General gynaecology: Effect of uterotonics on intra-operative blood loss during laparoscopy-assisted vaginal hysterectomy: a randomised controlled trial

DOI: 10.1111/j.1471-0528.2005.00804.x www.blackwellpublishing.com/bjog

General gynaecology

Effect of uterotonics on intra-operative blood loss during laparoscopy-assisted vaginal hysterectomy: a randomised controlled trial Fung-Wei Chang,a Mu-Hsien Yu,a Chih-Hung Ku,b Chi-Huang Chen,a Gwo-Jang Wu,a Jah-Yao Liua a Department of Obstetrics and Gynecology, Tri-service General Hospital, National Defense Medical Center, National Defense University, Taipei, Taiwan b School of Public Health, National Defense Medical Center, National Defense University, Taipei, Taiwan Correspondence: Dr J-Y Liu, Department of Obstetrics and Gynecology, Tri-Service General Hospital, National Defense Medical Center, No. 325, Sec 2, Cherng-Kong Road, Neihu, 114 Taipei, Taiwan.

Accepted 28 September 2005.

Objective To investigate the effectiveness of uterotonics

misoprostol and oxytocin on reducing blood loss during laparoscopy-assisted vaginal hysterectomy (LAVH). Design Randomised, double-blind placebo-controlled trial. Setting University hospital. Population One hundred and seventy-three women underwent

LAVH for symptomatic uterine myomas and were randomly allocated to uterotonics (n = 91) or placebo (n = 82). Methods Women underwent LAVH for symptomatic uterine

myomas and were randomly assigned to receive either rectal misoprostol (400 mg) and intravenous oxytocin (10 IU/hour) or placebo during LAVH. Parameters related to surgical outcome were compared.

Results Patient characteristics and indications for LAVH were similar in both groups. Mean [SD] for all continuous data estimated weight of blood loss (198.1 [123.2] vs 396 [337.6] g; P < 0.0001), mean operation time (106.2 [39.4] vs 116.6 [34.6] minutes; P = 0.02), mean change in haemoglobin (1.5 [1.0] vs 1.9 [1.2] g/dL; P = 0.02) and haematocrit levels (4.8 [2.9]% vs 5.8 [3.6]%; P = 0.04) and mean hospitalisation period (3.3 [0.8] vs 3.9 [1.1] days; P < 0.0001), which were significantly less in the group given rectal misoprostol and intravenous oxytocin than in the placebo group, respectively. There was no significant difference in complications and side effects between the two groups (P > 0.05). Conclusion Combined rectal misoprostol and intravenous

oxytocin is a feasible and effective method of reducing blood loss and operation time in LAVH.

Main outcome measures The main outcome measure was

intra-operative blood loss. Please cite this paper as: Chang F-W, Yu M-H, Ku C-H, Chen C-H, Wu G-J, Liu J-Y. Effect of uterotonics on intra-operative blood loss during laparoscopyassisted vaginal hysterectomy: a randomised controlled trial. BJOG 2006; 113:47–52.

Uterine leiomyomas are the most common benign tumours among women. The majority of hysterectomies are performed to manage symptomatic uterine leiomyomas.1 Patients frequently undergo laparoscopy-assisted vaginal hysterectomy (LAVH), which is one of the recommended treatments to experience faster recoveries, fewer complications and shorter hospital stays when compared with patients undergoing abdominal hysterectomy.2 Nonetheless, the complication rate from hysterectomy increases with increasing uterine weight, mainly due to increased blood loss during surgery, especially

for a large myomatous uterus. Many gynecologic surgeons use a vasopressin injection pericervically at time of vaginal hysterectomy to decrease blood loss.3 However, this method still may not effectively decrease haemorrhage during LAVH for a large uterus. Traditionally, misoprostol and oxytocin have been used as uterotonic agents for postpartum haemorrhage, primarily as part of active management. These agents have strong uterotonic properties and are used widely in obstetric and gynecological practice, with safe therapeutic ranges. The mechanisms of action of misoprostol and oxytocin are different. Misoprostol, a prostaglandin E1 analogue, apparently reduces uterine artery blood flow in early pregnancy,4 increases

