Total percutaneous access for endovascular aortic aneurysm repair (“Preclose” technique)

Methods 

All patients who underwent EVAR and TEVAR at a single tertiary care university medical center between October 2003 and August 2006 were identified by a retrospective review of a prospectively maintained endovascular database and medical records. A cohort of 183 consecutive patients who underwent percutaneous femoral access requiring 12F to 24F (4.4-mm to 8.6-mm outer diameter) managed with the Preclose technique using the Proglide device were compared with a consecutive, contemporaneous series of 154 patients who underwent open surgical exposure of their femoral arteries. The decision on which method of femoral access was used depended on anatomic factors and surgeon preference. Two of the four operators who contributed to the current case series preferentially used surgical exposure for their femoral access.

The distribution of devices is given in Table I. Although there were some differences in the frequency of devices between the two groups, the proportion of EVAR and TEVAR cases (P = .13) was similar. A variety of anesthetic techniques were used, depending on patient comorbidities and operator preferences, but the distribution of techniques for the Preclose and surgical groups, respectively, were similar: general, 49% vs 55%; regional, 45% vs 44%; and local, 5% vs 1% (P = .10). The two cohorts were analyzed on a per groin basis. The Preclose group was further divided into a group consisting of smaller 12F to 16F sheaths and a group consisting of larger 18F to 24F sheaths, which are the typical profiles required for thoracic and abdominal endograft iliac limbs and main devices, respectively.

Table I. Distribution of abdominal and thoracic endovascular devices between Preclose and surgical exposure groups

	Device	Preclose	Surgical	P
	EVAR
	Zenith⁎	81	63	.58
	Excluder†	28	42	.01
	AneuRx‡	2	3	.38
	Other	3	0	.25
	TEVAR
	TAG†	67	27	.0001
	Talent‡	1	11	.002
	TX2⁎	1	8	.01
	Total cases	183	154

EVAR, Endovascular repair; TEVAR, thoracic endovascular repair.

⁎ Cook, Bloomington, Ind. † W. L. Gore & Associates, Flagstaff, Ariz. ‡ Medtronic, Minneapolis, Minn.

Perioperative outcomes, procedure times, and operating room usage costs (exclusive of any devices or disposables) were examined. Procedure time was defined as the period of time from either incision (surgical exposure) or skin puncture (Preclose technique) to final dressing application. At our institution, the usage cost of the operating room is (US) $3935 for the first 60 minutes (not prorated for shorter periods) and then $50/min thereafter.

Technical success (in-hospital or 30-day) of the Preclose technique was defined as closure of the arteriotomy without the need for any access site–related adjunctive surgical or endovascular procedures stemming from hemorrhagic, infectious, or ischemic complications. Similarly, technical success for surgical exposures was defined as primary suture repair of the arteriotomy without the need for additional local reconstruction such as endarterectomy, patch angioplasty, or interposition grafting.

For both groups, access site–related complications such as seromas, infections, wound dehiscence requiring dressing changes, hematomas with or without transfusions, and prolongation of hospitalization were counted as failures of the therapy regardless of whether they required additional postoperative surgical or endovascular intervention. Moderate or large (>2 cm) asymptomatic or subclinical hematomas or seromas and arteriovenous fistulas were included as complications.

The first postoperative computed tomography (CT) scan was typically obtained at the 1-month follow-up visit. This and all subsequent CT scans always included the femoral vessels.

Statistical analyses were performed using the Student t test for continuous variables and the Fisher exact test for categoric variables. All aggregate values are given as mean ± standard deviation. Significance was achieved at P < .05. This study was approved by the Institutional Review Board.

Results 

Preclose technique group 

Between December 2004 and August 2006, 262 endovascular aortoiliac repairs were performed (137 EVAR, 118 TEVAR, 7 iliac). Of the 381 femoral arteries accessed for insertion of 12F to 24F sheaths, 279 (73.2%) were managed with 559 Proglide devices using the Preclose technique in 183 patients (105 EVAR, 71 TEVAR, 7 iliac). Four femoral arteries (1.4%) required only one device, 270 (96.8%) were closed with two devices, and five arteries (1.8%) required a third device. The four femoral arteries that used only one device all involved 12F sheaths and occurred early in our experience. Although all four cases were successful, two devices were routinely used for all subsequent cases. The distribution of sheath sizes is given in Fig 1. Large size sheaths (≥18F) comprised >63% (175/279) of the sheaths in the Preclose cohort.

