Key Points

QuestionÌý Does routine protamine administration at the conclusion of transfemoral transcatheter aortic valve implantation (TAVI) enhance successful hemostasis?

FindingsÌý In the ACE-PROTAVI randomized clinical trial, which included 410 patients at 3 Australian hospitals, patients receiving routine protamine after TAVI had a higher rate of successful hemostasis and a lower risk of vascular complications vs placebo.

MeaningÌý Routine use of protamine improved hemostasis after TAVI.

ImportanceÌý Vascular complications after transfemoral transcatheter aortic valve implantation (TAVI) remain an important cause of procedure-related morbidity. Routine reversal of anticoagulation with protamine at the conclusion of transfemoral TAVI could reduce complications, but data remain scarce.

ObjectiveÌý To evaluate the efficacy and safety of routine protamine administration after transfemoral TAVI.

Design, Setting, and ParticipantsÌý The ACE-PROTAVI trial was an investigator-initiated, double-blind, placebo-controlled randomized clinical trial performed at 3 Australian hospitals between December 2021 and June 2023 with a 1-year follow-up period. All patients accepted for transfemoral TAVI by a multidisciplinary heart team were eligible for enrollment.

InterventionsÌý Eligible patients were randomized 1:1 between routine protamine administration and placebo.

Main Outcomes and MeasuresÌý The coprimary outcomes were the rate of hemostasis success and time to hemostasis (TTH), presented as categorical variables and compared with a χ² test or as continuous variables as mean (SD) or median (IQR), depending on distribution. The major secondary outcome was a composite of all-cause death, major and minor bleeding complications, and major and minor vascular complications after 30 days, reported in odds ratios (ORs) with 95% CIs and P values.

ResultsÌý The study population consisted of 410 patients: 199 patients in the protamine group and 211 in the placebo group. The median (IQR) patient age in the protamine group was 82 (77-85) years, and 68 of 199 patients receiving protamine (34.2%) were female. The median (IQR) patient age in the placebo group was 80 (75-85) years, and 89 of 211 patients receiving the placebo (42.2%) were female. Patients receiving up-front protamine administration had a higher rate of hemostasis success (188 of 192 patients [97.9%]) than patients in the placebo group (186 of 203 patients [91.6%]; absolute risk difference, 6.3%; 95% CI, 2.0%-10.6%; P = .006); in addition, patients receiving up-front protamine had a shorter median (IQR) TTH (181 [120-420] seconds vs 279 [122-600] seconds; P = .002). Routine protamine administration resulted in a reduced risk of the composite outcome in the protamine group (10 of 192 [5.2%]) vs the placebo group (26 of 203 [12.8%]; OR, 0.37; 95% CI, 0.1-0.8; P = .01). This difference was predominantly driven by the difference in the prevalence of minor vascular complications. There were no adverse events associated with protamine use.

Conclusions and RelevanceÌý In the ACE-PROTAVI randomized clinical trial, routine administration of protamine increased the rate of hemostasis success and decreased TTH. The beneficial effect of protamine was reflected in a reduction in minor vascular complications, procedural time, and postprocedural hospital stay duration in patients receiving routine protamine compared with patients receiving placebo.

Trial RegistrationÌý anzctr.org.au Identifier:

Degenerative aortic valve disease is the most prevalent valvular heart disease in older individuals. If left untreated after the appearance of symptoms, the expected mortality after 2 years is almost 50%.¹ In the last 10 to 15 years, transcatheter aortic valve implantation (TAVI) has become the preferred treatment option across the entire risk spectrum in older patients with degenerated tricuspid aortic valve stenosis, due to excellent clinical results and rapid improvement in quality of life.^2-8 Notwithstanding its excellent outcomes, vascular complications related to the access site remain a major concern in transfemoral TAVI.

