Evaluation of Poly-N-Acetyl Glucosamine as a Hemostatic Agent in Patients Undergoing Cardiac Catheterization: A Double-Blind, Randomized Study

Abstract

Background: 

This is the first blinded, randomized, placebo-controlled clinical trial to evaluate the efficacy of poly-N-acetyl glucosamine (p-GLcNAc) in improving hemostasis in patients undergoing cardiac catheterization.

Methods: 

Patients were randomly assigned to have either a placebo-treated (n = 17) or a p-GlcNAc–treated (n = 16) 3 × 3-cm patch topically placed at the femoral insertion site at the completion of their catheterization procedure with a mechanical pressure clamp applied over it. The amount of pressure was measured.

Results: 

Although the placebo group had slightly higher clamp pressure applied to the femoral arterial puncture site at the end of the catheterization procedure (189 ± 47 vs. 149 ± 49 mm Hg, p = 0.042), the time to effective hemostasis (16 ± 7 vs. 10 ± 3 minutes, p = 0.01) was decreased in the p-GlcNAc group by 37%.

Conclusion: 

The application of p-GlcNAc patches improved hemostasis at the arterial puncture site in patients undergoing cardiac catheterization.

With the aging population and improving methods of revascularization, cardiac catheterization has grown to encompass more than 1.5 million procedures per year in the United States alone.¹ Complications that may occur after cardiac interventions are costly, increase the patient’s length of stay in the hospital, and affect morbidity.² Inadequate hemostasis can lead to significant blood loss, vessel occlusion, thrombosis, formation of arteriovenous fistulas, pseudoaneurysm formation, and patient discomfort.³ The incidence of major vascular complications requiring surgery after femoral artery catheterization is 0.2% to 1.0% and may be significantly higher after complex interventions.^4,5

Hemostasis of the femoral artery puncture site after catheterization can be achieved using manual compression, compression devices, and, more recently, percutaneous arterial closure devices. Although compression is associated with very low complication rates for bleeding, infection, and thrombosis, it requires patients to be immobile and supine for long periods of time.^6,7 Several technical modifications have been advocated to help minimize the negative impact of femoral artery puncture. Collagen plugs and percutaneous surgical closure devices have been developed to speed hemostasis. These devices allow vascular control of access sites by suture closure of arteriotomy, with or without luminal plugs, thus enabling earlier mobilization and ambulation, shortened observation time, and earlier discharge.⁸ However, there have been several reports of increases in thrombotic complications, especially when devices with luminal collagen plugs are used,⁹ and increases in the severity of infectious complications¹⁰ with percutaneous arterial closure devices compared with manual compression. Complications at the access site caused by arterial cannulation occur in 1% to 8% of cases, but the rate may be as high as 14% with some cardiac interventional procedures; these include hematomas, pseudoaneurysms, arteriovenous fistulas, and even the need for transfusion.^11–15 Given the complications, much interest has been directed toward the development of vascular closure devices that will combine compression pressure with a hemostatic agent to reduce bleeding complications.

The SyvekPatch developed by Marine Polymer Technologies (Danvers, MA) consists of a solid membrane formulation of the polysaccharide polymer, poly-N-acetyl glucosamine (p-GlcNAc). It has been approved by the U.S. Food and Drug Administration approved since 1997 for “use in the promotion of rapid control of bleeding in patients following hemodialysis in patients on anticoagulation therapy” and since 1998 “for use in the local management of bleeding wounds such as vascular site, percutaneous catheters or tubes and surgical debridement.” The effectiveness of this material in reducing the time to hemostasis has been reported in different clinical settings. As a topical hemostatic agent in the swine splenic hemorrhage model¹⁶ and later in human mesenteric bleeding,¹⁷ p-GlcNAc was shown to possess greater hemostatic efficacy than collagen- or oxidized cellulose-based products. Several observational studies have documented the ease, safety, and efficacy of the patch in relatively large numbers of patients undergoing cardiac catheterization and percutaneous cardiac intervention.^18–22 Most of these studies were either uncontrolled or poorly controlled, and their results have not been confirmed by blinded, randomized studies. Thus, this study was designed to evaluate, under controlled conditions, in a blinded randomized fashion, the effectiveness of poly-N-acetyl glucosamine as a hemostatic agent in patients undergoing cardiac catheterization.

