. J Clin Med 2023; 12: 945. doi:10.3390/jcm12030945

¹⁶ Kappetein AP, Head SJ, Généreux P, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. J Am Coll Cardiol 2012; 60: 1438–54. doi:10.1016/j.jacc.2012.09.001

¹⁷ Généreux P, Piazza N, Alu MC, et al. Valve Academic Research Consortium 3: Updated Endpoint Definitions for Aortic Valve Clinical Research. J Am Coll Cardiol 2021; 77: 2717–46. doi:10.1016/j.jacc.2021.02.038

¹⁸ Useini D, Haldenwang P, Schlömicher M, et al. Mid-Term Outcomes after Transapical and Transfemoral Transcatheter Aortic Valve Implantation for Aortic Stenosis and Porcelain Aorta with a Systematic Review of Transfemoral versus Transapical Approach. Thorac Cardiovasc Surg 2020; 68: 623–32. doi:10.1055/s-0039-1692719

¹⁹ Kapadia SR, Makkar R, Leon M, et al. Cerebral Embolic Protection during Transcatheter Aortic-Valve Replacement. N Engl J Med 2022; 387: 1253–63. doi:10.1056/NEJMoa2204961

²⁰ McInerney A, Vera-Urquiza R, Tirado-Conte G, et al. Pre-dilation and Post-dilation in Transcatheter Aortic Valve Replacement: Indications, Benefits and Risks. Interv Cardiol 2021; 16: e28. doi:10.15420/icr.2020.35

²¹ Kramer B, Vekstein AM, Bishop PD, et al. Choosing transcatheter aortic valve replacement in porcelain aorta: outcomes versus surgical replacement. Eur J Cardiothorac Surg 2023; 63: ezad057. doi:10.1093/ejcts/ezad057

²² Ribichini F, Ferrero V, Pesarini G. Choosing transcatheter aortic valve replacement in porcelain aorta: outcomes versus surgical replacement. Eur J Cardiothorac Surg 2023; 63: ezad156. doi:10.1093/ejcts/ezad156