ª RCOG 2006 BJOG An International Journal of Obstetrics and Gynaecology

47

Introduction

Chang et al.

myometrial contractions in uterine-atonic postpartum haemorrhage and reduces blood loss during caesarean section and myomectomy.5,6 However, instability of rectal absorption is a disadvantage with use of misoprostol. An additional intravenous uterotonic agent, oxytocin, is required to provide fast availability and to increase myometrial contractions, leading to a reduction in myometrial blood flow.7 Conventionally, to our knowledge, oxytocin receptors are not abundant in the non-pregnant uterus. However, recent evidence shows that uterine myomas are similar to pregnancy-related adaptation in expressing more oxytocin receptors that bind to oxytocin, leading to myometrial contractility in animal models.8 Clinically, for rapid onset and sustained effect of uterotonic agents, the combination of intravenous oxytocin and rectal misoprostol to induce uterine contraction and decrease intrauterine blood flow appears to be reliable and practical. Based on the hypothesis that uterotonic agents may decrease intraoperative haemorrhage in LAVH for uterine myoma, we used a double-blind randomised controlled trial to investigate the effectiveness of pre-operative rectal misoprostol and intravenous oxytocin on reducing blood loss in LAVH and their secondary impact on hospital stay and complications.

Materials and methods A randomised controlled study was set up to test the following null hypothesis: There is no significant difference in efficacy of rectal misoprostol and intravenous oxytocin for reduction of intra-operative blood loss. Our two-year trial was conducted between 11 January 2002 and 10 January 2004 at Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan. In deciding which treatment to opt for symptomatic uterine myomas, patients were given full information concerning alternative treatments available to manage their specific problems, including endometrial ablation/transcervical resection of the endometrium (TCRE), gonadotrophin-releasing hormone (GnRH) analogues treatment and uterine artery embo-lisation and myomectomy. Those who chose to undergo hysterectomy after counseling were recruited. Women scheduled to undergo LAVH were invited to participate. Patients who had hypertension, cardiac or pulmonary disease, known adnexal masses and had known adverse reactions to misoprostol and oxytocin were excluded in the screening. None of the patients included in this study had received pre-operative hormone therapy such as a GnRH agonist therapy or oral contraceptive pills, which could affect intra-operative bleeding. An extremely large uterus is a contraindication for LAVH (uterine size similar to more than 18 gestational weeks). In the absence of any of these, allocation of the following treatment was generated using a table of randomised numbers before the procedure. Assignment was concealed by placement in consecutively numbered,

48

opaque, sealed envelopes, with numbers drawn in consecutive order by a manager. The investigators were blind to allocation. Only the study statisticians and the data monitoring committee saw actual acquired data, but none had any contact with study participants. Patients in the study group were given 400 mg of rectal misoprostol (Cytotec, 200 mg; Searle, Chicago, USA) 60 minutes before surgery and 10IU of intravenous oxytocin (Piton-S, 10IU; Organon, Oss, Holland) diluted in 500 mL 5% dextrose was given 10 minutes before surgery and infused throughout the procedure at a rate of 10 IU/hour. Patients in the control group were given starch tablets without misoprostol and ampoules of 1 mL 5% dextrose in the same way as the study group. Pelvic adhesions were exclusion criteria identified during the procedure. The study protocol was reviewed and approved by the Institutional Review Board of National Defense Medical Center. The same team performed all of the operations in the standard manner. Patient demographic parameters, uterine weight, operation time, haemogram (haemoglobin and haematocrit), blood loss, blood transfusion, vital signs, morbidity and period of hospitalisation were recorded before and after surgery. To accurately measure blood loss, the amount of blood accumulated in the aspiration equipment was taken as the basis for calculating each patient’s blood loss during surgery, in which suction only was used to clean the operative field. Complete blood count was obtained before surgery and 17 [3] hours post-operatively. Single-dose prophylactic antibiotic was given 30 minutes pre-operatively. The experimental hypotheses were tested using statistical procedures available in SAS for Windows version 9.1. Continuous and ordinal data are represented as means [SDs] and other data, as the normal scale (present 1, absent 0). x2 and Fisher’s exact tests were used for ordinal and continuous data, and the Kruskal–Wallis test was used for continuous data in comparisons between groups. P < 0.05 was considered statistically significant. Multiple linear regression analysis was used to determine the relationships between numeric independent operative variables (uterine weight, operation time, hospital stay, changes in haemoglobin and haematocrit) and the numeric dependent variable, intra-operative blood loss. Multiple linear regression with stepwise selection was used to assess the association of interests, as well as to adjust for potential confounding variables.