METHODS

Between August 2000 and March 2002, after local institutional review board approval was obtained, consent for this study was obtained from 117 patients scheduled to undergo cardiac catheterization at the VA Boston Healthcare System. The other entry eligibility criteria were weight of at least 60 kg and age between 18 and 80 years. The exclusion criteria were medically significant obesity, defined as 25% greater than ideal body weight according to Metropolitan Life Standards;²³ immunosuppression caused by chemotherapy and/or radiation; existing renal or hepatic disease; uncontrolled hypertension (diastolic blood pressure > 100 mm Hg or systolic blood pressure > 180 mm Hg); and blood coagulopathy or dyscrasia evidenced by clinical history and abnormal blood coagulation tests (bleeding time, platelet count, prothrombin time, or activated partial thromboplastin time). After consent had been obtained, a standardized template bleeding time²⁴ measurement was performed on each patient and blood was drawn for routine complete blood count analysis. Patients requiring an interventional procedure during the same catheterization were subsequently excluded from the study before randomization (n = 84). The remaining patients, who underwent only a diagnostic cardiac catheterization and whose postcatheterization activated clotting time was < 350 seconds, were randomized into the study at the end of the procedure, just before the application of a mechanical clamp onto the femoral arterial puncture site. No heparin was administered during the catheterization procedures. After the sheath was pulled, a randomized test patch was applied to the skin at the puncture site. A 5 × 3-cm rubber balloon, sandwiched between two pieces of gauze, was placed over the patch, and a mechanical clamp was applied on top of it (Fig. 1A). The balloon was connected through rubber tubing to a pressure gauge and a pressure transducer (Fig. 1B). The latter was connected to a MacLab (ADInstruments, Milford, MA) for the continuous recording of the balloon pressure throughout the duration of the study. The nature of the study prevented complete blinding of the treatment strategy for either the patient or the treating physician. However, most outcomes were evaluated without knowledge of the assigned technique.

Fig. 1.: Compression pressure measurement system.

The clamp was operated by the cardiology fellow and not the research assistant directly involved in the study. The cardiology fellow, who was blinded to the clamp pressure, tightened the clamp to a level that the patient could tolerate. The clamp was released at 5-minute intervals so that the femoral puncture site could be inspected for bleeding without disrupting the patch. Bleeding was defined as formation of a drop of blood within 5 seconds of clamp removal. If bleeding was observed, the clamp was reapplied and another inspection was performed after 5 minutes. The sterile 3 × 3-cm test patches, which were either the p-GlcNAc (SyvekPatch) or placebo, were indistinguishable from one another. The test patches were supplied for the study by Marine Polymer Technologies. The packages were number coded by the supplier so that the persons conducting the study were blinded to which type of patch was being applied. The patient’s heart rate and arterial blood pressure were also documented during the study period. When bleeding ceased, the time was recorded and the patch was kept in place so as not to dislodge an underlying fresh clot. The absence or presence and the size of any hematoma were also noted. Additional relevant clinical data on the randomized patients were obtained from the hospital’s computerized record system. Data analysis was performed by using Student’s t tests for all variables except for the clamp pressure determinations. Analysis of the clamp pressure was performed by repeated-measures analysis (SAS Statistical Software, Cary, NC). The significance level was set at p < 0.05, and the cumulative data were expressed as means ± sd.

RESULTS

The baseline patient characteristics before cardiac catheterization are shown in Table 1. There were no significant differences in precatheterization characteristics between the group of patients receiving the experimental patch and the group receiving the placebo patch. The pressure applied to the groin during the study (along with the blood pressure and heart rate) and the time to cessation of bleeding in the two patient groups are shown in Table 2. There was a slight difference in the compression pressure, but no significant difference in the arterial pressure or the heart rate between the two groups. However, hemostasis at the puncture site was achieved significantly earlier in the p-GlcNAc (SyvekPatch) group than in the placebo group (an average of 10.0 vs. 15.8 minutes). The use of p-GlcNAc reduced the time to hemostasis by 37%. No evidence of infection or rebleeding was noted during the follow-up period. Three patients, one in the p-GlcNAc group and two in the placebo group, developed small hematomas that were inconsequential and required no intervention.

