Triple vs. single uterine tourniquet to reduce hemorrhage at myomectomy: a randomized trial

This study aimed to compare the effect of triple uterine tourniquet and single tourniquet on intraoperative blood loss during open myomectomy. Women were randomized to undergo open myomectomy with a triple (n = 30) or single uterine tourniquet (n = 30). All symptomatic women aged 18–48 who had three or more myomas or at least one myoma greater than 8 cm if there were less than three myomas were eligible for the study. The primary outcome variable was the volume of intraoperative blood loss. The sample size was set to detect a 240 ml difference in blood loss with 80% power at α = 0.05, with an effect size of 0.8. The rate of transfusions, change in hemoglobin, volume of drains, operation time, tourniquet time, and perioperative complications were secondary outcomes. We found no significant difference in intraoperative blood loss between triple and single uterine tourniquets (527 [102–2931]) ml vs. 508 [172–2764] ml, p = 0.238). Between the single and triple tourniquet groups, the median weight of myoma (379 [136–3850] vs. 330 [140–1636] g, p = 0.451) and median number (1 [1–18] vs. 2 (1–13), p = 0.214), total operation time (84 ± 31 min vs. 79 ± 27 min, p = 0.503), ischemia time (35 ± 21 min vs. 30 ± 14 min., p = 0.238), drain volume at 48th hour (196 ± 89)ml vs. 243 ± 148 ml, p = 0.144) and decrease in hemoglobin (2.3 ± 1.8 g/dl vs. 2.8 ± 1.4 g/dl, p = 0.437) were similar. Eight (27%) patients in the triple tourniquet group and 12 (40%) patients in the single tourniquet group were transfused (p = 0.273). One patient underwent hysterectomy 6–8 h after myomectomy in a single tourniquet group. There was no clinically significant difference in intraoperative blood loss between triple and single uterine tourniquets during open myomectomy. ClinicalTrials.gov ID: NCT02392585, 03/13/2015.


Introduction
Physical blockage of the uterine blood supply is one of the oldest and most effective interventions for reducing intraoperative blood loss during myomectomy [1][2][3]. To ensure that, either only uterine arteries or both uterine and ovarian arteries are blocked by different methods to achieve occlusion of vessels but often with rubber tourniquets.
During open myomectomy, uterine tourniquets, either single tourniquet on uterine arteries or triple tourniquets on uterine and ovarian arteries, are classically advised as an effective way to control bleeding [4]. However, it is unclear whether applying a tourniquet to the infundibulopelvic ligament to block ovarian circulation is safe. First, the possible effects of ischemia due to temporary blocking of ovarian circulation on ovarian function raise concerns. There is no evidence-based data on the safe ischemia time or effect of this procedure on ovarian survival. Second, during the * Ragıp A. Al atakanal@gmail.com 1 Faculty of Medicine, Department of Obstetrics and Gynecology, Ataturk University, 25240 Erzurum, Turkey procedure, an injury necessitating oophorectomy may occur in the infundibulopelvic ligament [5]. No information is available in the literature regarding the frequency of such injuries.
In addition to safety concerns, there is no evidence that a triple tourniquet reduces intraoperative blood flow better than a single tourniquet. We previously compared the efficacy of double triple tourniquet and single tourniquets and found no difference in intraoperative blood loss between these methods [5]. Women with a uterus greater than 12 gestational weeks were included in this study. Intraoperative blood loss during myomectomy is highly variable, depending on the number, size, location of the myomas, and the duration of the operation [6]. The myoma burden can be better defined by ultrasonography. In the current study, to confirm the results of the previous study, we conducted a randomized controlled study to compare the effects of single and triple tourniquets on intraoperative blood loss during myomectomy in women whose inclusion criteria were determined by ultrasound.