Results The CONSORT scheme of study reporting was adhered to and Fig. 1 shows the flow of patients through the trial. Initially, 225 patients were screened in the study. Forty-six were initially excluded: 18 refused, 7 had history of hypertension, cardiac or pulmonary disease, 5 had history of pre-operative hormone therapy, 7 had adnexal mass and in 9 the uterus was too large to perform LAVH. A total of 179 patients were

ª RCOG 2006 BJOG An International Journal of Obstetrics and Gynaecology

Uterotonics use in hysterectomy

Initial screening of patients intended for LAVH (n = 225)

Screening exclusions (n = 46): Refused to participate (n = 18) History of hypertension, cardiac or pulmonary disease (n = 7) History of pre-operative hormone therapy such as a GnRH analogue or oral contraceptive pills (n = 5) Adnexal mass (n = 7) Extremely large uterus (n = 9) (>18-week gestational size)

Recruited and randomised before laparoscopy (n = 179)

Misoprostol plus oxytocin (n = 94)

Laparoscopy

Excluded during laparoscopy Pelvic adhesions (n = 3)

Analysed (n = 91)

Placebo group (n = 85)

Laparoscopy

Excluded during laparoscopy Pelvic adhesions (n = 3)

Analysed (n = 82)

Figure 1. Flow diagram of participants throughout trial.

recruited. Six patients were excluded due to pelvic adhesions during laparoscopy. Overall, among 173 patients eligible for inclusion in our study, 91 were allocated to misoprostol plus oxytocin and 82 were allocated to placebo (Fig. 1). The demographic data and operative indications of patients undergoing LAVH in both groups did not differ significantly (Tables 1 and 2; P > 0.05, Kruskal–Wallis and x2 tests). Regarding primary indications, most patients had

undergone hysterectomy to manage menorrhagia with anaemia in both study (54.9%) and placebo (57.3%) groups. In our study, both groups had reached the proposed sample size of 80. Assuming a standard deviation of blood loss of 350 g, this trial had 95% power at the 5% level of significance, to show a difference in intra-operative blood loss of 200 g (between 150 and 350 g). A power study is calculated to be 0.99 and would be sufficient to show a true mean difference of

ª RCOG 2006 BJOG An International Journal of Obstetrics and Gynaecology

49

Chang et al.

Table 1. Demographics and characteristics of cases (mean [SD]) Characteristics

Study group Placebo group P (n 5 91) (n 5 82)

Mean age [SD] (years) Range Mean body mass index [SD] (kg/m2) Mean parity [SD] (number) Mean gravida [SD] (number) Mean uterine weight [SD] (g)

45.9 [3.6] 37–54 23.6 [5.4]

46.1 [4.8] 34–63 24.7 [3.9]

2.0 [1.1] 2.3 [1.0] 2.9 [1.3] 3.7 [1.7] 393.7 [284.7] 335.4 [251.5]

NS

Table 3. Multiple linear regression assessed for uterotonics effects on blood loss and other parameters. Model was assessed by PROC REG with step-wise selection (SAS 9.1), also adjusted for age, haemoglobin and operation duration Variable

Parameter estimate

SE

F

P

Intercept Uterotonics Uterine weight Hematocrit change

83.06 2203.04 0.52 24.16

32.97 27.59 0.052 4.31

6.35 54.16 98.59 31.41

0.0127 ,0.0001 ,0.0001 ,0.0001

NS NS NS NS

P . 0.05, statistically insignificant difference, based on Kruskal–Wallis test.