Major bleeding leads to an increased length of stay and is associated with increased morbidity and mortality. Preventing bleeding and selecting the appropriate treatment for vascular complications are pivotal in improving patient outcomes after transfemoral TAVI. Anticoagulation is needed during the procedure, with surgeons typically aiming for an activated clotting time (ACT) longer than 250 seconds. Heparin remains the most widely used anticoagulant worldwide during transfemoral TAVI. Heparin reversal by administration of protamine, prior to removal of the sheath and closure of the arteriotomy site, can be considered. However, surgical practice varies widely among centers and operators.

Although protamine is routinely used in cardiac surgery, data regarding its safety and efficacy in preventing bleeding complications in transfemoral TAVI are scarce. Most published data come from observational trials and the PS TAVI trial, a small, single-center randomized clinical trial, which found no difference in major or life-threatening bleeding with the administration of protamine vs placebo.⁹ On the other hand, protamine can cause allergic reactions (including anaphylaxis), hypotension, bronchospasm, and skin reactions in 1% to 10% of patients.¹⁰ Moreover, protamine may have significant anticoagulant and antiplatelet adverse effects at higher doses.¹¹ Therefore, we performed a multicenter randomized clinical trial on routine protamine administration after transfemoral TAVI to inform clinical practice.

We hypothesized that routine protamine administration at the conclusion of the transfemoral TAVI procedure, prior to femoral sheath removal, would reduce the risk of major bleeding and vascular complications without increasing the risk of ischemic events and would reduce time to hemostasis (TTH). Our study aimed to compare a routine vs selective protamine administration strategy during transfemoral TAVI.

The ACE-PROTAVI trial was an investigator-initiated, double-blind, placebo-controlled randomized clinical trial performed at 3 Australian hospitals. The trial protocol is available in Supplement 1. The trial was registered on ANZCTR.org.au () and adhered to the Consolidated Standards of Reporting Trials () guidelines. All patients provided written informed consent prior to the transfemoral TAVI procedure.

The trial protocol was approved by the Alfred Hospital ethics committee (project 182/21), and research and governance committees at each participating site. An independent data and safety monitoring board provided oversight. Adjudication of reported outcomes was performed by an independent committee that evaluated all reported clinical events in a masked manner.

All patients accepted for transfemoral TAVI by a multidisciplinary heart team were eligible for enrollment in the trial. The main exclusion criteria were documented protamine allergy or anaphylaxis, planned arterial access via surgical cutdown, or percutaneous coronary intervention (PCI) with drug-eluting stent implantation within the past 3 months.

After providing informed consent, patients were randomized by computer-generated code. The randomization outcome was numbered and sealed in opaque envelopes that concealed the treatment designation.

The anesthetic team or nursing staff prepared two 10-mL syringes: 1 with protamine (10 mg/mL) and 1 with a placebo solution of sodium chloride, 0.9%. For patients randomized to the protamine group, syringe A contained 100 mg of protamine-sulfate (10 mg/mL), and syringe B contained 10 mL of a solution of sodium chloride, 0.9%. For patients randomized to the placebo group, syringe A contained 10 mL of a solution of sodium chloride, 0.9%, and syringe B contained 100 mg of protamine-sulfate (10 mg/mL). The primary operators and the patient were masked to the actual contents of syringes A and B. Protamine dosage was set at a ratio of 1 mg of protamine to 100 IU of heparin of the total heparin dose, not exceeding 100 mg.

Prior to the removal of the large femoral sheath, the ACT was assessed. Syringe A was then injected, and sheath removal or arteriotomy closure, per the operator’s discretion, was performed. If hemostasis was achieved, the time between sheath removal and confirmed arterial hemostasis was recorded. If hemostasis was not achieved after 10 minutes, syringe B was injected per the operator’s request. This preserved the masking of primary operators but ensured that if ongoing bleeding was noted, protamine was injected to improve hemostasis, as in routine management. The need for syringe B was documented on the study registration form. A sticker on arterial access site dressing was applied, with a notification if compression on the ward was required, so delayed hemostasis failure could be documented. If clinically indicated, a femoral ultrasound was performed 3 to 48 hours afterwards to assess hematoma size and vascular complications. Time to discharge and any occurrence of secondary end points during the in-hospital stay were documented.