Table 1: Patient Characteristics Precatheterization

Table 2: Postcatheterization Variables (Mean ± sd)

DISCUSSION

In this study, we have demonstrated that rapid hemostasis can be achieved at the wound site by applying a p-GlcNAc-based (SyvekPatch) bandage post-procedure in patients who underwent cardiac catheterization. The bleeding times and hematologic assays performed confirmed that none of the patients had pre-existing thrombolytic abnormalities, although several patients were on coumadin or heparin therapy and most were receiving aspirin. The p-GlcNAc group required statistically less time for hemostasis, less compression time and had less hematoma formation. To negate the effect of pressure alone on hemostasis results, we designed a measurable pressure application device that allowed us to record the pressure applied on top of the bandage at the wound site. Our choice of using a C-clamp was based on the fact that the mechanical clamp, used extensively in many medical centers, was shown to have a shorter mean compression time and lower frequency of hematoma formation when compared to manual pressure.² Also, since the amount of direct pressure on a wound is a very important component of hemostasis, we attempted to control that variable by utilizing a C-clamp device to measure the pressure applied (Figure 1). Clamp pressure for the p-GlcNAc group was statistically less than that of placebo, raising question of “if equal” would hemostasis have occurred even sooner? Also, since the wound sites were only checked every five minutes, if hemostasis occurred between the time checks its exact time could not be ascertained. In spite of the ability to only check for hemostasis every five minutes, we observed a significant decrease of 38% in the time to control bleeding at the wound site with the use of the p-GlcNAc patch in comparison to the placebo patch. The effective hemostasis achieved by the p-GlcNAc patch was even more apparent, as we did not observe any difference between the placebo and p-GlcNAc group in their pre-cath characteristics (Table 1).

Glowacki, et al. reported the use of the SyvekPatch in 800 cases undergoing diagnostic cardiac catheterization.²⁵ They applied the patch with 10 minutes of manual compression with no apparent hematomas, in contrast to the 35–40 minutes of manual compression required previously in their cases where SyvekPatch had not been used. Both our study and theirs found that an average time of 10 minutes sufficient to establish hemostasis, regardless of the type of compression applied. Local vascular complications were limited to minor (<2cm) hematomas. No bleeding, arterial thrombosis, or infections were noted. Similar to our experience with anticoagulated patients, it has been reported that only a 1% hematoma rate in an observational study on the effectiveness of the patch in 621 patients using anticoagulants or anti-platelet medications.²⁶ The SyvekPatch offers advantages over other vascular closure devices in that it is easy to use, is fast acting, can be applied by the nursing staff, leaves no subcutaneous matter behind that may be a source of infection,¹⁰ has a low complication rate, and is relatively inexpensive.

We hypothesized that the rapid hemostatic action of p-GlcNAc patch must be a result of the interaction between the blood cells and p-GlcNAc, as the patch first encounters the blood at the wound site. Subsequent experiments in our laboratory have validated these assumptions. Red blood cells rapidly aggregate in the presence of p-GlcNAc in a time- and concentration (p-GlcNAc)-dependent manner.²⁷ The half-life for this aggregation was approximately 5 to7 minutes. Maximum aggregation was achieved at 10 to 15 minutes, corresponding well with the time of p-GlcNAc patch-mediated hemostasis observed in the Cath lab. Similarly, p-GlcNAc was also found to cause rapid aggregation and activation of platelets in vitro, similar to the in vitro kinetics of the red blood cells and hemostasis observed in the cath lab.²⁸ The in vitro action of p-GlcNAc was equally effective on whole blood, blood treated with heparin, EDTA and/or protamine, supporting our in vivo observation of rapid hemostasis by p-GlcNAc patch in anticoagulated patients in the cath lab. We have shown that the mechanism of p-GlcNAc -mediated aggregation and activation of blood cells in vitro was achieved via ionic interactions, and involved cell surface receptors,^27,28 and also by the activation of coagulation pathways²⁹ that may lead to the formation of a clot at the wound site. The cellulose-based patches that were used as placebo in these study do not demonstrate such global aggregation and activation of blood cells, thus leading to prolonged hemorrhaging at the wound site. As pressure was applied for both, cellulose and p-GlcNAc-based bandages, pressure alone had no effect on the hemostasis at the wound site. Our results demonstrate that, the primary mechanism of p-GlcNAc-mediated hemostasis is based on its interaction with and activation of the blood cells and coagulation pathways at the wound site. The SyvekPatch has been shown to be effective in significantly reducing the time to hemostasis in this study.

ACKNOWLEDGMENTS

We thank C. Robert Valeri, MD, Naval Blood Research Laboratory (Boston, MA), for encouragement and useful discussions.