Methods
This study was a randomized clinical trial conducted at Ataturk University Hospital between March 2015 and March 2016. This study compared intraoperative blood loss during open myomectomy with either a single or triple uterine tourniquet. Approval was obtained from the Clinical Research Ethics Committee of the Faculty of Medicine. The trial was registered on ClinicalTrials.gov (ID: NCT02392585, date of registration: 03/13/2015).
All symptomatic women between the ages of 18 and 48 years who were offered an abdominal myomectomy procedure with a leiomyoma burden meeting the following criteria based on ultrasound were eligible for the study: (1) three or more uterine leiomyomas and (2) at least one myoma greater than 8 cm if there were less than three leiomyomas. Women with pedunculated or broad ligament myomas were excluded from this study.
All eligible women were invited to participate in the study and those who provided written informed consent were consecutively enrolled. The women were randomly assigned 1:1 to undergo myomectomy using triple or single tourniquets. Randomization was performed using a computer random number generator with a permutated block size of 10. This procedure was performed by a physician who was not included in the study, and a letter indicating the group assignment was placed inside opaque envelopes, the outer cover of which was sequentially numbered, and the envelope was pasted. When a woman was included in the study, she was given the first study number in the line. Group allocation was performed immediately before surgery by the principal investigator using sealed and sequentially numbered opaque envelopes containing group allocation. Patients, investigators, and staff involved in the trial were not blinded to group allocation.
All interventions and procedures were performed as described in our previous study [5]. In this study, we used a Rummel-type tourniquet, as described by Te Linde's operative gynecology by Rock and Breech [4]. An opening was made in the avascular part of the broad ligament, on both sides of the uterine isthmus. If a single tourniquet would be placed, a No. 8 French Foley catheter in which the port site was cut was passed around the lower segment of the uterus through the openings. If a triple tourniquet were used, a similar catheter was additionally passed around each infundibulopelvic ligament through ipsilateral openings in the broad ligament. Both ends of the Foley catheter were passed through a 2-3 cm long catheter that was cut from the sterile aspirator tube. The tourniquet was tightened as much as possible and secured with a hemostat placed perpendicular to the distal end of the catheter. The catheters were not released until all myomas were removed and the uterus was completely sewn. In the surgery, no additional method other than the tourniquet was used to reduce blood loss.
In all cases, abdominal entry was performed using a suprapubic transverse incision. A 30-degree Trendelenburg position was then given, and the uterus was exteriorized. In most cases, surgical retractors were not used and abdominal packing was not performed. All myomas identified by palpation were removed during surgery. The uterine defects were closed with polyglactin sutures in two or more layers. The serosal layer was approximated using baseball sutures. During the procedure, the blood in the surgical field was removed by aspiration with a suction device, and surgical sponges were not used as much as possible. Before the abdomen was closed, all visible blood was aspirated, the abdomen was washed with sterile saline, and the washing fluid was aspirated to a different suction canister. The volumes of the delivered and aspirated washing fluids were measured, and the missing amount was subtracted from the total drain volume at the 48th hour. A hemovac drain was placed in the pouch of the Douglas. All surgical procedures were performed by senior surgeons.
The primary outcome variable of the study was the estimated intraoperative blood loss. Blood loss was calculated by combining the aspiration volume and weight of the sponges used. For a more precise measurement, the fluid collected in the aspirator canister was measured using 50 ml syringes graduated from 2 ml. Wet gauze and wet abdominal sponges were not used during surgery. The used sponges were collected and weighed every 2 min. The average dry weight of the sponges was subtracted from the wet weight, and the remaining was divided by 1.06 to estimate the blood volume, assuming that one milliliter of blood weighed 1.06 g [7]. The average dry weight of a gauze sponge and abdominal sponge was calculated by calculating the average weight of 100 pieces of each one at the beginning of the study. All weighing procedures were repeated three times using an electronic weighing scale with a measurement accuracy of 1 g, and the average of the measurements was calculated. All weighing and measurement procedures were performed by the same operating room technician.
The secondary outcomes of the study were the amount of transfusion, change in hemoglobin levels, volume of drains, total operation time, tourniquet time, number and weight of myomas, and perioperative complications. Packed red blood cell transfusions were made when hemoglobin was < 8 g/dl and included transfusion made intraoperatively or postoperative within 7 days. The change in hemoglobin level was calculated by subtracting the hemoglobin measured at the 48th hour postoperatively from the preoperative hemoglobin level. All fluid coming from the drains within 48 h was calculated as the drain volume, and the drains were removed at the 48th hour. Operation time was calculated as the skin-toskin surgery time. All women were examined at an appointment visit 6-8 weeks after surgery, and all adverse events occurring within 6-8 weeks after surgery were recorded as perioperative complications.
In this study, AMH levels were planned to be measured at the beginning of the study and at 6 and 12 months later. However, most women did not visit their 6-month-or 12-month appointments. This part of the study protocol was excluded from the analysis as it could not be completed.
The sample size was calculated based on the amount of blood loss during surgery. Because mean blood loss and standard deviation vary significantly during myomectomy, a large effect size that would be clinically meaningful was chosen in this study. The effect size was estimated from a previous report in which the effects of triple tourniquet and single tourniquet were compared on intraoperative blood loss in myomectomy [5]. In this study, the mean blood loss was 370 ± 300 ml in all the patients. Accordingly, when the mean blood loss was 370 ml, 26 women in each group were required to detect a 240 ml difference in blood loss between the two groups with 80% power and 5% significance. The effect size of this study was calculated as 0.8. An effect size of 0.80 or more is generally regarded as a large effect size [8]. Considering possible dropouts, we calculated a sample size of 30 women in each group.
The statistical analysis was based on the intention-to-treat principle. The statistical software used for analysis was G Power 3.1 and SPSS 20.0 (IBM Corp.) [9]. The Kolmogorov-Smirnov test was used to check the normality of the distributions. The data were analyzed using the Student's t-test and Mann-Whitney U test for continuous variables and Pearson Chi-Square test or Fisher's exact test for categorical variables. All tests were 2-sided and p values less than 0.05 were considered statistically significant.

