According to the anatomical studies of the distribution of myometrial blood vessels, arcuate arteries run transversely from lateral to medial, to join at the median plane. They give off radial arteries and spiral arterioles, which course centripetally towards the endometrium in also a transverse plane [7,23,24]. Thus, at least theoretically, a longitudinal uterine incision will transect a large number of transversely running arteries and arterioles. In comparison, a transverse uterine incision will parallel their course, leading to the transection of a fewer number [9].
Moreover, suturing of a vertical incision places the sutures parallel to the course of the transversely running vessels, which hinders adequate ligation of all transected arteries and veins, together with their branches and tributaries respectively. On the contrary, suturing transverse incisions places the sutures perpendicular to them, and thus, provides an effective hemostatic seal that surround all transected vessels [9].
However, all these assumptions were based upon anatomical studies of normal uteri. Few studies addressed the distortion of myometrial vascular anatomy by the growing myomas. Studies of the vascular pattern of myomatous uteri by injection, micro-radiographic and histologic techniques revealed absence of an intrinsic vascular network to supply the fibroid mass. However, vascular supply depends upon localized expansion of the normal myometrial vasculature, where surrounding displaced and distorted arcuate and radial arteries give off a number of small arteries that penetrate the myoma anywhere along its circumference [25]. Angiographic study by Discepola and his colleagues, of the course of the arterial blood vessels encircling uterine myomas revealed that the neo-vessels, developing around myomas, do not follow the normal anatomy of myometrial vasculature. Instead, on the surface of anterior and posterior wall myomas, blood vessels run mostly diagonally, e.g. 49.2% of the blood vessels on the surface of posterior wall myomas run diagonally with an angle 30-60◦ with the horizontal plane, and 21.7% run mostly horizontally with an angle 0-30◦; whereas 29.2% run mostly vertically with an angle 60-90◦. In addition, nearly 40% of the vessels crossed the myoma midline [26].
Based on these findings, Discepola and his colleagues concluded that arterial vessels on the surface of myomas can be injured whatever the direction of uterine incision[26]. However, it seems that this conclusion stems more from a radiological perspective, rather than a surgical one. Since the inherent nature of surgery implies injury of some blood vessels, and it is the art of surgery that strives for respecting anatomy and minimizing tissue damage, especially to large vessels and vital structures. In fact, it is logically plausible from a surgical perspective that if a major portion of blood vessels runs with an angle between 0-60◦, a transverse incision would be less traumatic than a longitudinal incision transecting all horizontally- and transversely-directed arterial vessels. Moreover, the conclusion of Discepola and his colleagues is further limited by the fact arteriographic techniques usually fail to visualize veins and small arterial vessels. They are usually limited to arterial vessels with a diameter exceeding that of the 5-F catheter (1.67 mm).[26]
Micro-radiographic and histologic studies revealed that the venous drainage of fibroids is sparse compared to the arterial supply, with only few moderate-sized draining veins are occasionally seen. However, intramural fibroids cause significant compression and distention of the surrounding arcuate venous plexus. These distended venous channels contribute substantially to the heavy bleeding during myomectomy [25]. Thus, a transverse uterine incision would be least traumatic to the transversely directed myometrial arteries, aretrioles and venous channels, and to the diagonally directed fibroid feeding vessels.
This theory was previously questioned by Morita and his colleagues in the setting of laparoscopic myomectomy. Transverse uterine incisions were associated with significant decline in operative blood loss compared to longitudinal incisions (137.6 ± 88.1 vs. 235.8 ± 169.4 mL respectively). This decline was more evident in large myomas (> 7 cm in diameter); with an estimated blood loss 158.9 ± 87.1 mL in transverse incisions, compared to 362.3 ± 147.3 mL in longitudinal incisions. Whereas no statistically significant differences were found between either incisions in smaller myoma (< 7 cm).[17] Logically, as the uterine incision is extended longitudinally, its traumatic effect is more pronounced, with the greater transection of traversing vessels; in contrast to the transection of vessels of the same axial plane with extension of the transverse incision. This effect might not be evident with smaller myomas. However, validation by further studies is needed, as the proposed effect of myoma size in the study of Morita [17] is confounded by the large difference in operative time with myomas > 7 cm between longitudinal incision (165.4 min) and transverse incision (129 min), owing to the easier suturing technique of transverse incisions relative to the longitudinal ones.
Despite the disparities between the laparoscopic and open approaches for myomectomy, in addition to pretreatment with gonadotrophin-releasing hormone analogs and intra-myometrial vasopressin injection in the study of Morita; however, the underlying surgical theory of transection of more transversely running vessels with longitudinal incisions might also be held true in case of open myomectomies.
In consistence with the same theory, the results of this study report the absence of statistically significant differences in intra-operative blood loss using either incisions. It should be noted that only myomas < 10 cm in diameter were included in our study, with an average myoma size of approximately 8 cm.
Postoperative pyrexia following myomectomy has been linked to the development of myoma-bed hematoma [21], which is linked with the subsequent weakening of the developing scar [12]. In this study, postoperative pyrexia developed in one patient in the transverse incision group and in two patients in the longitudinal incision group. However, being a secondary outcome, our sample size is underpowered to detect a significant difference in the incidence of such uncommon complication.
To sum up, based on our results, it is evident that blood loss is comparable using transverse or longitudinal incisions, at least with myomas ranging < 10 cm in size. This provides the base for further studies investigating the impact of incision direction on blood loss in larger myomas, incidence of myoma-bed hematoma, obstetric-quality of the ensuing scar, and formation of pelvic adhesions.
Among the strengths of this study: (1) up to our knowledge, this is the first study to explore the effect of uterine incision direction in open myomectomy; (2) homogeneity of patients within the two groups regarding the number, size and site of myomas; (3) estimation of intraoperative blood loss both directly and indirectly. Limitations include: (1) although we tried to restrict the range of myoma sizes, we failed to statistically adjust blood loss to varying myoma size due to limited sample size; (2) different surgeons were involved in the study, although all had similar adequate experience.