intra-operative blood loss. After adjusting for age, haemoglobin and operation duration, blood loss was found to be 203 g less in the study group (uterotonics) than in the placebo group (Table 3; P < 0.0001). In addition, uterine weight and peri-operative haematocrit changes were significantly correlated with blood loss (Table 3; P < 0.0001). Linear regression showed a significant positive correlation between uterine weight and blood loss (Fig. 2); significantly less blood loss occurred in the study group than in the placebo group (P < 0.05). The greater the uterine weight, the greater the intra-operative blood loss in both groups (trend test = 0.0003 for the study group and 0.0001 for placebo group). Multiple linear regression analysis revealed that uterine weight correlated with the amount of operative blood loss, irrespective of other operative outcomes (operation duration, hospital stay, haemoglobin and haematocrit). Uterine weight was the best predictor of operative blood loss during LAVH, after adjusting for the other operative outcomes. Comparisons of the operative outcomes are shown in Table 4. The study group had better operative parameters such as blood loss, post-operative haemogram, operation time and hospital stay than did the placebo group. Postoperative complications included vomiting (17 patients, 18.7%) and fever (17 patients, 18.7%) in the study group and 16 patients [19.5%] and 17 patients [20.7%], respectively,

in the placebo group (P > 0.05 for all, Fisher’s exact test). No statistically significant differences in complications were found between the study and placebo groups. Complications were self-limiting. One case (1.1%) had shivering and one had diarrhoea in the study group. No patients required blood transfusion or readmission to the hospital.

Conclusion Over 50% of women studied who had symptomatic uterine myomas complained of heavy menstrual flow (Table 2). Furthermore, it was common to present complex symptoms including palpable pelvic mass and extrinsic compression of adjacent pelvic organs, which increased surgical risk. However, these indications for hysterectomy are not universally accepted and have been the subject of debate. Numerous conservative approaches, such as GnRH agonist therapy, uterine artery embolisation, endometrial ablation/TCRE and others, have been used to reduce and stop the growth of uterine myomas and heavy menstrual bleeding prior to a definitive surgical intervention. LAVH is an alternative

Table 2. Indications for LAVH Indications

Study group Placebo group P (n 5 91) (n 5 82)

Menorrhagia with anaemia 50 (54.9%) Compression symptoms 17 (18.7%) Pelvic pain 17 (18.7%) Uterine growth after 7 (7.7%) menopause

47 (57.3%) 13 (15.9%) 12 (14.6%) 10 (12.2%)

P . 0.05, statistically insignificant difference, based on x2 test.

50

NS NS NS NS

Figure 2. Blood loss vs uterine weight in the study and placebo groups.

ª RCOG 2006 BJOG An International Journal of Obstetrics and Gynaecology

Uterotonics use in hysterectomy

treatment modality, leading to a definitive treatment for uterine myomas, and thus an optimal solution for patients who do not wish to preserve fertility and for whom other therapeutic approaches have failed.2 LAVH not only offers the same advantages of both abdominal and vaginal hysterectomy procedures but also has the same disadvantages as vaginal hysterectomy. Heavy menstrual flow and anaemia are common symptoms of patients with myomatous uterus. Reducing blood loss during surgery decreases the need for blood transfusion and decreases postoperative morbidity. Our report addresses the effectiveness of uterotonics on decreasing haemorrhage during LAVH, especially for the large myomatous uterus. The uterotonic agents, misoprostol and oxytocin, have been used for prevention of postpartum haemorrhage, reduction of intra-operative blood loss during caesarean sections5,7 and abdominal myomectomy.6 We demonstrated that the use of uterotonic agents in LAVH resulted in a statistically significant reduction in intra-operative blood loss. Blood-loss-reduction effects of prostaglandin analogues and oxytocics have been demonstrated in vitro and in animal models.5–7 Misoprostol and oxytocin induce potent uterine contractility regardless of their different mechanisms.9 The major effect of misoprostol is on the myometrium and the cervix, whereas oxytocin, mediated by oxytocin receptors in the uterine muscle and myoma, stimulates synthesis and release of contractile prostaglandins. Increased uterine contractility directly affects uterine vascular structures that stem from both the uterine artery and the utero-ovarian anastomosis, decreasing blood supply to the uterus and myomas. Decreased blood volume in the uterus and constricted uterine vasculature due to uterine contraction and vasoconstrictive impact of misoprostol and oxytocin results in reducing intraoperative blood loss.10,11 Our study demonstrated a statistically significant reduction of intra-operative blood loss with the use of misoprostol and oxytocin during LAVH for uterine myomas. It is worth mentioning that the reduced intra-

