Safeguard™ Manual Assist Technique: The SMAT Study

1Anthony Roberts, RN, MBA, MSN, CCRN, 2John Collins, MD, 3Edward Harlamert, MD, 4Judy Champion, RN, CCRN, 4Karen Cronin RN, CCRN, 2Jamie Avram, RN, BSN, 2Colleen Reou, RN, 2Kathleen Mostek, RN, CCRC
1Anthony Roberts, RN, MBA, MSN, CCRN, 2John Collins, MD, 3Edward Harlamert, MD, 4Judy Champion, RN, CCRN, 4Karen Cronin RN, CCRN, 2Jamie Avram, RN, BSN, 2Colleen Reou, RN, 2Kathleen Mostek, RN, CCRC


Femoral sheath removal is generally a nursing responsibility in many acute and critical care settings. Protocols and physician orders for duration of bedrest and manual compression vary widely across and within health care centers in the U.S.1 While manual pressure is often still considered the gold standard of sheath removal,2 the method chosen may be dependent on the individual removing the sheath.3 Access site complications are reported anywhere from 0.5% to 27%.4 There is lack of sufficient data to construct universal guidelines to minimize complications, improve efficiency, reduce cost, and reduce patient discomfort.1
Manual pressure often requires up to 30 minutes of pressure to stop bleeding and requires a great deal of technique. Inconsistent pressure due to hand and arm fatigue can lead to complications and can necessitate increased numbers of nursing staff due to limits on the number of sheaths an individual can pull.5 Staff-related repetitive action musculoskeletal work injuries that are related to sheath removal continue to be a problem and contribute to medical costs in the U.S.6 There is, therefore, significant interest in the clinical usefulness of compression devices which can minimize nursing time and alleviate fatigue associated with manual compression.

The SMAT (Safeguard Manual Assist Technique) study was a prospective, non-randomized, multi-center trial of the safety and efficacy of the Safeguard Manual Assist Technique (Datascope Corporation, Mahwah, NJ) in the management of bleeding following diagnostic and interventional cardiac catheterization via the percutaneous femoral approach. Informed consent, study protocol, and the data collection form were approved by the respective study center hospital IRB. All patients or legally acceptable representatives signed informed consent.
Patients greater than 18 years scheduled for diagnostic or interventional catheterization procedure via common femoral artery with an activated clotting time (ACT) that did not exceed 180 or meet other criteria for delayed sheath pull were eligible. Sheath pull for patients that received bivalirudin (The Medicines Company, Parsippany, NJ) was delayed two hours post infusion by hospital protocol. Patient exclusion criteria included known allergy to adhesives, excoriation, redness, or inflammation of the skin at the proposed Safeguard application site, pregnant or lactating, had a major catheterization procedure related complication at the femoral artery access site pre-sheath removal including hematoma ≥ 6 cm, pseudoaneurysm, or arterio-venous (AV) fistula, or had a femoral artery that was suspected to have experienced a back wall puncture or that underwent more than one arterial puncture during the catheterization procedure. Also excluded were patients whom in the investigator’s opinion, were not suitable for inclusion in the trial (e.g., morbid obesity with body mass index ≥ 40 kg/m2, bleeding disorder, or were felt not able to adhere to the protocol). The Safeguard 24 cm pressure-assisted dressing was applied according to the manufacturer’s instructions. Time to ambulation and discharge were determined by hospital protocol and clinical judgment.
The safety endpoints of major and minor complications before or at 24-48 hours for the device were compared to the historical data control group, a subset of data of the manual compression arm of the previously published RACE (Reduced vascular complications after percutaenous coronary interventions with a non-mechanical suture device) study.7 Major complications were defined as access site bleeding requiring transfusion, vascular repair or the need for vascular repair, any new ipsilateral lower extremity ischemia, surgery for access site-related nerve injury, new access site nerve injury, access site-related infection requiring IV antibiotics or extending hospitalization. Minor complications included bleeding requiring > 30 minutes to re-establish hemostasis, non-treated pseudoaneurysm, non-treated arteriovenous (AV) fistula documented by ultrasound, access site hematoma > 6 cm (after sheath pull), skin excoriation at the site after removal of the dressing, skin erythema, allergic reaction to adhesive, ipsilateral lower extremity arterial emboli, transient loss of lower extremity pulse, deep vein thrombosis, access site-related vessel laceration, transient access-site related nerve injury, access site wound dehiscence, and localized access-site infection treated with intramuscular or oral antibiotics. The mean active compression time for the historical control was 28 minutes. The acceptance criteria was defined as the proportion of Manual Assist Technique patients experiencing a major adverse event to be less than or equal to the serious adverse event rate of the historical control patients at the 95% upper confidence level.

