Author + information
- Received July 16, 2013
- Revision received October 20, 2013
- Accepted October 24, 2013
- Published online February 1, 2014.
- Chiara Bernelli, MD∗,
- Alaide Chieffo, MD∗∗ (, )
- Matteo Montorfano, MD∗,
- Francesco Maisano, MD†,
- Gennaro Giustino, MS∗,
- Gill Louise Buchanan, MBChB∗,
- Jaclyn Chan, MBBS∗,
- Charis Costopoulos, MD∗,
- Azeem Latib, MD∗,
- Filippo Figini, MD∗,
- Ermelinda De Meo, MS∗,
- Francesco Giannini, MD∗,
- Remo Daniel Covello, MD†,
- Chiara Gerli, MD†,
- Annalisa Franco, MD†,
- Eustachio Agricola, MD‡,
- Pietro Spagnolo, MD†,
- Micaela Cioni, MD†,
- Ottavio Alfieri, MD†,
- Paolo Guido Camici, MD‡ and
- Antonio Colombo, MD∗
- ∗Interventional Cardiology Unit, San Raffaele Scientific Institute, Milan, Italy
- †Cardiothoracic Department, San Raffaele Scientific Institute, Milan, Italy
- ‡Cardiology Department, San Raffaele Scientific Institute, Milan, Italy
- ↵∗Reprint requests and correspondence:
Dr. Alaide Chieffo, Interventional Cardiology Unit, San Raffaele Scientific Institute, Via Olgettina 60 20132 Milan, Italy.
Objectives This study sought to evaluate the impact of baseline activated clotting time (ACT)–guided heparin administration on major bleeding after transfemoral transcatheter aortic valve implantation (TAVI).
Background Bleeding after TAVI is frequent and associated with unfavorable prognosis. Proper intraprocedural heparin dose administration may reduce the risk of potential overdosing in this frail study group.
Methods Of the patients who underwent transfemoral TAVI in our center from November 1, 2007 to June 31, 2012, 362 were retrospectively analyzed. Because abnormally high baseline ACT values were noted, heparin was administered at the operator's discretion, according to baseline ACT (ACT-guided, n = 174) or patient's body weight (non–ACT-guided, n = 188). The primary study objective was 30-day major bleeding as defined by the Valve Academic Research Consortium criteria. Secondary objectives were any life-threatening, and minor bleeding, and other Valve Academic Research Consortium outcomes at 30 days.
Results Bleeding occurred in 167 (46.1%) patients; of these, 76 (21.0%) had major bleeding. The ACT-guided group had a significantly lower occurrence of major (7.5% vs. 33.5%, p < 0.001), life-threatening (12.1% vs. 20.2%, p = 0.04), and any bleeding (25.9% vs. 64.9%, p < 0.001). Conversely, no differences were noted in the other study objectives. After adjustment for potential confounders, the protective odds ratio for ACT-guided therapy on major bleeding was 6.4 (95% confidence interval: 2.3 to 17.9; p < 0.001) at 30 days.
Conclusions In our experience, heparin administration according to baseline ACT was correlated with a significantly lower occurrence of major bleeding in transfemoral TAVI. This strategy might be a useful tool in reducing bleeding in this high-risk study group.
Transcatheter aortic valve implantation (TAVI) has been demonstrated to be a valid therapeutic option for high-risk patients with symptomatic severe aortic stenosis (AS) (1,2). However, a wide range of potential complications might mitigate the beneficial effect of TAVI procedures. Particularly, TAVI-related bleeding still remains an important complication demanding careful consideration. Bleeding after TAVI is relatively frequent, ranging from 22.8% to 77.0%, and has been correlated with increased mortality (3–5). This is of particular concern in the TAVI study group mainly composed of elderly patients who are more prone to bleeding. Furthermore, patients with severe AS may have abnormalities in coagulation pathways as indicated by a higher incidence of spontaneous bleeding (6). The presence of severe comorbidities in this patient group, and the large sheath sizes required for arterial access are the main causes of bleeding complications after TAVI procedures. However, a common cause of bleeding in this study group might be represented by excessive intraprocedural anticoagulation therapy with heparin. According to the most recent expert consensus statement on TAVI, weight-based intravenous unfractionated heparin (UFH) is the recommended adjunctive intraprocedural antithrombotic therapy during TAVI with a suggested activated clotting time (ACT) >300 s throughout the procedure (7). On the other hand, an ACT target range of 250 to 300 s has been reported in the published data (8). Nevertheless, to the best of our knowledge, no study has specifically evaluated appropriate heparin dosing in this clinical setting. Because of the risk of potential overdosing with consequent hemorrhagic complications, strategies that provide a more accurate heparin dosing need to be investigated.