operative blood loss improved clarity of the operative field and reduced operation time. Administration of misoprostol by the rectal route in LAVH may allow the drug to be absorbed without interfering in vaginal surgery and may avoid adverse effects associated with the oral route. The absorption and pharmacokinetics of rectal misoprostol include a lower peak serum concentration and longer half-life than oral misoprostol. Misoprostol acid concentration reaches a peak in plasma at between 15 and 60 minutes (mean = 40.5 minutes) after rectal administration and then declines slowly. The same dose of oral misoprostol reaches a peak concentration earlier (mean = 18 minutes), but falls steeply by 60 minutes, and then remains lower than when it is rectally administered.12 The longer half-life of rectally administered misoprostol could prolong uterine contraction, thus reducing intra-operative haemorrhage. No pharmacokinetic differences have been found with misoprostol between pregnant and non-pregnant women.13 However, the threshold serum level leading to clinical effects is unknown and may be low; absolute serum levels may not necessarily be correlated with quantifiable clinical superiority. And, in our experience, it was not uncommon to find that the residuum of misoprostol was still not completely dissolved several hours after rectal administration. An intravenous injection of oxytocin appears in the circulation within 15 seconds and reaches the maximum level in 60 seconds, with a half-life of 3 minutes after a single intravenous injection.14 Therefore, it is advisable to combine the use of intravenous oxytocin to achieve faster initial effects followed by rectal misoprostol for sustained uterine contraction. Clinically, the addition of intravenous oxytocin helps to assure that uteri are well contracted before surgery. Uterotonic agents may be used at different dosages and administered by different routes. We used 400 mg of intrarectal misoprostol and 10 IU/hour of intravenous oxytocin to achieve uterine contraction and decrease intra-operative blood loss. We chose this method because the route of administration and dosage of drugs had a low incidence of side effects and because these drugs are commonly used to prevent postpartum haemorrhage and reduce intra-operative blood loss during caesarean section, showing them to be clinically convenient and feasible. Furthermore, the pharmacokinetics of these misoprostol and oxytocin dosages and routes are well known.5–7 The most common side effects of misoprostol are nausea, diarrhoea, pyrexia and shivering. The risk of hypertension, nausea and vomiting is less with oxytocin alone.5 Their synergistic effect, however, may achieve more effective uterine contractility. In our study, although the side effects of fever and vomiting were seen after LAVH, the difference between the two groups did not reach statistical significance. The reason for this might have been that while we administered the uterotonic agents during surgery, the patients were either still anesthetised or under the effects of an analgesic. In the literature, side effects of misoprostol were seen within

ª RCOG 2006 BJOG An International Journal of Obstetrics and Gynaecology

51

Table 4. Comparisons of operative events Characteristics

Study group Placebo group P (n 5 91) (n 5 82)

Mean measured blood loss [SD] (g) Mean haemoglobin change [SD] (g/dL) Mean haematocrite change [SD] (%) Mean operation time [SD] (minutes) Mean hospital stay [SD] (days)

198.1 [123.2] 396.5 [337.6]

,0.0001

1.5 [1.0]

1.9 [1.2]

0.02

4.8 [2.9]

5.8 [3.6]

0.04

106.2 [39.4]

116.6 [34.6]

0.02

3.3 [0.8]

3.9 [1.1]

,0.0001

P , 0.05, statistically significant difference, based on Kruskal–Wallis test.

Chang et al.

