A 22-year-old nulliparous woman, who had undergone a dilatation and evacuation (D & E) at 16 weeks of gestation (mid-trimester termination of pregnancy) in a private clinic, presented to the gynaecological emergency room. A referral letter stated that she had sustained a uterine perforation with suspected bowel injury during the operative procedure. She was resuscitated with intravenous fluids before transfer and was generally stable on admission. On examination, she was found to have a soft abdomen, with a firm bulky 16-week size uterus and mild suprapubic tenderness. Her bowel sounds were normal. Speculum examination revealed mild active bleeding through the cervical os and bimanual examination revealed slight cervical excitation.
Further resuscitation was continued, and she was immediately transferred to theatre. A midline laparotomy revealed a 2-l haemoperitoneum of fresh blood and clots. There was an 8-cm uterine wall laceration extending from the left cornua across the fundus with active bleeding (Figure 1). The fallopian tubes and ovaries appeared normal. The sigmoid mesocolon was completely avulsed over an 8-cm section. The bowel serosa was bruised but not perforated. Fetal parts were found in the abdominal cavity.
The uterine and peritoneal cavities were evacuated and the uterine defect closed in two layers using number 1 vicryl sutures (Ethicon®, NJ, USA), with particular attention to secure the uterine angles. However, despite the suturing and the use of intravenously administered syntocinon, intramuscular administration of 500 μg ergometrine and 800 μg rectal misoprostol, haemostasis was not achieved. Bimanual compression, however, controlled the bleeding. As there was persistent moderate vaginal and uterine wound bleeding upon compression release and given that this was the woman’s first pregnancy, we opted to preserve her uterus if at all possible. Therefore, it was decided to apply a modified B-Lynch suture without opening the uterus.
In brief, the bladder was reflected and a number 1 vicryl suture passed transversely through the anterior cervix, then upwards through the fundus of the uterus at the edge of the perforation, down the back of the uterus and transversely through the uterosacral ligaments close to the uterus. It was then passed upwards through the other side of the uterine fundus and then downwards to the anterior cervix (Figure 2). The uterus was then compressed and the suture tightened. This controlled the bleeding. The intraoperative coagulation profile was normal. With an estimated intraoperative blood loss of around 2 l, she was initially transfused with 2 units of O negative blood and 2 units of fresh frozen plasma followed by 4 units of cross-matched blood.
In view of the trauma to the sigmoid colon, the surgical team performed a Hartman’s procedure resecting the traumatised section of bowel and creating an end colostomy in the left iliac fossa.
The woman made an excellent postoperative recovery and was discharged on day 7. She is currently awaiting a surgical reversal of her colostomy, a follow-up hysteroscopy to rule out the presence of uterine synechiae and a hysterosalpingography to determine tubal patency. Although her uterus was preserved, fertility function is yet to be confirmed. She was however advised to avoid another spontaneous pregnancy in the immediate future. A caesarean section would be required if she would eventually conceive.
This is the first report of the B-Lynch suture1 being used to control uterine atony following trauma after termination of pregnancy (TOP).
D & E has become one of the most frequently used methods of TOP in the second trimester in the non-NHS sector. It is a blind procedure and literature suggests that risk of perforation with this procedure is 1:1000 in the first trimester and 3:1000 in the third trimester2 with a likelihood of unintended abdominal surgery. Cervical trauma, haemorrhages requiring transfusion, sepsis or incomplete evacuation requiring further surgery have been associated as well. Some studies suggest that rates of psychological sequelae are higher among women who have had a TOP with a small increase in the risk of preterm delivery.
Uterine atony is treated medically at first, but when medical management fails, surgical intervention is necessary. Syntocinon is generally ineffective in maintaining sustained uterine contractility in mid-trimester gestations. Although misoprostol, which is a prostaglandin E1 analogue, was used in this case after the usual oxytocics, syntocinon and ergometrine,3 it still did not control the uterine bleeding. Gemeprost, another prostaglandin E1 analogue is more commonly used in therapeutic TOP and has been proven to be effective in arresting primary postpartum haemorrhage. However, misoprostol is cost-effective with fewer side effects and quicker pharmacokinetic action,4 whereas gemeprost requires refrigeration and should be allowed to warm at room temperature prior to administration. We may be criticised for not using intramuscular or intramyometrial hemabate in this woman as it is recommended for the management of postpartum haemorrhage and known to cause prolonged uterine contractions in women undergoing medical termination. There was, however, a reluctance to use it because hemabate is known for its long list of contraindications and possible side effects and this patient was asthmatic and had a history of heart murmur.
Surgical intervention may include ligating the uterine vessels or more commonly a hysterectomy. Other interventions include tamponade using a pack or specially designed balloons, internal iliac artery ligation or arterial embolisation under imaging.
In 1997, B-Lynch et al.1 introduced a suturing technique, also known as the ‘brace suture’, which resulted in satisfactory haemostasis in patients with postpartum haemorrhage due to uterine atony. The method seems to be effective and particularly useful because of its simplicity of application, life-saving potential, relative safety and capacity for preserving the uterus and subsequent fertility. Only case reports are available to evaluate the efficacy of the procedure so far. Cho’s square suture and Hayman’s5 modification of the B-Lynch suture technique have been described as other uterine compression suture techniques. The theory behind each technique is the same: a mechanical compression of the uterine vascular sinuses that prevents further engorgement with blood and subsequent haemorrhage.
Although originally described using number 2 chromic catgut, a vicryl 1 (Ethicon®) suture was used in this case. It is recommended to use absorbable, as opposed to delayed absorbable suture material to avoid bowel complications. Care should be taken to avoid major vessels and the ureters. The round-bodied needle is inserted about 3 cm from the right lateral border through the anterior uterine wall which exits the same wall 2–3 cm just above the first entry point. The suture material is wrapped over the fundus 3–4 cm medial to the corneal area of the fallopian tube, and the needle is reinserted through the right posterior uterine wall at the same level of entry as anteriorly. The needle exits the same wall horizontally about 3 cm from the left lateral border and the suture is taken vertically over the fundus in the contralateral corneal area. The suture should be pulled tight so that it hugs the uterine surface. The needle is then reinserted and withdrawn in through the anterior uterine wall about 3 cm from the left. The suture is then pulled, while the uterus is being compressed manually and tied anteriorly to secure the tension.
The B-Lynch suture and its modifications are simple procedures, which can be used to preserve fertility in patients of childbearing age group. The procedure is most successful with uterine hypotonia when medical as well as local measures fail to achieve haemostasis.