Clin. Cardiol. 20,981-992 (1997)

Review

Rapid Hemostasis of Arterial Puncture Sites with Collagen After Diagnostic and Interventional Cardiac Catheterization.

Sigmund Silber, M.D., FACC

Dr. Müller Hospital, Munich, Germany

Summary: Despite the continuous reduction of sheath sizes in diagnostic and interventional cardiac catheterizations and the discontinuation of coumadin use after coronary stent implantation, a challenging role remains for hemostatic devices in the sealing of femoral puncture sites. Since the introduction of the vascular hemostatic device (VHD) in 1991 and the hemostatic puncture closing device (HPCD) in 1992, numerous studies were published investigating these devices. The deployment success rates reported in 2292 patients for VHD is 97%, ranging from 88 to 100%. For HPCD, the mean deployment success rate resulting from 622 published patients leads to an identical result of 97%, ranging between 91 and 100%. For time to hemostasis, data have been analyzed according to the four different clinical situations, depending on level of anticoagulation (none or full) and the time of sheath removal (immediately or delayed). In randomized studies, when compared with the manual control groups both devices revealed a statistically significant reduction in time to hemostasis: 12 to 16 minutes less for diagnostic catheterization and 14 to 30 minutes less for PTCA. As for minor local complications, no clinically relevant differences seem to exist. None of these devices has proven to reduce major local complications. Prospective trials addressing early mobilization after percutaneous transluminal coronary angioplasty and the cost effectiveness of arterial closure devices in defined subgroups are warranted.

Key Words: collagen, hemostasis, complication, cardiac catheterization, PTCA

Introduction:

Despite the continuous reduction of sheath sizes in diagnostic and interventional cardiac catheterizations and despite the discontinuation of coumadin use after coronary stent implantation, a challenging role for hemostatic devices in sealing femoral puncture sites remains: patients undergoing diagnostic coronary angiography may be ambulated almost immediately and discharged many hours earlier than currently practiced in most centers utilizing a supine restriction period of 6 hours after diagnostic catheterization [1]. On the other hand, patients undergoing PTCA by the femoral approach (which is the access used far more frequently than the brachial or radial approach [2]) are usually immobilized overnight, which may result in significant discomfort with increased back pain and need for analgesics [3]. In patients with low-risk procedures, when prolonged vascular access does not seem to be needed, sheath pulling immediately after PTCA increases patients' comfort (returning to their rooms without a sheath), decreases burden for the medical staff and may reduce hospital costs by shortening the length of stay. Hemostatic devices may allow patients to walk 2 to 3 hours after the end of the procedure and hence further increase their comfort. In addition, even with the current stent protocols using ASA and ticlopidine, major local bleeding complications may still occur in 1% (STRESS III [Stent Restenosis Study], ISAR [Intra-coronary Stenting Antithrombotic Regimen]) up to 2.2% (STARS [Stent Anticoagulation Re-gi-men Study]) and 2.4 % (FANTASTIC [Full Anticoagulation versus Ticlopidine plus Aspirin after Stent Implantation] [4-7]). Furthermore, Glycoprotein IIb/IIIa inhibitors are in-creasing-ly used in high and also low risk patients: although the increased rate of bleeding complications in the EPIC study [Evaluation of c7E3 for the Prevention of Ischemic Complications] could be significantly reduced by decreasing the concomitant heparin dosage [8], the EPILOG [Evaluation of PTCA to Improve Long-term Outcome by c7E3 GP IIb/IIIa receptor blocker] and RESTORE [Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis] trials still revealed a rate of major bleeding of 1.8% and 2.5% respectively in the patients treated with Glycoprotein IIb/IIIa inhibitors and even of 2.3% to 3.1% in the placebo groups [8, 9]. The use of low molecular weight heparin in patients at high risk for stent thrombosis may also be associated with a higher bleeding risk [10].
Since there has been no published overview summarizing and analyzing the results for collagen devices, it is the purpose of this paper to review the data and to provide a differentiated analysis of success and local complication rates for patients undergoing diagnostic or interventional cardiac catheterizations.

Characterization of Protocols and Patients Enrolled

To analyze the studies, it is important to differentiate the protocols investigated. With regard to hemostasis, four different clinical situations may be encountered: 1) immediate sheath pulling after diagnostic catheterization is usually related to smaller sheath sizes and to no or a low level of anticoagulation; 2) after PTCA (or other coronary interventions), which are usually performed using larger catheters, sheaths may have been pulled with delay and without continued anticoagulation, at a time when little or no anticoagulation is effective; 3) on the other hand, delayed sheath-pulling in patients under continued full anticoagulation (prolonged heparin administration or on coumadin according to previous stent protocols) must be stric-t-ly discerned; and 4) sheath pulling immediately after PTCA is - of course - always performed under full anticoagulation. Unfortunately, many of the pub-lished studies did not differentiate between these clinical settings and thus reported a mixture of overall results.
The exclusion criteria used in most of the studies were quite homogeneous: inadvertent penetration of the dorsal arterial wall with the puncture needle, previous application of collagen sealing of the femoral access site, known allergy to collagen, clinical signs of or known peripheral artery disease, patients with acute myocardial infarction, status post thrombolytic therapy, known coagulation defects or known platelet dysfunction, severe and uncontrolled arterial hypertension (systolic > 220 mmHg or diastolic > 120 mmHg), preexisting hematoma or hematoma developed during the procedure or patients with a venous femoral sheath.

