Safety and Complications Associated With the Use of Protamine in Percutaneous Coronary Intervention

© 2025 HMP Global. All Rights Reserved.
Any views and opinions expressed are those of the author(s) and/or participants and do not necessarily reflect the views, policy, or position of the Journal of Invasive Cardiology or HMP Global, their employees, and affiliates. 

J INVASIVE CARDIOL 2025. doi:10.25270/jic/24.00336. Epub January 29, 2025.

Objectives. There is a paucity of data on the use of protamine after percutaneous coronary intervention (PCI). The purpose of this study was to assess the incidence of thrombotic complications of protamine after high-risk PCI.

Methods. The authors conducted a retrospective analysis of 168 patients. All patients received protamine intra- or immediately post-index PCI. Baseline characteristics and procedural characteristics including heparin dosing, protamine dosing, and bleeding and thrombotic complications were evaluated. The primary outcome was the incidence of acute stent thrombosis (ST), subacute ST, and ‘other’ thrombotic complications. Secondary outcomes included mortality within 24 hours and within 30 days of the index procedure.

Results. A total of 168 patients were included. The majority of patients received dual anti-platelet therapy prior to the index procedure (85%). The average procedure time was 202 ± 103 minutes, and an average of 2.59 (± 1.38) stents were deployed. An average protamine dose of 32mg was administered, and the median dose was 20mg (IQR 20). Seventy-three (43%) had a coronary perforation and five (3%) had access site related bleeding requiring transfusion. Four (2%) patients had acute ST, no patients experienced subacute ST, and 2 (1%) patients developed non-coronary arterial thrombosis. Eight (5%) died within 24 hours of their PCI and 14 (8%) patients died within 30 days after PCI.

Conclusions. In our cohort, administration of protamine was well tolerated in the majority of patients, however, 3.6% of patients did experience coronary or peripheral arterial thrombosis warranting caution when using protamine in these challenging scenarios.

The number of complex coronary interventions has steadily increased over the past decade, and inherently carries higher rates of death, acute myocardial infarction (MI), and major adverse cardiovascular events (MACE) than non-complex coronary interventions.¹ Several variables have been implicated in contributing to increased procedural risk in percutaneous coronary intervention (PCI) and can be compartmentalized into patient- and lesion-related variables. The latter includes bifurcation disease, left main stenosis, saphenous vein graft disease, severe calcification, ostial lesions, and chronic total occlusions (CTO).^2-6 Patient-related variables include advanced age, impaired left ventricular function, and prior cardiovascular disease (such as prior MI and peripheral arterial disease), along with diabetes mellitus, chronic lung disease, and chronic kidney disease.^7-14

Included in these procedural risks are bleeding complications, which have been shown to be the most common adverse periprocedural event in PCI in the United States.^15,16 The prognostic profile of these complications is broad, with the inpatient incidence of major bleeding correlating significantly with stent thrombosis and 30-day mortality.¹⁷ Closure devices have demonstrated variable reduction in bleeding complications compared with manual compression, but quicker hemostasis is observed when closure devices are used.^18,19 Irrespective of the approach for closure, the anticoagulant state required for a PCI to avoid thrombotic and embolic sequelae presents a challenge to achieving the required quality of closure. 

A prolonged activated clotting time (ACT) during PCI is correlated with increased bleeding.^20,21 Protamine-sulfate, a neutralizer of the effects of heparin, reduces ACT levels and thus has been used to minimize bleeding events. Barriers to its use include anaphylactic reactions, paradoxically promoting fibrinolysis in hemostatic sites, acute pulmonary hypertension, and acute in-stent thrombosis.^22,23 In the last 2 decades, several studies have been published on the outcomes of protamine use post-PCI.^18,24-27 These studies have demonstrated a reduced risk of bleeding complications without any significant difference in mortality or in-stent thrombosis after protamine administration.  In this study, we sought to add to the current literature regarding the use of protamine after PCI using a large single-center retrospective data set.

Design and population

We performed a retrospective analysis of patients from a single-center healthcare system between who underwent PCI 2015 and 2021 and received protamine due to various complications. We identified 168 patients who received protamine either intra- or post-procedurally. The study and waiver of informed consent were approved by the Henry Ford Health Institutional Review Board.

Data collection, study definitions, outcomes, and analysis

All data were selected based on the above inclusion criteria and gathered through chart review and manual abstraction of the electronic health record. Baseline demographics and underlying comorbidities were collected by researchers, including age, gender, body mass index, ejection fraction (EF), history of smoking, renal disease, diabetes mellitus and type of insulin used, atrial fibrillation, prior use of antiplatelet agents, prior MI, prior PCI, prior coronary artery bypass graft (CABG), and pre- and post-procedural laboratory workup. Detailed intraprocedural characteristics were analyzed, including indications of PCI, duration of procedure, access site, targeted coronary vessels, types and sizes of stents deployed, dosing of heparin, peri-procedural ACT, dosing and indication of protamine administration, use of mechanical circulatory support (MCS), use of intravascular ultrasound (IVUS), type of atherectomy (if performed), and the use of a closure device.

