Incidence, Predictive Factors, and Prognostic Value of Myocardial Injury Following Uncomplicated Transcatheter Aortic Valve Implantation
Author + information
- Josep Rodés-Cabau, MD* (josep.rodes{at}criucpq.ulaval.ca),
- Marcos Gutiérrez, MD,
- Rodrigo Bagur, MD,
- Robert De Larochellière, MD,
- Daniel Doyle, MD,
- Mélanie Côté, MSc,
- Jacques Villeneuve, MD,
- Olivier F. Bertrand, MD, PhD,
- Eric Larose, MD,
- Juan Manazzoni, MD,
- Philippe Pibarot, PhD and
- Eric Dumont, MD
- ↵*Reprint requests and correspondence:
Dr. Josep Rodés–Cabau, Quebec Heart & Lung Institute, Laval University, 2725 Chemin Ste-Foy, G1V 4G5 Quebec City, Quebec, Canada
Abstract
Objectives This study sought to: 1) determine the incidence, degree, and timing of the rise in serum cardiac markers of myocardial injury associated with uncomplicated transcatheter aortic valve implantation (TAVI); and 2) evaluate the predictive factors and prognostic value of myocardial injury associated with TAVI.
Background Very few data exist on the occurrence and clinical relevance of myocardial injury during TAVI procedures.
Methods A total of 101 patients who underwent successful TAVI (transfemoral [TF] approach, n = 38; transapical [TA] approach, n = 63) were included. Creatine kinase-MB (CK-MB) and cardiac troponin T (cTnT) levels were determined at baseline and at 6 to 12, 24, 48, and 72 h following TAVI.
Results TAVI was associated with some degree of myocardial injury in 99% of the patients (TF: 97%, TA: 100%) as determined by a rise in cTnT (maximal value, 0.48 μg/l, interquartile range [IQR]: 0.24 to 0.82 μg/l) and in 77% of the patients (TF: 47%, TA: 95%) as determined by a rise in CK-MB (maximal value, 18.6 μg/l; IQR: 11.0 to 27.4 μg/l). TA approach and baseline renal dysfunction were associated with a higher increase in biomarkers of myocardial injury (p < 0.01 for both). A larger myocardial injury was associated with a smaller improvement of left ventricular ejection fraction (LVEF) (p < 0.01). The degree of rise in cTnT was an independent predictor of cardiac mortality at 9 ± 10 months of follow-up (hazard ratio: 1.14 per each increase of 0.1 μg/l, 95% confidence interval: 1.02 to 1.28, p = 0.028).
Conclusions TAVI was systematically associated with some degree of myocardial injury, with TA approach and baseline renal dysfunction determining a higher increase in biomarkers of myocardial injury. A greater degree of myocardial injury was associated with less improvement in LVEF and a higher cardiac mortality at follow-up.
- cardiac troponin T
- CK-MB
- myocardial infarction
- myocardial injury
- percutaneous aortic valve implantation
- transcatheter aortic valve implantation
Periprocedural myocardial infarction is one of the potential complications associated with transcatheter aortic valve implantation (TAVI) (1). However, most reports on TAVI to date have not provided a definition for periprocedural myocardial infarction (1–4), and some studies have used an arbitrary definition of periprocedural myocardial infarction such as an increase in creatine kinase-MB (CK-MB) >2 to 10 times the upper normal limit (5–7), or the same definition as that used for spontaneous myocardial infarction (8). Apart from the rare occurrence of coronary complications, such as coronary ostia occlusion following valve prosthesis implantation that would lead to myocardial damage (9), uncomplicated TAVI procedures could also be associated with a rise in cardiac markers of myocardial injury. The procedure involves some degree of myocardial tissue compression caused by the balloon and valve prosthesis as well as several short episodes of extreme hypotension and global ischemia (balloon valvuloplasty, valve implantation) that might induce some degree of myocardial injury. Also, the transapical (TA) approach, which is an alternative to the transfemoral (TF) approach (2,3,8), involves the puncture of the ventricular apex and the introduction of large catheters through it. The objectives of this study were: 1) to determine the incidence, degree, and timing of the rise in serum markers of myocardial injury (CK-MB and cardiac troponin T [cTnT]) associated with uncomplicated TAVI (TA and TF); and 2) to evaluate the predictive factors and prognostic value of myocardial injury associated with TAVI.
