Myocardial Injury Common with TAVI, But Clinical Import Less Certain

Download this article's Factoid (PDF & PPT for Gold Subscribers)

Some myocardial injury appears to be endemic to transcatheter aortic valve implantation (TAVI). Transapical procedures appear particularly susceptible to this complication, producing biomarker increases two- to fourfold higher than transfemoral procedures, according to data published in the May 17, 2011, issue of the Journal of the American College of Cardiology. In addition, a higher degree of myocardial injury is associated with diminished improvement in left ventricular function and increased cardiac mortality at midterm follow-up.

To examine the incidence and degree of myocardial injury as evidenced by rises in cardiac markers, investigators led by Josep Rodés-Cabau, MD, of Laval University (Quebec City, Canada), analyzed single-center data from 101 patients who underwent successful TAVI, 38 with a transfemoral approach and 63 with a transapical approach. All procedures were performed with either Edwards Sapien or Sapien XT valves (Edwards Lifesciences, Irvine, CA). Biomarker levels were measured at baseline and multiple time points within 72 hours of the procedure. 

Overall, mean aortic gradient decreased from 42 ± 17 mm Hg at baseline to 9 ± 3 mm Hg after TAVI, while aortic valve area increased from 0.62 ± 0.19 cm2 to 1.65 ± 0.24 cm2. Some degree of residual aortic regurgitation (AR) was seen in 74% of patients; AR was trivial or mild in 68% and moderate in 6%.

A Tale of 2 Biomarkers 

CK-MB levels were within normal limits (defined as under 10 mg/L for CK-MB and 0.03 µg/L for troponin T) for the entire cohort at baseline, but at 24 hours they had increased to above that threshold in 77% of patients. They reached a median peak of 18.6 µg/L at 24 hours then returned to baseline values at 72 hours.

Stratified by approach, 95% of transapical patients showed values above the upper limit of normal compared with 47% of transfemoral patients (P < 0.0001). The degree of increase was also higher in the transapical group at all time points (P < 0.0001), with a median maximal value in the transapical group at 72 hours of 22.6 µg/L vs. 9.9 µg/L in the transfemoral group (P < 0.0001).

The pattern was similar, if exaggerated, for cardiac troponin T values. At baseline, 85% of patients were within normal limits, while after TAVI all but 1% had increased to above normal limit. The median peak value was 0.48 µg/L at 48 hours, which persisted through 72 hours.

In the transfemoral group, the peak value for cardiac troponin T occurred at 6 to 12 hours after the procedure, whereas in the transapical group it was recorded at 48 hours. In addition, measurements increased to above normal values in 100% of transapical patients and 97% of transfemoral patients (P = 0.38). However, the degree of increase was higher in the transapical compared with the transfemoral group at all time points (P < 0.0001). Likewise, the median maximal value within 72 hours in the transapical group exceeded that of the transfemoral group (0.74 µg/L vs. 0.19 µg/L; P < 0.0001).    

One patient developed new Q waves in the inferior ECG leads following the procedure, but no life-threatening ventricular arrhythmias occurred during hospitalization.

In multivariate analysis, the transapical approach was the only independent predictor of higher rise in CK-MB following TAVI (P < 0.0001), while both a transapical approach and baseline renal dysfunction predicted a greater increase in troponin T (P < 0.0001 and P = 0.003, respectively).

Interestingly, patients with CAD had no greater increases in cardiac markers than those without coronary disease. Moreover, among CAD patients, biomarker levels were similar between those who did or did not undergo complete revascularization prior to TAVI.

Prognostic Value Confirmed

Six patients died within 30 days. They showed greater increases in troponin T levels compared with survivors but no difference in CK-MB levels. 

At a mean follow-up of 9 months, 19 patients had died, 9 of cardiac causes. Multivariate analysis found that the degree of increase in troponin T independently predicted cardiac mortality (HR 1.14 for each 0.1 µg/L increase; 95% CI 1.02-1.28; P = 0.028). A troponin T level greater than 0.60 µg/L was determined to be the cutoff with the best combined sensitivity (67%) and specificity (62%).

LVEF as determined by echocardiography tended to be higher at 6- to 12-month follow-up than at baseline. Biomarker increases were inversely related to changes in LVEF (P = 0.009 for CK-MB and P = 0.003 for troponin T). A CK-MB level greater than 26 µg/L and a troponin T level greater than 0.48 µg/L were identified as the cutoffs with the best sensitivity (72% and 64%, respectively) and specificity (64% and 79%, respectively) for predicting LVEF decline after uncomplicated TAVI.  

A Myocardial Injury-Prone Procedure 

The authors call attention to several procedural characteristics that predispose TAVI toward increased myocardial injury, including:

  • Episodes of extreme hypotension leading to short periods of myocardial ischemia
  • Stretching of the basal myocardial septum by the balloon expanding the prosthesis and by the prosthesis itself
  • Systematic oversizing of balloon-valve prostheses with respect to the aortic annulus
  • Coronary embolism secondary to embolization of microparticles from the native aortic valve following retrograde crossing of the valve, especially during balloon valvuloplasty and valve implantation
  • Introduction of large catheters through the ventricular apex in the transapical approach

According to Dr. Rodés-Cabau and colleagues, until now either no definition or arbitrary definitions have been used to diagnose periprocedural MI in TAVI. “Our results showed that a CK-MB increase of about [greater than] 3 times the upper normal limit and a [troponin T] increase of about [greater than]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 [MI] following TAVI.”

