- Email: [email protected]
Prosthesis–Patient Mismatch After Transcatheter Aortic Valve Implantation
Prosthesis–Patient Mismatch After Transcatheter Aortic Valve Implantation Hisato Takagi, MD, PhD, and Takuya Umemoto, MD, PhD; for the ALICE (All-Literature Investigation of Cardiovascular Evidence) Group Department of Cardiovascular Surgery, Shizuoka Medical Center, Shizuoka, Japan
Background. We reviewed currently available studies that investigated prosthesis–patient mismatch (PPM) in transcatheter aortic valve implantation (TAVI) with a systematic literature search and meta-analytic estimates. Methods. To identify all studies that investigated PPM in TAVI, MEDLINE and EMBASE were searched through August 2015. Studies considered for inclusion met the following criteria: the study population included patients undergoing TAVI and outcomes included at least post-procedural PPM prevalence. We performed three quantitative meta-analyses about (1) PPM prevalence after TAVI, (2) PPM prevalence after TAVI versus surgical aortic valve replacement (SAVR), and (3) late allcause mortality after TAVI in patients with PPM versus patients without PPM. Results. We identified 21 eligible studies that included data on a total of 4,000 patients undergoing TAVI. The first meta-analyses found moderate PPM prevalence of 26.7%, severe PPM prevalence of 8.0%,
and overall PPM prevalence of 35.1%. The second meta-analyses of six studies, including 745 patients, found statistically significant reductions in moderate (p [ 0.03), severe (p [ 0.0003), and overall (p [ 0.02) PPM prevalence after TAVI relative to SAVR. The third meta-analyses of five studies, including 2,654 patients, found no statistically significant differences in late mortality between patients with severe PPM and patients without PPM (p [ 0.44) and between patients with overall PPM and patients without PPM (p [ 0.97). Conclusions. Overall, moderate, and severe PPM prevalence after TAVI was 35%, 27%, and 8%, respectively, which may be less than that after SAVR. In contrast to PPM after SAVR, PPM after TAVI may not impair late survival.
P
31% increase in late all-cause mortality over PPM absence (hazard ratio [HR] for overall [moderate and severe] PPM: 1.31; 95% confidence interval [CI]: 1.16 to 1.48; p < 0.00001; HR for moderate PPM: 0.99; 95% CI: 0.92 to 1.07; p ¼ 0.78; HR for severe PPM: 1.27; 95% CI: 1.11 to 1.46; p ¼ 0.0008) [5]. After transcatheter aortic valve implantation (TAVI) for severe AS, however, PPM prevalence, predictors for PPM, and impacts of PPM on late echocardiographic and clinical outcomes remain unclear. For instance, moderate and severe PPM prevalence was respectively 6.3% and 0% in a study [6], whereas another study reported moderate PPM prevalence of 46.3% and severe PPM prevalence of 14.8% [7]. We reviewed currently available studies that investigated PPM after TAVI with a systematic literature search and meta-analytic estimates.
rosthesis–patient mismatch (PPM) describes a state in which the effective orifice area (EOA) of a normally functioning heart valve prosthesis is too small in relation to the patient’s body size, which results in high transvalvular pressure gradients [1]. For aortic valve prostheses, PPM is considered to be severe when the EOA index (EOAI) is less than 0.65 cm2/m2 and moderate when this value is 0.65 to 0.85 cm2/m2, and PPM is classed as not clinically meaningful when the EOAI is greater than 0.85 cm2/m2 [2]. In surgical aortic valve replacement (SAVR) for severe aortic stenosis (AS), patients with PPM have worse functional class and exercise capacity, reduced regression of left ventricular (LV) hypertrophy, inferior recovery of coronary flow reserve, impaired blood coagulation status, and more adverse cardiacrelated events compared with patients without PPM [3, 4]. In addition, PPM may be associated with an increased risk of both perioperative and late (generally 1 year or more) mortalities. Our previous meta-analysis of 32 studies in patients undergoing SAVR suggests that PPM (especially severe PPM) presence is associated with a Accepted for publication Nov 17, 2015. Address correspondence to Dr Takagi, Department of Cardiovascular Surgery, Shizuoka Medical Center, 762-1 Nagasawa, Shimizu-cho, Suntogun, Shizuoka 411-8611, Japan; email: [email protected].
Ó 2016 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2016;-:-–-) Ó 2016 by The Society of Thoracic Surgeons
Material and Methods All studies that investigated PPM after TAVI were identified with a two-level search strategy. First, databases, including MEDLINE and EMBASE, were searched The Appendix can be viewed in the online version of this article [http://dx.doi.org/10.1016/j.athoracsur.2015. 11.048] on http://www.annalsthoracicsurgery.org.
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.11.048
2
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
through August 2015 with the use of Web-based search engines (PubMed and OVID). Second, relevant studies were identified through a manual search of secondary sources that included references of initially identified articles and a search of reviews and commentaries. All references were downloaded for consolidation, elimination of duplicates, and further analysis. Search terms included aortic valve; percutaneous, transcatheter, transluminal, transarterial, transapical, transaortic, transcarotid, transsubclavian, transaxillary, transiliac, or transfemoral; and mismatch, mismatching, or mismatched. Studies considered for inclusion met the following criteria: the study population included patients undergoing TAVI and outcomes included at least post-procedural PPM prevalence. Data about detailed inclusion criteria, TAVI device type, duration of follow-up, PPM prevalence, and late echocardiographic and clinical outcomes were abstracted (as available) from each individual study. We performed three quantitative meta-analyses about (1) PPM prevalence after TAVI, (2) PPM prevalence after TAVI versus SAVR, and (3) late all-cause mortality after TAVI in patients with PPM versus patients without PPM. Study-specific estimates were combined in a random-effects model. All analyses were conducted with Review Manager version 5.3.5 (available from http://tech. cochrane.org/revman) and OpenMeta[Analyst] version 0.1503 (available from http://www.cebm.brown.edu/open_ meta/). In addition to the quantitative meta-analyses, we reviewed predictors for PPM and impacts of PPM on late echocardiographic and clinical outcomes after TAVI.
Results Of 80 potentially relevant publications screened initially, we identified 21 eligible studies that reported at least
Ann Thorac Surg 2016;-:-–-
post-procedural PPM prevalence [6–26], which included data on a total of 4,000 patients undergoing TAVI. The patient characteristics and valve types were summarized in Appendix Table E1.
PPM Prevalence After TAVI After TAVI, moderate PPM prevalence was 6.3% [6] to 46.3% [7] with arithmetic mean of 29.0%, severe PPM prevalence was 0% [6, 22, 23] to 21.3% [18] with mean of 11.6%, and overall (moderate and severe) PPM prevalence was 6.3% [6] to 61.1% [7] with mean of 40.1% (Appendix Table E2). The first meta-analyses found moderate PPM prevalence of 26.7% (95% CI: 20.8% to 32.7%; Fig 1), severe PPM prevalence of 8.0% (95% CI: 4.9% to 11.1%; Fig 2), and overall PPM prevalence of 35.1% (95% CI: 26.8% to 43.4%; Fig 3).
Predictor for PPM After TAVI The following factors were substantially associated with PPM occurrence after TAVI (Appendix Table E3): (1) larger weight [17, 19], height [19], body surface area (BSA) [7, 9, 17, 19, 21], body mass index (BMI) [17, 21], LV mass [21], and prosthesis/annulus ratio [19]; (2) younger age [19, 21]; (3) smaller aortic valve area [14], aortic valve area index [9, 14], and LV ejection fraction [19, 21]; and (4) obesity, hypertension, diabetes, chronic obstructive pulmonary disease, oxygen dependence [21], peripheral vascular disease [26], moderate mitral regurgitation [19], major arrhythmia [21], acute myocardial infarction [25], and prior coronary artery bypass grafting [21]. Independent predictors (identified with the use of multivariate logistic regression) for severe PPM occurrence were BSA (p ¼ 0.00002) and LV outflow tract (LVOT) diameter (p ¼ 0.04) [17]. However, preprocedural mild or greater aortic regurgitation (AR) existed in 56.0% of patients
Fig 1. Moderate prosthesis–patient mismatch prevalence in transcatheter aortic valve implantation. (CI ¼ confidence interval; Ev ¼ event; PARTNER ¼ Placement of Aortic Transcatheter Valve; Trt ¼ treatment.)