View Large Image Download to PowerPoint

Fig 1. Distribution of sheath sizes in the Preclose (striped) and surgical (solid) cohort. There were significantly higher proportions of 16F sheaths in the Preclose group (*P = .03) and 18F sheaths in the surgical group (#P < .0001).

The overall technical success rate of the Preclose technique was 94.3% (263/279 femoral arteries). There were 16 complications (ie, failures) requiring open repair of 13 femoral arteries and emergency placement of a covered stent in two arteries for severe retroperitoneal hemorrhage. One case of necrotizing arteritis with a mycotic pseudoaneurysm presented on postoperative day 27 that required a common femoral artery replacement with autogenous femoral vein (Table II). Between the subsets of smaller and larger size sheaths, the technique was significantly more successful in the former group at 99% (103/104) vs 91.4% (160/175, P < .01; Fig 2). There were no other hematomas, pseudoaneurysms, or other access-related complications that required additional surgical or endovascular interventions. All cause mortality was 2.2% (4/183), but access-related mortality was 0%.

Table II. Detail of reasons for Preclose failure⁎

	No.	Sheath (F)	Reason(s) for failure	Treatment
	1	24	Coagulopathy, persistent bleeding	Open repair
	2	18	Coagulopathy, persistent bleeding	Open repair
	3†	24	High (suprainguinal) puncture
			Retroperitoneal hemorrhage	Endo
	4†	20	High (suprainguinal) puncture	Endo
			Retroperitoneal hemorrhage
			Tortuous iliac artery
	5	22	Focal dissection, limb ischemia	Open repair
	6	18	Obesity, subcutaneous deployment	Open repair
	7	16	Sutures pulled through artery	Open repair
	8	20	Focal dissection, limb ischemia	Open repair
	9	18	Low SFA deployment	Open repair
			Focal dissection
			Limb ischemia
	10	20	Severely scarred groin	Open repair
	11	24	Persistent bleeding	Open repair
	12	20	Suture pulled through artery	Open repair
	13	20	Suture pulled through artery	Open repair
	14	18	Suture pulled through artery	Open repair
	15	20	Suture pulled through artery	Open repair
	16	20	Mycotic pseudoaneurysm (POD 27)	Open repair

Endo, Endovascular repair; SFA, superficial femoral artery; POD, postoperative day.

⁎ 15 of the 16 failures occurred intraoperatively; only one patient (patient 16) required postoperative surgical repair, which was for a necrotizing arteritis using autogenous conduit. † A Viabahn (W. L. Gore & Associates, Flagstaff, Ariz) covered stent was used for the repair.

View Large Image Download to PowerPoint

Fig 2. Technical success rate by sheath size for the Preclose technique.

Surgical exposure group 

Between October 2003 and August 2006, 154 consecutive endovascular aortic repairs (108 EVAR and 46 TEVAR) that only involved femoral introduction of the devices were identified (total of 258 femoral exposures). Cases during this period that involved alternative access, such as iliac or aortic sites, were excluded. The distribution of sheath sizes among the 258 surgical femoral exposures was largely similar to the Preclose group except where noted (Fig 1). A total of 16 intraoperative and early postoperative complications resulted in a technical success rate of 93.8% (242/258; P = .86, compared with Preclose technique). The 16 complications consisted of 10 endarterectomies with patch angioplasties, 3 wound infections, 2 infected seromas requiring incision and drainage, and 1 severe arteritis that required débridement and replacement of the common femoral artery with an autogenous femoral vein. All cause mortality was 1.3% (2/154; P = .69, compared with Preclose technique) with 0% access-related mortality (Table III).

Table III. Comparison between subsets of Preclose and surgical exposure groups

	Technique	EVAR			TEVAR
		Preclose	Surgical	P	Preclose	Surgical	P
	Patients (n)	101	108		48	46
	Femoral arteries (n)	173	212		48	46
	Age (years)	71 ± 8	72 ± 9	.84	68 ± 15	72 ± 11	.11
	Male:Female	91:10	98:10	1.00	39:9	35:11	.62
	EIA (mm)	8.8 ± 1.3	8.8 ± 1.5	.96	9.4 ± 1.4	9.5 ± 1.7	.59
	Complication (%)⁎	11(11)	10(9.2)	.82	5(10)	6(13)	.76
	Mortality (%)†	1(1.0)	2(1.9)	1.00	3	0	.24
	Procedure time‡	115 ± 45	128 ± 31	<.001	80 ± 34	112 ± 48	.019
	OR cost	$6697 ± $2240	$7351 ± $1557	.01	$5087 ± $1533	$6556 ± $2399	<.001
	OR cost + Proglide§	$7881	$7351		$5679	$6556

EVAR, Endovascular repair; TEVAR, thoracic endovascular repair; EIA, external iliac artery diameter; OR, operating room.

