Rethinking uterine compression suture for atonic postpartum hemorrhage.

Abstract

Postpartum hemorrhage (PPH) is one of the leading causes of maternal morbidity and mortality worldwide. An atonic uterus is the most common cause of severe PPH.1 The reference standard for treating atonic PPH is adequate uterotonics and uterine fundal massage, optionally using bimanual compression of the uterus, intrauterine balloon tamponade, and correction of coagulopathy. Patients who resist these treatments require arterial embolization or surgical interventions such as uterine compression sutures (UCS) or hysterectomy. In 1997, Christopher Balogun-Lynch et al reported an innovative hemostatic technique called the B-Lynch suture.2 They reported five cases of B-Lynch sutures (two after vaginal delivery and three after cesarean section) and stated that the B-Lynch suture is an alternative to hysterectomy.2 The B-Lynch suture is the original form of the various UCS available for intractable atonic PPH, and has saved women's lives and uteri. B-Lynch and Javaid explain that there are two mechanisms by which UCS can lead to hemostasis: one is to increase the contractions of the uterine muscle and the other is to block the vascular space of the uterus.3 Several modifications of the B-Lynch suture have been developed, and the success rates of UCS range from 61% to 100%, with no significant difference in suture methods.4-6 However, evidence for the benefits of UCS is scarce,7 and UCS have been reported with complications such as partial uterine necrosis, pelvic pain, synechiae, pyometra, endometriosis, and uterine rupture in a subsequent pregnancy.4 As the reasons for the success or failure of hemostasis with UCS have not been fully explored, we revisit UCS from two perspectives: what kind of atonic PPH really needs UCS and what surgical interventions are effective. First, what type of atonic PPH should UCS be performed for is a clinical conundrum. Atonic PPH remains a diagnosis of exclusion. In many cases, hemostasis can be achieved by increasing the contractions of the uterus, but there are some cases in which even if the uterus is well contracted and coagulopathy is corrected, hemostasis cannot be obtained.8 In the latter case, the uterine bleeding persists, causing blood clots to accumulate in the uterus, which results in seemingly atonic uteri despite the ability to contract.8 We have used dynamic CT in the evaluation of severe PPH over the past decade.8 Most of the cases were refractory to conventional treatment and were transferred to our hospital. In our study, no case of atonic PPH required arterial embolization or laparotomy when contrast extravasation was not observed in the arterial phase (n = 16).8 On the other hand, when the contrast agent was found to be leaking in the arterial phase (n = 14), arterial embolization was required in 50% of cases, and in 36% of cases, hemostasis was obtained by direct compression of the bleeding point with an intrauterine balloon.8 This indicates that there is a subgroup of atonic PPH in which it is difficult to stop bleeding with conventional treatment, which we have named PRACE (postpartum hemorrhage resistant to treatment showing arterial contrast extravasation on dynamic CT).8 These findings suggest that surgical interventions, such as UCS or hysterectomy, are not necessary in the absence of PRACE, though we have to acknowledge that CT has limited availability, particularly in the developing world, and that CT scans are not suitable in hemodynamically unstable patients. The question arises as to whether conventional UCS are effective for hemostasis for PRACE. We previously identified the exact locations of the bleeding site in women with PRACE.8 The study revealed that the site of arterial bleeding is usually a single point, not a plane.8 Blood is supplied to the uterus mainly from the uterine arteries, but also from the ovaries and vaginal arteries.9 These vessels anastomose with each other, branching from the lateral side of the uterus to the myometrium and running to the opposite side (Figure 1A).9 Considering the blood flow to the uterus and the location of the sutures, it is likely that UCS can achieve hemostasis if the site of bleeding is close to the center of the uterine cavity, but if the site of bleeding is more lateral than the sutures, then the UCS is likely to be ineffective for PRACE. Assuming that hemostasis is possible for bleeding from the central two-thirds of the uterine cavity, we investigated the 14 cases of PRACE previously reported8 to see if UCS can achieve hemostasis. Three of the eight PRACE cases (37.5%) from the upper uterus (upper PRACE) and all six of the PRACE cases (100%) from the lower uterus (lower PRACE) were bleeding from the outer third of the lumen, with nine of the 14 cases (64.3%) considered to have failed to achieve hemostasis even after UCS was performed (Figure 1B). These findings suggest that upper, not lower, PRACE is a relatively favorable candidate for UCS. Finally, we would like to address our opinion on the role of UCS when dealing with atonic PPH after vaginal and cesarean delivery, respectively. Vaginal delivery is reported to have the highest odds as a factor requiring hysterectomy despite having undergone UCS.5 This is probably because the bleeding points tend to be on the lateral side.8 Indeed, all but one case of lower PRACE was following a vaginal delivery,8 whereas more than half of the upper PRACE were after cesarean section.8 Hence, the success rate of hemostasis would be low if UCS were performed after vaginal delivery without knowing the bleeding point. The significance of UCS in cesarean delivery is not well understood. The number of UCS performed after cesarean delivery is 10 times greater than after vaginal delivery,5 probably because the procedure can be performed simultaneously at the time of delivery. This indicates that many of the UCS performed in cesarean deliveries are prophylactic interventions, and the effectiveness of prophylactic UCS for uterine atony is questionable in cesarean delivery.10 When the bleeding site is around the center of the upper uterus (Figure 1C), hemostasis is achieved with conventional UCS. If a dynamic CT scan reveals that a bleeding point is on the lateral side of the uterine cavity, a minimally invasive technique, such as simple suture targeting a single bleeding point using blunt straight needles (Figure 1D), rather than the conventional UCS, will be more effective in achieving hemostasis regardless of the mode of delivery. In summary, UCS are still being performed for atonic PPH despite the fact that there is little scientific evidence of their usefulness because of the apparently high success rate of UCS. Very few cases of PPH actually need UCS, and the effectiveness of UCS depends on the location of the bleeding point. The hemostatic suture targeting a single arterial bleeding point may be a more effective method for intractable atonic PPH requiring a hysterectomy. We believe that it is time to reconsider UCS aiming to achieve hemostasis “blindly”. We thank Yoshitsugu Chigusa, Kaoru Kawasaki, and Masaki Mandai for constructive feedback on this manuscript.