Ann Thorac Surg 2016;-:-–-
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
3
Fig 2. Severe prosthesis–patient mismatch prevalence in transcatheter aortic valve implantation. (CI ¼ confidence interval; Ev ¼ event; PARTNER ¼ Placement of Aortic Transcatheter Valve; Trt ¼ treatment.)
with no PPM, 47.3% of patients with moderate PPM, and 43.4% of patients with severe PPM (p < 0.001) [21]. In addition, 44.8% of patients with no PPM and 39.5% of patients with PPM underwent prior percutaneous coronary intervention (p ¼ 0.03). These findings suggest that
preprocedural mild or greater AR and prior percutaneous coronary intervention are negatively associated with PPM occurrence [21]. Furthermore, optimal position (defined as 5 to 10 mm below the native aortic annulus measured fluoroscopically) was also inversely correlated to PPM
Fig 3. Overall (moderate and severe) prosthesis–patient mismatch prevalence in transcatheter aortic valve implantation. (CI ¼ confidence interval; Ev ¼ event; PARTNER ¼ Placement of Aortic Transcatheter Valve; Trt ¼ treatment.)
4
Patient (n)
Prosthesis–Patient Mismatch Moderate
Study
Adjustment
Clavel 2009 [8] Matchinga Finkelstein 2013 [10] Unadjusted
TAVI SAVR 50 86
100 49
Diagnosis Timing
Threshold TAVI SAVR (cm2/m2) (%) (%)
Univariate (OR [95% CI])
Severe TAVI SAVR (%) (%)
Univariate (OR [95% )CI]
TAVI SAVR ( (%) %)
Univariate (OR [95% CI])
At discharge NR Before 0.85 discharge At discharge 0.85 At discharge 0.85
0.65 0.65
NR 17.4 63.3
Unavailable 0.12 [0.05, 0.27]b
10.0 5.8
26.0 26.5
0.32 [0.11, 0.88]b 0.17 [0.06, 0.52]b
23.3
NR 89.8
Unavailable 0.03 [0.01, 0.10]b
0.65 0.65
NR 20.5 45.0
Unavailable 0.32 [0.12, 0.85]b
8.6 10.3
29.3 22.5
0.23 [0.08, 0.67]b 0.39 [0.11, 1.41]b
30.8
NR 67.5
0.63 [0.43, 0.92]b
46.4
60.0
Unavailable 0.21 [0.08, 0.55] (multivariate, 0.26 [0.08, 0.82]) 0.58 [0.41, 0.80]b
Unavailable
17.9
30.6
0.49 [0.18, 1.32]b
Giannini 2011 [12] Kamperidis 2015 [16]d
Matchingc Propensity-scoree matching
58 39
58 40
Pibarot (PARTNER) 2014 [21]
Unnecessary because of randomization Unadjusted
304
270
At 7 days
0.85
0.65
26.6
31.9
0.78 [0.54, 1.11]b
19.7
28.1
56
36
At 30 days
0.9
0.6
17.9
30.6
0.49 [0.18, 1.32]b
0
0
Sherif 2010 [22]
Overall
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
Table 1. Prosthesis–Patient Mismatch Prevalence in Transcatheter Aortic Valve Implantation Versus Surgical Aortic Valve Replacement
a
Each TAVI patient was matched 1:1 with both a SAVR with a stented and a stentless valve patient for sex (exact match), aortic annulus diameter (within 0.05 mm), and LVEF (within 5%) as determined by b c d Calculated by us. Matched for sex, BMI, aortic annular diameter, and LVEF. Transcatheter versus surgical sutureless aortic echocardiography, BSA (within 0.3 m2), and BMI (within 5 kg/m2). e valve. Created with age, sex, BSA, logistic European System for Cardiac Operative Risk Evaluation I, aortic annulus, mean transvalvular pressure gradient, aortic valve area index, stroke volume index, and LVEF at baseline. BMI ¼ body mass index; BSA ¼ body surface area; CI ¼ confidence interval; LVEF ¼ left ventricular ejection fraction; Aortic Transcatheter Valve; SAVR ¼ surgical aortic valve replacement; TAVI ¼ transcatheter aortic valve implantation.
NR ¼ not reported;
OR ¼ odds ratio;
PARTNER ¼ Placement of
Ann Thorac Surg 2016;-:-–-
Ann Thorac Surg 2016;-:-–-
occurrence (optimal position: 61.8% of patients without PPM vs 25.0% of patients with PPM; p ¼ 0.015) [14].
PPM Prevalence After TAVI Versus SAVR Six studies [8, 10, 12, 16, 21, 22], including a total of 745 patients (399 TAVI vs 346 SAVR), reported direct comparison of PPM prevalence after TAVI versus SAVR (Table 1). The second meta-analyses found statistically significant reductions in moderate (odds ratio [OR] 0.36; 95% CI: 0.14 to 0.90; p ¼ 0.03), severe (OR 0.36; 95% CI: 0.20 to 0.63; p ¼ 0.0003), and overall PPM prevalence (OR 0.23; 95% CI: 0.07 to 0.79; p ¼ 0.02) after TAVI relative to SAVR (Fig 4).
Impact of PPM After TAVI on Late Echocardiographic Outcomes Statistically significant impacts of PPM after TAVI on 6-month echocardiographic outcomes (Appendix Table E4) included larger mean aortic gradient [7, 9] and smaller reduction in LV mass regression, LV filling pressure, LV diastolic function grade, and left atrial volume index [9]. In addition, PPM was independently associated with less LV mass regression (b coefficient –13 5; p ¼ 0.012) and percentage of LV mass regression (p ¼ 0.017) at 1 year [21].
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
5
Impact of PPM After TAVI on Late Clinical Outcome Five studies [9, 17, 19, 21, 26], including a total of 2,654 patients, reported impacts of PPM after TAVI on late survival (Table 2). The third meta-analyses found no statistically significant differences in late mortality between patients with severe PPM and patients without PPM (HR 1.32; 95% CI: 0.65 to 2.67; p ¼ 0.44), and between patients with overall PPM and patients without PPM (HR 1.01; 95% CI: 0.80 to 1.27; p ¼ 0.97) (Fig 5). A proportion of patients, who did not demonstrate an improvement in New York Heart Association (NYHA) functional class status at 6 months, was significantly larger in patients with PPM than in patients without PPM (36.7% vs 1.5%; p < 0.001) [9].
Comment The present pooled analysis found that overall (moderate and severe), moderate, and severe PPM prevalence after TAVI was 35.1%, 26.7%, and 8.0%, respectively, which appears somewhat lower than that after SAVR. In 27 studies with 21,802 patients included in a meta-analysis by Head and colleagues [27], the EOAI threshold of 0.85 cm2/m2 was used, and 44.2% of patients undergoing
Fig 4. PPM prevalence in TAVI versus surgical AVR. (AVR ¼ aortic valve replacement; CI ¼ confidence interval; IV ¼ inverse variance; PARTNER ¼ Placement of Aortic Transcatheter Valve; PPM ¼ prosthesis–patient mismatch; TAVI ¼ transcatheter aortic valve implantation.)
6
OR ¼ odds ratio; a
Calculated by us.
b
Combined in a meta-analysis.
28.6 110 . Van Linden 2013 [26]
CI ¼ confidence interval; HR ¼ hazard ratio; NR ¼ not reported; NRCA ¼ nonrandomized continued access; PPM ¼ prosthesis–patient mismatch; RCT ¼ randomized controlled trial.