The coprimary end points of this study were to determine if routine protamine administration, compared with selective protamine administration (ie, placebo injection up-front in syringe A and, if required, protamine in syringe B), increased the rate of procedural hemostasis success and reduced the TTH. Hemostasis success was defined as hemostasis achieved within 20 minutes, without further (manual) compression required. TTH was defined as elapsed time after sheath removal and first observed and confirmed arterial hemostasis.

The secondary end points were the risk of the composite outcome of all-cause death, major and minor bleeding complications, and major and minor vascular complications after 30 days; the risk of each component of the secondary composite outcome; the need for transfusion for bleeding related to the access site; the hematoma size on vascular ultrasound 3 to 48 hours after the procedure, if clinically indicated; the risk of myocardial infarction; the risk of stroke; length of postprocedure stay; and the rate of delayed hemostasis failure. The outcomes were primarily defined according to the Valve Academic Research Consortium definitions.¹²

We hypothesized that routine protamine administration would reduce the risk of hemostasis failure and reduce TTH. The trial was powered for a 25% reduction in baseline TTH after sheath removal of 13 plus or minus 10 minutes. We estimated this baseline TTH on limited published data, including a randomized clinical trial¹³ comparing Perclose ProGlide and ProStar XL PVS devices (Abbott Cardiovascular) after large-bore arteriotomy. In this trial, mean (SD) TTH was 9.8 (17) minutes after closure using ProGlide and 13 (19) minutes after closure using ProStar. In addition, we used a prospective registry of the MANTA (Teleflex) device.¹⁴ Based on this estimation, a sample size of 400 enrolled patients would give the trial at least 80% power to show the superiority of routine protamine administration in reducing TTH, with a 2-sided α of .05 and a dropout rate of 15%. Baseline characteristics and procedural data were presented for all randomized patients, while outcome data were only presented for patients who did not withdraw. Reasons for withdrawal will be explicitly reported. Categorical data were presented as numbers and corresponding percentages, and compared using a χ² test or a Fisher exact test if expected cell count was less than 5. Continuous data were presented as mean (SD) or median (IQR), depending on the distribution, which was evaluated by visual inspection of histograms, P-P plots, and Kolmogorov-Smirnov tests. The primary composite clinical end point was reported in odds ratios (ORs) with 95% CIs and P values. A time-to-event analysis was also presented to appreciate the time-dependency of complications, reported in hazard ratios (HRs) and 95% CIs. For sensitivity purposes, both analyses of the composite end point were also performed in a mixed model with adjustment for center as a random variable. The rate of missing data is reported in eTable 1 in Supplement 2. For all statistical analysis, a P value less than .05 was considered statistically significant. All tests were 2-tailed. Analyses were performed in SPSS Statistics, version 29.0 (IBM).

The study population consisted of 410 patients, enrolled between December 2021 and June 2023, who were randomized in a 1:1 ratio between routine and selective protamine administration (placebo). A total of 15 randomized patients did not receive the study drug, either because they withdrew consent (n = 2), per operator discretion (n = 3), due to a planned procedure with surgical cutdown (n = 1), or due to randomization error (n = 9). Figure 1 presents the flowchart of patient inclusion and eTable 2 in Supplement 2 presents differences between patients included and not included in the final analysis. Consequently, 395 patients were analyzed for the eventual end points: 192 patients in the protamine group and 203 patients in the placebo group. Baseline characteristics for patients in the protamine and placebo groups are presented in Table 1. The median (IQR) patient age in the protamine group was 82 (77-85) years, and 68 of 199 patients receiving protamine (34.2%) were female. The median (IQR) patient age in the placebo group was 80 (75-85) years, and 89 of 211 patients receiving the placebo (42.2%) were female. There was no difference in the incidence of peripheral arterial disease between the groups (14 patients in the protamine group [7.0%] vs 14 patients in the placebo group [6.6%]).