In our clinical practice, we systematically collected baseline ACT values during TAVI procedures, so we investigated a strategy where baseline ACT values were used for guiding subsequent heparin administration. Accordingly, the aim of the present study was to evaluate the impact of the baseline ACT-guided versus a non–ACT-based “weight-adjusted” heparin administration strategy on bleeding in transfemoral (TF) TAVI.
From November 2007 to June 2012, all consecutive patients with symptomatic severe AS treated with TF-TAVI in our center (San Raffaele Scientific Institute, Milan, Italy) were retrospectively analyzed. Patients were considered eligible for TAVI if they had a high or prohibitive risk for conventional surgery after being reviewed by a dedicated heart team, according to our current practice previously reported (9,10).
Procedures and devices
Depending on clinical conditions and evaluation of the patient by the cardiac anesthesiologist, the procedure was performed under either general or local anesthesia with conscious sedation. Both self-expanding and balloon-expandable prostheses were implanted. Initially, in November 2007, the Sapien THV (Edwards Lifesciences, Irvine, California) was used; the Medtronic CoreValve ReValving System (Medtronic, Minneapolis, Minnesota) valves became available in July 2008, and the Sapien XT replaced the Sapien THV in April 2010 (10). Details of the procedure were previously reported (9,10). In cases of full percutaneous TF approach, all therapeutic femoral access sites were closed with the Prostar (Abbott Vascular, Abbott Park, Illinois) pre-closure device. Moreover, the arterial access and closure of the access sites were performed with a “crossover technique” (11,12).
Heparin administration strategy: ACT versus non–ACT-guided group
As a standard practice used by our cardiac anesthetists, a baseline ACT measurement before administering heparin was performed in all patients undergoing TAVI procedures. ACT samples were drawn through the arterial sheath immediately after the insertion. To clear the sample from the flush solution contaminated by heparin, 10 ml of blood was withdrawn before taking the 4-ml ACT sample. ACT was measured using the Medtronic ACT Plus System.
Interestingly, we noticed abnormally high values of baseline ACT in our study group. This observation raised the awareness of some operators who have started to administer the dosage of heparin on the basis of the baseline ACT values. Accordingly, on the basis of the operator's decision, in our series, 2 heparin administration strategies were performed: 1) an ACT-guided strategy in which heparin was adapted according to the baseline ACT; and 2) a non–ACT-guided strategy in which the heparin dosing was weight-adjusted. Details regarding the 2 heparin administration strategies are illustrated in Figure 1.
In both groups, the intraprocedural anticoagulant effect of the UFH was monitored and subsequently adjusted to maintain the ACT between 200 and 300 s. Heparin anticoagulation was reversed with protamine at neutralizing doses, if necessary, according to operator's preference.
All patients received acetylsalicylic acid 100 mg before the procedure and lifelong thereafter. A 300-mg loading dose of clopidogrel was administered the day before the procedure followed by 75 mg daily for 6 months. With regard to patients on oral anticoagulation therapy (OAT), this was interrupted at least 3 days prior to the TAVI procedure. In this group of patients, when necessary, subcutaneous low-molecular-weight heparin was used and was interrupted at least 12 h prior to the procedure. In patients with indication to OAT, dual antiplatelet therapy at discharge was prescribed according to the bleeding and thrombotic risk of each patient (13).
Study objectives and data collection
The primary study objective was 30-day occurrence of major bleeding. Secondary study objectives were life-threatening, minor, and any bleeding; need for red blood cell (RBC) transfusions; and the other Valve Academic Research Consortium (VARC) outcomes at 30 days.
Study objectives were defined according to the VARC criteria (3).