2y
4y 38.4 9.8
7.7 9.0 13.6 19.7 NR 26.1 34.9 30.2 26.6 165 272 166 1637 304 . . . NRCA RCT Ewe 2011 [9] Kukucka 2013 [17] Mu~ noz-García 2013 [19] Pibarot (PARTNER) 2014 [21]
Ann Thorac Surg 2016;-:-–-
PARTNER ¼ Placement of Aortic Transcatheter Valve;
[0.69, 1.29] [0.81, 1.78]b [0.76, 1.44]b [0.67, 1.80] [030, 1.13]b [0.55, 1.31]b
[0.52, 2.97]b [1.84, 21.42]b
Overall Severe Severe Overall Moderate Severe Overall Moderate Severe Overall Overall 17.6 7.0 mo 30 mo 40 mo 36 mo 1y 18.2 33.8 44.0 43.8 46.4
Multivariate-Adjusted
NR 1.25 6.28 NR 0.94 1.20 1.05 1.10 0.58 0.85 NR OR 0.88 [0.04, 18.70] NR 10.05 [2.54, 39.74]a 1.14 [0.46, 2.82]a,b 0.97 [0.72, 1.31] 1.23 [0.85, 1.79] 1.05 [0.85, 1.28] 0.92 [0.57, 1.49] 0.51 [0.27, 0.98] 0.74 [0.48, 1.13] 1.22 [0.59, 2.54]a,b
a,b
Univariate-Unadjusted PPM Follow-Up Overall Severe Moderate Patient (n) Subgroup Study
PPM (%)
Table 2. Impact of Prosthesis–Patient Mismatch in Transcatheter Aortic Valve Implantation on Late Survival
HR [95% CI] for Overall All-Cause Mortality
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
SAVR were found to have with PPM. In addition, seven studies found that 34.2% of patients had moderate PPM and 9.8% had severe PPM [27]. Our pooled analysis of six studies with direct comparison found statistically significant reductions in moderate, severe, and overall PPM prevalence after TAVI relative to after SAVR. These six studies, however, included a relatively small number of patients (n ¼ 745). Only one randomized controlled trial of them, the PARTNER [21], reported a statistically nonsignificant reduction in moderate PPM prevalence (OR 0.78; 95% CI: 0.54 to 1.11) but a significant reduction of severe (OR 0.63; 95% CI: 0.43 to 0.92) and overall (OR 0.58; 95% CI: 0.41 to 0.80) PPM prevalence after TAVI relative to after SAVR. This difference may be related to the superior hemodynamic performance of transcatheter versus surgical valves. Although the transcatheter valves are stented valves, the stent is thinner and no sewing ring occupies the annular space, which causes less obstruction to blood flow and a difference that would be more important when implanted in a small aortic annulus [8, 15, 20]. In the present analysis, although larger weight, height, BSA, and BMI (all of which are directly related to EOAI defining PPM) were associated with higher PPM prevalence, only BSA was reported to be an independent predictor for severe PPM [17]. It is unclear why aforementioned comorbidity identified in our analysis is nonindependently associated with PPM. As a matter of course, comorbidity associated with PPM cannot be changed. A radical solution of PPM after SAVR for small anulli must be aortic root enlargement to accommodate a larger prosthesis. Meanwhile, possible prevention of PPM after TAVI may be only optimal position, defined as 5 to 10 mm below the native aortic annulus [14]. After SAVR, PPM is associated with an increase in allcause and cardiac-related mortality over long-term follow-up. The meta-analysis by Head and colleagues [27] recognized that PPM was associated with a statistically significant increase in all-cause mortality (HR 1.34; 95% CI: 1.18 to 1.51) and a trend toward an increase in cardiac-related mortality (HR 1.51; 95% CI: 0.88 to 2.60). Both moderate and severe PPM was associated with increased all-cause mortality (HR for moderate PPM: 1.19; 95% CI: 1.07 to 1.33; HR for severe PPM: 1.84; 95% CI: 1.38 to 2.45) and cardiac-related mortality (HR for moderate PPM: 1.32; 95% CI: 1.02 to 1.71; HR for severe PPM: 6.46; 95% CI: 2.79 to 14.97) [27]. In addition, in a recent metaanalysis by Chen and colleagues [28], PPM after SAVR was significantly associated with increased mid-term (OR 1.42; 95% CI: 1.19 to 1.69) and long-term (OR 1.52; 95% CI: 1.26 to 1.84) all-cause mortality. Particularly, severe PPM is a meaningful predictor of reduced long-term survival in all populations undergoing AVR. Severe PPM was associated with reduced survival (adjusted HR 1.50; 95% CI: 1.24 to 1.80), whereas moderate PPM was not (adjusted HR 0.96; 95% CI: 0.86 to 1.07) [28]. In addition, in younger patients, women, and patients with preoperative LV dysfunction, PPM after AVR tends to be associated with increased long-term all-cause mortality. The metaanalysis by Chen and colleagues [28] showed that PPM
Ann Thorac Surg 2016;-:-–-
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
7
Fig 5. Impact of PPM in transcatheter aortic valve implantation on late survival. (CI ¼ confidence interval; IV ¼ inverse variance; NRCA ¼ nonrandomized continued access; PARTNER ¼ Placement of Aortic Transcatheter Valve; PPM ¼ prosthesis–patient mismatch; RCT ¼ randomized controlled trial.)
was associated with higher mid- and long-term mortality rates only in younger populations (10-year OR 1.55; 95% CI: 1.24 to 1.93) and predominantly female populations (10-year OR 1.48; 95% CI: 1.12 to 1.96). Furthermore, PPM had a negative effect on survival in patients with impaired LV ejection fraction (adjusted HR 1.26; 95% CI: 1.09 to 1.47) [28]. Despite aforementioned evidence about negative effects of PPM after SAVR on late survival, there may be no impacts of PPM after TAVI on late mortality. The present meta-analysis of five studies, including a total of 2,654 patients, found that overall PPM, even severe PPM, was not associated with worse late survival. The following factors may explain the differential impact of PPM on survival after TAVI and SAVR. First, after TAVI, patients with PPM had greater BMI [17, 21] and higher prevalence of obesity [21] than patients without PPM, whereas these differences between the PPM and no-PPM groups were not present in SAVR [21]. Greater BMI is a powerful independent predictor of better survival (ie, obesity paradox) after TAVI [29], and the indexation of the prosthetic valve EOA to the patient’s BSA may overestimate PPM severity in obese patients [30]. These two phenomena may contribute to the absence of negative impact of PPM after TAVI on survival [21]. Second, moderate or severe AR is common after TAVI and an adverse prognostic indicator of short- and long-term survival, and mild AR may be also associated with increased long-term mortality [31]. Patients with PPM have less post-procedural AR than patients without PPM after TAVI, whereas paravalvular regurgitation is rare in SAVR regardless of PPM status [21]. Patient without PPM
after SAVR have an optimal valve hemodynamic performance (ie, no residual AS and no paravalvular regurgitation), whereas patients without PPM after TAVI have no residual stenosis but often have paravalvular regurgitation that may impair LV mass regression and adversely affect survival. Paravalvular regurgitation, hence, may confound or mask the effect of PPM on LV mass regression and survival after TAVI, which is supported because severe PPM became an independent predictor of 1-year mortality in the TAVI-nonrandomized continued access cohort of the PARTNER study when excluding patients with post-procedural AR [21]. Only one study included in the present analysis suggests that PPM after TAVI may affect 6-month improvement in NYHA functional class status [9]. In a study by Bleiziffer and colleagues [7], however, no differences were found in NYHA functional status (p ¼ 0.482) or selfassessed health state (p ¼ 0.541) between the PPM and no-PPM groups, and the improvement in self-assessed health state was less apparent in patients with severe PPM (from 45% 21% to 47% 22%; p ¼ 0.830). Tzikas and colleagues [25] also found no difference in functional status between patients with severe PPM and patients without it at 6 months (proportion of patients with NYHA class I or II, 50% and 69%, respectively; p ¼ 0.45). In addition, in a study by Van Linden and colleagues [26], no difference was found in the proportion of patients with NYHA class I or II at 1-year follow-up (71.7% and 71.4% in patients without PPM and patients with PPM, respectively; p ¼ 1.000). Furthermore, during the 17.6 7.0month follow-up period, no significant difference was found between patients with and without PPM in terms of
8
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
major adverse valve-related and cardiovascular events (log-rank p ¼ 0.82) [9]. Taken together, PPM after TAVI may not affect late clinical status. Our analysis must be viewed in the context of its limitations. First, LVOT measurements after TAVI are difficult. Errors can occur in estimating prosthetic valve EOA by Doppler echocardiography, particularly in patients with transcatheter aortic valves in whom the measurement of stroke volume in the LVOT is challenging [21]. For the SAPIEN prosthesis, this measurement is usually underneath the prosthesis stent, whereas for the CoreValve prosthesis it is usually above the lower edge of the stent [7]. The LVOT velocity should be measured underneath the stent, which seems difficult in patients with a CoreValve prosthesis [32]. Second, the stent of any current transcatheter prosthesis is partially located in the LVOT and thus may lead to geometric changes and potential changes in blood flow characteristics [26]. Third, the results of TAVI have evolved over the timeframe investigated here; thus, the present results must be interpreted with caution. In conclusion, overall (moderate and severe), moderate, and severe PPM prevalence after TAVI was 35%, 27%, and 8%, respectively, which may be less than that after SAVR. In contrast to PPM after SAVR, PPM after TAVI may not impair late survival.