In a telephone interview with TCTMD, Dr. Rodés-Cabau said, “The main limitation of our study is the number of patients, especially when it comes to determining prognostic value. But I think we made an important step in the right direction, suggesting that it merits further evaluation with more patients and longer follow-up.” The small cohort also precluded a separate analysis of the myocardial risks of the transapical and transfemoral procedures, he added, “but that, I think, will be the next step.”

In the paper, the authors acknowledge that because the decision about whether to revascularize obstructive lesions prior to TAVI depended on clinical judgment rather than protocol, some bias may have been introduced regarding a possible connection between complete revascularization and myocardial injury. “It’s likely that the patients who received revascularization were the ones who needed it most, so I don’t think we can draw any definitive conclusions, but at least from this first [small study] the data seem to be reassuring,” Dr. Rodés-Cabau noted.

Interesting—Just Not Very Practical

Peter C. Block, MD, of Emory University Hospital (Atlanta, GA), said that although the study was scientifically sound, it offered little help in clinical decision making.

In a telephone interview with TCTMD, he described the finding of a strong link between biomarker increases and transapical procedures as “sort of duh,” explaining, “After all, you’re putting a hole in the heart and destroying myocardial cells.” Of far greater interest are transfemoral procedures, he said, in which the biomarker release is likely due to multiple factors.

“One thing I didn’t see, though they tried to [address] it with the multivariate analysis, is whether CK-MB loss is associated with the length of pacing or the number of paces,” Dr. Block said. “The longer the pacing, the more time the myocardium is not receiving any [blood] flow.” This is an area where technique may play a role, but the paper sheds little light on it, he noted.  

Another important but unanswered question is the impact of balloon valvuloplasty itself on biomarkers, Dr. Block commented. “I don’t think anyone has studied that, but [the information] would be important for understanding whether or not [biomarker elevation] is a true valve-related phenomenon,” he said. “If it is, that would raise some serious questions about future valve design.”

Dr. Rodés-Cabau largely agreed about potential design implications: “I’m not sure that operator experience can contribute much to attenuating biomarker rises, but technological improvements—for example, reduction of catheter size in transapical procedures—might make a difference.”

Asked how the data might have differed if CoreValve (Medtronic, Minneapolis, MN) had been used in place of Edwards prostheses, Dr. Rodés-Cabau responded, “With CoreValve there would be no transapical procedures. But it would be interesting to compare the 2 devices in transfemoral TAVI. The main difference with the self-expanding CoreValve is the absence of the need for rapid pacing during valve expansion, but we don’t know if this can make a difference in myocardial injury.”

As to the usefulness of the prognostic cutoffs presented in the paper, Dr. Block said he was “not at all impressed” given the wide scatter of data points relating biomarker levels to outcomes. “Statistically, they can [derive cutoff values], but what it means clinically for the individual patient is still very much up in the air,” he said, noting that “some patients with very high CK-MB or troponins had LVEFs that actually got better.”

These data, though interesting, don’t provide any clinical guidance, Dr. Block said. “It does not help me going in to advise a patient whether or not they should have a valve replacement. And it doesn’t tell me anything about managing a patient after a procedure. Maybe with these data I may think about putting a patient with a high CK-MB on ACE inhibitors—but that’s a whole other study,” he added.

Dr. Rodés-Cabau agreed about the lack of clinical implications, at least until further data are accumulated, although he noted that “at the least patients with high biomarker rises should be followed more closely.”

In the end, the study’s biggest contribution may lie in encouraging future researchers to look carefully at their data to try to pin down the relationship between biomarkers and outcomes, Dr. Block observed. In that regard, he noted, it would be interesting to compare the current findings with the data that emerge from the ongoing PARTNER trials.

Study Details

Compared with transfemoral patients, transapical patients had more peripheral arterial disease (44% vs. 18%; P = 0.009), tended to have a higher incidence of porcelain aorta (29% vs. 11%; P = 0.05), and underwent more rapid pacing runs (6 ± 2 vs. 5 ± 2; P = 0.0001). However, transfemoral patients were older (82 ± 6 vs. 78 ± 9 years; P = 0.006).

Selection of the transfemoral or transapical approach was based on the appropriateness of the iliofemoral arteries.


Rodés-Cabau J, Gutiérrez M, Bagur R, et al. Incidence, predictive factors, and prognostic value of myocardial injury following uncomplicated transcatheter aortic valve implantation. J Am Coll Cardiol. 2011;57:1988-1999.



Related Stories:

Acute Kidney Injury Quadruples Mortality Risk After Percutaneous Aortic Valve Replacement
Consensus TAVI Endpoint Definitions Raise Standard of Research in New Field
Impressive Transapical TAVI Results Reported in Very High-Risk Patients


Myocardial Injury Common with TAVI, But Clinical Import Less Certain

Some myocardial injury appears to be endemic to transcatheter aortic valve implantation (TAVI). Transapical procedures appear particularly susceptible to this complication, producing biomarker increases two to fourfold higher than transfemoral procedures, according to data published in the May 17, 2011,
  • Dr. Rodés-Cabau reports serving as a consultant for Edwards Lifesciences.
  • Dr. Block reports being an investigator for the PARTNER trial.