⁎ Access-related failure/complication only. † All-cause mortality. ‡ Duration in minutes of endovascular procedure. § Mean only.

Discussion 

Percutaneous access during endovascular aortic repairs has been difficult because of the large sizes of the delivery systems. Avoidance of surgical femoral exposure may result in shorter procedure time, fewer wound complications, and increased patient comfort. The practical size limit of achieving hemostasis with manual compression alone is likely 12F (sheath), although this has never been formally studied. Among the various percutaneous arterial closure devices, suture-mediated devices offer the purported advantages of a permanent suture, the least amount of intravascular and extravascular foreign material, and similarity with conventional arterial repair.

In this study, we retrospectively examined the safety and effectiveness of percutaneous closure of femoral arteries after introduction of 12F to 24F sheaths using a suture-mediated closure device. Although the successful closure was slightly higher for 12F to 16F sheaths, the overall rate of technical success was >94% in a consecutive series of nearly 300 femoral arteries. Furthermore, almost all of the complications occurred intraoperatively and were amenable to treatment with surgical or endovascular methods, and there was no access-related mortality.

Several authors have previously described the Preclose technique of percutaneous femoral closure after large sheath access.1, 2, 3, 4, 5, 6, 7 The entire reported experience, however, has solely involved the use of the Perclose Prostar XL device, the technical success rates have varied widely from 62% to 100%, and similar with the current study, closures of smaller sheaths (<18F) were better than larger ones (Table IV). In contrast with the Proglide device used in the current study, the Prostar XL:

Table IV. Review of previously reported experience on percutaneous closures of access sites after endovascular repair (all using the Prostar XL)

	First author	Date	Vessels (N)	Sheath (F)	Success rate
	Haas1	1999	13	16,22	100%
	Traul2	2000	29	16,22,24	Overall,62% 16F,71% 22F,75% 24F,25%
	Teh4	2001	82	16-20,22	85%
	Howell3	2001	148	16	94%
	Kennedy6	2003	15	16	80%
	Torsello7	2003	27	14-24	87%
	Morasch5	2004	94	12,18	81%

Admittedly, all of these relative disadvantages can be overcome with proper technique and sufficient experience. The only advantage that the Prostar XL technique offers over the Proglide is that it typically requires only one device per femoral artery because there are already two sutures oriented in a cross-pattern and, therefore, there is a cost benefit ($425 per device vs $590 for 2 Proglide devices, a difference of $165).

Although review of the time points at which each of the complications occurred did not demonstrate a clear learning curve, the cases serve to illustrate a few key points about their management and the apparent risk factors for technical failure. Proper initial femoral puncture (ie, anterior aspect of the mid-common femoral artery) is critical. In the case of a suprainguinal puncture of the external iliac artery, the inguinal ligament can impede complete laying-down of the slipknot and the access site is too high for manual compression, leading to uncontrolled hemorrhage. This occurred in two cases. In both, this was not immediately recognized from the small amount of bleeding at the puncture site and the guidewire was removed. In both of these instances, the proximal superficial femoral artery was rapidly exposed through a medial longitudinal incision, guidewire access was re-established, and a short covered Viabahn stent (W. L. Gore & Associates, Flagstaff, Ariz) was deployed.

These cases notwithstanding, the ability to maintain guidewire access from the beginning to the end of the closure allows management of most hemorrhagic complications by reinsertion of a sheath or large dilator, restoration of hemostasis, and unhurried surgical repair of the arteriotomy. In contrast with a suprainguinal puncture, a low femoral puncture (ie, superficial femoral artery) can result in a flow-limiting dissection with limb ischemia due to the small size of the entry vessel.