References 1. Rahimtoola SH. The problem of valve prosthesis–patient mismatch. Circulation 1978;58:20–4. 2. Pibarot P, Dumesnil JG. The relevance of prosthesis–patient mismatch after aortic valve replacement. Nat Clin Pract Cardiovasc Med 2008;5:764–5. 3. Pibarot P, Dumesnil JG. Prosthesis–patient mismatch: definition, clinical impact, and prevention. Heart 2006;92:1022–9. 4. Mohty D, Malouf JF, Girard SE, et al. Impact of prosthesis– patient mismatch on long-term survival in patients with small St Jude Medical mechanical prostheses in the aortic position. Circulation 2006;113:420–6. 5. Takagi H, Yamamoto H, Iwata K, Goto SN, Umemoto T. A meta-analysis of effects of prosthesis-patient mismatch after aortic valve replacement on late mortality. Int J Cardiol 2012;159:150–4. 6. Unbehaun A, Pasic M, Drews T, et al. New 29-mm balloonexpandable prosthesis for transcatheter aortic valve implantation in large annuli. Ann Thorac Surg 2013;95:1982–90. 7. Bleiziffer S, Hettich I, Hutter A, et al. Incidence and impact of prosthesis-patient mismatch after transcatheter aortic valve implantation. J Heart Valve Dis 2013;22:309–16. 8. Clavel MA, Webb JG, Pibarot P, et al. Comparison of the hemodynamic performance of percutaneous and surgical bioprostheses for the treatment of severe aortic stenosis. J Am Coll Cardiol 2009;53:1883–91. 9. Ewe SH, Muratori M, Delgado V, et al. Hemodynamic and clinical impact of prosthesis-patient mismatch after transcatheter aortic valve implantation. J Am Coll Cardiol 2011;58: 1910–8. 10. Finkelstein A, Schwartz AL, Uretzky G, et al. Hemodynamic performance and outcome of percutaneous versus surgical stentless bioprostheses for aortic stenosis with anticipated patient-prosthesis mismatch. J Thorac Cardiovasc Surg 2014;147:1892–9. 11. Freeman M, Webb JG, Willson AB, et al. Multidetector CT predictors of prosthesis-patient mismatch in transcatheter aortic valve replacement. J Cardiovasc Comput Tomogr 2013;7:248–55.
Ann Thorac Surg 2016;-:-–-
12. Giannini C, Petronio AS, Nardi C, et al. Left ventricular reverse remodeling in percutaneous and surgical aortic bioprostheses: an echocardiographic study. J Am Soc Echocardiogr 2011;24:28–36. 13. Gotzmann M, Lindstaedt M, Bojara W, et al. Hemodynamic results and changes in myocardial function after transcatheter aortic valve implantation. Am Heart J 2010;159: 926–32. 14. Jilaihawi H, Chin D, Spyt T, et al. Prosthesis-patient mismatch after transcatheter aortic valve implantation with the Medtronic-Core valve bioprosthesis. Eur Heart J 2010;31: 857–64. 15. Kalavrouziotis D, Rod es-Cabau J, Bagur R, et al. Transcatheter aortic valve implantation in patients with severe aortic stenosis and small aortic annulus. J Am Coll Cardiol 2011;58:1016–24. 16. Kamperidis V, van Rosendael PJ, de Weger A, et al. Surgical sutureless and transcatheter aortic valves: hemodynamic performance and clinical outcomes in propensity scorematched high-risk populations with severe aortic stenosis. JACC Cardiovasc Interv 2015;8:670–7. 17. Kukucka M, Pasic M, Dreysse S, et al. Patient-prosthesis mismatch after transapical aortic valve implantation: incidence and impact on survival. J Thorac Cardiovasc Surg 2013;145:391–7. 18. Laflamme J, Puri R, Urena M, et al. Incidence and risk factors of hemolysis after transcatheter aortic valve implantation with a balloon-expandable valve. Am J Cardiol 2015;115: 1574–9. 19. Mu~ noz-García AJ, Mu~ noz-García M, Carrasco-Chinchilla F, et al. Incidence and clinical outcome of prosthesis-patient mismatch after transcatheter aortic valve implantation with the CoreValve prosthesis. Int J Cardiol 2013;167:1074–6. 20. Nombela-Franco L, Ruel M, Radhakrishnan S, et al. Comparison of hemodynamic performance of self-expandable CoreValve versus balloon-expandable Edwards SAPIEN aortic valves inserted by catheter for aortic stenosis. Am J Cardiol 2013;111:1026–33. 21. Pibarot P, Weissman NJ, Stewart WJ, et al. Incidence and sequelae of prosthesis-patient mismatch in transcatheter versus surgical valve replacement in high-risk patients with severe aortic stenosis: a PARTNER trial cohort–a analysis. J Am Coll Cardiol 2014;64:1323–34. 22. Sherif MA, Abdel-Wahab M, Awad O, et al. Early hemodynamic and neurohormonal response after transcatheter aortic valve implantation. Am Heart J 2010;160:862–9. 23. Spangenberg T, Budde U, Schewel D, et al. Treatment of acquired von Willebrand syndrome in aortic stenosis with transcatheter aortic valve replacement. JACC Cardiovasc Interv 2015;8:692–700. 24. Tamborini G, Fusini L, Gripari P, et al. Feasibility and accuracy of 3DTEE versus CT for the evaluation of aortic valve annulus to left main ostium distance before transcatheter aortic valve implantation. JACC Cardiovasc Imaging 2012;5: 579–88. 25. Tzikas A, Piazza N, Geleijnse ML, et al. Prosthesis-patient mismatch after transcatheter aortic valve implantation with the medtronic CoreValve system in patients with aortic stenosis. Am J Cardiol 2010;106:255–60. 26. Van Linden A, Kempfert J, Blumenstein J, et al. Prosthesispatient mismatch after transcatheter aortic valve implantation using the Edwards SAPIENÔ prosthesis. Thorac Cardiovasc Surg 2013;61:414–20. 27. Head SJ, Mokhles MM, Osnabrugge RL, et al. The impact of prosthesis-patient mismatch on long-term survival after aortic valve replacement: a systematic review and metaanalysis of 34 observational studies comprising 27 186 patients with 133 141 patient-years. Eur Heart J 2012;33:1518–29. 28. Chen J, Lin Y, Kang B, Wang Z. Indexed effective orifice area is a significant predictor of higher mid- and long-term mortality rates following aortic valve replacement in patients with prosthesis-patient mismatch. Eur J Cardiothorac Surg 2014;45:234–40.
Ann Thorac Surg 2016;-:-–-
29. Kodali SK, Williams MR, Smith CR, et al, PARTNER Trial Investigators. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med 2012;366: 1686–95. 30. Mohty D, Dumesnil JG, Echahidi N, et al. Impact of prosthesis-patient mismatch on long-term survival after aortic valve replacement: influence of age, obesity, and left ventricular dysfunction. J Am Coll Cardiol 2009;53: 39–47.
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
9
31. Athappan G, Patvardhan E, Tuzcu EM, et al. Incidence, predictors, and outcomes of aortic regurgitation after transcatheter aortic valve replacement: meta-analysis and systematic review of literature. J Am Coll Cardiol 2013;61: 1585–95. 32. Zamorano JL, Badano LP, Bruce C, et al. EAE/ASE recommendations for the use of echocardiography in new transcatheter interventions for valvular heart disease. Eur Heart J 2011;32:2189–214.
