Article Text


Original research
Validation of an extrinsic compression and early ambulation protocol after diagnostic transfemoral cerebral angiography: a 5-year prospective series
  1. Daniel A Tonetti1,2,
  2. Christopher Ferari3,
  3. Jennifer Perez1,
  4. Alp Ozpinar1,
  5. Ashutosh P Jadhav2,
  6. Tudor G Jovin2,
  7. Bradley A Gross1,2,
  8. Brian Thomas Jankowitz1,2
  1. 1 Department of Neurological Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
  2. 2 Stroke Institute, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA
  3. 3 West Virginia University School of Medicine, Morgantown, West Virginia, USA
  1. Correspondence to Dr Brian Thomas Jankowitz, Department of Neurological Surgery, University of Pittsburgh, Pittsburgh, PA 15213, USA; jankbt{at}


Background and purpose Access-site complications constitute a substantial portion of the morbidity associated with transfemoral cerebral angiography, yet no standardized protocol exists for femoral closure and practice patterns vary widely. The objective of this single-arm prospective cohort study was to validate the efficacy and safety of a standardized femoral closure strategy for all diagnostic angiography, regardless of antiplatelet regimen.

Methods A single-arm, prospective study was designed enrolling consecutive patients undergoing diagnostic transfemoral cerebral angiography by a single neurointerventional surgeon from March 2013 – March 2018. The closure protocol consisted of 20 minutes of manual compression to the site of arterial access and 2 hours of bedrest. The primary outcome was hematoma or oozing after manual compression. Demographic, clinic, and laboratory data were collected and analyzed, and patients were stratified by antiplatelet use.

Results Of 525 angiograms, 263 (50.1%) were on patients taking antiplatelet medication, with 66 (12.6%) on dual antiplatelet regimens. Five patients (0.95% of all patients) met the primary outcome: in all five cases, there was no further oozing or enlarging hematoma after the additional compression period. There were not significant differences in primary outcome in groups stratified by antiplatelet use, and there were no instances of delayed hematoma, pseudoaneurysm, or arteriovenous fistula.

Conclusion In this single-arm cohort study of 525 consecutive transfemoral angiograms with a standardized extrinsic compression protocol, hemostasis was achieved without complication in >99% regardless of antiplatelet strategy. This protocol is effective and safe for diagnostic transfemoral angiography regardless of a patient’s antiplatelet use.

  • standards
  • technique
  • angiography

Statistics from


Transfemoral cerebral angiography has become a cornerstone in the diagnosis and treatment of cerebrovascular disease. A substantial proportion of the morbidity associated with these procedures is related to access-site complications. Closure of the arteriotomy site is usually achieved by extrinsic manual compression or arterial closure devices. Despite the widespread use of femoral cerebral arteriography, there is no gold standard closure protocol or technique and practice varies widely.

Several large reports have been published previously1 2 demonstrating the safety of manual compression after transfemoral cerebral arteriography followed by early ambulation. None have focused on the use of antiplatelet agents, which are becoming increasingly common among neuroendovascular patients in the era of flow diversion, and which may predispose patients to a higher risk of access site complications. No standardized protocol exists for femoral closure for patients undergoing routine diagnostic angiography.

The objective of this single-arm prospective cohort study was to validate the efficacy and safety of a standardized femoral closure strategy consisting of 20 minutes of manual compression followed by ambulation and hospital discharge 2 hours after closure, particularly in patients on aspirin or dual antiplatelet regimens.


A single-arm, prospective study was designed enrolling all patients undergoing diagnostic cerebral angiography by a single neurointerventional surgeon at one of four tertiary care institutions in an academic practice from March 2013 – March 2018. The study was approved by our local Institutional Review Board. All patients were included if femoral access was intended, regardless of indication for angiography and regardless of medical history, coagulation parameters, or medications. Patients were excluded if there was a planned intervention, if a non-femoral arterial approach was anticipated, or if a sheath size larger or smaller than 5-French was planned. In all cases, common femoral artery access was obtained via fluoroscopic confirmation of femoral artery puncture site over the femoral head and a femoral arteriogram was performed for verification.