The following procedural complications were evaluated: coronary perforation, coronary dissection, pericardial effusion/tamponade requiring pericardiocentesis, access site complications, urgent cardiac surgery, acute or subacute stent thrombosis (ST), and mortality. Acute ST was defined as thrombosis that occurred within 24 hours of PCI, while subacute ST was defined as thrombosis that occurred within 30 days post-PCI. The primary outcome was the incidence of acute ST, subacute ST, and other thrombotic complications. Secondary outcomes included mortality within 24 hours and 30 days of the index procedure. All study variables are presented descriptively, and no statistical comparisons were conducted. Categorical variables are reported using frequency and proportion, and continuous variables are reported using mean, median, and SD.

Of the study population, the mean age of patients was 72 ± 12.1 years, and 61 (36%) were female. Seventy-seven (46%) patients had prior PCI, 55 (33%) underwent prior CABG, 144 (85%) were on dual antiplatelet therapy (DAPT) prior to the procedure, and 1 (0.5%) patient used neutral protamine Hagedorn (NPH) insulin for diabetes mellitus. The mean EF was 50 ± 14.3%, the mean international normalized ratio (INR) was 1.79, and the mean platelet counts were 202 ± 72.4 ×10⁹/L. Baseline demographics and clinical characteristics are listed in Table 1.

The majority of PCIs were performed electively, with 53 (32%) performed for acute coronary syndrome (ACS). The average procedure time was 202 ± 103 minutes; femoral artery access was used in 129 (77%) of cases, and 113 (67%) underwent use of a closure device. The majority of interventions were performed on a single coronary vessel in 113 (67%) patients, with 47 (28%) interventions on 2 vessels and 8 (4%) on 3 vessels. An average of 2.59 ± 1.38 stents were deployed. IVUS was used in 96 (57%) patients. Fifty-nine (35.1%) patients received atherectomy, either rotational, orbital, or laser (27, 23, and 9 patients, respectively), and 40 (23.8%) patients had an MCS device placed prior to or during PCI. The mean total heparin dose administered interprocedurally was 12 269 ± 5773 units and the average protamine dose was 32 mg with a median dose of 20 mg (IQR 20). Procedural and angiographic characteristics are listed in Table 2. Indications for the use of protamine are shared in the Figure.

In our cohort, 73 (43%) patients experienced a coronary perforation, 19 (11%) of whom had a pericardial effusion requiring pericardiocentesis, and 3 (1.8%) underwent urgent cardiac surgery. In total, 31 (18.4%) patients had access site complications; of these, 13 sustained a hematoma and 16 had bleeding, 5 of whom required blood transfusions.

The primary outcome occurred in 6 (3.6%) patients. Acute ST was noted in 4 patients (2%); no patients experienced subacute ST, and 2 (1.2%) patients developed arterial thrombosis. All 4 patients who sustained acute ST were on DAPT prior to their procedure. Large doses of heparin were administered in all 4 cases, varying between 16 000 and 34 000 units cumulatively throughout the procedure, and none had access site complications. The protamine dose administered at the conclusion of the procedure varied between 20 mg to 50 mg. Of the 4 patients, 3 had definite stent thrombosis, and 1 had probable stent thrombosis.

The first patient presented with ACS and underwent PCI to the proximal left anterior descending artery (LAD). The patient developed a coronary perforation, which was managed conservatively, but subsequently developed acute ST, which was thought to be due to undersized stenting. The ACT prior to the administration of 25 mg of protamine was 282. The second patient underwent an elective LAD-CTO PCI using retrograde collaterals via the left circumflex artery (LCx) and developed an intramural hematoma to the proximal LAD with stenting of the LCx to treat a retrograde dissection; the final ACT was 344 prior to receiving 50 mg of protamine. The third patient presented with ACS and, during PCI to the LCx, developed a perforation to the distal obtuse marginal; this was treated with coiling, with a final ACT of greater than 400 prior to receiving 20 mg of protamine. Several hours later, the patient developed cardiac arrest and thrombosis within the LCX. The final patient developed a mid-LAD perforation, which was treated with a covered stent, then developed post-procedural cardiac arrest without a coronary angiogram, which was presumably due to stent thrombosis. Protamine was administered, although there was no documented ACT prior to administration. In all 4 patients, the original complication prompted the administration of protamine. Two of these patients died.

Of the 2 patients who developed arterial thrombosis, the first patient underwent Impella (Abiomed)-assisted PCI via the transcaval technique. The device was removed post-procedurally and protamine was administered, but within 24 hours the patient was found to have acute occlusions of the superior mesenteric, hepatic, and splenic arteries. The second patient also underwent Impella-assisted PCI, and, upon completion of the procedure, had failure of the closure devices, which required the use of manual pressure for hemostasis and administration of protamine. This later led to the development of thrombosis in the ipsilateral femoral artery.

The secondary outcome occurred in 14 (8%) patients. The clinical outcome and post procedural complications are listed in Table 3.