Background. We reviewed currently available studies that investigated prosthesis–patient mismatch (PPM) in transcatheter aortic valve implantation (TAVI) with a systematic literature search and meta-analytic estimates. Methods. To identify all studies that investigated PPM in TAVI, MEDLINE and EMBASE were searched through August 2015. Studies considered for inclusion met the following criteria: the study population included patients undergoing TAVI and outcomes included at least post-procedural PPM prevalence. We performed three quantitative meta-analyses about (1) PPM prevalence after TAVI, (2) PPM prevalence after TAVI versus surgical aortic valve replacement (SAVR), and (3) late allcause mortality after TAVI in patients with PPM versus patients without PPM. Results. We identified 21 eligible studies that included data on a total of 4,000 patients undergoing TAVI. The first meta-analyses found moderate PPM prevalence of 26.7%, severe PPM prevalence of 8.0%,
and overall PPM prevalence of 35.1%. The second meta-analyses of six studies, including 745 patients, found statistically significant reductions in moderate (p [ 0.03), severe (p [ 0.0003), and overall (p [ 0.02) PPM prevalence after TAVI relative to SAVR. The third meta-analyses of five studies, including 2,654 patients, found no statistically significant differences in late mortality between patients with severe PPM and patients without PPM (p [ 0.44) and between patients with overall PPM and patients without PPM (p [ 0.97). Conclusions. Overall, moderate, and severe PPM prevalence after TAVI was 35%, 27%, and 8%, respectively, which may be less than that after SAVR. In contrast to PPM after SAVR, PPM after TAVI may not impair late survival.
P
31% increase in late all-cause mortality over PPM absence (hazard ratio [HR] for overall [moderate and severe] PPM: 1.31; 95% confidence interval [CI]: 1.16 to 1.48; p < 0.00001; HR for moderate PPM: 0.99; 95% CI: 0.92 to 1.07; p ¼ 0.78; HR for severe PPM: 1.27; 95% CI: 1.11 to 1.46; p ¼ 0.0008) [5]. After transcatheter aortic valve implantation (TAVI) for severe AS, however, PPM prevalence, predictors for PPM, and impacts of PPM on late echocardiographic and clinical outcomes remain unclear. For instance, moderate and severe PPM prevalence was respectively 6.3% and 0% in a study [6], whereas another study reported moderate PPM prevalence of 46.3% and severe PPM prevalence of 14.8% [7]. We reviewed currently available studies that investigated PPM after TAVI with a systematic literature search and meta-analytic estimates.
rosthesis–patient mismatch (PPM) describes a state in which the effective orifice area (EOA) of a normally functioning heart valve prosthesis is too small in relation to the patient’s body size, which results in high transvalvular pressure gradients [1]. For aortic valve prostheses, PPM is considered to be severe when the EOA index (EOAI) is less than 0.65 cm2/m2 and moderate when this value is 0.65 to 0.85 cm2/m2, and PPM is classed as not clinically meaningful when the EOAI is greater than 0.85 cm2/m2 [2]. In surgical aortic valve replacement (SAVR) for severe aortic stenosis (AS), patients with PPM have worse functional class and exercise capacity, reduced regression of left ventricular (LV) hypertrophy, inferior recovery of coronary flow reserve, impaired blood coagulation status, and more adverse cardiacrelated events compared with patients without PPM [3, 4]. In addition, PPM may be associated with an increased risk of both perioperative and late (generally 1 year or more) mortalities. Our previous meta-analysis of 32 studies in patients undergoing SAVR suggests that PPM (especially severe PPM) presence is associated with a Accepted for publication Nov 17, 2015. Address correspondence to Dr Takagi, Department of Cardiovascular Surgery, Shizuoka Medical Center, 762-1 Nagasawa, Shimizu-cho, Suntogun, Shizuoka 411-8611, Japan; email: [email protected].
Ó 2016 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2016;-:-–-) Ó 2016 by The Society of Thoracic Surgeons
Material and Methods All studies that investigated PPM after TAVI were identified with a two-level search strategy. First, databases, including MEDLINE and EMBASE, were searched The Appendix can be viewed in the online version of this article [http://dx.doi.org/10.1016/j.athoracsur.2015. 11.048] on http://www.annalsthoracicsurgery.org.
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.11.048
2
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
through August 2015 with the use of Web-based search engines (PubMed and OVID). Second, relevant studies were identified through a manual search of secondary sources that included references of initially identified articles and a search of reviews and commentaries. All references were downloaded for consolidation, elimination of duplicates, and further analysis. Search terms included aortic valve; percutaneous, transcatheter, transluminal, transarterial, transapical, transaortic, transcarotid, transsubclavian, transaxillary, transiliac, or transfemoral; and mismatch, mismatching, or mismatched. Studies considered for inclusion met the following criteria: the study population included patients undergoing TAVI and outcomes included at least post-procedural PPM prevalence. Data about detailed inclusion criteria, TAVI device type, duration of follow-up, PPM prevalence, and late echocardiographic and clinical outcomes were abstracted (as available) from each individual study. We performed three quantitative meta-analyses about (1) PPM prevalence after TAVI, (2) PPM prevalence after TAVI versus SAVR, and (3) late all-cause mortality after TAVI in patients with PPM versus patients without PPM. Study-specific estimates were combined in a random-effects model. All analyses were conducted with Review Manager version 5.3.5 (available from http://tech. cochrane.org/revman) and OpenMeta[Analyst] version 0.1503 (available from http://www.cebm.brown.edu/open_ meta/). In addition to the quantitative meta-analyses, we reviewed predictors for PPM and impacts of PPM on late echocardiographic and clinical outcomes after TAVI.
Results Of 80 potentially relevant publications screened initially, we identified 21 eligible studies that reported at least
Ann Thorac Surg 2016;-:-–-
post-procedural PPM prevalence [6–26], which included data on a total of 4,000 patients undergoing TAVI. The patient characteristics and valve types were summarized in Appendix Table E1.
PPM Prevalence After TAVI After TAVI, moderate PPM prevalence was 6.3% [6] to 46.3% [7] with arithmetic mean of 29.0%, severe PPM prevalence was 0% [6, 22, 23] to 21.3% [18] with mean of 11.6%, and overall (moderate and severe) PPM prevalence was 6.3% [6] to 61.1% [7] with mean of 40.1% (Appendix Table E2). The first meta-analyses found moderate PPM prevalence of 26.7% (95% CI: 20.8% to 32.7%; Fig 1), severe PPM prevalence of 8.0% (95% CI: 4.9% to 11.1%; Fig 2), and overall PPM prevalence of 35.1% (95% CI: 26.8% to 43.4%; Fig 3).
Predictor for PPM After TAVI The following factors were substantially associated with PPM occurrence after TAVI (Appendix Table E3): (1) larger weight [17, 19], height [19], body surface area (BSA) [7, 9, 17, 19, 21], body mass index (BMI) [17, 21], LV mass [21], and prosthesis/annulus ratio [19]; (2) younger age [19, 21]; (3) smaller aortic valve area [14], aortic valve area index [9, 14], and LV ejection fraction [19, 21]; and (4) obesity, hypertension, diabetes, chronic obstructive pulmonary disease, oxygen dependence [21], peripheral vascular disease [26], moderate mitral regurgitation [19], major arrhythmia [21], acute myocardial infarction [25], and prior coronary artery bypass grafting [21]. Independent predictors (identified with the use of multivariate logistic regression) for severe PPM occurrence were BSA (p ¼ 0.00002) and LV outflow tract (LVOT) diameter (p ¼ 0.04) [17]. However, preprocedural mild or greater aortic regurgitation (AR) existed in 56.0% of patients
Fig 1. Moderate prosthesis–patient mismatch prevalence in transcatheter aortic valve implantation. (CI ¼ confidence interval; Ev ¼ event; PARTNER ¼ Placement of Aortic Transcatheter Valve; Trt ¼ treatment.)
