Submucous myomata frequently can lead to symptoms including abnormal uterine bleeding and infertility compared with other types.1–4 Hysteroscopic myomectomy has been considered the optimal surgical method. The surgical challenge is to achieve complete resection while avoiding complications, the complexity of the procedure being determined by the size, type, and location of the submucous myomas.2,4–6 The risks of hysteroscopic myomectomy include bleeding, cervical injury, uterine perforation, adhesion formation, and fluid intravasation, which can lead to life-threatening hyponatremia, heart failure, pulmonary, and cerebral edema.5,7
Preoperative gonadotropin-releasing hormone (GnRH) agonist has been used to reduce the risks associated with hysteroscopic myomectomy.8–10 A recent meta-analysis suggested that GnRH agonists may improve some outcomes, but there is insufficient evidence to support its routine use.11–13 Potential drawbacks of GnRH agonists include their high cost, menopausal side effects, extra clinic visits, delay in the surgical procedure, more difficult cervical dilatation, and the possibility of initial worsening of the hemorrhage. Vasopressin does not have these effects and has been used during abdominal and laparoscopic myomectomy.14 However, studies on the use of vasopressin in hysteroscopic myomectomy are limited. Two trials investigated intracervical injection of vasopressin during operative hysteroscopy and showed a reduction in fluid intravasation and blood loss.15,16 We have previously described a new technique of vasopressin injection under visualization during hysteroscopy.17
The aim of the present study is to investigate whether transcervical intralesional vasopressin injection can result in shorter operative time, less fluid intravasation, less blood loss, and better visual clarity during hysteroscopic myomectomy.
MATERIALS AND METHODS
Premenopausal women aged older than 18 years with symptomatic submucous myoma requiring hysteroscopic myomectomy in our university teaching hospital were invited to participate in the study. The exclusion criteria included menopausal women, use of preoperative GnRH agonists, submucous myoma more than 5 cm, endometrial premalignant or malignant pathologies, concomitant endometrial ablation, use of anticoagulant therapy, women with cardiovascular diseases, migraine, asthma, heart failure, epilepsy, and those who refused to participate.
The primary outcome was surgical time for myomectomy. The secondary outcomes were volume of inflow fluid, fluid deficit, surgeon's rating of intraoperative blood loss, and visual analog scale (VAS) assessment of intensity of discoloration of effluent and visual clarity. The sample size calculation was based on two previous studies by Perino et al and Muzii et al.13,18 A reduction of operative time of 50% with preoperative GnRH agonist treatment was observed. From the analysis of our 1-year data, the rate of patients whose duration of surgery exceeded 20 minutes without preoperative GnRH agonist treatment was 85%. If vasopressin is used with the aim to achieve a reduction of 50% in the operative time, this would require 20 patients in each arm with a power of 80% and the level of significance of .05.
The diagnosis of submucous myoma was made by transvaginal ultrasound examination and diagnostic hysteroscopy. The severity of abnormal uterine bleeding was assessed by pictorial blood loss assessment chart.19 The grade of submucous myoma was determined by hysteroscopy using the Wamsteker grading system adopted by the European Society for Gynecological Endoscopy.6 The size of the myoma was measured with ultrasonography. All eligible women had preoperative evaluation at the joint preoperative assessment clinic by gynecologists and anesthetists. The mode of anesthesia was determined by anesthetists not involved in the study. The preoperative hemoglobin level was measured and renal function test was performed.
Randomization to either vasopressin group or placebo group was carried out in the operating room by assigning each recruited patient a consecutively numbered, sealed nonopaque envelope, which contained the designated treatment as determined by a computer-generated series of random numbers. A block randomization was used with an allocation ratio of 4:4; the randomization was performed by an independent research nurse.
In the vasopressin group, one vial (20 units) of vasopressin was diluted to 50 mL with normal saline solution, and not more than 10 mL (4 units) of the diluted solution was used to infiltrate the submucous myomas. In the placebo group, 50 mL of normal saline was prepared. On the prescription charts, vasopressin was ordered for both groups and study labels were used to indicate participation in the clinical trial so that group allocation could not be determined from the charts.