Definition of Hemostasis

Basically, two parameters of measuring the success of hemostasis have been reported: the mean value of the times until complete hemostasis occurred in each individual patient (time to hemostasis) and the percentage of patients showing complete hemostasis after a specified time interval (hemostasis success rate). Although theoretically both parameters can be measured simultaneously, most of the studies reported only either one parameter or incomplete data.

Time to hemostasis

Time to hemostasis is defined as the time elapsed from initial compression at removal of the sheath until the completion of compression. The first time at which no bleeding occurs is taken as the time to hemostasis in this particular patient. However, the measurement of time to hemostasis is not standardized: results for time to hemostasis intrinsically depend on the time interval to the first and between the subsequent checks for bleeding; using a minimum time resolution of e.g. 15 minutes, one cannot expect to find a time to hemostasis of less than 15 minutes. Table 1 summarizes the time intervals between sheath removal and the first check as well as the subsequent time intervals used in studies determining time to hemostasis with collagen devices. As one can see, time intervals between 30 seconds and 15 minutes have been used for the first check interval and between 1 minute and 10 minutes for the subsequent intervals. Furthermore, not all studies clearly defined the time intervals for all the groups investigated; and in some studies even the time intervals within the same study varied. The choice of the time interval between the bleeding checks is an ambiguous decision: too short intervals may not give sufficient time for thrombus formation and may artificially increase - particularly in the manual control groups - the time to hemostasis.
In some studies, after deployment of a hemostatic device all patients automatically received a vascular C-clamp [21] or an air cushion device [22]. Therefore, in these studies the determination of time to hemostasis was not possible.

Abbreviations: diagn=diagnostic, P'TCA=percutaneous transluminal coronary angioplasty, VHD=vascular hemostatic device, HPCD=hemostatic puncture closing device.

TABLE I Published time intervals between sheath removal and checkings for hemostasis in collagen studies for determination of time to hemostasis. Some studies did not clearly define all time intervals ("?").

This parameter reveals the percentage of patients showing complete hemostasis at a specified point of time. The shorter the time interval defined, the lower the success rate. Thus, in addition to sheath size and level of anticoagulation, when comparing the results for hemostasis success rates one must consider possible differences of the time points at which the success rate was measured. The time intervals for determination of hemostasis success rates range from zero seconds (immediate hemostasis, [19, 23, 24]), to 2 to 5 minutes [12, 13, 17, 18, 22, 25, 26] and even up to one hour [21].

Vascular Complications

In this analysis, the definitions of major and minor vascular complications were used according to the US multicenter trial [11]. The following complications were classified as major: thrombosis or loss of distal pulses, large pseudoaneurysm or AV-fistula, bleeding with need for transfusion or vascular surgery. Bleeding from puncture site not needing transfusion and/or vascular surgery as well as a small pseudo-aneurysm treated medically were classified as a minor complication.

Fig. 1 Devices used for sealing arterial puncture sites. VHD=vascular hemostatic device, HPCD=hemostatic puncture closing device, GVSD=Gershony vascular hemostatic device.

Hemostatic Devices

Sandbags do not reduce vascular complications and even increase patients' discomfort [27]. Mechanical devices like C-clamps, stasis buttons or air cushions were used as a replacement for manual compression, but upon physical examination did not clearly show a reduction in hematoma formation [28, 29]. Recent data suggest a reduction in ultrasound-detected AV-fistulas and pseudoaneurysms with the C-clamp [30]. Mechanical devices, cannot, of course, reduce the time to hemostasis and therefore cannot decrease the minimum time required for bed rest.
Hemostatic devices for rapid closure of arterial puncture sites may be classified according to their mechanisms as illustrated in Figure 1. The vast majority of clinical experience has been gained using bovine collagen devices: the prototype was the VHD (Figure 2, VasoSeal®, Datascope Corp., Montvale, NJ), a pure collagen plug device, followed by the HPCD (Figure 3, Angio-Seal™, originally developed by the Kensey Nash Corporation, Exton, PA; it is now a trademark of Quinton Instrument Company, Bothell, WA, within the USA, and of Sherwood Davis & Geck, St. Louis, MO, outside the USA). In addition to the collagen, the HPCD applies an intraarterial anchor. Both devices are discussed in detail below.
Another approach is the installation of a fibrin sealant via the arterial sheath. Fibrin sealant is a well-known tissue adhesive which combines fibrinogen (from human plasma) and (bovine) thrombin to form fibrin. First encouraging results were obtained in animals and in 20 patients [31, 32]. Fibrin´s widespread application, however, may be limited by the need for use of human plasma products. Therefore, the GVSD (Gershony Vascular Sealing Device, Duet™, Vascular Solutions, Inc, Minneapolis, Minnesota) using a mixture of bovine thrombin and collagen may be more promising [33]. A different way of inducing hemostasis is the application of low-energy radiofrequency (30 to 35 watts) via a guide wire through subcutaneous tissue to the periarterial wall [34]. First clinical results in 55 patients have been reported; the success rate seems to be related to the level of anticoagulation [34].

The clinical findings using suture devices (Perclose, Inc) are controversial. The Prostar™ device uses 4 needles (2 sutures); the Techstar™ device uses 2 needles (1 suture). After predilation of the subcutaneous tissue (requiring a 21-F tissue track), the suture containing device is advanced into the artery, the needles are retracted and knots are tied against the arterial wall, facilitated by a knot-pushing tool [35]. Whereas preliminary results with the Prostar™ device were encouraging, the relatively high rate of local vascular complications in the US multicenter trial make further analysis necessary [35, 36]. Modifications of the Prostar plus™ and Techstar™ devices seem to provide more reliable results [37, 38]; the use of both devices in a single center with high volume experience was, however, associated with a need for vascular repair in 2.1% of the PTCA patients [39]. Furthermore, the deployment success rate of 89.6% (786/877 patients) with even the 6F-Techstar™ device appears relatively low [40].