Ann Thorac Surg 2016;-:-–-
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
3
Fig 2. Severe prosthesis–patient mismatch prevalence in transcatheter aortic valve implantation. (CI ¼ confidence interval; Ev ¼ event; PARTNER ¼ Placement of Aortic Transcatheter Valve; Trt ¼ treatment.)
with no PPM, 47.3% of patients with moderate PPM, and 43.4% of patients with severe PPM (p < 0.001) [21]. In addition, 44.8% of patients with no PPM and 39.5% of patients with PPM underwent prior percutaneous coronary intervention (p ¼ 0.03). These findings suggest that
preprocedural mild or greater AR and prior percutaneous coronary intervention are negatively associated with PPM occurrence [21]. Furthermore, optimal position (defined as 5 to 10 mm below the native aortic annulus measured fluoroscopically) was also inversely correlated to PPM
Fig 3. Overall (moderate and severe) prosthesis–patient mismatch prevalence in transcatheter aortic valve implantation. (CI ¼ confidence interval; Ev ¼ event; PARTNER ¼ Placement of Aortic Transcatheter Valve; Trt ¼ treatment.)
4
Patient (n)
Prosthesis–Patient Mismatch Moderate
Study
Adjustment
Clavel 2009 [8] Matchinga Finkelstein 2013 [10] Unadjusted
TAVI SAVR 50 86
100 49
Diagnosis Timing
Threshold TAVI SAVR (cm2/m2) (%) (%)
Univariate (OR [95% CI])
Severe TAVI SAVR (%) (%)
Univariate (OR [95% )CI]
TAVI SAVR ( (%) %)
Univariate (OR [95% CI])
At discharge NR Before 0.85 discharge At discharge 0.85 At discharge 0.85
0.65 0.65
NR 17.4 63.3
Unavailable 0.12 [0.05, 0.27]b
10.0 5.8
26.0 26.5
0.32 [0.11, 0.88]b 0.17 [0.06, 0.52]b
23.3
NR 89.8
Unavailable 0.03 [0.01, 0.10]b
0.65 0.65
NR 20.5 45.0
Unavailable 0.32 [0.12, 0.85]b
8.6 10.3
29.3 22.5
0.23 [0.08, 0.67]b 0.39 [0.11, 1.41]b
30.8
NR 67.5
0.63 [0.43, 0.92]b
46.4
60.0
Unavailable 0.21 [0.08, 0.55] (multivariate, 0.26 [0.08, 0.82]) 0.58 [0.41, 0.80]b
Unavailable
17.9
30.6
0.49 [0.18, 1.32]b
Giannini 2011 [12] Kamperidis 2015 [16]d
Matchingc Propensity-scoree matching
58 39
58 40
Pibarot (PARTNER) 2014 [21]
Unnecessary because of randomization Unadjusted
304
270
At 7 days
0.85
0.65
26.6
31.9
0.78 [0.54, 1.11]b
19.7
28.1
56
36
At 30 days
0.9
0.6
17.9
30.6
0.49 [0.18, 1.32]b
0
0
Sherif 2010 [22]
Overall
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
Table 1. Prosthesis–Patient Mismatch Prevalence in Transcatheter Aortic Valve Implantation Versus Surgical Aortic Valve Replacement
a
Each TAVI patient was matched 1:1 with both a SAVR with a stented and a stentless valve patient for sex (exact match), aortic annulus diameter (within 0.05 mm), and LVEF (within 5%) as determined by b c d Calculated by us. Matched for sex, BMI, aortic annular diameter, and LVEF. Transcatheter versus surgical sutureless aortic echocardiography, BSA (within 0.3 m2), and BMI (within 5 kg/m2). e valve. Created with age, sex, BSA, logistic European System for Cardiac Operative Risk Evaluation I, aortic annulus, mean transvalvular pressure gradient, aortic valve area index, stroke volume index, and LVEF at baseline. BMI ¼ body mass index; BSA ¼ body surface area; CI ¼ confidence interval; LVEF ¼ left ventricular ejection fraction; Aortic Transcatheter Valve; SAVR ¼ surgical aortic valve replacement; TAVI ¼ transcatheter aortic valve implantation.
NR ¼ not reported;
OR ¼ odds ratio;
PARTNER ¼ Placement of
Ann Thorac Surg 2016;-:-–-
Ann Thorac Surg 2016;-:-–-
occurrence (optimal position: 61.8% of patients without PPM vs 25.0% of patients with PPM; p ¼ 0.015) [14].
PPM Prevalence After TAVI Versus SAVR Six studies [8, 10, 12, 16, 21, 22], including a total of 745 patients (399 TAVI vs 346 SAVR), reported direct comparison of PPM prevalence after TAVI versus SAVR (Table 1). The second meta-analyses found statistically significant reductions in moderate (odds ratio [OR] 0.36; 95% CI: 0.14 to 0.90; p ¼ 0.03), severe (OR 0.36; 95% CI: 0.20 to 0.63; p ¼ 0.0003), and overall PPM prevalence (OR 0.23; 95% CI: 0.07 to 0.79; p ¼ 0.02) after TAVI relative to SAVR (Fig 4).
Impact of PPM After TAVI on Late Echocardiographic Outcomes Statistically significant impacts of PPM after TAVI on 6-month echocardiographic outcomes (Appendix Table E4) included larger mean aortic gradient [7, 9] and smaller reduction in LV mass regression, LV filling pressure, LV diastolic function grade, and left atrial volume index [9]. In addition, PPM was independently associated with less LV mass regression (b coefficient –13 5; p ¼ 0.012) and percentage of LV mass regression (p ¼ 0.017) at 1 year [21].
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
5
Impact of PPM After TAVI on Late Clinical Outcome Five studies [9, 17, 19, 21, 26], including a total of 2,654 patients, reported impacts of PPM after TAVI on late survival (Table 2). The third meta-analyses found no statistically significant differences in late mortality between patients with severe PPM and patients without PPM (HR 1.32; 95% CI: 0.65 to 2.67; p ¼ 0.44), and between patients with overall PPM and patients without PPM (HR 1.01; 95% CI: 0.80 to 1.27; p ¼ 0.97) (Fig 5). A proportion of patients, who did not demonstrate an improvement in New York Heart Association (NYHA) functional class status at 6 months, was significantly larger in patients with PPM than in patients without PPM (36.7% vs 1.5%; p < 0.001) [9].
Comment The present pooled analysis found that overall (moderate and severe), moderate, and severe PPM prevalence after TAVI was 35.1%, 26.7%, and 8.0%, respectively, which appears somewhat lower than that after SAVR. In 27 studies with 21,802 patients included in a meta-analysis by Head and colleagues [27], the EOAI threshold of 0.85 cm2/m2 was used, and 44.2% of patients undergoing
Fig 4. PPM prevalence in TAVI versus surgical AVR. (AVR ¼ aortic valve replacement; CI ¼ confidence interval; IV ¼ inverse variance; PARTNER ¼ Placement of Aortic Transcatheter Valve; PPM ¼ prosthesis–patient mismatch; TAVI ¼ transcatheter aortic valve implantation.)
6
OR ¼ odds ratio; a
Calculated by us.
b
Combined in a meta-analysis.
28.6 110 . Van Linden 2013 [26]
CI ¼ confidence interval; HR ¼ hazard ratio; NR ¼ not reported; NRCA ¼ nonrandomized continued access; PPM ¼ prosthesis–patient mismatch; RCT ¼ randomized controlled trial.