Procedural characteristics are also shown in Table 1. Sheath size, heparin dosage, peak ACT, and ACT at sheath removal were similar between both groups. Most patients received a self-expandable valve in both groups (118 patients in the protamine group [59.3%] and 116 patients in the placebo group [55.0%]). Femoral arteriotomy was closed with suture-based devices in most of the cases (Perclose ProGlide in 201 patients [49.0%] and ProStar in 61 patients [15.2%]). Cerebral protection devices were used in only 2 patients. The median (IQR) procedural time was reduced in the protamine group (58 [49-70] minutes) compared with the placebo group (60 [52-73] minutes; P = .007).

Patients receiving routine protamine administration had a higher rate of hemostasis success (188 of 192 patients [97.9%]) than patients receiving selective protamine administration (186 of 203 patients [91.6%]; absolute risk difference, 6.3%; 95% CI, 2.0%-10.6%; P = .006). Patients receiving routine protamine administration also had a significantly reduced median (IQR) TTH (181 [120-420] seconds vs 279 [122-600] seconds; P = .002) (Figure 2; Table 2). The masked administration of syringe B in cases of persistent bleeding was also reduced in the protamine group (10 [5.2%] vs 34 [16.7%]; P < .001) (Table 2).

Further postprocedural outcomes are shown in Table 3. There were no differences in the nadir of hemoglobin levels or estimated glomerular filtration rates. Median (IQR) length of stay postprocedure was marginally shorter in the protamine group (3 [2-3] days) vs the placebo group (3 [2-4] days; P = .02). There were 5 total ischemic strokes (1 in 199 patients in the protamine group [0.5%] and 4 in 211 patients in the placebo group [1.9%]; OR, 0.26; 95% CI, 0.03-2.39; P = .24), and there were no myocardial infarctions in either group.

Routine protamine administration resulted in a reduced risk of the composite outcome: 5.2% (10 of 192 patients) in the protamine group vs 12.8% (26 of 203 patients) in the placebo group (OR, 0.37; 95% CI, 0.17-0.79; P = .01) (Table 2). Sensitivity analysis in a generalized linear mixed model with adjustment for the stratification variable (center) for the primary clinical composite end point showed an adjusted OR of 0.20 (95% CI, 0.07-0.61; P = .005). The eFigure in Supplement 2 shows the time-to-event analysis for this composite outcome (HR, 0.40; 95% CI, 0.19-0.82). There were no protamine-associated adverse events.

As shown in Table 2, this reduction in the composite end point due to routine protamine administration is mainly driven by an increased risk of minor vascular complications in the placebo group (4 of 192 patients in the protamine group [2.1%] vs 17 of 203 patients in the placebo group [8.4%]; P = .01). Other end points were reduced significantly (hematoma: 1 [0.5%] vs 7 [3.4%], respectively; P = .04) or were numerically reduced in the protamine group (arterial dissections: 0 vs 4 [2.0%]; P = .051; pseudoaneurysms: 1 [0.5%] vs 6 [3.0%]; P = .07). These types of vascular complications are shown in eTable 3 in Supplement 2.

In the ACE-PROTAVI randomized clinical trial, we investigated the effect of routine protamine administration vs selective protamine administration prior to large sheath removal and arteriotomy closure after transfemoral TAVI. We found that routine use of protamine increased the rate of hemostasis success, with shorter THH. This beneficial effect was reflected in reductions in minor vascular complications, procedural time, and postprocedural hospital stay duration in patients receiving routine protamine compared with patients receiving the placebo.

Access site–related complications remain an important and potentially avoidable cause of procedure-related morbidity associated with transfemoral TAVI. These complications are correlated with increased length of stay, reduced quality-of-life outcomes, and increased 30-day mortality. Initial vascular complication rates in early TAVI trials ranged from 10% to 20%, but current literature shows a significant decrease. The TVT registry reports a major complications rate of approximately 5%, and data from the German Aortic Valve Registry (GARY) and Netherlands Heart Registry report rates of approximately 3%.^15-17 Contemporary Australian data report major complications rates ranging from 3% to 11%.^18,19 The overall vascular complication rate in this study was favorably comparable, with only 6 of 395 total patients (1.5%) experiencing major vascular or bleeding complications and 32 patients (8.1%) experiencing minor vascular or bleeding complications.