FIG.2 The VHD (Vascular Hemostatic Device, VasoSeal®) comprises a blunt-tipped 11-F dilator (center) which is inserted using the short guide wire, a short 11.5-F sheath (right) and two 90 mg collagen cartridges (between the above).

FIG.3 The HPCD (Hemostatic Puncture Closing Device, Angio-Seal™) consists of a short guide wire, a dilator with 2 lumina, an 8-F sheath (mounted together, center), the "carrier" device (containing the anchor, the collagen and the suture, above) and a tamper (below).

Comparing VHD and HPCD

The basis of VHD and HPCD is collagen. Purified bovine collagen has been used in vascular, abdominal and dental surgical procedures since late in 1960 as an adjunct to hemostasis when control of bleeding by ligature or other conventional methods was insufficient [41-43]. The bio-de-gradable collagen plug induces platelet activation and aggregation, releasing coagulation factors and resulting in the formation of fibrin and the subsequent generation of a thrombus [44]. It is assumed that anticoagulation with heparin or even antiaggregation with aspirin do not affect hemostasis induced by collagen [12, 45]. Collagen is ultimately degraded and resorbed by granulocytes and macro-phages. These cells, releasing their collagenase enzymes, invade the plug and selectively digest the collagen as a function of the physical properties of the different collagens [46]. The immunogenicity of collagen has been a subject of debate, focusing in particular on injectable collagen, which is used to correct dermal defects such as acne or wrinkles. The reports of a possible link of collagen to autoimmune disease have implicated only injectable collagen, which is quite different in structure and the degree of cross-linking than the collagen sponge used in VHD and HPCD. Antigenicity of purified collagen is considerably reduced and, although allergies to collagen are described [47], allergic reactions to VHD have not been a clinical problem [11-13].

Device Description and Deployment

The VHD consists of a purified collagen plug that induces the formation of a hemostatic cap directly over the arterial puncture site when inserted adjacent to the arterial wall. The method of its deployment is described in detail elsewhere [11, 12]. In brief, VHD deploys a collagen plug onto the external arterial wall after dilation of the skin and subcutaneous tissue to 11.5 F. It comprises four parts: a blunt-tipped 11-F dilator, one of seven differently sized 11.5-F sheaths selected by length using a pre-procedure needle depth measurement technique, and two 90 mg collagen cartridges. When the sheath is to be pulled, a short guide wire is inserted and the existing sheath removed while complete hemostasis is maintained with manual compression. Then the blunt-tipped 11-F dilator is inserted over the guide wire just down to the site of the arterial puncture. Guidance is obtained by feeling the resistance of the dilator against the outer surface of the artery as well as by the marker on the dilator. The 11.5-F sheath is then advanced over the dilator down to the arterial surface. While still holding pressure, the dilator and the guide wire are removed and the collagen cartridge deployed with a "push and pull" movement. We saw in a previous study that one collagen plug is as effective as two but is better tolerated [25, 48].

The HPCD provides a mechanical block of the arterial puncture site with an anchor from inside the artery, guiding and holding the collagen in the tract. It consists of four components within a single delivery device ("carrier") requiring an 8-F sheath: anchor, collagen plug, connecting suture and a tamper. All three components deployed into the patient (anchor, suture and collagen) are completely resorbable; the anchor and the suture are made from polyglycolic and polylactic acids. The small plug contains only about 14 mg collagen. The technique of its deployment has been described in detail elsewhere [18, 19]. In brief, a short guide wire is inserted and the existing sheath removed while hemostasis is maintained with manual compression. First, the location of the end of the 8-F sheath within the artery is determined by inserting a modified dilator having 2 lumina: one (distal) for the guide wire and one (proximal) at the end of the 8-F sheath. The location of the end of the 8-F sheath is determined by the presence of blood flow through the modified dilator. Our preferred location of the sheath is ca. 1 cm further down the puncture site inside the artery lumen. The dilator is then removed and the carrier device introduced into the 8-F sheath. The anchor is secured against the intraluminal arterial wall (we check three times at different angles) and the collagen plug is deployed on the outer arterial wall. A tamper is pushed downwards to compress the collagen against the outer arterial wall ("sandwich technique"). Finally, a spring is attached between the tamper and a metal tag fixed to the positioning suture, thus applying continuous pressure on the tamper [24]. Although the deployment technique may sound complicated, it usually takes less than 60 seconds to deploy the de-vice.

The material used for the intraarterial anchor of the HPCD is a 50:50 D,L polylactic-coglycolic acid copolymer and well established in medical use. It has a safe history, for it is widely used in sutures, bioresorbable meshes and sustained release drug delivery systems. Some concern about the concept of inserting a foreign body (although resorbable) into the lumen of the artery has been expressed. After initial experience in animals [49], ultrasound studies in patients have shown that HPCD did not cause more changes in flow pattern than observed in the control group with manual compression [19, 50]. Therefore, the described changes of flow pattern after intraarterial anchoring are related to the puncture procedure itself rather than to the hemostatic device [19, 50]. The anchor is absorbed in the majority of patients within 4 weeks [50]; after 2 months, complete absorption of the device was documented by ultrasound in all patients [23]. US and European single and multicenter trials have established the safety of the concept of intraarterial anchoring [18, 19, 51].