Methods
The study population consisted of 101 patients who underwent successful uncomplicated TAVI (TF approach: n = 38; TA approach: n = 63) in our center. Patients who died within the 24 h following the procedure precluding cardiac biomarker measurements (n = 5), and those with major procedural complications (need for hemodynamic support with extracorporeal circulation: n = 6; conversion to open heart surgery: n = 3; myocardial tears during left ventricular apex repair: n = 9; life-threatening arrhythmias during the procedure: n = 7) were excluded. Selection of TF or TA approaches was based on the appropriateness of the iliofemoral arteries (10). All patients were included in a registry, and baseline and procedural characteristics variables were prospectively recorded. Patients with coronary artery disease were divided into those with and those without complete revascularization before TAVI. Incomplete revascularization was defined as the presence of an obstructive lesion (≥70% diameter stenosis, visual estimation) in 1 of the 3 main coronary vessels and main branches that was not treated before TAVI. The decision on whether or not to revascularize a vessel with an obstructive coronary lesion depended on the criteria of the physician responsible for the patient and the TAVI team. An electrocardiogram (ECG) was performed within 24 h before TAVI, immediately after the procedure, and at hospital discharge. All ECGs were evaluated for the purpose of the study by a cardiologist blinded to clinical data. All patients had continuous ECG monitoring until hospital discharge. A Doppler echocardiography was performed before the procedure and at hospital discharge. All procedures were performed under a Compassionate Clinical Program approved by Health Canada, and all patients provided signed informed consent for the procedures.
The TAVI procedures have been extensively detailed in previous studies (10). All procedures were performed with the Edwards valve (Edwards SAPIEN or SAPIEN XT, Edwards Lifesciences Inc., Irvine, California).
Measurements of the serum markers of myocardial injury
Blood samples were collected at 6 to 12, 24, 48, and 72 h following the procedure. CK-MB mass and cTnT levels were measured at each point time. CK-MB and cTnT measurements were performed by electrochemiluminescence immunoassay (Roche, Minneapolis, Minnesota). Based on the 99th percentile in a healthy population and the requirement of a ≤10% coefficient variation, the upper normal limits for CK-MB and cTnT levels in our institution were 10 and 0.03 μg/l, respectively. Myocardial injury was defined as a CK-MB level >10 μg/l or a cTnT level ≥0.04 μg/l. In those patients with elevated CK-MB or cTnT levels at baseline, myocardial injury was defined as any increase >20% after the procedure (11).
Follow-up
Clinical follow-up was carried out by clinical visits or through telephone contact. Patients were followed at 1, 6, and 12 months after the procedure and annually thereafter. Doppler echocardiography was performed at 6-month to 1-year follow-up. Death at any time during the follow-up period was recorded and further classified as of cardiac or noncardiac origin.
Statistical analysis
Continuous variables are expressed as mean ± SD or median (25th to 75th interquartile range [IQR]) depending on variable distribution. Group comparisons were analyzed using the Student t test or Wilcoxon rank sum test for continuous variables, and chi-square test for categorical variables. An analysis of variance for repeated measures was performed to test for equal means at different times (baseline, 6 to 12, 24, 48, and 72 h) for the cardiac enzyme values, and a 2-way analysis of variance for repeated measures with interaction was used to compare the changes in cardiac enzyme levels at different time points between groups (TF-TAVI vs. TA-TAVI; complete vs. incomplete revascularization before TAVI). The covariance among repeated measures was modeled using a compound symmetry structure, which indicated that the correlations between all pairs of measures were the same. The variables associated with a greater myocardial injury were determined by univariate analysis, and those variables with a p value <0.10 were entered in a stepwise linear regression analysis. Relationships between cardiac enzyme and left ventricular changes were expressed with the Spearman correlation. The univariate normality assumptions were verified with the Shapiro-Wilk tests. The multivariate normality assumptions were verified with the Shapiro-Wilk tests after a Cholesky factorization. The Brown and Forsythe variation of the Levene test statistic was used to verify the homogeneity of variances. Cardiac enzyme values were log transformed to fulfill the normality and variances assumptions. Receiver-operating characteristic (ROC) curve analysis was performed to discriminate power of the cardiac enzyme rise for a left ventricular ejection fraction (LVEF) decrease at 6- to 12-month follow-up (Δ <0 vs. Δ ≥0). A multivariate Cox regression analysis was used to determine the variables predictive of cumulative mortality at follow-up, including all baseline and procedural variables with a p value <0.10 in the univariate analysis. The results were considered significant with p values ≤0.05. All analyses were conducted using the statistical package SAS version 9.2 (SAS Institute Inc., Cary, North Carolina).