The definition of clinically meaningful bleeding was on the basis of objective criteria, including an obvious source of bleeding and number of RBC transfusions.
A clear distinction between procedural and non-procedural-related blood loss as well as obvious and nonobvious source of hemorrhagic events were taken into account in defining bleeding complications. Furthermore, bleeding and the need for RBC transfusion were classified as intraprocedural, pre-discharge (i.e., during hospital stay), and post-discharge (3,14).
Follow-up data on clinical status and medication were prospectively collected in all patients at 30 days with additional review of the patients' case notes to enable all events to be confirmed. All serious adverse events were adjudicated independently by 2 interventional cardiologists and 1 cardiac surgeon. All patients provided written informed consent.
Continuous variables are presented as mean ± SD or as median (interquartile range [IQR]), and were compared using the Student unpaired t or Mann-Whitney rank-sum tests, on the basis of appropriate testing for a normal distribution. The normality assumption for continuous variables was evaluated by the 1-sample Kolmogorov-Smirnov test. Categorical variables are presented as frequencies and percents and were compared using the chi-square test when appropriate (expected frequency >5); otherwise the Fisher exact test was used. Univariate analysis was performed for all variables listed in Table 1. To determine independent predictors of 30-day major bleeding, variables with probability value ≤0.20 in the univariate analysis were entered in the multivariate model. To decrease the effect of selection bias and potential confounders on major bleeding outcomes, the variables such as the operator's decision in choosing the heparin administration strategy and the learning effect (the cohort was divided into tertiles according to procedural date to adjust for experience in TAVI accrued over time), were included in the logistic regression analysis. Moreover, to account for the confounding effect between ACT versus non–ACT-guidance groups and major bleeding, a propensity analysis, using a nonparsimonious logistic regression model including all variables listed in Table 1, was constructed. The score was incorporated in the model; its calibration and discrimination were assessed by the goodness of fit with the Hosmer-Lemeshow and C-statistics, respectively. Results are presented as odds ratio (OR) with 95% confidence intervals (CIs). For all analyses, a 2-sided p < 0.05 indicated statistical significance.
Statistical analysis was performed using the Statistical Package for Social Sciences (version 20.0, SPSS, Inc., Chicago, Illinois).
The patient flow chart is illustrated in Figure 2. A total of 362 patients underwent TF-TAVI during the study period. Baseline clinical and procedural characteristics of the study group are summarized in Table 1. Overall, baseline ACT values were 159.1 ± 28.3 s. Of note, baseline ACT ≥175 s was present in the 30.4% of the patients.
Thirty-day VARC outcomes in the overall study group
Bleeding occurred in 167 (46.1%) patients; of these, 76 (21.0%) had major bleeding. Fifty-nine (16.3%) and 32 (8.8%) patients had life-threatening and minor bleeding, respectively. Interestingly, in 15.0%, the source of bleeding was not obvious. Among 142 patients with obvious bleeding, procedure-related bleeding occurred in 82.4% of cases. Of these, 86.4% were secondary to major/minor vascular complications and access site bleeding/hematoma not defined as major/minor vascular complications. Furthermore, a significant proportion of patients with an obvious source of bleeding (17.6%) developed non-procedure-related hemorrhagic complications. In-hospital bleeding and their temporal relationships with the procedure are shown in Figure 3. All bleeding events occurred during the index hospital stay, mostly in the first 24 h, with no registered cases after discharge.
RBC transfusion was required in 140 (38.7%) patients. The median number of transfused RBC packs was 2 (IQR: 2 to 4). The reasons for in-hospital RBC transfusion are listed in Table 2. In the transfused patients, post-procedure hemoglobin level was 8.4 ± 0.93 g/dl. With regard to timing, 31 (22.1%) patients were transfused during the procedure and 109 (77.9%) patients during the index hospital stay. Eleven patients required further RBC transfusion after discharge in the absence of an obvious bleeding source.
The other 30-day VARC outcomes are illustrated in Table 3. Of note, patients who developed any bleeding had a significantly higher occurrence of 30-day all-cause mortality (7.2% vs. 2.2%, p = 0.04) as well as a trend toward higher cardiovascular mortality (6.6% vs. 2.2%, p = 0.05). Moreover, in-hospital stay was longer in the bleeding patient cohort (7 [IQR: 5.75, 11.0] days vs. 6 [IQR: 5.0, 9.0] days; p < 0.001).