Published Data Regarding Vascular Hemostatic and Hemostatic Puncture Closing Device

Deployment Success Rates

Table II lists the results reported for a successful deployment of either VHD or HPCD: In 2292 patients reported, the deployment success rate for VHD is 97%, ranging from 88% to 100%. For HPCD, the mean deployment success rate resulting from the 622 published patients leads to an identical result of 97%, ranging between 91% and 100%.

TABLE II Deployment success rates of collagen devices

Time to Hemostasis

The published data on time to hemostasis for VHD and HPCD in patients undergoing diagnostic cardiac catheterization with no systemic or little admi-ni-stration of heparin is listed in Table 3. The mean values for time to hemostasis vary between 1.7 and 4.3 minutes. In randomized studies, both devices revealed statistically significant reductions of time to hemostasis of about 12 to 16 minutes less than the manual control groups.
In PTCA patients, the level of anticoagulation at the time of sheath removal was reported by most but not all groups (Table 3). For VHD, a statistically significant reduction in time to hemostasis has been shown for immediate sheath removal and delayed sheath removal without prolonged anticoagulation (Table 3). The one study using VHD in patients with delayed sheath removal and prolonged anticoagulation had no control group [15]. For HPCD, a statistically significant reduction in time to hemostasis has been shown for immediate sheath removal and delayed sheath removal with prolonged anticoagulation (Table 3). For both devices, the average gain in time to hemostasis was approximately 14 to 30 minutes and therefore somewhat more than the gain in diagnostic patients. A gain in time to hemostasis for immediate sheath removal cannot be calculated, since there is no real control group for immediate sheath removal and manual compression.
The results of reports comprising a mixture of diagnostic and therapeutic interventions also showed an overall statistically significant reduction in time to hemostasis for VHD and HPCD (Table 3). Not all authors, however, revealed data on the level of anticoagulation at the time of sheath removal (Table 3).

Abbreviations: ACT=activated clotting time, PTT=partial thromboplastin time, INR=International Normalized Ratio. Other abbreviations as in Table I.

TABLE III Time to hemostasis using collagen devices. Differences can be attributed to different study designs regarding time of sheath removal and levels of anticoagulation.

Time to Ambulation

One of the primary goals using hemostatic devices is enabling patients to walk earlier. In most studies, however, early ambulation was not an endpoint, so the data of time to ambulation cannot be used for this purpose [11, 18]. To analyze the reported time to ambulation, the time of sheath removal is important to know (Table 4). Therefore, with delayed sheath removal after PTCA, a time to ambulation of 2 to 3 days is not surprising [15, 52]. Unfortunately, not all publications revealed the time of sheath removal (Table 4).

Abbreviations as in Table I.

TABLE IV Time to ambulation using collagen devices

For diagnostic cardiac catheterization, two studies addressed the primary endpoint of early ambulation: with immediate sheath removal, patients were successfully ambulated 30 minutes after the deployment of VHD [25] and even 20 minutes after HPCD [56]. For PTCA patients, no study specifically investigating early ambulation has been published. In PTCA-studies, the shortest time intervals to ambulation were in the range of 6 to 7 hours for VHD [26, 53], and 8 hours for HPCD [18]. In mixed patient populations, after diagnostic or therapeutic cardiac catheterization, attempts were made to ambulate patients "within 2 hours" after the deployment of VHD [14] or starting at 4 hours after HPCD [19, 51].

Local Complications

Complications after manual compression:
Vascular complications in patients undergoing diagnostic cardiac catheterizations occur in the order of 0.5% [60]. The incidence of major vascular complications (requiring blood transfusion or vascular surgery) in patients undergoing diagnostic cardiac catheterization has been reported to range from 0.35% to 5% [61-65]. While in earlier decades these incidences often included thrombosis and distal embolism, later studies reported lower vascular complication rates of < 0.5%, reflecting improved equipment and extensive operator experience [60, 66]. With the use of 5F- catheters for outpatient coronary angiography, major and minor complications were less frequent than earlier reported [66]. In recent studies, major vascular complications (as defined above) did not occur in patients undergoing diagnostic cardiac catheterization [11, 19, 26]. Small hematomas are only scarcely reported; many centers do not report them and/or consider them unimportant [2, 60]. Kern et al. reported 8% (23/287) minor hematomas in outpatients undergoing 5F-cath-e-teri-zations [66].
Complications after collagen devices:
The complication rates reported for collagen plugging are somewhat confusing, because several studies did not differentiate between diagnostic and interventional procedures and because various classifications of complications with different methods of measurement were used: in the European VHD multicenter trial, only the overall complications were reported for the 105 patients receiving the collagen plug after diagnostic and for the 140 patients after interventional procedure, despite significant differences in the doses of heparin (5715 ± 4615 U vs. 15378 ± 3025 U). ACTs or PTTs were not reported [12].
For VHD, some authors found a significant reduction in local complications [13, 59], whereas other findings were nonconclusive [11]. In contrast, some reported an increased incidence of only minor [26] or even major complications [16] and therefore described collagen plugging of arterial puncture sites with VHD as a "deep disappointment" [67].
In these trials, however, the collagen groups were not compared to the control groups under identical conditions: in the collagen groups the sheaths were pulled immediately, whereas in the control groups the sheaths were pulled several hours later [13, 22, 59, 68] or even the next day [14, 22]. Because the sheath-dwell time represents one of the risk factors for local complications [26, 69], this parameter should be kept constant in both groups to evaluate the possible influence of the collagen plug on local complications. We therefore performed a prospective, randomized trial with both groups having identical sheath-dwell times [17]. According to the classification of major and minor complications suggested by the US multicenter trial [11], one major complication needing vascular surgery occurred in the collagen group and none in the control group. When comparing minor complications in both groups, no statistical difference was found for the development of any hematoma. Significantly more patients assigned to collagen (4%) developed a large hematoma than patients assigned to conventional sheath pulling (0%) [17].
Tables 5 and 6 list the published reports on local complications using VHD or HPCD in detail.