The study protocol consisted of immediate post-procedure transfer from the angiography suite to the recovery unit followed by sheath removal by a neurointerventional fellow and 20 min of manual compression to the site of arterial access. Manual compression was applied using four-finger compression focused on the arteriotomy site, approximated as one inch above the skin puncture site, and was performed by either a neurointerventional fellow or a neurointerventional technician trained in femoral compression. After 20min, the femoral access site was then inspected and, if a hematoma or oozing were noted, another 10 min of manual pressure was held. A large series of patients undergoing coronary angiography have demonstrated that 10–15 min of manual femoral pressure results in low rates of hematoma formation.3 We sought to develop a protocol that could be used for any patient regardless of antiplatelet use and therefore selected 20 min of manual compression for this protocol. Furthermore, knee immobilizers were not used for any patient in this study. Patients remained supine for a total of 2 hours after sheath removal, after which patients ambulated with the assistance of a nurse and were discharged from the hospital if no hematoma was present. Patients were accompanied by a trained nurse at all times after sheath removal, and hematoma presence or absence was recorded both after completion of manual compression and after patient ambulation. Patients were discharged with instructions to avoid running, exercising, or lifting weights for 24 hours after the procedure: after 24 hours, they were allowed to return to normal activity levels. Clinical and demographic information collected included age, gender, body mass index (BMI), side of femoral access, sheath size, whether the patient was on antiplatelet medications, and laboratory values including coagulation parameters.

Outcome measures

The primary outcome was any hematoma or oozing from the access site after 20 min of extrinsic compression. Complications related to the procedure were also noted, including any additional intervention necessary for a femoral hematoma or pseudoaneurysm. All patients had serial access site examinations documented until hospital discharge.

Patients were stratified by primary outcome to determine whether any factors could predict failure of the extrinsic compression protocol. Comparisons of stratified groups were performed using Fisher’s Exact test and Chi-square test. Comparison of means testing was used to calculate significance for averaged continuous variables including age, platelet count, and international normalized ratio (INR). Statistical assessments were considered significant when P<0.05.


Patient demographics

During the 5-year study period, 490 patients undergoing a total of 525 diagnostic cerebral angiograms via the 5-French transfemoral technique met inclusion criteria and were enrolled in this prospective study cohort. The study population consisted of transfemoral arteriography performed on 363 females and 162 males with an average age of 54.9 years. The majority (n=514, 97.9%) were performed via a right femoral approach, while 11 patients (2.1%) had their left common femoral artery accessed for the procedure. All 525 patients underwent successful transfemoral arteriography with zero aborted cases.

Regarding antiplatelet use, 181 of 525 angiograms (34%) were performed on patients currently taking aspirin 81 mg, 16 (3%) on clopidogrel 75 mg, and 66 (13%) on both aspirin and clopidogrel. The average INR among all patients was 1.00. The average platelet count among all patients was 2 48 000 platelets/microliter, and no patients had thrombocytopenia (defined as platelets<1 00 000). The average BMI among all patients was 28.7, and 204 patients had BMI>30.

Primary outcome

Of 525 angiograms performed during the study period, five patients (0.95%) met the primary outcome of palpable hematoma or oozing after 20 min of extrinsic manual compression over the femoral artery, requiring an additional 10 min of compression (table 1). In all five cases, there was no further oozing or enlarging hematoma after the additional compression period. No patients required any additional intervention (eg, thrombin injection) for hematoma after protocol completion. There were no incidences of retroperitoneal hematoma, arteriovenous fistula, pseudoaneurysm, return to the hospital, or femoral artery dissection.

Table 1

Study cohort stratified by primary putcome

In 520 other patients (99.05%), there was no evidence of hematoma formation after completion of the manual pressure protocol. All patients had palpable distal pulses at the end of the study protocol.