The 30-day VARC outcomes according to heparin administration strategy: ACT vs. non–ACT-guided group
Baseline ACT values were similar in the 2 groups (158.8 ± 29.3 s in the ACT-guided vs. 158.9 ± 24.1 in the non–ACT-guided group; p = 0.89).
The choice of heparin administration strategy was fairly balanced among the 4 operators involved, with 2 operators more likely to choose the ACT-guided strategy and the other 2 inclined to choose the non–ACT-guided strategy. Notably, in the ACT-guided group, a lower dose of total heparin (4,000 international units [IU] [IQR: 3,375 to 5,000 IU] vs. 6,000 IU [IQR: 5,000 to 6,000 IU]; p < 0.001) was administered. In addition, in this group a lower weight-adjusted heparin dose (IU/kg) (58.8 [IQR: 44.2 to 74.7] vs. 81.7 [IQR: 71.4 to 88.1]; p < 0.001) was given. The mean intraprocedure post-heparin ACT in the ACT-guided group was consequently lower than in the non–ACT-guided group (238.4 ± 58.9 s vs. 267.7 ± 51.6 s; p < 0.001). Protamine was administered at neutralizing dose in 38 patients (20.2%) of the non–ACT-guided group and in 21 patients (12.1%) of the ACT group (p = 0.80).
Notably, in the ACT-guided group a significantly lower occurrence of major bleeding (7.5% vs. 33.5%, p < 0.001), as well as life-threatening (12.1% vs. 20.2%, p = 0.04) and any bleeding (25.9% vs. 64.9%, p < 0.001) was observed. Figure 4 reports bleeding and transfusion need in the ACT-guided versus non–ACT-guided groups.
Causes of overt bleeding in the 2 groups are illustrated in Figure 5. Interestingly, the occurrence of access site bleeding/hematoma not defined as major/minor vascular complications was significantly lower in the ACT-guided group than in the non–ACT-guided group (11.1% vs. 27.4%, p = 0.03). Moreover, there was a trend toward a lower incidence of non-procedure-related bleeding in the ACT-guided group (13.9% vs. 17.0%, p = 0.08). During in-hospital stay, no adjunctive anticoagulation therapies were required except in those who developed new-onset atrial fibrillation. However, administration of anticoagulation therapy, which might partially account for a higher rate of post-procedure bleeding, was similar in the 2 groups who developed atrial fibrillation (12.8% in ACT-guided vs. 11.5% in non–ACT-guided group, p = 0.49). Moreover, among 61 patients with an indication for OAT, 27.9% were discharged on OAT alone, 39.3% on OAT in combination with aspirin, and 32.8% on OAT plus clopidogrel, with a similar proportion in the 2 groups.
However, other secondary outcomes did not significantly differ between the 2 groups (Table 3). Notably, despite the fact that a lower heparin dose was administered in the ACT-guided group, the rate of thrombotic complications was similar in the 2 groups.
Learning curve effect
A potential learning curve effect across cases performed in the first tertile of the study period and those performed in the second and third tertiles was noted for whole bleeding at 30 days. Time distribution of bleeding is illustrated in Figure 6. Over the time course of our TAVI experience, bleeding was significantly reduced because of smaller sheath size and the learning curve effect.
A separate analysis was performed, excluding the cohort of the first 59 patients who had received heparin exclusively on the basis of body weight, leading to a final study group of 174 patients in the ACT-guided and 129 patients in non–ACT-guided groups. As expected, the occurrence of life-threatening bleeding was not statistically different between the 2 groups (12.1% vs. 14.7%, p = 0.49). However, there remained a significant difference in major bleeding (7.5% vs. 33.3%, p < 0.0001), as well as any bleeding (25.9% vs. 59.7%, p < 0.0001) and a trend toward a reduction in minor bleeding (6.3% vs. 11.6%, p = 0.07) in favor of the ACT-guided group.