Abbreviations as in Table I

TABLE V Detailed analysis of studies reporting minor or major complications using the vascular hemostatic device (VHD)

Abbreviations as in Table I

TABLE VI Detailed analysis of studies reporting minor or major complications using the hemostatic puncture closing device (HPCD)

Although both devices carry the inherent risk of inadvertent intraarterial collagen insertion, published reports on this device-related complication exist only for VHD. (Table 7).

TABLE VII Reports on intra-arterial complications (insertions or protrusions) using the vascular hemostatic device (VHD)

Cost/Effectiveness

Hemostatic closure devices are expensive and have to prove their cost/effec-tiveness.
Effectiveness can be defined as hemostatic success. So far, no study has clearly proven one hemostatic device to be clearly better than others. Therefore differences between these devices may be related to different costs.
Costs may be divided into procedural costs, hospital costs and follow-up costs. Procedural costs reflect the price of the device plus the time for deployment (cathlab time, doctors' time spent for deployment). In our facilities, VHD has the lowest price, HPCD is in the middle range and suture devices are the most expensive. The time for deployment of either VHD or HPCD is similar (less than 1 minute), whereas the suture devices take us usually 4 to 5 minutes. Hospital costs: in Germany, hospitals are paid by the day (length of stay). In diagnostic patients undergoing catheterization in the afternoon, all hemostatic devices may make an overnight stay unnecessary and therefore save the costs for 1 day. There is, however, a conflict of interest between health insurers and hospitals: saving overnight stay is important to the health insurers, but leads to a loss of income for the hospitals. In PTCA patients, usually 1 or 2 days may be saved, with the same conflicting interests as described for diagnostic procedures. In our study, the decision to discharge was left to the ward physicians. Nevertheless, patients with collagen sealing were discharged one day earlier than the control group [17]. This is in good agreement with a US study, which reported a significant decrease from 2.4 ± 0.98 days (control group, 56 patients) to 1.53 ± 0.8 days (collagen group, 47 patients) [74]. Follow-up costs are predominantly defined by local vascular complications; these complications may lead to substantial costs, even higher than that of the catheterization itself [1]. Hemostatic devices should at least not increase the rate of local vascular complications or - even better - lead to a decrease of complications. There is, however, no clear evidence that hemostatic devices decrease the rate of local complications; some devices even show a tendency to increase local complications.

Future Aspects

The data suggest that mechanical forces play a more important role in sealing arterial puncture sites with collagen than previously anticipated [12]. The collagen itself may not be that important for devices using intraarterial anchoring [56] and could perhaps be replaced by nonbiological, resorbable materials. Besides device changes (like 6F-devices for 6F-PTCA's or 4F-devices for 4F-diagnostics), it is important to fill the gaps of missing studies that have early ambulation as primary endpoint after PTCA. Newer antiplatelet protocols with ticlopidine pretreatment and/or the administration of IIb/IIIa inhibitors also need to be investigated. Furthermore, prospective trials addressing the cost/effectiveness of arterial closure devices in defined subgroups of patients are warranted.

References

1. Krause PB, Klein LW. Utility of a percutaneous collagen hemostasis device: To plug or not to plug? J Am Coll Cardiol. 1993;22:1280-1282

2. Krone RJ, Johnson L, Noto T and the Registry Committee of the Society for Cardiac Angiography and Interventions. Five year trends in cardiac catheterization: A report from the registry of the society for cardiac angiography and interventions. Cathet Cardiovasc Diagn. 1996;39:31-35

3. Waksman R, Scott NA, Ghazzal ZMB, Mays R, Frerichs FA, Petersen JY, King III SB. Randomized comparison of flexible versus nonflexible femoral sheaths on patient comfort after angioplasty. Am Heart J. 1996;131:1076- 1078

4. Stress III Investigators. Early outcome after coronary stent placement with high pressure inflation and antiplatelet therapy: Interim results of the STRESS III trial. Circulation. 1996;94:I-684

5. Schömig A, Neumann FJ, Kastrati A, Schühlen H, Blasini R, Hadamitzky M, Walter H, Zitzmann-Roth E-M, Richardt G, Alt E, Schmitt C, Ulm K. A randomized comparison of antiplatelet and anticoagulant therapy after the placement of coronary artery stents. N Engl J Med. 1996;334:1084-1089

6. Leon M, Baim DS, Giambartolomei A, Williams DO, Diver DJ, Senerchia C, Fitzpatrick M, Popma JJ, Kuntz RE. Clinical and angiographic results from the stent anticoagulation regimen study (STARS). Circulation. 1996;94: I-685

7. Bertrand M, Legrand V, Boland J, Fleck E, Bonnier J, Emmanuelson H, Vrolix M, Missault L, Chierchia S, Casaccia M, Niccoli L, Oto A. Full antocoagulation versus Ticlopidine plus Aspirin after stent implantation: A randomized multicenter European study: The FANTASTIC trial. Circulation. 1996;94:I-685