The use of protamine and its role in the reversal of heparin is well studied and established in cardiac surgery,^11,20,21 and protamine is used in other cardiac²² and vascular interventions.^23,24 Protamine use remains much more controversial in interventional cardiology. This could be due to the much lower need for heparin reversal postprocedure, the minimally invasive approach with smaller sheath sizes, and, perhaps, fear of thrombotic complications. A large meta-analysis demonstrated that protamine was associated with fewer bleeding events after PCI,²⁵ and there is no evidence of an increased risk of stent thrombosis.²⁶

The beneficial effects of protamine on hemostasis success and reduced vascular complications are in line with earlier observational studies and trials, such as the analysis performed by Al-Kassou et al.²⁷ They demonstrated that routine administration of protamine in 873 patients significantly reduced bleeding and vascular complications (4.1%) compared with no protamine (11.8%), and that there was no increase in ischemic events, such as stroke and myocardial infarction, with administration of protamine (1.8%) compared with no protamine (3.6%). However, the study was performed in a nonrandomized, nonmatched population with a potential for selection bias, especially in patients not receiving protamine. The patients receiving protamine did so in the most recent years of the patient cohort and were compared with a historical cohort of patients not receiving protamine; hence, outcomes could also be biased by advancements made in TAVI. In another observational study supporting the use of protamine, Kneizeh et al²⁸ compared 191 patients receiving half-dose protamine with 706 patients receiving full-dose protamine. After propensity score matching (including relevant clinical characteristics such as antiplatelet therapy), low-dose protamine predicted the combined end point (OR, 3.54; 95% CI, 1.36-9.17; P = .009). In addition, even in multivariable analysis, low-dose protamine continued to be a predictor for worse combined end points in the matched model (OR, 3.07; 95% CI, 1.17-8.08; P = .02).

To our knowledge, the PS TAVI trial was the first randomized clinical trial to compare protamine and placebo in patients who underwent TAVI.⁹ The study was limited by a small sample size (N = 100) that precluded any statistically significant differences between the protamine and placebo groups. In addition, the intervention was not masked, and there was a crossover rate of 9% of patients randomized to the placebo group who received protamine. These limitations were considered in the design of the ACE-PROTAVI trial with a larger sample size and the use of 2 syringes to maintain masking and reduce crossover, while maintaining patient safety in the event of bleeding.

The findings of these studies and the results of the ACE-PROTAVI study emphasize the potential benefit of heparin reversal using protamine on completion of transfemoral TAVI. In the past 20 years, major and minor steps have been undertaken to improve the safety and efficacy of the transfemoral TAVI procedure. Some of these adjustments have been adopted by most implanters, while others remain specific to operators, institutions, or even regions or countries. The use of protamine at the time of large sheath removal is highly dependent on individual operators and even varies within the same implanter teams. We anticipate that the results of this study will provide robust evidence to guide decision-making by TAVI teams.

Although we found a reduction in the composite of death and minor and major vascular complications, along with minor and major bleeding complications, we did not see a reduction in major or life-threatening bleeding. Heparin reversal might improve closure and hemostasis and reduce minor vascular complications, but it is unlikely to affect catastrophic vascular complications, such as the rupture or perforation of arteries. However, a reduction in minor bleeding and vascular complications will lead to earlier mobilization and less nursing requirement on the ward and is associated with reduced length of stay, as demonstrated in this trial.

There are some limitations to consider in the ACE-PROTAVI trial. The adverse effects of protamine are infrequent, with the risk of an anaphylaxis reaction ranging from 0.19% to 0.69%.²⁹ With 410 patients included in this trial and 226 patients receiving protamine (192 in the protamine group and 34 in the placebo group who required the use of syringe B), the sample size may have been too small to conclude that there was no additional risk involved with protamine use. On the other hand, protamine is well documented and established in cardiac surgery, and the rare risk of severe complications does not preclude using routine protamine in that setting. Another limitation was the exclusion of patients who underwent PCI less than 3 months before the TAVI procedure. Although current evidence does not show an increased risk of stent thrombosis post-PCI, caution with routine protamine administration in patients with recent PCI is warranted. Finally, although the primary end point of hemostasis success was defined at 20 minutes, syringe B could already have been administered after 10 minutes. This approach could have biased the results toward the placebo group.