Predictors of major bleeding
Table 4 reports the variables significantly associated with 30-day major bleeding in univariate and multivariate analyses. Operator preference in heparin administration strategy was not found to have a significant effect in our series. Moreover, no correlation was observed between major bleeding and sheath size nor between percutaneous versus surgical access closure and device implanted. After adjustment for baseline confounders, the absence of ACT guidance (OR: 5.89, 95% CI: 2.76 to 12.60; p < 0.001) and baseline estimated glomerular filtration rate (OR: 0.98, 95% CI: 0.97 to 1.00; p = 0.04) were the only predictors of 30-day major bleeding. Importantly, irrespective of the time the procedure was performed, the ACT-guided strategy was associated with a significantly lower occurrence of major bleeding. Moreover, after multivariate adjustment for potential confounders ACT-guided propensity, absence of ACT guidance was identified as an independent predictor of major bleeding at 30 days (OR: 6.4, 95% CI: 2.3 to 17.9; p < 0.001). The C-statistic of 0.73 from the model and the p value of 0.47 from the Hosmer-Lemeshow goodness-of-fit test indicate a good model fit.
The present investigation provides 2 main novel findings. 1) Baseline ACT levels were generally elevated in patients with severe AS. 2) Compared with the non–ACT-guided strategy, baseline ACT-guided heparin administration during TAVI significantly reduced the occurrence of major bleeding, as well as any bleeding. In addition, ACT guidance was found to be an independent predictor of major bleeding events.
TAVI-related bleeding complications: incidence, prognostic value, and potential mechanisms
The occurrence of bleeding and the need for transfusion are still common and remain an important issue in patients undergoing TAVI with an increased risk of subsequent adverse outcomes. Rates of major and life-threatening bleeding according to VARC definitions in the periprocedural period have been as high as 15% to 32% and 5% to 16%, respectively (15). A meta-analysis showed that any and major bleeding occurred at a pooled estimate rate of 41.4% (95% CI: 35.5% to 47.6%) and 22.3% (95% CI: 17.8% to 28.3%), respectively (4). This is in concordance with our experience, where we observed major bleeding in 76 (21.0%) patients. In addition, 59 (16.3%) patients experienced life-threatening bleeding, and 140 (38.7%) patients required RBC transfusion during the index hospital stay. Bleeding has a strong impact on prognosis in this study group. Indeed, the published data has suggested the harmful impact of severe bleeding and the need of RBC transfusion on short- and long-term survival (2,4,5,15). Similarly, in our experience, the occurrence of any bleeding was associated with a significantly higher 30-day overall mortality and a trend toward higher cardiovascular mortality, as well as a longer duration of hospital stay.
Several factors contribute to bleeding complications during TAVI. The majority of bleeding during TAVI is procedure-related. Indeed, the main cause is represented by vascular/access site complications. The use of large delivery catheters has been associated with a high rate of access site complications, which may in turn be associated with significant, clinically overt bleeding. In addition, the presence of diseased ilio-femoral axis in this old study group may account for the increased vulnerability to vascular complications (15).
Another important factor contributing to vascular and consequently to bleeding complications is represented by the learning curve process. Decreasing the size of TAVI delivery catheters has already been associated with reduced vascular access site complications. In addition, increased operator experience in the use of percutaneous closure devices and the management of vascular complications have reduced the rate and impact of vascular complications during TAVI (16). In accordance with previous reports, there was also a learning curve effect in our experience. Bleeding mainly related to vascular complications has dramatically reduced over time because of a smaller sheath profile and an improved operator experience.
Abnormalities of the coagulation pathways and risk for bleeding in severe AS
Both aging and severe AS are associated with an altered hemostatic state, leading to a tendency toward increased bleeding. Indeed, the frail, high-risk nature of these patients may increase the vulnerability to develop bleeding during and after TAVI. Moreover, abnormalities of coagulation pathways have been described in patients with severe AS with a higher incidence of spontaneous bleeding (6). The acquired type 2A von Willebrand syndrome has been advocated as the cause of higher occurrence of spontaneous bleeding in these patients.