Results
The clinical, echocardiographic, and procedural characteristics of the entire study population and the TF and TA groups are shown in Table 1. Mean aortic gradient decreased from 42 ± 17 mm Hg at baseline to 9 ± 3 mm Hg after TAVI, and aortic valve area increased from 0.62 ± 0.19 cm2 at baseline to 1.65 ± 0.24 cm2 after TAVI. Some degree of residual aortic regurgitation occurred in 75 patients (74%), and was trivial, mild, and moderate in 33%, 35%, and 6% of the cases, respectively.
Clinical, Echocardiographic, and Procedural Characteristics of the Study Population (n = 101)
Serum markers of myocardial injury following TAVI
The mean values of CK-MB and cTnT at each time point within the 72 h following the procedure for the entire study population and for the TF- and TA-TAVI groups are shown in Figure 1. CK-MB level was within normal limits in all patients at baseline and increased above the upper normal limit in 77% of the patients with a median peak of 18.6 μg/l (IQR: 11.0 to 27.4 μg/l) at 24 h following the procedure and returned to baseline values at 72 h after TAVI. In the TA-TAVI group, the CK-MB levels were above the upper normal values in 95% of the patients compared with 47% of the patients in the TF-TAVI group (p < 0.0001). The degree of CK-MB increase was also higher in the TA-TAVI group compared with the TF-TAVI group at all time points following the procedure (p < 0.0001) (Fig. 1B), and the median maximal value of CK-MB in the TA-TAVI group within the 72 h following the procedure was 22.6 μg/l (IQR: 16.3 to 32.4 μg/l) compared with 9.9 μg/l (IQR: 6.4 to 13.4 μg/l) in the TF-TAVI group (p < 0.0001). The increase in CK-MB levels grouped according to the degree of rise in CK-MB following TAVI in the entire study population and in the TF-TAVI and TA-TAVI groups is shown in Figure 2.
Changes in Serum Markers of Myocardial Injury Following TAVI
Changes in creatine kinase-myocardial band (CK-MB) and cardiac troponin T (cTnT) levels within the 72 h following transcatheter aortic valve implantation (TAVI) in all patients (A) and grouped according to the approach (transfemoral [TF], transapical [TA]) (B). Values are expressed as median (25th to 75th interquartile range).
Degree of Increase in CK-MB and cTnT Levels Following TAVI
Cardiac biomarker changes following TAVI are grouped according to the degree of CK-MB (A) and cTnT (B) increase in all patients and in the TF and TA approach groups. Abbreviations as in Figure 1.
The cTnT levels were within the normal limits at baseline in 85% of the patients and increased above the upper normal limit in all patients but 1 (1%) following TAVI, with a median peak of 0.48 μg/l (IQR: 0.24 to 0.82 μg/l) at 48 h following the procedure (Fig. 1A). The cTnT values continued to be above baseline values at 72 h following TAVI (Fig. 1A). In the TF-TAVI group, the peak of cTnT was at 6 to 12 h after the procedure, whereas in the TA-TAVI group, it occurred at 48 h after the procedure. cTnT levels increased to above the upper normal values in all patients in the TA-TAVI group compared with 97% of the patients in the TF-TAVI group (p = 0.38). The degree of cTnT increase was higher in the TA-TAVI group compared with the TF-TAVI group at all time points following the procedure (p < 0.0001) (Fig. 1B). The median maximal value of cTnT in the TF-TAVI group within the 72 h following the procedure was 0.19 μg/l (IQR: 0.13 to 0.31 μg/l) compared with 0.74 μg/l (IQR: 0.48 to 1.07 μg/l) in the TA-TAVI group (p < 0.0001). The increase in cTnT levels grouped according to the degree of rise in cTnT following TAVI in the entire study population and in the TF-TAVI and TA-TAVI groups is shown in Figure 2B.