Stratification by antiplatelet use

Of all angiograms, 263 (50.1%) were performed on patients taking antiplatelet medications (table 2). Of these, 197 (74.9%) were on one antiplatelet agent and 66 patients (25.1%) were on dual antiplatelets. The other 262 patients (49.9%) were not taking antiplatelet medications. Patients taking antiplatelet agents were more likely to be older (58.1 vs 51.7 years, P<0.0001) and had higher rates of obesity (47.1% vs 30.2%, P=0.0001). There were three hematomas (1.1%) after the manual compression protocol for those patients undergoing angiography while actively using antiplatelet agents, compared with two hematomas (0.8%) in those not on antiplatelet agents.

Table 2

Study cohort stratified by antiplatelet agent use


Femoral artery closure techniques vary widely in the United States. Two previous prospective reports have demonstrated that manual compression after transfemoral cerebral angiography with 2 hours of supine bedrest has the potential to be safe and effective.1 2 However, major differences exist between the patient populations examined in the present study and prior reports. Zuckerman et al2 evaluated 107 patients undergoing cerebral angiography with a 2-hour supine time after 10 min of manual compression and noted a 4.7% hematoma rate. Notably, antiplatelet or anticoagulation use was not reported, and all patients in that study underwent 4-French catheterization. Wagenbach et al 1 retrospectively analyzed a group of 295 patients undergoing cerebral angiography and prospectively collected data on ambulation times. The authors reported on a very heterogenous conglomeration of patients including both inpatients and outpatients undergoing diagnostic or interventional radiology with a range of catheter sizes, with various compression times, and sporadically with the use of hemostatic sponges. The minority of patients in that study were on antiplatelet agents and subgroups are not reported. There remains no standard of care for manual compression time after transfemoral angiography or for time to mobilization.

In this 5-year prospective single-arm study, we sought to prospectively evaluate and validate the efficacy and safety of a uniform protocol of extrinsic compression for femoral artery closure followed by early ambulation and compare outcomes among all outpatient diagnostic angiograms in a modern neurointerventionalist practice with prevalent antiplatelet use. We then used those patients not prescribed antiplatelet medications as an internal control to determine whether rates of extrinsic compression failure were higher in patients taking antiplatelet agents. In a consecutive series of 525 outpatients who underwent 5-French transfemoral diagnostic angiography followed by this protocol, we report a 0.95% overall protocol failure rate.

Only five patients (0.95% of 525 total patients) failed the protocol and required additional manual compression. Three of these patients had palpable hematomas, which triggered additional compression. In the two other patients, there was ongoing oozing from the access site. It is important to note that no patient in this study, including the five which required additional manual compression, developed a delayed hematoma, pseudoaneurysm, or arteriovenous fistula, leg ischemia, or required any further intervention related to an access site complication. All patients in this study ambulated 2 hours after closure regardless of the amount of manual compression time. No patients developed a hematoma after ambulation, a finding that is consistent with prior reports regarding ambulation 2 hours after arterial closure.1

Comparing the 263 patients who underwent angiography while taking antiplatelet medications versus those 262 patients who were not taking antiplatelet agents, those on aspirin or clopidogrel were more likely to be older (P<0.0001) and had higher rates of obesity (47.1% vs 30.2%, P=0.0001). Despite these differences, there were not significant differences in the primary outcome in this study, largely owing to the very low rate of manual compression failure (1.1% in patients taking antiplatelet agents versus 0.8% for those not on an antiplatelet regimen). This failure rate is very similar to failure rates of transfemoral vascular closure devices in large registries.4 Given the rarity of failure, we believe this standardized protocol is safe and efficacious to be applied to all patients undergoing diagnostic transfemoral angiography regardless of antiplatelet regimen.

Femoral closure devices are widely-used, convenient for the interventionalist, and have been largely demonstrated to be efficacious and safe across patient populations.5–7 However, there are several important potential disadvantages to these devices including vessel dissection, stenosis, or occlusion. In addition to the primary complication, re-entry site complications can present a secondary challenge if the patient ultimately requires multiple interventions. Serious complications associated with closure devices including device entrapment requiring surgical cutdown for removal and delayed vascular claudication are rare but have been reported.8 9 In large series, primarily in cardiac literature, rates of small hematomas associated with vascular closure devices is in the order of 2%–7%,10 11 though not all of these are clinically relevant.