Results
During the study period, 93 patients were evaluated, of which 75 women who were recommended myomectomy were invited to participate in the study. Sixty women agreed to participate in this study (Fig. 1). All the patients completed the study and were included in the analysis.
Age, gravida, parity, rate of nulliparous women, frequency of previous laparotomy, and indications for surgery were similar in both groups (Table 1). Packed red blood cell transfusion was administered to seven (23%) patients in the single tourniquet group and two (7%) patients in the triple tourniquet group preoperatively. The preoperative hemoglobin level was higher in the triple tourniquet group.
Intraoperative blood loss was similar in both groups (single tourniquet 508 [172-2764] ml vs. triple tourniquet 527 [102-2931] ml, p = 0.238) ( Table 2). The weight and number of myomas removed during the operation, ischemia time, operation time, postoperative hemoglobin level, change in hemoglobin level, and volume in drains were similar in both groups.
Surgery was generally well-tolerated. One patient in the single tourniquet group developed a hematoma in the broad ligament, which was controlled by suturing. One patient in the single tourniquet group was reopened 6-8 h after myomectomy due to bleeding and a hysterectomy was performed.

Discussion
We found no difference in intraoperative bleeding when single or triple tourniquets were used to reduce blood loss in open myomectomy, as reported previously [5].
Occlusion of the ovarian artery does not seem to have a significant effect on intraoperative bleeding during myomectomy. Although more intraoperative bleeding due to ovarian anastomoses is expected in more complex and long-lasting operations when a single tourniquet is applied, it is unlikely that the type of tourniquet will cause clinically significant differences in these cases where the expected amount of bleeding is high.
Our study has several strengths. First, the number and total weight of myomas removed, ischemia time, and operation time, all known to be associated with the amount of blood loss in open myomectomy, were found to be similar between the two groups. Second, we set a lower limit for the size and number of myomas when enrolling patients in the study. Thus, myomas that are small and can be removed quickly, and even those that do not require an intervention to prevent bleeding, were not included in the study. Third, we calculated the amount of intraoperative blood loss by measuring the volume of suction and the weight of the surgical sponges, which both are objective and accurate methods of measuring blood loss at surgery. Wet gauze and abdominal sponges were not used in this study to improve precision.
However, there are several limitations of the study. First, the study was underpowered to detect a difference in blood loss < 240 ml. However, we think that the smaller bleeding differences between the methods of open myomectomy are clinically insignificant. Second, whether the flow in the ovarian arteries was completely blocked by the tourniquets was  not checked with intraoperative Doppler velocimetry examination. We think that the possibility of flow in the ovarian arteries is unlikely because the tourniquets tightened as much as possible. However, since no measurement has been made, this possibility cannot be definitively excluded. Third, the study was not specifically powered to assess the secondary outcomes. One woman underwent a hysterectomy, and there was no infundibulopelvic ligament injury. However, it was not possible to draw any conclusions from these results because the sample size was not sufficient. Packed red blood cell transfusion rates were high in both groups, despite the low estimated blood loss in the groups. We used a restrictive threshold of 8 g/dl for packed red blood cell transfusion. Probably, a lower hemoglobin threshold of 7 g/dl would be tolerated by patients in this study. However, the risk of blood transfusion is strongly related to the preoperative hemoglobin level and myoma burden on an individual basis, as well as the transfusion strategy. Red blood cell transfusion rates should be interpreted with caution, as this study was not designed to have red blood cell transfusion rates as the primary outcome variable.
In conclusion, there was no significant difference in intraoperative blood loss between patients who underwent triple uterine tourniquet and single uterine tourniquet on open myomectomy. Given the potential complications associated with the triple tourniquet procedure, we recommend not routinely performing this procedure.
Author contributions NG, OEY, and RAA contributed to the study conception, design, material preparation, data collection, and analysis. The manuscript was written by RAA and SO. All authors read and approved the final manuscript.

Funding
The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
Data availability No data available.