Notably, the baseline ACT value in our TAVI study group was 159.1 ± 28.3 s. This value was higher than the ACT value (129 ± 26 s) of a historical group of patients treated with percutaneous coronary intervention in our center (17). Moreover, the baseline ACT in a cohort of patients undergoing percutaneous coronary intervention for stable coronary artery disease during the same study period at our institution and who were pretreated with dual antiplatelet therapy was 136.1 ± 23.4 s.
Apart from the possible differences in the intrinsic coagulation pathway and platelet function in patients with AS, we could not find any other reasons to explain why these patients have a higher baseline ACT.
Furthermore, a significant proportion of patients undergoing TAVI developed bleeding not directly related to the procedure or received blood transfusions after the procedure despite having no obvious source of hemorrhage. Généreux et al. (4) also reported that among the 25% of patients who needed RBC after TAVI, 57% of the transfusions given were not directly related to the procedure (gastrointestinal/genitourinary bleeding, or no obvious source). Similarly, despite the fact that the majority of bleeding with an obvious source in our series was procedure-related and was driven by access site/related bleeding, accounting for 86.4% of the cases, about one-fifth of our study group experienced an obvious spontaneous bleeding not directly procedure-related. Gastrointestinal, genitourinary, and soft tissue bleeding were responsible for the majority of these non-procedure-related bleeding events during the hospital stay. Moreover, procedure-related blood loss requiring RBC transfusion was observed in only 88 (24.3%) patients.
Heparin administration according to baseline ACT-guided strategy to reduce bleeding in TAVI patients
Another important factor that may account for an increased vulnerability to bleeding in this study group is excessive anticoagulation therapy with heparin. TAVI requires an optimal adjunctive periprocedural antithrombotic therapy. Vessel injury itself, the introduction of catheters and wires during endovascular procedures set off the coagulation cascade. Furthermore, during TAVI, procedural embolic events may occur during balloon valvuloplasty, manipulation of the catheters across the aortic valve, and valve implantation (15). Therefore, during TAVI, the thrombotic risk needs to be carefully balanced with the bleeding hazard.
Administration of heparin according to body weight and its monitoring by intraprocedural arterial ACT is the method mostly used in percutaneous coronary intervention, and this methodology has been adopted in all endovascular procedures. Currently, heparin is the recommended anticoagulation therapy during TAVI with a targeted ACT ranging between 200 and 300 s throughout the procedure (7,8). The cost and availability of a rapid “point of care” test for dose adjustment as well as the presence of an antagonist makes heparin a valid therapeutic option for this procedure.
The concept of an overdose of heparin in this study group as the potential cause of bleeding has been recently perceived by many TAVI centers. Although in the PARTNER (Placement of Aortic Transcatheter Valves) trial, the intraprocedural anticoagulation regimen was a UFH 5,000-IU bolus weight-adjusted-based dose to keep ACT ≥250 s, most TAVI centers moved toward a lower heparin dosing ranging between 50 and 100 IU/kg (18). Similarly, in our non–ACT-guided group, the median dose of UFH was 81.7 IU/kg.
Nonetheless, a simple weight-adjusted UFH administration does not take into account factors that may affect the individual patient response to heparin, risking potential overdosing in these particularly frail patients. The elderly TAVI study group is generally more vulnerable to the adverse effect of antithrombotic drugs, particularly bleeding. Factors that may affect therapeutic agents (e.g., renal function, hepatic metabolism, body mass distribution) as well as factors more specific to hemostasis (e.g., platelet dysfunction, coagulation disorders) may partially account for such higher vulnerability to the side effects of antithrombotic medications. Furthermore, the TAVI study group often presents with impaired renal and liver function due to the high prevalence of concomitant heart failure and is therefore at risk of an exaggerated heparin response and prolonged dose elimination.
Minimizing the bleeding risk correlated to a higher heparin dose might be of utmost clinical relevance in this study group. In the present study, we evaluated the added benefit of baseline ACT-guided heparin administration compared with the standardized weight-adjusted antithrombotic treatment during TF-TAVI on bleeding events and outcomes. Importantly, despite the potential learning curve effect, the baseline ACT-guided strategy allows a more accurate and tailored heparin administration with a significant reduction in UFH dosing. This leads to a significant reduction in 30-day major and any bleeding during TF-TAVI. Our results strongly suggest the existence of a possible relationship between bleeding and total/pro kilogram heparin dose as well as intraprocedural ACT values during TAVI. In addition, we observed a temporal relationship between the heparin administration and bleeding. In fact, the majority (88.7%) of bleeding occurred within the first 24 h after the TAVI procedure.
