The reconstruction of a distal finger injury is challenging for hand surgeons due to the complicated structure of the distal finger as well as the scarcity of the available surrounding tissue. Several surgical techniques, such as cross-finger flaps, thenar flaps, homodigital flaps, and free flaps, have been proposed to reconstruct distal dorsal finger defects [1]. However, these surgical techniques have several limitations. For example, a cross-finger flap has the advantages of easy harvesting and reliable anatomy but requires two operative stages and an immobilization period, which causes stiffness of the fingers. Free flaps can cover any size of a finger defect and can be used in any site of finger defects, but they require a long operation time and are difficult surgically.

Dorsal metacarpal artery (DMA) flaps, such as the first DMA (FDMA) flap and the reverse DMA (RDMA) flap, have been used successfully for distal dorsal finger defects [2,3]. They can cover relatively large wounds and can enable the early mobilization of the affected finger, and they also have consistent anatomy [4]. However, Sebastin et al. [5] noted that in their clinical study, the caliber of the DMAs decreased and the incidence of congenital absence of the DMA increased from the first dorsal metacarpal space to the fourth dorsal metacarpal space. Although there have been several anatomic studies of DMAs, they were evaluated in cadavers from the Western [6,7]. There have been no studies on the anatomic variation of DMAs in Asian cadavers. Therefore, the purposes of this study were to evaluate the anatomy of DMA using Korean fresh cadavers and to report the clinical outcomes of DMA flaps.

In our study, five flaps had complete healing, and one flap had partial necrosis, which healed by the second intention. Several studies that evaluated the use of thumb reconstruction with FDMA flaps reported a 90% to 100% flap survival rate [8-10], and several studies that evaluated RDMA flaps also reported that the flap survival rates were nearly 88% to 93% [2,11]. These favorable results support the hypothesis that FDMA and RDMA flaps are reliable reconstruction methods for relatively large distal dorsal finger defects. We could identify that the DMAs have constant anatomy, running over the epimysium of the dorsal interosseous muscles, reliable as a source of pedicle flaps.

However, in our cadaver study, the ulnar branch of the FDMA was not identified in two of the four cadavers (50.0%). This rate was higher than that in other FDMA cadaveric studies. Murakami et al. [12] reported that the ulnar branch of the FDMA was not observed in 14 of 62 cadaveric hands (22.5%). Sherif [13] reported that in their clinical study, the ulnar branch of the FDMA was not observed in four of 18 patients (22.2%). Therefore, the higher rate in our cadaver study may be attributed to the small number of specimens. Since the FDMA could be congenitally absent, it is very important to check for the presence of the ulnar branch of the FDMA preoperatively by using Doppler ultrasound. In our study, the second and third DMAs were not identified in two of the four cadavers (50.0%). Beldame et al. [14] reported that in their cadaver study, the third and fourth DMAs were not observed in 15.3% and 57.2% of the cadavers, respectively [14], and Dauphin and Casoli [6] reported that in their cadaver study, the third and fourth DMAs were not observed in 3% and 14% of the cadavers, respectively. Sebastin et al. [5] and Couceiro et al. [15] also noted that a preoperative Doppler examination is mandatory when considering the use of DMA flaps. As we mentioned earlier, in our study, the Doppler ultrasound was performed before every operation and the success rate of the operation was also high. If there is no ulnar branch of FDMA at preoperative check, it is recommended to consider a free flap.

Ghoraba and Mahmoud [8] noted that the FDMA has two venae comitantes that usually have a connection with the large cutaneous superficial veins. In many of the studies that have used FDMA flaps, one or two dorsal cutaneous veins were included in the subcutaneous pedicle to prevent venous pathologies. However, while some articles have proposed that the superficial dorsal vein should be included in the pedicle [16,17], in other studies, it was not essential [10,18]. In our two patients that had FDMA flaps, the superficial dorsal vein was included in the pedicle, and in the other two patients, it was not included. In cases that did not include the superficial dorsal vein in the pedicle, the flaps healed well without venous pathologies. Thus, venous complications may not be associated with the inclusion of the superficial dorsal veins with the subcutaneous pedicle. Meanwhile, since venae comitantes are very small and thin, they can be easily damaged when dividing the superficial dorsal veins from the subcutaneous pedicle if the superficial dorsal vein has the same course as the pedicle. Therefore, in such a case, it would be better to include the dorsal vein within the pedicle.

Generally, the RDMA flap’s pivot point is usually 0.5 to 1 cm proximal to the adjacent MCP joint, where there are distal connections between the palmar metacarpal arteries and the dorsal metacarpal arteries [19]. Omokawa et al. [20] reported that the DMAs were consistently connected to the palmar arterial system at the level of the metacarpal head; however, some articles have reported that the connection of the DMAs and the palmar arterial system was observed to be more proximal to the level of the MCP joint. Yoon et al. noted that the fourth DMA anastomoses with the vessels branching from the deep palmar arch at the level of the metacarpal shaft [21]. Dauphin and Casoli [6] reported that palmar perforators anastomose with the DMA at the proximal part of the intermetacarpal space. In our cadaver study, we also observed the connection between the third DMA and the deep palmar arch vessels at the level of the metacarpal shaft. The connecting vessels are very small and can be easily torn during dissection. When the branching vessels are injured, the vessel pedicle can also be damaged during hemostasis. Therefore, when elevating an RDMA flap, careful dissection is necessary during elevating the flap proximal to the distal direction, until it reaches the MCP joint.

There are several limitations to our study. First, it was a retrospectively designed study, and second, our study was a small-scale study. Third, the mean follow-up period was 6 months, which was very short to evaluate any long-term complications such as scar contracture. Despite these disadvantages, to the best of our knowledge, this is the first study to evaluate the DMA anatomy using Asian fresh cadavers.

The good surgical results that were found in this study showed that DMA flaps are a reliable flap for the reconstruction of relatively large soft tissue defects of the dorsal finger. The absence of the DMAs in some cadavers was noted in our cadaver study. Therefore, a preoperative Doppler examination is mandatory when considering the use of DMA flaps.