Overall schema of mechanically coupled end-to-side anastomosis via a microvascular anastomotic coupling device. Adapted from Ragnarsson et al.10
Tattooed radial forearm fasciocutaneous free flap in situ.
Flap after microvascular anastomosis before insetting into pharyngeal defect. Pedicle passed beneath McFee incision. (Arterial end-to-end anastomosis in midneck.)
Close-up view of mechanical end-to-side anastomosis into internal jugular vein stump in neck base.
Radial forearm fasciocutaneous free-flap pharyngeal reconstruction. Mechanical end-to-side anastomosis to midcervical internal jugular vein.
Close-up view of mechanical end-to-side anastomosis into midcervical internal jugular vein.
DeLacure MD, Kuriakose MA, Spies AL. Clinical Experience in End-to-Side Venous Anastomoses With a Microvascular Anastomotic Coupling Device in Head and Neck Reconstruction. Arch Otolaryngol Head Neck Surg. 1999;125(8):869-872. doi:10.1001/archotol.125.8.869
Microvascular anastomosis remains one of the most technically sensitive aspects of free-tissue transfer reconstructions. Despite the availability of various mechanical anastomotic coupling systems for human clinical use during the last 8 years, reported clinical series remain rare.
To describe a clinical experience in applying a mechanical microvascular anastomotic coupling device (MACD) to end-to-side anastomotic configurations in head and neck free-flap reconstruction.
The MACD is a readily available high-density polyethylene ring–stainless steel pin system that has been found to be highly effective in clinical studies of end-to-end arterial and venous anastomosis and in laboratory studies of end-to-side anastomosis of rabbit arteries.
Thirty-seven end-to-side venous anastomoses were attempted, of which 33 (89%) were completed. Of these, 9 patients had critical anastomoses (only 1 venous anastomosis per patient). In patients undergoing parallel venous anastomoses, 6 had both anastomoses performed using the MACD; in the remaining 12 patients, 1 of the anastomoses was performed using the MACD. A variety of donor flaps and clinical contexts were encountered. Flap survival in the MACD series was 100%. Four anastomoses were converted to conventional suture technique intraoperatively.
The MACD is well suited to end-to-side venous anastomosis when carefully and selectively used by experienced microvascular surgeons. The most common clinical situation requiring this configuration and technique was the lack of ipsilateral recipient veins for end-to-end anastomosis or a ligated internal jugular vein stump that required this approach for device application. Previous radiation therapy does not appear to be a contraindication to its use.
DESPITE numerous refinements in microsurgical technique and instrumentation, the microvascular anastomosis remains one of the most technically sensitive aspects of free-tissue transfer reconstructions. Concurrent with the development of these techniques, various anastomotic coupling systems have been introduced in an effort to facilitate the performance and reliability of microvascular anastomoses. Introduced in 1962, the Nakayama device consisted of 2 metallic rings and 12 interlocking pins and holes.1 This device mechanically approximated vessel ends that were passed through the opposing rings and everted over the device's pins with the metallic ring-pin device remaining in situ as a permanent implant. In 1979, Ostrup and Berggren2 introduced an adaptation of this device (Unilink) that subsequently evolved into a microvascular anastomotic coupling device (MACD) (3M Precise; 3M Healthcare, St Paul, Minn). The device was subsequently acquired in 1996 by the Medical Companies Alliance Corporation (Bessemer, Ala).
The MACD is a generally available, high-density, polyethylene ring–stainless steel pin system that has been found to be highly effective in laboratory animal and human clinical studies. Histologic evaluation of mechanical anastomoses performed with the MACD has documented the microvascular repair process to be analogous to sutured vessels.3,4 Tensile burst strength studies have shown mechanically coupled anastomoses to be 50% stronger than sutured vessels 16 weeks postoperatively.5 Laboratory animal studies have documented equal or greater patency in mechanically coupled anastomoses as well as faster performance6,7 in normal and irradiated vessels.6- 10
Although the MACD has been available for human clinical use for the last 8 years, reported clinical use in human subjects remains rare.11- 17 Nylander et al11 reported a favorable experience in hand surgery. Sasson et al15 reported a general experience with the 2.5-mm device in 9 patients. Ahn et al12 presented a general experience with the device in 100 patients. DeLacure et al17 reported on clinical experience with the MACD in end-to-end arterial and venous anastomosis of 29 head and neck free-tissue transfers.
With the increased application of free-tissue transfer techniques in head and neck reconstruction, previously treated patients (surgery and/or radiotherapy) constitute an increasing proportion of individuals undergoing free-flap surgery. Such patients often exhibit fibrosis, scar tissue encasement, and ligation or harvest of potential recipient vessels, seriously limiting the choice of vessels with regard to caliber, length, and anastomotic configuration. Due to caliber mismatch or to the unavailability of recipient ipsilateral veins other than the internal jugular, end-to-side anastomotic configuration is often necessary (Figure 1).