In the ACE-PROTAVI double-blind placebo-controlled randomized clinical trial, we investigated the effect of routine protamine administration vs selective protamine administration prior to large sheath removal and arteriotomy closure after transfemoral TAVI. We found that routine use of protamine increased the rate of hemostasis success, with a shorter TTH. This beneficial effect was reflected in reductions in minor vascular complications, procedural time, and postprocedural hospital stay duration in patients receiving routine protamine.

Accepted for Publication: June 21, 2024.

Published Online: August 14, 2024. doi:10.1001/jamacardio.2024.2454

Corresponding Authors: Antony S. Walton, MBBS, Heart Centre, The Alfred Hospital, 55 Commercial Rd, Melbourne, VIC 3004, Australia (a.walton@alfred.org.au); Pieter A. Vriesendorp, MD, PhD, MSc, Heart+Vascular Center, Department of Cardiology, Maastricht University Medical Center, P. Debyelaan 25, Maastricht 6229 HX, the Netherlands (pvriesendorp@gmail.com).

Author Contributions: Dr Vriesendorp and Mr Walton had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Stub and Walton are co–senior authors.

Concept and design: Vriesendorp, Nanayakkara, Htun, McGaw, Soon, Zimmet, Stub, Walton.

Acquisition, analysis, or interpretation of data: Vriesendorp, Nanayakkara, Heuts, Ball, Chandrasekhar, Dick, Haji, Htun, Noaman, Palmer, Cairo, Shulman, Lin, Hastings, Waldron, Proimos, Yudi, Zimmet, Stub, Walton.

Drafting of the manuscript: Vriesendorp, Heuts, Noaman, Palmer, Stub.

Critical review of the manuscript for important intellectual content: Nanayakkara, Ball, Chandrasekhar, Dick, Haji, Htun, McGaw, Noaman, Palmer, Cairo, Shulman, Lin, Hastings, Waldron, Soon, Proimos, Yudi, Zimmet, Stub, Walton.

Statistical analysis: Vriesendorp, Nanayakkara, Heuts, Ball, Noaman, Palmer, Stub.

Obtained funding: Vriesendorp, Dick, Stub, Walton.

Administrative, technical, or material support: Nanayakkara, Haji, Htun, McGaw, Noaman, Palmer, Shulman, Waldron, Soon, Yudi, Zimmet, Walton.

Supervision: Nanayakkara, Noaman, Palmer, Soon, Stub, Walton.

Conflict of Interest Disclosures: Dr Vriesendorp reported personal fees from Abbott Vascular and Boehringer Ingelheim outside the submitted work. Dr Chandrasekhar reported speaking fees from Boston Scientific, Medtronic, and Amgen outside the submitted work. Dr Stub reported grants from the Cabrini Foundation during the conduct of the study; grants from the Australian National Health and Medical Research Council (NHMRC) and the National Heart Foundation of Australia (NHF) outside the submitted work; and serving on medical advisory boards for Edwards Lifesciences, Medtronic, and Anteris. Dr Walton reported grants from the Fox Foundation outside the submitted work. No other disclosures were reported.

Funding/Support: This study was supported by the Epworth Research grant (Richmond, Australia), Fox Family Foundation grant (Melbourne, Australia), and Cabrini Research Foundation grant (Malvern, Australia). Dr Stub is supported by National Health and Medical Research Council and National Heart Foundation of Australia fellowships.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Meeting Presentation: This paper was presented at the 2023 meeting of Transcatheter Cardiovascular Therapeutics; October 23, 2023; San Francisco, California.

Data Sharing Statement: See Supplement 3.

Additional Contributions: We acknowledge the structural heart and research coordinators across Alfred, Cabrini, and Epworth hospitals.