A total of 273 patients were screened prior to catheterization and 101 patients were enrolled in the study. Reasons for failure to enroll following catheterization were physician choice to utilize an alternative method, lack of data collection personnel, multiple access needle puncture sites or double wall punctures, presence of venous sheath on the same side as arterial sheath, inadequate anatomy to support dressing, body mass index > 40 kg/m2, skin sensitivity to tapes, history of femoral stents, bleeding complications, right and left heart catheterization, low platelet count, large abdominal apron, scar tissue, pseudoaneurysm, laboratory abnormalities and physician discretion. The majority of patients were male (65%) with a mean age of 63.0 years, mean body mass index (BMI) of 28.4 kg/m2. Slightly more than half (59%) had a history of prior procedure and 23% had a history of diabetes. The demographic profiles of patients at the two sites was balanced (Table 1).

Medications administered prior to and during the procedure included anticoagulants, aspirin, non-steroidal anti-inflammatory drugs (NSAIDs), and GP IIb/IIIa inhibitors (Table 2). One patient was withdrawn prior to completion of the Manual Assist Technique, per clinician judgment, after 15 minutes of active compression due to oozing at the site.

Device removal was accomplished in all patients without incident or need for solvent application. The device remained fully adhered until planned removal in all patients. The median adhesion time in the overall population was 3.9 hours (range 0.3-23.9) with median adhesion time of 3.6 hours (range 2.1-23.9) and 16.4 (0.3-23.9) in the diagnostic and interventional patient subgroups, respectively.
One patient was withdrawn prior to completion of the manual assist technique after 15 minutes of active compression due to oozing at the site. Therefore, the efficacy evaluation population consisted of 100 patients. Mean active compression time was significantly shorter (p<0.0001) than that reported in the control population for all patients. The mean active compression time for all patients (n = 100), diagnostic patients (n = 72) and interventional patients (n = 28) was 7.7 ± 3.3 minutes, 6.6 ± 2.9 minutes, and 10.5 ± 2.5 minutes, respectively.

Additional measures of efficacy included passive compression time, total compression time, time to ambulation and time to discharge (Table 3). For each measurement, mean times were, as expected, lower in the diagnostic group than in the interventional group. Mean passive compression times were approximately 10 minutes less than total compression times in both treatment groups. The mean times to ambulation and discharge were 3.4 and 4.3 hours, respectively, in the diagnostic group, and were 9.5 and 15.1 hours, respectively, in the interventional group. Time to ambulation and discharge were determined by hospital protocol.

One major complication (successful injection of thrombin for pseudoaneurysm in a 72-year-old male) was reported in 101 patients (1%) compared to 2.4% incidence of major complications reported in 85 patients evaluated in the control study.
Minor complications consisted of 2 cases (2%) of access site hematoma > 6 cm after sheath pull. Other reported events not included in this table due to their low frequency rate and/or severity level are hematoma 2–6 cm: 5.9% (N=6/101), bruising or ecchymosis 4.9% (N=5/101) and swelling 1.9% (N=2/101). One patient complained of right facial weakness and memory loss; head CT was negative and symptoms resolved by time of discharge from procedure. There were no reports of major bleeding.

Although the sample used in the study was prospective, it is limited by being non-randomized. Because of the low major complication rates found in this study, it may be necessary to conduct larger trials at multiple institutions to evaluate generalizability of results to larger populations. Manual compression data is well-established in the literature; however, using historical manual does have the possibility of a limitation by creating a sampling bias.

The purpose of the Safeguard Manual Assist Technique (SMAT) prospective study was to evaluate the safety and effectiveness of the Safeguard Manual Assist Technique in reducing active compression time in patients undergoing femoral cannulation.
The results of the SMAT study demonstrated that the Safeguard 24 cm pressure-assisted device was effective in reducing active compression time as compared to a historical control population without increasing major risk in patients undergoing diagnostic and interventional procedures. The study also showed that the device was effective in assisting in obtaining and maintaining hemostasis in these patients.

This study was funded by Datascope Corp., Mahwah, NJ.

The authors can be contacted at arobert3@clarian.org




1. Chatriwalla AK, Bhatt DL. You can't keep a good man (or woman) down. Journal of Invasive Cardiology March 2006;18(3):109.
2. Galli A, Palatnick A. Ask the experts. Critical Care Nurse April 2005;25(2):88-95.
3. Benson LM, Wunderly D, Perry B, et al. Determining best practice: Comparison of three methods of femoral sheath removal after cardiac interventional procedures. Heart & Lung March/April 2005;34(2):115-121.
4. Allie DE, Herbert CJ, Walker CM. Vascular access site hemostasis: "An endovascular surgeon's perspective" Manual compression may not be benign, Part 1. Cath Lab Digest October 2005;12(9): 6-14.
5. Jones T. Effectiveness of mechanical compression devices in attaining hemostasis after femoral sheath removal. American Journal of Critical Care 2002;11:155-162.
6. Holton M. Ergonomics revisited: Carpal tunnel syndrome. Cath Lab Digest October 2005;13(3):48-53.
7. Sanborn TA, Ogilby JD, Ritter JM, et al. Reduced vascular complications after percutaenous coronary interventions with a nonmechanical suture device: Results from the randomized RACE study. Catheterization and Cardiovascular Interventions 2004;61:327-332.