In 1989, Ragnarsson et al10 reported their laboratory experience with end-to-side arterial anastomosis in rabbits. In their model, a balloon angioplasty was used to facilitate the technique. To our knowledge, a significant human clinical series of end-to-side anastomoses performed with the MACD has yet to be reported. Herein, we report our specific clinical experience with the MACD in 33 end-to-side venous anastomoses in 27 head and neck free-tissue transfers.
We retrospectively reviewed operative reports, microsurgical records, and hospital charts for 27 patients undergoing free-tissue transfer reconstruction of head and neck defects using the MACD at the Roswell Park Cancer Institute, Buffalo, NY, from June 1, 1995, through April 30, 1998.
There were 5 women and 22 men; average age was 53 years (range, 29-86 years). Indications for free-tissue transfer reconstruction included defects subsequent to primary and recurrent cancer ablation (n=26) and cervical esophageal reconstruction (n=1). Eight patients had received full-dose external beam therapeutic radiation to the recipient field in the head and neck before reconstruction.
Free-flap donor sites included radial forearm (n=22), fibula (n=2), rectus abdominus (n=1), jejunum (n=1), and latissimus dorsi (n=1).
Thirty-seven venous anastomoses were attempted with the MACD. Recipient vessels included the ligated trunk of the internal jugular system (n=2) (Figure 2, Figure 3, and Figure 4) and the internal jugular vein proper (n=35) (Figure 5 and Figure 6). Vein grafts were not used in this series.
Several technical modifications need to be considered in adapting the MACD for end-to-side configuration. Adequate mobilization of the recipient vessel should be performed for a tension-free anastomosis. Implant size is determined by the outer diameter of the donor vessels. A transverse venotomy without the removal of any vessel wall is performed on the recipient vessel with an 11-blade scalpel or adventitia scissors to the internal diameter of the implant ring selected. It is advantageous to begin by placing the recipient (side) vessel on the implant pins first, followed by the donor (end) vessel. Pin placement should be started at the ends of the venotomy, followed by the sides of the transverse incision. In vessels with thick and noncompliant walls, the sequence of pin placement should start from the sides of transverse venotomy, followed by the ends of the venotomy.
Thirty-three (89%) of 37 attempted end-to-side venous anastomoses were completed using the MACD. Four anastomoses were abandoned intraoperatively due to thickened or friable vessel walls that were less pliable and did not evert well over the device pins. These 4 anastomoses were successfully completed using conventional suture technique.
This series included 27 patients with 33 successfully completed end-to-side venous anastomoses using the MACD. Of the 27 patients, 9 had critical anastomoses (9 MACDs); 6 patients had 2 parallel anastomoses performed with the MACD (12 MACDs); and the remaining 12 patients had 2 parallel anastomoses, of which 1 was performed using the MACD (12 MACDs). Critical anastomoses were those with only 1 venous anastomosis and that were performed using the MACD. The patency of these anastomoses was absolutely critical for flap survival. The overall success rate at the anastomotic level in patients with critical anastomoses was 100% (9/9). In patients undergoing parallel venous anastomoses, it is impossible to determine clinically whether one of the veins had thrombosed, and these flaps are, therefore, excluded from anastomotic patency statistics. Flap survival in the 27 patients was 100%.
The 2.5-mm coupling device implant was used most frequently (24 anastomoses), followed by the 2.0-mm (6 anastomoses) and 1.5-mm rings (3 anastomoses). The average time to complete an anastomosis was 5 minutes. There were no cases of anastomotic disruption of the device, leaks, or hematomas in this series.
The MACD system for microvascular anastomosis has been effective in a diverse patient population undergoing free-tissue transfer reconstruction of head and neck defects. Technical advantages have included an overall simplification of microsurgical anastomotic technique, shorter anastomotic time, decreased flap ischemia time, and lower requirements for working space. Facility in application of the device, as with any new technique, involves a learning curve, even for the experienced microsurgeon.
Our overall experience has underlined the need for experience and judgment in selective application of the device. General principles include tension-free anastomotic geometry, proper incident angle (45°-90°) of the donor vessel in relation to the recipient one, the avoidance of torquing when impaling the donor vessel on the device, and satisfactory mobilization of a segment of the recipient vein for application of a curved vascular clamp if needed. This last point is particularly important in the internal jugular system with its relatively high-flow volume and pressure, which often required a Satinsky or curved bulldog-type vascular clamp for adequate occlusion.