Each surgical procedure involved two surgeons (Surgeon A and Surgeon B). Surgeon A was A.S.W.W., C.W.C., S.W.Y., or H.L.F. Surgeon B was either A.S.W.W. or C.W.C. At the time of induction of anesthesia, the sealed envelope was opened to reveal the randomization to only Surgeon A. Surgeon A then prepared the study solution in the operating room and performed the injection to the submucous myoma. Surgeon B was not allowed to enter the operating room until after injection was completed and was kept blinded to the allocated treatment. Surgeon B performed the resection of submucous myoma and then documented the estimated blood loss and VAS assessment of the intensity of discoloration of the effluent and visual clarity.
Surgeon A performed diagnostic hysteroscopy with an operating channel and a 30-cm long, 20-gauge, single-lumen ovum aspiration needle connected to a 10-mL syringe with the study fluid used to allow direct infiltration of the submucous myomas under visual control. After systematic examination of the uterine cavity, the ovum aspiration needle was inserted through the working channel of the hysteroscope until the injection tip was visualized. The needle tip was then inserted into the submucous myoma and the study fluid was injected. If multiple leiomyomas were noted, each was injected according to decreasing size until visual blanching was observed or until the maximum 10 mL of solution had been used. Surgeon A could choose either to leave or remain present and observe the resection procedure but was not allowed to communicate with or assist Surgeon B.
Surgeon B performed cervical dilatation with Hegar dilators. A 26-Fr monopolar resectoscope equipped with a loop electrode was used. The electrosurgical unit was set at 60-W cutting energy in blend mode. The myoma was resected using a standard slicing technique and 1.5% glycine was used as the distending solution. The inflow pressure was maintained by gravity and the fluid bags were kept at 90–100 cm. The inflow and outflow volume of the distension solution were monitored continuously during the procedure by nurses not involved in the study. The outflow fluid was collected into measuring bottles by connecting tubing to the outflow port of the resectoscope and by using a pouch placed beneath the patient's buttocks. Each glycine bag contained 2 L of fluid and the unused fluid was emptied into a measuring bottle to calculate the volume of the inflow fluid used. The volumes of the fluid used during injection of the study drug and myomectomy were recorded separately. The time for cervical dilatation, myomectomy, polypectomy, and total surgical time were recorded separately. The duration of hysteroscopic myomectomy was determined from the time of insertion of the resectoscope to when the procedure was completed. After completing the procedure, Surgeon B completed a record sheet to document estimated blood loss, rating of bleeding (1=minimal, 2= moderate, 3=excessive), VAS on the intensity of discoloration of effluent (VAS 0–10, 0 clear, 10 unclear), and visual clarity (VAS 0–10, 0 worst visual clarity, 10 best visual clarity). The completeness of resection and intraoperative and postoperative complications were documented. Patients were scheduled for follow-up at 4 weeks at which time transvaginal ultrasound scans were performed to check for residual myoma and at 3 months to review the pictorial blood loss assessment charts.
Continuous variables are presented as median and range, whereas qualitative variables are presented as absolute frequency and percentage. Comparisons between categorical variables were tested by the use of contingency tables and calculation of the χ2 test or Fisher's exact test. Mann–Whitney U test was used for comparisons between continuous variables. P<.05 was considered statistically significant. Statistical analysis was performed using Statistics Package for Social Science 20.
An approval certificate was obtained from the Hong Kong Pharmacy and Poisons Board for the use of vasopressin in the clinical trial (certificate number 002961). Ethical approval was obtained from the Institutional Clinical Research Ethics Review Board (CRE reference number 2011.018-T). The protocol has been registered with the center for clinical trials, clinical trials registry (CCTCTR number CUHK_CCT00291). Written consent was obtained from all the recruited patients.
From November 2011 to February 2014, 48 patients were assessed for eligibility. Eight patients were excluded from the study as a result of medical conditions, prior use of GnRH agonists, or refusal to participate.
Forty patients fulfilled the study recruitment criteria and were randomized; 20 patients received the allocated treatment in each arm. One patient randomized to the placebo group was found to have the myoma protruded out from the cervical os at the time of surgery and was therefore excluded from the statistical analysis (Fig. 1).
The clinical characteristics of the patients are shown in Table 1. The two groups were comparable in age; parity; history of vaginal delivery; grade, size, and number of submucous myoma; type of anesthesia used; and surgeon involved. The median time for cervical dilatation was similar for both groups. The median duration for myomectomy was 19.9 minutes in the vasopressin group compared with 29.0 minutes in the placebo group. The difference was not statistically significant (P=.14). The percentage of cases with time for myomectomy exceeding 20 minutes was 50.0% in the vasopressin group and 68.4% in the placebo group.