One patient (1%) developed new Q waves in the inferior leads in the ECG following the procedure. No life-threatening ventricular arrhythmia occurred during the hospitalization period.
Predictors of myocardial injury following TAVI
The degree of myocardial injury depending on baseline and procedural characteristics of the entire study population is shown in Table 2. In the multivariate analysis, TA-TAVI was the only independent predictor of a higher rise in CK-MB following the procedure (r2 = 0.24, p < 0.0001), and TA-TAVI and baseline renal dysfunction were the 2 independent predictors of a higher rise in cTnT following the procedure (r2 = 0.30, p < 0.0001, r2 = 0.07, p = 0.003, for TA-TAVI and renal dysfunction, respectively).
Degree of CK-MB and cTnT Increase Following TAVI Depending on Baseline and Procedural Characteristics
The rise in CK-MB and cTnT levels at each time point within the 72 h following the procedure grouped according to the presence of coronary artery disease and the completeness of coronary revascularization before TAVI is shown in Figure 3. Patients with prior coronary artery disease exhibited a rise in cardiac markers similar to those without coronary artery disease (maximal CK-MB levels: 18 μg/l [IQR: 10.9 to 26.6 μg/l] vs. 19.3 μg/l [IQR: 11.3 to 31.6 μg/l], p = 0.61; maximal cTnT levels: 0.49 μg/l [IQR: 0.25 to 0.81 μg/l] vs. 0.43 μg/l [IQR: 0.21 to 0.87 μg/l], p = 0.99). Among the patients with a history of coronary artery disease, those who had complete revascularization prior to TAVI exhibited a rise in cardiac markers similar to those without complete revascularization (maximal CK-MB levels: 18.6 μg/l [IQR: 11.0 to 26.7 μg/l] vs. 16.3 μg/l [IQR: 10.8 to 25.9 μg/l], p = 0.99; maximal cTnT levels: 0.51 μg/l [IQR: 0.24 to 0.81 μg/l] vs. 0.37 μg/l [IQR: 0.27 to 0.81 μg/l], p = 0.67).
Coronary Artery Disease, Coronary Revascularization Completeness Prior to TAVI, and Degree of Myocardial Injury
Changes in CK-MB and cTnT levels within the 72 h following TAVI grouped according to the presence of coronary artery disease and coronary revascularization completeness before TAVI. Values are expressed as median (25th to 75th interquartile range). (A) All patients. (B) TF approach group. (C) TA approach group. Abbreviations as in Figure 1.
Prognostic value of myocardial injury
A total of 6 patients (5.9%) died within the 30 days following the procedure. These patients tended to exhibit a greater increase in cTnT levels (0.98 μg/l [IQR: 0.38 to 1.33 μg/l]) following TAVI than those who survived (0.48 μg/l [IQR: 0.23 to 0.82 μg/l], p = 0.19). There were no differences in CK-MB rise between patients who had died and those who had survived at 30-day follow-up (15.1 μg/l [IQR: 6.2 to 26.5 μg/l] vs. 12.9 μg/l [IQR: 6.9 to 20.3 μg/l], p = 0.83).
At a mean follow-up of 9 ± 10 months, a total of 19 patients had died. Of these, 10 patients died during the follow-up period of noncardiac causes (respiratory = 4, cancer = 3, renal failure = 2, sepsis = 1). Baseline and procedural characteristics of the patients grouped according to the occurrence of global (cardiac and noncardiac) and cardiac cumulative death at follow-up are shown in Tables 3 and 4, respectively. In the multivariate analysis, the degree of increase in cTnT was identified as the independent predictor of cardiac mortality at follow-up (hazard ratio: 1.14 for each increase in cTnT of 0.1 μg/l, 95% confidence interval [CI]: 1.02 to 1.28, p = 0.028). A cTnT level >0.60 μg/l was identified as the cutoff point with the best sensitivity and specificity for the prediction of cumulative cardiac death following uncomplicated TAVI, with an area under the ROC curve of 0.65 (95% CI: 0.40 to 0.89, p = 0.015). The sensitivity and specificity of a cTnT >0.60 μg/l for the prediction of cardiac mortality following TAVI were 67% and 62%, respectively.