2y
4y 38.4 9.8
7.7 9.0 13.6 19.7 NR 26.1 34.9 30.2 26.6 165 272 166 1637 304 . . . NRCA RCT Ewe 2011 [9] Kukucka 2013 [17] Mu~ noz-García 2013 [19] Pibarot (PARTNER) 2014 [21]
Ann Thorac Surg 2016;-:-–-
PARTNER ¼ Placement of Aortic Transcatheter Valve;
[0.69, 1.29] [0.81, 1.78]b [0.76, 1.44]b [0.67, 1.80] [030, 1.13]b [0.55, 1.31]b
[0.52, 2.97]b [1.84, 21.42]b
Overall Severe Severe Overall Moderate Severe Overall Moderate Severe Overall Overall 17.6 7.0 mo 30 mo 40 mo 36 mo 1y 18.2 33.8 44.0 43.8 46.4
Multivariate-Adjusted
NR 1.25 6.28 NR 0.94 1.20 1.05 1.10 0.58 0.85 NR OR 0.88 [0.04, 18.70] NR 10.05 [2.54, 39.74]a 1.14 [0.46, 2.82]a,b 0.97 [0.72, 1.31] 1.23 [0.85, 1.79] 1.05 [0.85, 1.28] 0.92 [0.57, 1.49] 0.51 [0.27, 0.98] 0.74 [0.48, 1.13] 1.22 [0.59, 2.54]a,b
a,b
Univariate-Unadjusted PPM Follow-Up Overall Severe Moderate Patient (n) Subgroup Study
PPM (%)
Table 2. Impact of Prosthesis–Patient Mismatch in Transcatheter Aortic Valve Implantation on Late Survival
HR [95% CI] for Overall All-Cause Mortality
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
SAVR were found to have with PPM. In addition, seven studies found that 34.2% of patients had moderate PPM and 9.8% had severe PPM [27]. Our pooled analysis of six studies with direct comparison found statistically significant reductions in moderate, severe, and overall PPM prevalence after TAVI relative to after SAVR. These six studies, however, included a relatively small number of patients (n ¼ 745). Only one randomized controlled trial of them, the PARTNER [21], reported a statistically nonsignificant reduction in moderate PPM prevalence (OR 0.78; 95% CI: 0.54 to 1.11) but a significant reduction of severe (OR 0.63; 95% CI: 0.43 to 0.92) and overall (OR 0.58; 95% CI: 0.41 to 0.80) PPM prevalence after TAVI relative to after SAVR. This difference may be related to the superior hemodynamic performance of transcatheter versus surgical valves. Although the transcatheter valves are stented valves, the stent is thinner and no sewing ring occupies the annular space, which causes less obstruction to blood flow and a difference that would be more important when implanted in a small aortic annulus [8, 15, 20]. In the present analysis, although larger weight, height, BSA, and BMI (all of which are directly related to EOAI defining PPM) were associated with higher PPM prevalence, only BSA was reported to be an independent predictor for severe PPM [17]. It is unclear why aforementioned comorbidity identified in our analysis is nonindependently associated with PPM. As a matter of course, comorbidity associated with PPM cannot be changed. A radical solution of PPM after SAVR for small anulli must be aortic root enlargement to accommodate a larger prosthesis. Meanwhile, possible prevention of PPM after TAVI may be only optimal position, defined as 5 to 10 mm below the native aortic annulus [14]. After SAVR, PPM is associated with an increase in allcause and cardiac-related mortality over long-term follow-up. The meta-analysis by Head and colleagues [27] recognized that PPM was associated with a statistically significant increase in all-cause mortality (HR 1.34; 95% CI: 1.18 to 1.51) and a trend toward an increase in cardiac-related mortality (HR 1.51; 95% CI: 0.88 to 2.60). Both moderate and severe PPM was associated with increased all-cause mortality (HR for moderate PPM: 1.19; 95% CI: 1.07 to 1.33; HR for severe PPM: 1.84; 95% CI: 1.38 to 2.45) and cardiac-related mortality (HR for moderate PPM: 1.32; 95% CI: 1.02 to 1.71; HR for severe PPM: 6.46; 95% CI: 2.79 to 14.97) [27]. In addition, in a recent metaanalysis by Chen and colleagues [28], PPM after SAVR was significantly associated with increased mid-term (OR 1.42; 95% CI: 1.19 to 1.69) and long-term (OR 1.52; 95% CI: 1.26 to 1.84) all-cause mortality. Particularly, severe PPM is a meaningful predictor of reduced long-term survival in all populations undergoing AVR. Severe PPM was associated with reduced survival (adjusted HR 1.50; 95% CI: 1.24 to 1.80), whereas moderate PPM was not (adjusted HR 0.96; 95% CI: 0.86 to 1.07) [28]. In addition, in younger patients, women, and patients with preoperative LV dysfunction, PPM after AVR tends to be associated with increased long-term all-cause mortality. The metaanalysis by Chen and colleagues [28] showed that PPM
Ann Thorac Surg 2016;-:-–-
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
7
Fig 5. Impact of PPM in transcatheter aortic valve implantation on late survival. (CI ¼ confidence interval; IV ¼ inverse variance; NRCA ¼ nonrandomized continued access; PARTNER ¼ Placement of Aortic Transcatheter Valve; PPM ¼ prosthesis–patient mismatch; RCT ¼ randomized controlled trial.)
was associated with higher mid- and long-term mortality rates only in younger populations (10-year OR 1.55; 95% CI: 1.24 to 1.93) and predominantly female populations (10-year OR 1.48; 95% CI: 1.12 to 1.96). Furthermore, PPM had a negative effect on survival in patients with impaired LV ejection fraction (adjusted HR 1.26; 95% CI: 1.09 to 1.47) [28]. Despite aforementioned evidence about negative effects of PPM after SAVR on late survival, there may be no impacts of PPM after TAVI on late mortality. The present meta-analysis of five studies, including a total of 2,654 patients, found that overall PPM, even severe PPM, was not associated with worse late survival. The following factors may explain the differential impact of PPM on survival after TAVI and SAVR. First, after TAVI, patients with PPM had greater BMI [17, 21] and higher prevalence of obesity [21] than patients without PPM, whereas these differences between the PPM and no-PPM groups were not present in SAVR [21]. Greater BMI is a powerful independent predictor of better survival (ie, obesity paradox) after TAVI [29], and the indexation of the prosthetic valve EOA to the patient’s BSA may overestimate PPM severity in obese patients [30]. These two phenomena may contribute to the absence of negative impact of PPM after TAVI on survival [21]. Second, moderate or severe AR is common after TAVI and an adverse prognostic indicator of short- and long-term survival, and mild AR may be also associated with increased long-term mortality [31]. Patients with PPM have less post-procedural AR than patients without PPM after TAVI, whereas paravalvular regurgitation is rare in SAVR regardless of PPM status [21]. Patient without PPM
after SAVR have an optimal valve hemodynamic performance (ie, no residual AS and no paravalvular regurgitation), whereas patients without PPM after TAVI have no residual stenosis but often have paravalvular regurgitation that may impair LV mass regression and adversely affect survival. Paravalvular regurgitation, hence, may confound or mask the effect of PPM on LV mass regression and survival after TAVI, which is supported because severe PPM became an independent predictor of 1-year mortality in the TAVI-nonrandomized continued access cohort of the PARTNER study when excluding patients with post-procedural AR [21]. Only one study included in the present analysis suggests that PPM after TAVI may affect 6-month improvement in NYHA functional class status [9]. In a study by Bleiziffer and colleagues [7], however, no differences were found in NYHA functional status (p ¼ 0.482) or selfassessed health state (p ¼ 0.541) between the PPM and no-PPM groups, and the improvement in self-assessed health state was less apparent in patients with severe PPM (from 45% 21% to 47% 22%; p ¼ 0.830). Tzikas and colleagues [25] also found no difference in functional status between patients with severe PPM and patients without it at 6 months (proportion of patients with NYHA class I or II, 50% and 69%, respectively; p ¼ 0.45). In addition, in a study by Van Linden and colleagues [26], no difference was found in the proportion of patients with NYHA class I or II at 1-year follow-up (71.7% and 71.4% in patients without PPM and patients with PPM, respectively; p ¼ 1.000). Furthermore, during the 17.6 7.0month follow-up period, no significant difference was found between patients with and without PPM in terms of
8
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
major adverse valve-related and cardiovascular events (log-rank p ¼ 0.82) [9]. Taken together, PPM after TAVI may not affect late clinical status. Our analysis must be viewed in the context of its limitations. First, LVOT measurements after TAVI are difficult. Errors can occur in estimating prosthetic valve EOA by Doppler echocardiography, particularly in patients with transcatheter aortic valves in whom the measurement of stroke volume in the LVOT is challenging [21]. For the SAPIEN prosthesis, this measurement is usually underneath the prosthesis stent, whereas for the CoreValve prosthesis it is usually above the lower edge of the stent [7]. The LVOT velocity should be measured underneath the stent, which seems difficult in patients with a CoreValve prosthesis [32]. Second, the stent of any current transcatheter prosthesis is partially located in the LVOT and thus may lead to geometric changes and potential changes in blood flow characteristics [26]. Third, the results of TAVI have evolved over the timeframe investigated here; thus, the present results must be interpreted with caution. In conclusion, overall (moderate and severe), moderate, and severe PPM prevalence after TAVI was 35%, 27%, and 8%, respectively, which may be less than that after SAVR. In contrast to PPM after SAVR, PPM after TAVI may not impair late survival.