8. Lincoff AM, Tcheng JE, Miller DP, Booth JE, Montague EA, Topol EJ. Marked enhancement of clinical efficacy of platelet GP IIb/IIIa blockade with c7E3Fab (Abciximab) linked to reduction in bleeding complications: Outcome in the EPILOG and EPIC trials. Circulation. 1996;94:I-375

9. Anderson HV, Bertrand M, Whitworth HB, Sax FL, Willerson JT for the RESTORE Investigators. Bleeding risk with platelet inhibition using Tirofiban: The RESTORE Trial. Circulation. 1996;94:I-553

10. Goods CM, Liu MW, Jain SP, Mathur A, Yadav JS, Al-Shalbi KF, Dean LS, Iyer SS, Parks JM, Roubin GS. Low molecular weight heparin versus standard heparin in patients at high risk for stent thrombosis: Clinical outcomes. Circulation. 1996;94:I-684

11. Sanborn TA, Gibbs HH, Brinker JA, Knopf WA, Kosinski EJ, Roubin GS. A multicenter randomized trial comparing a percutaneous collagen hemostasis device with conventional manual compression after diagnostic angiography and angioplasty. J Am Coll Cardiol. 1993;22:1273-1279

12. Ernst SMPG, Tjonjoegin RM, Schräder R, Kaltenbach M, Sigwart U, Sanborn TA, Plokker HWT. Immediate sealing of arterial puncture sites after cardiac catheterization and coronary angioplasty using a biodegradable collagen plug: Results of an international registry. J Am Coll Cardiol. 1993;21:851-855

13. Schräder R, Steinbacher S, Burger W, Kadel C, Vallbracht C, Kaltenbach M. Collagen application for sealing of arterial puncture sites in comparison to pressure dressing: A randomized trial. Cathet Cardiovasc Diagn. 1992;27: 298-302

14. Foran JPM, Patel D, Brookes J, Wainwright RJ. Early mobilisation after percutaneous cardiac catheterisation using collagen plug (VasoSeal) haemostasis. Br Heart J. 1993;69:424-429

15. Bartorelli AL, Sganzerla P, Fabbiocchi F, Montorsi P, De Cesare N, Child M, Tavasci E, Passaretti B, Loaldi A. Prompt and safe femoral hemostasis with a collagen device after intracoronary implantation of Palmaz-Schatz stents. Am Heart J. 1995;130:26-32

16. v.Hoch F, Neumann F-J ,Theiss W , Kastrati A, Schömig A. Efficacy and safety of collagen implants for haemostasis of the vascular access site after coronary balloon angioplasty and coronary stent implantation. Europ Heart. J 1995;16:640-646

17. Silber S, Björvik A, Rösch A, Mühling H. Advantages of sealing arterial puncture sites after PTCA with a single collagen plug: A randomized, prospective trial. J Am Coll Cardiol. 1995;23:262A

18. Kussmaul WG 3rd, Buchbinder M, Whitlow PL, Aker UT, Heuser RR, King SB, Kent KM, Leon MB, Kolansky DM, Sandza JG Jr. Rapid arterial hemostasis and decreased access site complications after cardiac catheterization and angioplasty: results of a randomized trial of a novel hemostatic device. J Am Coll Cardiol. 1995;25:1685-1692

19. de Swart H, Dijkman L, Hofstra L, Bär FW, Van Ommen V, Tordoir J, Wellens HJJ. A new hemostatic puncture closure device for the immediate sealing of arterial puncture sites. Am J Cardiol. 1993;72:445-449

20. Murray CR, Mortazavi A, Warth DC, Buchbinder M. A randomized controlled prospective study of the Angio-Seal device versus manual pressure for rapid arterial puncture closure. Circulation. 1996;94:I-376

21. Webb JG, Carere RA, Dodek AA. Collagen plug hemostatic closure of femoral arterial puncture sites following implantation of intracoronary stents. Cathet Cardiovasc Diagn. 1993;30:314-316

22. Camenzind E, Grossholz M, Urban P, Dorsaz PA, Didier D, Meier B. Collagen application versus manual compression: a prospective randomized trial for arterial puncture site closure after coronary angioplasty. J Am Coll. Cardiol 1994;24:655-662

23. Kussmaul WG, Buchbinder M, Whitlow PL, Aker UT, Heuser RR, King SB, Kent KM, Leon MB, Kolansky DM, Sandza JG. Femoral artery hemostasis using an implantable device (AngioSeal™) after coronary angioplasty. Cathet Cardiovasc Diagn. 1996;37:362-365

24. Silber S, Dörr R, Mühling H, König U. Sheath pulling immediately after PTCA: Comparison of two different deployment techniques for the hemostatic puncture closure device. A prospective, randomized study. Cathet Cardiovasc Diagn. 1997;41:378-383

25. Silber S, Haentsch C, Seidel N, Mühling H. Early ambulation (30 minutes) after cardiac catheterization using the collagen plug. Efficacy and follow- up comparing two different plug dosages in 660 cases. J Am Coll Cardiol. 1993;21:150A

26. Slaughter PM, Chetty R, Flintoft VF, Lewis S, Sykora K, Beattie DM, Schwartz L. A single center randomized trial assessing use of a vascular hemostasis device vs. conventional manual compression following PTCA: what are the potential resource savings? Cathet Cardiovasc Diagn. 1995;34:210-214

27. Christensen B, Lacarella C, Manion R, Bruhn-Ding B, Meyer S, Wilson R. Sandbags do not prevent complications after catheterization. Circulation. 1994;90:I-205