Baseline and Procedural Characteristics of the Study Population, According to the Occurrence of Cumulative Mortality (Cardiac and Noncardiac) at Follow-Up
Baseline and Procedural Characteristics of the Study Population, According to the Occurrence of Cumulative Cardiac Mortality
The LVEF as determined by echocardiography at 6- to 12-month follow-up was available in 39 of the 46 patients (85%) with a follow-up ≥6 months. The LVEF tended to be higher at follow-up compared with baseline (58 ± 13% vs. 56 ± 13%, p = 0.13). The correlation between the increase (Δ) in CK-MB and cTnT and the changes in LVEF between baseline and follow-up are shown in Figure 4. The increase in CK-MB and cTnT levels following the procedure exhibited a significant correlation with the changes in LVEF between baseline and follow-up (r = −0.41, p = 0.009 for CK-MB; r = −0.46, p = 0.003 for cTnT). A CK-MB level >26 μg/l was identified as the cutoff point with the best sensitivity and specificity for the prediction of LVEF decrease following uncomplicated TAVI, with an area under the ROC curve of 0.74 (95% CI: 0.579 to 0.90, p = 0.026). The sensitivity and specificity of a CK-MB level >26 μg/l for the prediction of LVEF deterioration at follow-up were 72% and 64%, respectively. A cTnT level >0.48 μg/l was identified as the cutoff point with the best sensitivity and specificity for the prediction of LVEF decrease following TAVI, with an area under the ROC curve of 0.76 (95% CI: 0.61 to 0.92, p = 0.018). The sensitivity and specificity of a cTnT level >0.48 μg/l for the prediction of LVEF deterioration at follow-up were 64% and 79%, respectively.
Myocardial Injury and LVEF Changes Following TAVI
Relationship between the maximal increase in CK-MB (A) and cardiac troponin T (B) levels and the changes in left ventricular ejection fraction (LVEF) following transcatheter aortic valve implantation (TAVI).
Discussion
TAVI was systematically associated with some degree of myocardial injury as determined by CK-MB and cTnT release following the procedure. TA-TAVI was associated with an increase in CK-MB and cTnT levels about 2 and 4 times higher, respectively, than that observed after TF-TAVI. TA-TAVI and baseline renal dysfunction predicted a higher increase in biomarkers of myocardial injury after the procedure. The presence of prior coronary artery disease was not associated with a higher degree of myocardial injury, and no differences were observed between those patients with and without complete revascularization before the procedure. Finally, a higher degree of myocardial injury was associated with a lesser improvement in LVEF and a higher cardiac mortality at midterm follow-up.
Very few previous studies on TAVI have reported data on the occurrence of myocardial injury following TAVI. Grube et al. (5) and Buellesfeld et al. (6) reported an incidence of 1.5% to 1.8% of CK-MB elevation >2 times the upper normal limit following TF-TAVI. Svensson et al. (8) showed that 17% of the patients who underwent TA-TAVI had a periprocedural myocardial infarction as determined by using the definition of spontaneous myocardial infarction (11). Also following TA-TAVI, we have previously reported an increase in CK-MB values >5 times the upper normal limit in 6% of the patients (12). The present study, which is the first to our knowledge to systematically evaluate the degree of myocardial injury associated with uncomplicated TAVI, has shown that the procedure was associated with some degree of myocardial injury in 99% of the patients. The reasons for the occurrence of myocardial injury in most patients undergoing TAVI are probably multifactorial. The present study has shown that the main procedural factor associated with a significant increase in cardiac biomarker levels was the TA approach, which involves the introduction of large catheters through the ventricular apex. Also, the TAVI procedure is associated with several episodes of extreme hypotension leading to short periods of myocardial ischemia that could contribute to the increase in CK-MB and cTnT levels after the procedure. The use of a self-expandable valve such as the CoreValve (Medtronic, Minnesota) avoids the rapid pacing and extreme hypotension during valve expansion, and this might contribute to a reduction in myocardial injury (4–6). On the other hand, TAVI involves some direct stretching of the basal myocardial septum by the balloon expanding the prosthesis and by the prosthesis itself. The balloon-valve prosthesis is systematically overdimensioned with respect to the aortic annulus, and interestingly, a higher degree of overdimensioning was associated with a higher increase in cTnT in our study, although this procedural variable was not found to be an independent predictor of a higher degree of myocardial injury following TAVI. Also, coronary embolism secondary to the embolization of small microparticles from the native aortic valve following the retrograde crossing of the valve and especially during balloon valvuloplasty and valve implantation may be another potential cause of myocardial injury in these patients. Among the baseline characteristics, renal dysfunction was the only predictor of a higher cTnT rise following the procedure. It has been well demonstrated that renal dysfunction can be associated with higher levels of cTnT (13), and it has been shown that elevated cTnT levels and its changes (increase) remain an important prognostic factor despite the presence of renal dysfunction (14,15). Interestingly, we did not find any influence of the presence of prior coronary artery disease and pre-procedural completeness of coronary revascularization on myocardial injury following TAVI. Masson et al. (16) recently showed the lack of prognostic value of completing coronary revascularization before TAVI. There has been some debate about the usefulness of achieving a complete revascularization in coronary patients undergoing TAVI with no symptoms of angina. Although the very few data available to date, including the present study, do not support performing coronary interventions prior to the TAVI procedure, future randomized studies will have to determine the potential benefits of this strategy.