Any anatomic configuration that necessitates a significant amount of pushing or torquing of the sheath or delivery catheter can increase the risk of failure of the Preclose technique. This occurs most commonly in patients with small or diseased iliac arteries, or severe iliac tortuosity, or both. The increased pressure and torque applied to the sheath may extend the size of the arteriotomy made by the sheath profile alone. Furthermore, it may also cause the Proglide sutures to actually pull out of the vessel altogether or reduce their purchase such that they are insufficient to reappose the arteriotomy.

We believe that the ability to completely reverse the anticoagulation is important to the technique. Similar to open surgery, formation of clot is essential for hemostasis. Two patients were coagulopathic despite infusions of plasma, platelets, and protamine. There was persistent bleeding and the artery repaired surgically. Direct examination of the vessel showed the two Proglide sutures deployed correctly and sutures securely tied, but pulsatile bleeding between the sutures and needle holes. We now consider coagulopathy a contraindication to this technique.

Severe calcifications, groin scarring, and obesity can also lead to technical complications (Fig 3). Groin scarring from prior catheterizations and, ironically, from previously placed percutaneous closure devices or surgery can cause misdeployment of the Proglide sutures due to the inability of the needles to penetrate through the arterial wall or the overlying scar tissue. This increased resistance can lead to the separation of the suture from the back end of the needle. Although obesity as a single measure is not necessarily a risk factor, as it relates to the depth of the subcutaneous tissue at the groin, it can negatively affect the ability to properly insert the Proglide device into the artery.

View Large Image Download to PowerPoint

Fig 3. Computed tomography images illustrate anatomic contraindications to the Preclose technique. A, Morbid obesity with thick pannus. B, Severe iliofemoral calcifications.

To summarize, we would consider contraindications to the procedure to be:

It was encouraging that there were so few infectious complications in the Preclose group. We attribute this to the sterile technique and environment of the operating room. Indeed, the only such complication in our series involved a polytrauma victim with a thoracic aortic transection that was repaired with a stent graft. The emergency conditions and a potentially unrecognized break in sterile technique may have contributed to the Perclose infection and resultant mycotic femoral pseudoaneurysm. As with any procedure involving implantable endovascular devices, the highest level of aseptic technique is critical to avoid limb and life-threatening infectious complications.

The main limitation of this study is its retrospective design and its inherent selection bias. The criteria for deciding who undergoes surgical exposures or the Preclose technique were not standardized. Arguably, the surgical group may have been anatomically disadvantaged owing to many of the same factors that made the Preclose technique unsuitable, such as obesity, scarring, and severe atherosclerotic disease as mentioned. Conversely, the only way we can truly eliminate this would be through a prospective, randomized study, which may be ultimately required before the merits of this technique can be scientifically validated.

The economics of percutaneous access for EVAR and TEVAR deserve mention. Measuring only the time-based “usage” cost of the operating room is artificial and not reflective of all the fiscal complexities that enter into the in-hospital cost of this expensive therapy. It seemed appropriate, however, because operating room time was the only potential cost end point affected for EVAR, with the obvious exception of the device cost itself. We purposely did not consider the cost of the Coons dilators, which being part of the Preclose technique, should be factored into the overall device cost. Conversely, this would not have materially altered the final conclusions of the cost analysis, which was that the Preclose technique was not necessarily cost-effective.

Currently, percutaneous closures are not reimbursed, regardless of what setting or device. The only justification from the hospital’s standpoint involves some aggregate qualitative and quantitative measures of time-savings, fixed-resource utilization, and patient satisfaction. From a physician’s standpoint, surgical femoral exposure (Current Procedural Terminology code 34812) during endovascular procedures is reimbursed at $373.82 (2006 Medicare Fee Schedule for Florida, Locality 1) per groin. This means a loss of nearly $750 per patient for EVAR. Unfortunately, this discrepancy may have implications in the current setting of declining physician reimbursement.

Conclusion 

This study represents, to our knowledge, the largest series of percutaneous endovascular aortic repairs using the Preclose technique and the first using the Proglide device. The results indicate that the technique is a safe and effective method of percutaneous arteriotomy closure after introduction of large sheaths in a select group of patients, with a higher technical success rate than historically reported with the Prostar XL device.

The technique is well tolerated by patients, with almost no postoperative discomfort typical of a groin incision and rapid return to normal activities. Although to date we have not had any late ischemic events due to secondary development of occlusive disease at the site of Perclose deployment, long term outcomes of femoral arteries closed with this technique remains unknown at this time and clinical vigilance is warranted.

Author contributions