28. Bogart DB, Bogart MA, Miller JT, Farrar MW, Barr WK, Montgomery MA. Femoral artery catheterization complications: a study of 503 consecutive patients. Cathet Cardiovasc Diagn. 1995;34:8-13

29. Semler HJ: Transfemoral catheterization. Mechanical versus manual control of bleeding. Radiology. 1985;154:234-235

30. Pracyk JB, Wall, TC, Longabaugh JP, Tice FD, Hochrein J, Cox GC, Lee KL, Tcheng JE. Femoral vascular complications following coronary intervention: Is mechanical clamp hemostasis better than hand pressure ? Circulation. 1996;94:I-377

31. Ismail S, Combs MJ, Goodman NC, Teotia SS, Teates CD, Abbott RD, Fechner RE, Nolan SP, Powers ER, Spotnitz WD. Reduction of femoral arterial bleeding post catheterization using percutaneous application of fibrin sealant. Cathet Cardiovasc Diagn. 1995;34:88-95

32. Kipshizde N, Ferguson JJ, Macris MP, Clubb F, Cloy M, Horn J, Cummins F, Keane S, Nikolaychik V, Baker JE. Percutaneous delivery of a biosealant to achieve peripheral artery hemostasis: Experimental and clinical studies. Circulation. 1995;92:I-410

33. Gershony G, Kasprzyk DJ, Hussain HM, Powell J, Horzewski J. A Novel vascular sealing device for closure of arterial puncture sites. Circulation. 1995;92:I-409

34. Goy JJ, Debbas N, Depairon M, Mische H, Bottum P, Desombre L. Preliminary results with a new arterial sealing device. Circulation. 1995;92: I-56

35. Carere RG, Webb JG, Ahmed T, Dodek AA. Initial experience using Prostar™: A new device for percutaneous suture-mediated closure of arterial puncture sites. Cathet Cardiovasc Diagn. 1996;37:367-372

36. Vetter J, Ribeiro E, Hinohara T, Carrozza Jr.J, Simpson J. Suture mediated percutaneous closure of femoral artery access sites in fully anticoagulated patients following coronary interventions. Circulation. 1994;90:I-621

37. Cattelaens N, Gerckens U, Müller R, Staberock M, Grzan O, Lampe EG, Simpson J, Grube E. The Prostar plus percutaneous closure device versus manual compression following coronary interventions. Circulation. 1996;94:I-484

38. Gerckens U, Cattelaens N, Müller R, Staberock M, Grzan O, Lampe EG, Simpson J, Grube E. Early ambulation following elective diagnostic coronary angiography using a percutaneous arterial closure device (Techstar): A randomized trial versus manual compression. Circulation. 1996;94:I-484

39. Loubeyre Ch, Fajadet J, Karam C, Jordan Ch, Cassagneau B, Laurent J-P, Marco J. Prospective use of a percutaneous vascular closure device in patients undergoing PTCA and stenting. Circulation. 1996;94:I-376

40. Gerckens U, Cattelaens N, Müller R, Lampe EG, Grube E. Perkutaner Nahtverschluß von Femoralarterienzugängen nach diagnostischer Herzkatheteruntersuchung oder Koronarintervention. Dtsch med Wschr. 1996;121:1487-1491

41. Abbott WM, Austen WG. The effectiveness and mechanism of collagen- induced topical hemostasis. Surgery. 1975;78:723-729

42. Peacock E, Siegler H, Biggers P. Use of tanned collagen sponges in the treatment of liver injuries. An Surg. 1965;161:238-243

43. Silverstein ME, Chvapil M. Experimental and clinical experiences with collagen fleece as a hemostatic agent. J Trauma. 1981;21:388-393

44. Chvapil M, Holubec H. Effect of hemostatic fleece on 14C-Serotonin release by human platelets. Jpn Pharmacol Ther. 1990;18:57(2913)- 62(2918)

45. Chvapil M, Chvapil TA. Hemostatic effectiveness of hemostatic collagen fleece (Novacol) in heparinized and Aspirin treated rabbits. Jpn Pharmacol. Ther 1990;18:43(2899)-48(2904)

46. Chvapil M. Tissue reaction and biodegradation of implanted hemostatic collagen fleece in rats. Jpn Pharmacol Ther. 1990;18:179 (3927)- 192(3940)

47. Kitamura K, Yasuoka R, Ohara M, Shimotsuma M, Hagiwara A, Yamane T, Yamaguchi T, Takahashi T. How safe are the xenogeneic hemostats? - Report of a case of severe systemic allergic reaction. Surg Today. 1995;25: 433-435

48. Silber S, Dörr R, Rösch A, Mühling H. Complications after collagen plugging for hemostasis following femoral puncture: Experience in over 3000 patients. Circulation. 1995;92:I-56

49. Kensey KR, Evans DE, McGill LD, Nash JC. In vivo feasibility testing of a bioresorbable hemostatic puncture closure device. J Am Coll Cardiol. 1991;17:263A

50. de Swart H, Dijkman L, van Ommen V, Bär F, Wellens H. The hemostatic puncture closure device causes similar changes in femoral artery flow as a conventional pressure bandage: Results of a randomized study. J Am Coll Cardiol. 1994;22:286A

51. Aker UT, Kensey KR, Heuser RR, Sandza JG, Kussmaul WG 3rd. Immediate arterial hemostasis after cardiac catheterization: initial experience with a new puncture closure device. Cathet Cardiovasc Diagn. 1994;31: 228-232