It has been extensively shown that the occurrence of myocardial injury during cardiac procedures such as percutaneous coronary intervention or cardiac surgery is associated with worse acute and long-term outcomes (11,17). The present study is the first to our knowledge to suggest that significant periprocedural myocardial injury has also a prognostic value following TAVI procedures. The degree of myocardial injury correlated with the changes in LVEF at midterm follow-up, with a CK-MB increase of >26 μg/l and cTnT of >0.48 μg/l best determining LVEF deterioration following TAVI. Also, the degree of myocardial injury as determined by the increase in cTnT was associated with a higher cardiac mortality at midterm follow-up, with a troponin cutoff >0.60 μg/l best determining the occurrence of cardiac death following uncomplicated TAVI. Although magnetic resonance imaging studies have shown that any increase in cardiac biomarkers following coronary interventions is associated with some degree of irreversible myocardial damage (18), the definition of myocardial infarction following cardiac procedures has been based on the cardiac enzyme threshold (CK-MB >3 and >5 times the upper normal limit for percutaneous coronary intervention and cardiac surgery, respectively), which best predicted a poor clinical outcome. Either no definition or arbitrary definitions have been used to date for the diagnosis of periprocedural myocardial infarction associated with TAVI (1–8). Our results showed that a CK-MB increase of about >3 times the upper normal limit and a cTnT increase of about >15 times the upper normal limit following TAVI might be clinically relevant, and suggest that these cardiac biomarker rise thresholds might be appropriate to define the occurrence of periprocedural myocardial infarction following TAVI.
Study limitations
In patients with prior coronary artery disease, the decision to revascularize coronary obstructive lesions prior to TAVI depended on the criteria of the physician responsible for the patient and the TAVI team, with no pre-specified protocol. This might have introduced some bias regarding the association between revascularization completeness prior to and myocardial injury following TAVI. The small sample size limits the conclusions regarding the results of the different subgroup analyses presented in this study. Therefore, these results will have to be confirmed by larger studies.
Conclusions
Some degree of myocardial injury as determined by CK-MB and cTnT release occurred in the vast majority of patients undergoing TAVI, irrespective of the approach. TA-TAVI and baseline renal dysfunction, but not the presence of coronary artery disease and revascularization completeness prior to TAVI, determined a higher rise in biomarkers of myocardial injury following the procedure. A higher degree of myocardial injury was associated with a smaller improvement or even a deterioration of LVEF and a higher cardiac mortality at midterm follow-up, suggesting that CK-MB and especially cTnT values post-TAVI might become important biomarkers in the evaluation of TAVI procedures. Future studies including a larger number of patients and a longer follow-up will have to confirm these results and further determine the cardiac enzyme rise cutoff that should be used for the diagnosis of periprocedural myocardial infarction following TAVI.
Acknowledgments
The authors thank Jacinthe Aubé and Nathalie Boudreau for their exceptional work on data collection and patients' follow-up, and Serge Simard, MSc, for statistical analyses.
Footnotes
Continuing Medical Education (CME) is available for this article.
Drs. Rodés-Cabau, Doyle, and Dumont are consultants for Edwards Lifesciences Inc. All other authors have reported that they have no relationships to disclose.
- Received September 16, 2010.
- Revision received November 22, 2010.
- Accepted November 23, 2010.
- American College of Cardiology Foundation