90 minutes, and for oxytocin 3 minutes, after they were administered in patients who were not anesthetised.13,14 Whether general anaesthesia modifies the rate of misoprostol and oxytocin side effects remains unknown. Fortunately, all side effects were self-limiting. Based on our experience, combining rectal misoprostol and intravenous oxytocin as uterotonic agents for use in LAVH appears to be safe. Combined use of misoprostol and oxytocin appeared to be beneficial in reducing blood loss during LAVH for large uterine myomas; they seemed to be equally effective during LAVH as they are during caesarean section. Based on the wide application of uterotonics in obstetrics, our study on their use in LAVH indicates that the use of rectal misoprostol and intravenous oxytocin to reduce haemorrhage is both reliable and effective, especially when the uterus is large. In our study, intra-operative blood loss increased with increasing uterine weight. The effect of reduced blood loss might be more obvious in conditions when heavy haemorrhage and/or long operation time were expected, such as in large myomatous uterus.

Acknowledgements The study was supported by grant TSGH-C92-72 from Tri-service General Hospital and C.Y. Foundation, Taipei, Taiwan. j

References 1 Lepine LA, Hillis SD, Marchbanks PA, et al. Hysterectomy surveillanceUnited States. CDC surveillance summaries, August 8, 1997. Morb Mortal Wkly Rep 1997;46(ss-4):1–15. 2 Carter JE, Ryoo J, Katz A. Laparoscopic-assisted vaginal hysterectomy: a case control comparative study with total abdominal hysterectomy. J Am Assoc Gynecol Laparosc 1994;1:116–121.

52

3 Kammerer-Doak DN, Rogers RG, Maybach JJ, Mickelson MT. Vasopressin as an etiologic factor for infection in gynecologic surgery: a randomized double-blind placebo-controlled trial. Am J Obstet Gynecol 2001;185(6):1344–1347 [discussion 1347–1348]. 4 Yip SK, Tse AO, Haines CJ, Chung TK. Misoprostol’s effect on uterine arterial blood flow and fetal heart rate in early pregnancy. Obstet Gynecol 2000;95:232–235. 5 Acharya G, Al-Sammarai MT, Patel N, Al-Habib A, Kiserud T. A randomized, controlled trial comparing effect of oral misoprostol and intravenous syntocinon on intra-operative blood loss during cesarean section. Acta Obstet Gynecol Scand 2001;80:245–250. 6 Celik H, Sapmaz E. Use of a single pre-operative dose of misoprostol is efficacious for patients who undergo abdominal myomectomy. Fertil Steril 2003;79(5):1207–1210. 7 Ramsey P. Rectally administered misoprostol for the treatment of postpartum hemorrhage unresponsive to oxytocin and ergometrine: a descriptive study. Obstet Gynecol 1999;93(1):157–158. 8 Cesen-Cumming K, Houston KD, Copland JA, Moorman VJ, Walker CL, Davis BJ. Uterine leiomyomas express myometrial contractileassociated proteins involved in pregnancy-related hormone signaling. J Soc Gynecol Investig 2003;10(1):11–20. 9 Wray S. Uterine contraction and physiological mechanisms of modulation. Am J Physiol 1993;264:1–18. 10 Baxter GS, Clayton JK, Coleman RA, Marshall K, Sangha R, Senior J. Characterization of the prostanoid receptors mediating constriction and relaxation of human isolated uterine artery. Br J Pharmacol 1995;116:1692–1696. 11 Chen YL, Shepherd C, Spinelli W, Lai FM. Oxytocin and vasopressin constrict rat isolated uterine resistance arteries by activating vasopressin V1a receptors. Eur J Pharmacol 1999;376:45–51. 12 Khan RU, El-Refaey H. Pharmacokinetics and adverse-effect profile of rectally administered misoprostol in the third stage of labor. Obstet Gynecol 2003;101:968–974. 13 Zieman M, Fong SK, Benowitz NL, Banskter D, Darney PD. Absorption kinetics of misoprostol with oral or vaginal administration. Obstet Gynecol 1997;90:88–92. 14 Gibben D, Boyd NRH, Crocker S, Baumber S, Chard T. The circulating levels of oxytocin following intravenous and intramuscular administration of syntometrine. J Obstet Gynaecol Br Commonw 1972; 79:644–646.

ª RCOG 2006 BJOG An International Journal of Obstetrics and Gynaecology