52. Gibbs JSR, Slade AKB, Blake J, Nordrehaug JE, Rickards AF, Buller NP, Sigwart U. Femoral arterial hemostasis using a collagen plug after coronary artery stent implantation. J Interven Cardiol. 1992;5:85-88

53. Kühn C, Sümpelmann D, Geiger B, Siems R, Kunze KP, Geiger M, Mathey D, Schofer J. Frühzeitige Blutstillung nach koronartherapeutischen Eingriffen durch Anwendung von Kollagenplugs. Z Kardiol. 1993;82:515- 520

54. Chevalier B, Lancelin B, Berthaux X. Hemostatic puncture closure device versus regular compression: A randomized study. J Am Coll Cardiol. 1995;23:93A

55. Blengino S, Hann B, Maiello L, Nakamura S, Hall P, Biagi G, Finci L, Colombo A. A randomized study of the 8 French hemostatic puncture closure device vs manual compression after coronary interventions. J Am Coll Cardiol. 1995;25:262A

56. Silber S, Luckas C, Dörr R, Mühling H, Zindler G. Intra-arterial anchoring is superior to plain collagen plugging for sealing arterial puncture sites and ambulating after 20 minutes. Circulation. 1995;92: I-225

57. Condon JV, Elsner GB, Nootens MT, Gowan SA, Coverdale J, Linnemeier TJ, Rothbaum DA. Comparison of hemostasis time and vascular complications between the AngioSeal hemostatic puncture closure device and manual pressure. Circulation. 1995;92:I-599

58. de Swart H, Dijkman L, van Ommen V, Bär F, Wellens H. The hemostatic puncture closure device shortens time to hemostasis and ambulation after arterial catheterization: Results of a randomized study. J Am Coll Cardiol. 1994;22:355A

59. Nagtegaal EM, Schalij MJ, Buis B. Routine use of collagen to seal the femoral artery puncture site after percutaneous transluminal coronary angioplasty in fully anticoagulated patients; a clinical evaluation. J Am Coll Cardiol. 1993;21:231A

60. Johnson LW, Krone R. Cardiac Catheterization 1991: A report of the registry of the Society for Cardiac Angiography and Interventions. Cathet Cardiovasc Diagn. 1993;28:219-220

61. Bourassa MG, Noble J. Complication rate of coronary angiography: a review of 5250 cases studied by a percutaneous femoral technique. Circulation. 1976;53:106-114

62. Davis K, Kennedy JW, Kemp HG, Judkins MP, Gosselin AJ, Killip T. Complications of coronary arteriography from the collaborative Study of Coronary Artery Surgery (CASS). Circulation. 1979;59:1105-1112

63. Gersh BJ, Kronmal RA, Frye RL, Schaff HV, Ryan TJ, Gosselin AJ, Kaiser GC,
Killip T 3d. Coronary arteriography and coronary artery bypass surgery: morbidity and mortality in patients aged 65 years or older. Circulation. 1983;67:483-491

64. Green GS, McKinnon CM, Rosch J, Judkins MP. Complications of selective percutaneous transfemoral coronary arteriography and their prevention: a review of 445 consecutive examinations. Circulation. 1972;45: 552-557

65. Wyman RM, Safian RP, Portway V, Skillman JJ, McKay RG, Baim DS. Current complications of diagnostic and therapeutic cardiac catheterization. J Am Coll Cardiol. 1988;12:1400-1406

66. Kern MJ, Cohen M, Talley JD, Litvack F, Serota H, Aguirre F, Deligonul U, Bashore TM. Early ambulation after 5 French diagnostic cardiac catheterization: Results of a multicenter trial. Am J Cardiol. 1990;15:1475- 1483

67. Geschwind HJ. Percutaneous arterial approach revisited. Europ Heart J. 1995;16:579-580

68. Legrand V, Doneux P, Tilman Chu S. Comparison of puncture site closure with collagen plug (Vasoseal) or by early manual compression following PTCA. Circulation. 1993;88:I-72

69. Muller DWM, Shamir KJ, Ellis SG, Topol EJ. Peripheral vascular complications after conventional and complex percutaneous coronary interventional procedures. Am J Cardiol. 1992;69:63-68

70. Kadel C, Burger W, Skupin M, Kaltenbach M, Schräder R. Sealing of femoral artery puncture site with percutaneously applied collagen in 1000 patients: Complications requiring surgical repair. Circulation. 1988;88: I-251

71. Stiel GM, Beythien C, Kalkowski H, Hamper K, Rohwer HD, Nienaber CA, Hamm CW. Periphere Embolie von Hämostasekollagen (VasoSeal). Z Kardiol. 1992;81:543-545

72. Kiemeneij F, Laarman GJ. Improved anticoagulation management after Palmaz Schatz coronary stent implantation by sealing the arterial puncture site with a vascular hemostasis device. Cathet Cardiovasc Diagn. 1993;30: 317-322

73. Carere R., Webb J, Dodek A. Collagen plug closure of femoral arterial punctures. Are complications excessive? Circulation. 1994;90:I-621

74. Spokojny AM, Fahey F, Gibbs HH, Molloy T, Shaftel PA, Barra L, Sanborn TA. Use of collagen plug immediately post angioplasty in heparinized patients allows earlier discharge. Circulation. 1993;88:I-252

Address for reprints:
Sigmund Silber, MD
Associate Professor of Medicine
Dr. Müller Hospital
Am Isarkanal 36
81379 Munich, Germany
Tel.: 0049-89-2908310
Fax: 0049-89-2904202
email: silber@med.de