References 1. Rahimtoola SH. The problem of valve prosthesis–patient mismatch. Circulation 1978;58:20–4. 2. Pibarot P, Dumesnil JG. The relevance of prosthesis–patient mismatch after aortic valve replacement. Nat Clin Pract Cardiovasc Med 2008;5:764–5. 3. Pibarot P, Dumesnil JG. Prosthesis–patient mismatch: definition, clinical impact, and prevention. Heart 2006;92:1022–9. 4. Mohty D, Malouf JF, Girard SE, et al. Impact of prosthesis– patient mismatch on long-term survival in patients with small St Jude Medical mechanical prostheses in the aortic position. Circulation 2006;113:420–6. 5. Takagi H, Yamamoto H, Iwata K, Goto SN, Umemoto T. A meta-analysis of effects of prosthesis-patient mismatch after aortic valve replacement on late mortality. Int J Cardiol 2012;159:150–4. 6. Unbehaun A, Pasic M, Drews T, et al. New 29-mm balloonexpandable prosthesis for transcatheter aortic valve implantation in large annuli. Ann Thorac Surg 2013;95:1982–90. 7. Bleiziffer S, Hettich I, Hutter A, et al. Incidence and impact of prosthesis-patient mismatch after transcatheter aortic valve implantation. J Heart Valve Dis 2013;22:309–16. 8. Clavel MA, Webb JG, Pibarot P, et al. Comparison of the hemodynamic performance of percutaneous and surgical bioprostheses for the treatment of severe aortic stenosis. J Am Coll Cardiol 2009;53:1883–91. 9. Ewe SH, Muratori M, Delgado V, et al. Hemodynamic and clinical impact of prosthesis-patient mismatch after transcatheter aortic valve implantation. J Am Coll Cardiol 2011;58: 1910–8. 10. Finkelstein A, Schwartz AL, Uretzky G, et al. Hemodynamic performance and outcome of percutaneous versus surgical stentless bioprostheses for aortic stenosis with anticipated patient-prosthesis mismatch. J Thorac Cardiovasc Surg 2014;147:1892–9. 11. Freeman M, Webb JG, Willson AB, et al. Multidetector CT predictors of prosthesis-patient mismatch in transcatheter aortic valve replacement. J Cardiovasc Comput Tomogr 2013;7:248–55.
Ann Thorac Surg 2016;-:-–-
12. Giannini C, Petronio AS, Nardi C, et al. Left ventricular reverse remodeling in percutaneous and surgical aortic bioprostheses: an echocardiographic study. J Am Soc Echocardiogr 2011;24:28–36. 13. Gotzmann M, Lindstaedt M, Bojara W, et al. Hemodynamic results and changes in myocardial function after transcatheter aortic valve implantation. Am Heart J 2010;159: 926–32. 14. Jilaihawi H, Chin D, Spyt T, et al. Prosthesis-patient mismatch after transcatheter aortic valve implantation with the Medtronic-Core valve bioprosthesis. Eur Heart J 2010;31: 857–64. 15. Kalavrouziotis D, Rod es-Cabau J, Bagur R, et al. Transcatheter aortic valve implantation in patients with severe aortic stenosis and small aortic annulus. J Am Coll Cardiol 2011;58:1016–24. 16. Kamperidis V, van Rosendael PJ, de Weger A, et al. Surgical sutureless and transcatheter aortic valves: hemodynamic performance and clinical outcomes in propensity scorematched high-risk populations with severe aortic stenosis. JACC Cardiovasc Interv 2015;8:670–7. 17. Kukucka M, Pasic M, Dreysse S, et al. Patient-prosthesis mismatch after transapical aortic valve implantation: incidence and impact on survival. J Thorac Cardiovasc Surg 2013;145:391–7. 18. Laflamme J, Puri R, Urena M, et al. Incidence and risk factors of hemolysis after transcatheter aortic valve implantation with a balloon-expandable valve. Am J Cardiol 2015;115: 1574–9. 19. Mu~ noz-García AJ, Mu~ noz-García M, Carrasco-Chinchilla F, et al. Incidence and clinical outcome of prosthesis-patient mismatch after transcatheter aortic valve implantation with the CoreValve prosthesis. Int J Cardiol 2013;167:1074–6. 20. Nombela-Franco L, Ruel M, Radhakrishnan S, et al. Comparison of hemodynamic performance of self-expandable CoreValve versus balloon-expandable Edwards SAPIEN aortic valves inserted by catheter for aortic stenosis. Am J Cardiol 2013;111:1026–33. 21. Pibarot P, Weissman NJ, Stewart WJ, et al. Incidence and sequelae of prosthesis-patient mismatch in transcatheter versus surgical valve replacement in high-risk patients with severe aortic stenosis: a PARTNER trial cohort–a analysis. J Am Coll Cardiol 2014;64:1323–34. 22. Sherif MA, Abdel-Wahab M, Awad O, et al. Early hemodynamic and neurohormonal response after transcatheter aortic valve implantation. Am Heart J 2010;160:862–9. 23. Spangenberg T, Budde U, Schewel D, et al. Treatment of acquired von Willebrand syndrome in aortic stenosis with transcatheter aortic valve replacement. JACC Cardiovasc Interv 2015;8:692–700. 24. Tamborini G, Fusini L, Gripari P, et al. Feasibility and accuracy of 3DTEE versus CT for the evaluation of aortic valve annulus to left main ostium distance before transcatheter aortic valve implantation. JACC Cardiovasc Imaging 2012;5: 579–88. 25. Tzikas A, Piazza N, Geleijnse ML, et al. Prosthesis-patient mismatch after transcatheter aortic valve implantation with the medtronic CoreValve system in patients with aortic stenosis. Am J Cardiol 2010;106:255–60. 26. Van Linden A, Kempfert J, Blumenstein J, et al. Prosthesispatient mismatch after transcatheter aortic valve implantation using the Edwards SAPIENÔ prosthesis. Thorac Cardiovasc Surg 2013;61:414–20. 27. Head SJ, Mokhles MM, Osnabrugge RL, et al. The impact of prosthesis-patient mismatch on long-term survival after aortic valve replacement: a systematic review and metaanalysis of 34 observational studies comprising 27 186 patients with 133 141 patient-years. Eur Heart J 2012;33:1518–29. 28. Chen J, Lin Y, Kang B, Wang Z. Indexed effective orifice area is a significant predictor of higher mid- and long-term mortality rates following aortic valve replacement in patients with prosthesis-patient mismatch. Eur J Cardiothorac Surg 2014;45:234–40.
Ann Thorac Surg 2016;-:-–-
29. Kodali SK, Williams MR, Smith CR, et al, PARTNER Trial Investigators. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med 2012;366: 1686–95. 30. Mohty D, Dumesnil JG, Echahidi N, et al. Impact of prosthesis-patient mismatch on long-term survival after aortic valve replacement: influence of age, obesity, and left ventricular dysfunction. J Am Coll Cardiol 2009;53: 39–47.
TAKAGI AND UMEMOTO PROSTHESIS–PATIENT MISMATCH AFTER TAVI
9
31. Athappan G, Patvardhan E, Tuzcu EM, et al. Incidence, predictors, and outcomes of aortic regurgitation after transcatheter aortic valve replacement: meta-analysis and systematic review of literature. J Am Coll Cardiol 2013;61: 1585–95. 32. Zamorano JL, Badano LP, Bruce C, et al. EAE/ASE recommendations for the use of echocardiography in new transcatheter interventions for valvular heart disease. Eur Heart J 2011;32:2189–214.