The implant is sized to the outer diameter of the donor vessel. The choice of a ring that is smaller than optimal for a particular vessel diameter may result in gross uneven redundancy and "pleating" of the vessel on the ring, predisposing to anastomotic failure. Conversely, the choice of a device that is larger than optimal for a given vessel diameter may result in traumatic tearing of the vessel wall as attempts fail to accommodate it to the ring pins.
Several departures from end-to-end technical applications and standard vascular surgical principles are recommended. Although the lumen of the side vessel is slightly narrowed with the end-to-side application of this device, functionally significant coarctation of the vessel is not observed in this context (internal jugular vein). Additional mobilization of the recipient vessel is required to accommodate placement of a curved vascular clamp or of 2 straight vascular clamps far enough away from the device to allow tension-free application of the device to the recipient vessel. Otherwise a taut trampoline mat or drumheadlike situation is set up with inability to execute the anastomosis with the device in this configuration. A transverse venotomy (as opposed to axial venotomy in suture technique), without the removal of any vessel wall tissue, is performed. This is to minimize the tendency for vessel wall tearing and to maximize vessel wall available for device application while minimizing coarctation of the recipient vein. Thick recipient vessel walls are not uncommon in retreatment situations and may be a relative contraindication to the use of the MACD. For those contemplating application to the arterial side, significant atherosclerotic disease may be a relative contraindication to its use in an end-to-side configuration.
The sequence of pin placement is also important, with vessel wall impaled onto the pins at 1 end of the transverse venotomy followed by the other end, and then completing the procedure by impaling the vessel onto the pins at the sides of the transverse incision. Veins should be pliable and soft to allow eversion onto the device pins. In this aspect of device application, unsatisfactory vessel wall quality may result in traumatic tearing of the vessel edge. This will be aggravated by the choice of a device whose diameter exceeds that which may be required to avoid excessive stretching of the vessel walls. Vessels with thickened and noncompliant walls are commonly encountered in fields undergoing previous irradiation or operation. Such vessels may also exhibit increased friability and tear on attempts to impale the wall on the device pins. In such cases, the sequence of pin placement may be reversed (ie, pinning the vessel wall first to pins at the sides of the transverse venotomy, followed by those at 1 end of the venotomy and then the other end).
Although our study did not capture specific anastomotic completion times, it is observed generally that 3 to 5 minutes is required with the use of MACDs. This is in contrast to suture technique, which may require 2 to 3 times as long to complete. It is important to recognize the multiple factors that contribute to such measurements, regardless of technique. We have observed that use of the coupling device has resulted in decreased overall flap ischemia times.
Most of the mechanical technical difficulties occurred early in our application of the device; however, additional clinical experience in these contexts has allowed us to use the device selectively. Overall, a multifactorial confluence of smaller technical errors may result in vessel torquing after apparent successful completion of an anastomosis, predisposing, again, to anastomotic failure.
Our experience with the MACD has underlined the fundamental requirement for formal microsurgical training, experience, and mastery of conventional suture techniques, as such a device might appear to be an attractive alternative to the occasional microsurgeon. One should also be expert in application of the device to all types of end-to-end arterial and venous anastomoses, as the end-to-side configuration is significantly more demanding technically. The cost of the device compares favorably with suture technique when additional operative time is considered. Our application of the device to venous end-to-side anastomosis in head and neck reconstruction compares favorably with previously reported success with the MACD (100% in our series; 98% per the manufacturer; and 97% and 99% in previous reports12,17).
Additional refinements of such devices should extend their application in clinical microsurgery. Ring-pin devices constructed of alternative nonparamagnetic metal pins (eg, titanium) and bioresorbable rings may be considered in the future. We plan on increasing our experience in application of the device in end-to-side anastomosis on the arterial side. When carefully and selectively used by an experienced microvascular surgeon, the MACD can be a safe, fast, and reliable adjunct in clinical reconstructive microsurgery of the head and neck region.
Accepted for publication July 27, 1998.
Presented as a poster at the 1998 Combined Meeting of the American Society of Head and Neck Surgeons and Society of Head and Neck Surgeons, Palm Beach, Fla, May 14-16, 1998.
We thank Wesley L. Hicks, Jr, DDS, MD, for the contribution of his patients to this study and for creating many of the challenging reconstructive situations requiring this technical approach, and Thom R. Loree, MD, for his technical assistance in the execution of this technique.
Reprints: Mark D. DeLacure, MD, Department of Otolaryngology, New York University School of Medicine, 530 First Ave, Suite 7U, Skirball Building, New York, NY 10016 (e-mail: firstname.lastname@example.org).