Notably, other important findings observed were: 1) a significantly lower occurrence (11.1% vs. 27.4%, p = 0.03) of hematoma/access site bleeding not defined as major/minor vascular complications in the ACT-guided group; 2) a trend toward lower incidence (13.9% vs. 17.0%, p = 0.08) of non-procedure-related bleeding in the ACT-guided group; and 3) the absence of an increased rate of thrombotic events in the ACT-guided group, despite the lower dose of UFH administered.
Moreover, ACT guidance was shown to be a protective factor against major bleeding that was independent from the learning curve effect and operator's preference of heparin administration strategy.
Alternative anticoagulant therapies
Currently, alternative anticoagulation regimens are under investigation in the setting of TAVI. The role of bivalirudin versus heparin in TF-TAVI will be assessed in the randomized BRAVO 2/3 (Effect of BivaliRudin on Aortic Valve Intervention Outcomes 2/3) study. Primary endpoints will be 48-h/hospital discharge major bleeding, and net adverse clinical events including 30-day major bleeding, death, myocardial infarction, and stroke. This randomized study will provide further insight into the optimal anticoagulation regimen during TF-TAVI.
The most important limitation of our study is the nonrandomized comparison of 2 treatment strategies, even if propensity regression adjustment was used to overcome the potential confounders. Another caveat is that the ACT-guided heparin administration was given according to the operator's discretion, and thus it was not administered in a standardized fashion. Furthermore, age, comorbidities, platelet function, and dual antiplatelet therapy may have interfered with baseline ACT. Moreover, there was no uniformity on protamine administration or documentation on timing of arterial sheaths removal. Therefore, we cannot exclude that potential selection bias and unmeasured and residual confounders may have affected the results. Finally, the low rate of stroke seen in our group may have allowed lower procedural ACT levels. Despite these considerations, this simple strategy on the basis of baseline ACT collection to guide UHF dosing may benefit the reduction in bleeding rate during TF-TAVI.
Bleeding remains a frequent complication after TAVI and is associated with poor outcomes. In our study, baseline ACT-guided strategy for heparin administration in TF-TAVI was correlated with lower 30-day major and any bleeding. Our findings suggest that baseline ACT might be an effective tool in guiding heparin administration during TF-TAVI. Further randomized studies are warranted to determine whether heparin should be given according to baseline ACT guidance and to confirm the usefulness of this strategy in reducing the bleeding in TAVI study group.
Dr. Montorfano is a proctor for Edwards Lifesciences. Dr. Maisano has received consulting fees from Medtronic, Abbott Vascular, St. Jude Medical, and Valtech Cardio; has received royalties from Edwards Lifesciences; and is a cofounder of 4tech. Dr. Latib serves on the Medtronic Advisory Board; and has received consulting fees from Direct Flow Medical. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- activated clotting time
- aortic stenosis
- confidence interval
- interquartile range
- international units
- oral anticoagulation therapy
- odds ratio
- red blood cell
- transcatheter aortic valve implantation
- unfractionated heparin
- Valve Academic Research Consortium
- Received July 16, 2013.
- Revision received October 20, 2013.
- Accepted October 24, 2013.
- 2014 American College of Cardiology Foundation
- Leon M.B.,
- Piazza N.,
- Nikolsky E.,
- et al.
- Généreux P.,
- Head S.J.,
- Van Mieghem N.M.,
- et al.
- Tchetche D.,
- Van der Boon R.M.,
- Dumonteil N.,
- et al.
- Holmes D.R. Jr..,
- Mack M.J.,
- Kaul S.,
- et al.
- Davis E.M.,
- Friedman S.K.,
- Baker T.M.
- Godino C.,
- Maisano F.,
- Montorfano M.,
- et al.
- Kappetein A.P.,
- Head S.J.,
- Généreux P.,
- et al.
- Rodés-Cabau J.,
- Dauerman H.L.,
- Cohen M.G.,
- et al.
- Ruiz C.E.