The present retrospective study included 168 patients who underwent PCI followed by protamine administration and evaluated the incidence of acute and subacute ST. In clinical practice, anticoagulation reversal using protamine post-PCI remains controversial.²⁸ In our study, the primary outcome occurred in 6 patients. Specifically, acute ST occurred in 4 patients, and there were no patients who sustained subacute ST. These findings suggest that protamine is generally tolerated but warrants caution in its usage in high-risk PCI.

Our patient cohort represented high-risk coronary artery disease (CAD) requiring complex interventions as illustrated by lengthy procedural times, large cumulative heparin doses, use of various atherectomy devices, high rates of coronary perforation, and periods of hemodynamic instability evidenced by the need for MCS devices. These interventions further included left main and bifurcation stenting along with CTO revascularization, all of which are known to increase procedural risk. Furthermore, our cohort commonly had multiple co-morbidities, including advanced age, multi-vessel disease, and chronic kidney disease, all of which are also established risk factors for significant complications post-PCI.²⁹ Our patient demographics, combined with the complexity of CAD requiring specialized tools and techniques, were likely a large contributor to the frequency of complications, including coronary perforations and bleeding, prompting the use of protamine. The most common reason for protamine administration post-PCI was coronary perforation, followed by access site complications, such as hematoma or bleeding.

Of the 4 patients who developed IST, 2 of the cases involved coronary perforations with additional therapies including placement of a covered stent or coil occlusion, either of which may have contributed to stent thrombosis. The remaining stent thrombosis events occurred in patients with either an intramural hematoma or undersized stent, both of which may have resulted in stent thrombosis, independent of the use of protamine. Therefore, the direct role of protamine in the occurrence of the primary endpoint is unclear. Regarding the 2 patients who developed arterial thrombosis, either the use of transcaval access or the implementation of prolonged manual pressure may have contributed to the development of arterial thrombosis, but it was likely exacerbated by the concomitant use of protamine.

Although protamine use was well tolerated in most patients, the rate of IST in our study was 2.4%, which is considerably higher than the established rate of 0.7%.³⁰ Furthermore, a large recent meta-analysis showed no association between stent thrombosis and the use of protamine.³¹ However, the discrepancy between our rates of IST vs those previously reported is likely due to the high-risk anatomy in our cohort, resulting in an increased incidence of complications. These complications likely precipitated the thrombotic events seen in our cohort, which may have been exacerbated by the use of protamine. Thus, the direct role of protamine with stent thrombosis in our cohort is unclear.  

Limitations

Our study has several limitations. First, this is a retrospective single center study and therefore may be prone to patient selection bias. Second, our study had a relatively small sample size and limited follow-up period. Third, this was a single-arm study and had no comparison arm evaluating the risk of ST without the use of protamine.

In our cohort, patients received protamine mostly because of PCI-related complications. Of the 168 patients in our study, the incidence of acute ST within 24 hours was 2% and peripheral arterial thrombosis was noted in 1% of cases.

Hussayn Alrayes, DO¹; Ayman Alsaadi, MD²; Ahmad Alkhatib, MD³; Dhruvil Ashishkumar Patel, MD⁴; Mohammad Alqarqaz, MD¹; Tiberio Frisoli, MD¹; Brittany Fuller, MD¹; Akshay Khandelwal, MD⁵; Gerald Koenig, MD, PhD^1,6; Brian P. O’Neill, MD¹; Pedro Villablanca, MD¹; Mohammad Zaidan, MD⁷; William O’Neill, MD¹; Khaldoon Alaswad, MD¹; Mir Basir, DO¹

From the ¹Department of Cardiovascular Medicine, Henry Ford Health Systems, Detroit, Michigan; ²Department of Internal Medicine, Henry Ford Health System, Detroit, Michigan; ³Department of Internal Medicine, University of Kansas Medical Center, Kansas City, Kansas;  ⁴Department of Internal Medicine, University of Chicago Medical Center, Chicago, Illinois; ⁵Department of Cardiovascular Medicine, Allegheny Health Network, Pittsburgh, Pennsylvania; ⁶Wayne State University School of Medicine, Detroit, Michigan; ⁷Department of Cardiovascular Medicine, American Hospital Dubai, Dubai, United Arab Emirates.

The abstract was accepted by TCT in 2022.

Disclosures: Dr Frisoli is a proctor for Edwards Lifesciences, Abbott, Boston Scientific, and Medtronic. Dr B.P. O’Neill is a consultant to and receives research support from Edwards Lifesciences. Dr Villa Blanca is a consultant for Edwards Lifesciences, Medtronic, Shockwave, Abiomed, and Angiodynamics. Dr Basir is a consultant for Abiomed, Boston Scientific, Chiesi, Saranas, and Zoll. The remaining authors report no financial relationships or conflicts of interest regarding the content herein.

Address for correspondence: Hussayn Alrayes, DO, 2799 W Grand Blvd, Detroit, MI, 48202, USA. Email: halraye1@hfhs.org