Furthermore, there is significant cost associated with femoral closure devices – at our institution, cost to the hospital is $250 per MynxGrip (CardinalHealth, Dublin, OH) device. In total, for 525 patients, extrinsic manual compression rather than a MynxGrip has constituted a direct savings of approximately $1 31 250 for the patients enrolled in this study: this is in addition to cost savings from the prospective avoidance of potential complications related to closure device failure, occurring in approximately 1%–2% of MynxGrip deployments in large cardiovascular device safety registries.4 Non-direct costs reduced by our closure protocol include minimizing the need for post-procedural nursing care via early ambulation and hospital discharge. Furthermore, our protocol took place entirely in the recovery suite, thereby reducing both operational cost and turnover time by eliminating any excess time the patient must spend in the angiography suite. On the other hand, the protocol presented here requires 20 min of compression time from a neurointerventionalist or neurointerventional technician. These indirect costs are difficult to quantify but should be considered: in our 5-year experience, this protocol has significantly hastened turnover time while eliminating excess risk to the patient.

We report a 0.95% rate of oozing or hematoma formation in our study of 525 consecutive outpatient angiograms after 20 min of extrinsic manual compression. Subgroup analysis of one previously-reported prospective cohort demonstrated ambulation 2 hours post-procedure without hematoma formation occurs in >90% of outpatients,1 though importantly that analysis did not standardize the closure technique. We report an even lower rate of 0.95% in this consecutive series of outpatients followed by a standardized closure protocol. It is not surprising that outpatients fare well as they are more likely to be ambulatory prior to the procedure and less likely to have active comorbidities. Though this single-arm study is not a valid comparison between a standardized femoral closure technique and patient-specific closure techniques, it is possible that simply standardizing closure protocol has significant benefits, though patient selection remains key.

Study limitations

This study is a single-institutional analysis in one region of the country, subjecting it to the biases inherent to our study population. Our study enrolled only outpatients undergoing elective diagnostic cerebral angiography without intervention at one of four institutions in our practice. A higher complication rate with this closure protocol is likely in patients who undergo intervention, given the potential need for intraoperative antiplatelet or anticoagulation infusion, and with increased catheter and/or sheath manipulation. None of the patients in our study had thrombocytopaenia, which has been associated with transfemoral bleeding complications:12 thus, the results presented here may not be generalizable to patients with low platelet count. It is likely that we are selecting for a group of patients at the lowest risk of complication and/or hematoma formation. Additionally, though we instructed patients to immediately return to our emergency department in the event of post-discharge swelling at the access site, we did not routinely call patients at home post-procedure to gather data on post-procedural bruising. Therefore, it is possible that some patients with bruising or small hematomas could be missed with this protocol, though prior studies have not demonstrated meaningful rates of delayed hematoma formation after diagnostic angiography.

It remains possible that antiplatelet use contributes to a higher rate of hematoma formation in this population. However, only five patients in this study failed the protocol (and thus achieved the primary outcome of hematoma/oozing after manual compression): the rate of complication utilizing this protocol is so low that factors predisposing to extrinsic compression failure cannot be ascertained from this study.


In a single-institution prospective single-arm cohort study of 525 consecutive outpatients undergoing transfemoral diagnostic cerebral angiography using a standardized extrinsic compression protocol,>99% achieved hemostasis without complication regardless of antiplatelet strategy. All remaining patients achieved hemostasis with 10 additional min of manual compression. No patients required additional intervention, and all patients ambulated 2 hours post-procedure and were successfully discharged from the hospital. This protocol is effective and safe for diagnostic transfemoral angiography regardless of a patient’s antiplatelet status.


View Abstract


  • Patient consent for publication Not required.

  • Contributors Conception: BTJ. Gathering of data: all authors. Data analysis: DAT, CF, JP, AO. Final report drafting: all authors. Approval of final report: all authors.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.