Although the primary outcome result was negative, our prespecified secondary outcomes did show some important clinical and statistical differences including volume of inflow fluid, fluid deficit, blood loss, and visual clarity. There were significant reductions in the median volume of fluid used (4,200 mL compared with 9,800 mL) and fluid intravasation (0 mL compared with 300 mL) in the vasopressin group. One participant in the placebo group had a negative fluid deficit (−150 mL) as a result of heavy bleeding during the resection. The intraoperative blood loss was significantly different in both estimated blood loss and surgeon's rating. Up to 95% in the vasopressin group were rated as “minimal” compared with 36.8% in the placebo group (P=.001). The surgical field and the effluent fluid were clearer in the vasopressin group (Table 2) (see the Video, available online at http://links.lww.com/AOG/A566).
No patients in the vasopressin group had experienced any drug-related complications. There was one case of uterine perforation in the placebo group during cervical dilatation. In three participants receiving placebo, the surgeon commented that there was incomplete resection. These three cases were stopped as a result of a large volume of inflow fluid or significant fluid deficit. The surgeons terminated the procedure when the inflow volume reached 20 L or when the inflow volume reached 10 L with a fluid deficit of more than 1,000 mL.
Transvaginal ultrasound scans were performed at the 4-week follow-up, except for one patient in the placebo group who did not come to her appointment and had an ultrasound scan at 3 months. From the scans, one case of residual submucous myoma was confirmed in one of the participants with incomplete resection. Thirty-seven patients came to the follow-up at 3 months; all except two patients in the placebo group who reported improvement in their pictorial blood loss assessment chart score. Two of the three patients who failed to come to the scheduled appointment were contacted by phone and reported symptom improvement. One patient in the placebo group could not be contacted.
The histopathologic examination confirmed leiomyoma in all but three patients. One case in the vasopressin group was confirmed as a uterine sarcoma and two cases were endometrial polyps. The data were reanalyzed after excluding these three cases and showed similar statistically significant differences between groups. It was shown, however, that the participant with sarcoma accounted for the upper range of blood loss, VAS of intensity of discoloration of effluent, and poor visual clarity (after exclusion of the participant with sarcoma: blood loss 5 mL [range 1–20 mL], surgeon rating of bleeding minimal=100%, VAS of intensity of discoloration of effluent 1 [range 0–2], and VAS of visual clarity 9 [range 7–10]).
Hysteroscopic myomectomy is the preferred surgical treatment for submucous myoma. Complete resection may be difficult, the constrain being the amount of fluid intravasation that can be life-threatening.2,3,5,6 Vasopressin has been used in other gynecologic surgeries to control blood loss by its vasoconstrictive effects.13 The potential mechanisms to facilitate hysteroscopic myomectomy are that the constricted blood vessels reduce blood loss allowing for a clearer surgical field, less fluid is required for irrigation, and less fluid can enter the circulation through the constricted vessels. We have previously reported a simple technique of direct injection of vasopressin by diagnostic hysteroscopy with an ovum aspiration needle.17 We evaluated this method in the present randomized trial.
Our analysis indicated there was only 27% reduction in the cases with surgical time more than 20 minutes. This may be the result of the relatively high proportion of larger sized (44% more than 3 cm) myoma recruited into the trial. Also, the median time for myomectomy was 9 minutes less in the vasopressin group, but this did not reach statistical significance because our study was powered to detect a larger difference than actually observed. We postulate that reduction in operative time was not as marked as anticipated possibly as a result of inadvertent injection of vasopressin close to the myometrium, which may have caused uterine contraction, which needed time for relaxation and distension.
Although the primary outcome was negative, our study showed that by using vasopressin, the median inflow volume was 5,600 mL less and the median fluid deficit was zero. This is clinically useful because it would reduce the risks associated with fluid intravasation that is considered to be the most dangerous complication of hysteroscopic myomectomy.
Excluding the participant with sarcoma, all the participants receiving vasopressin were graded with “minimal” bleeding compared with 36.8% in the placebo group. The surgical field was clearer as reflected by the clearer effluent and higher VAS scores on visual clarity.
The use of vasopressin in hysteroscopic myomectomy has been studied. Corson et al15 injected 2–4 units of diluted vasopressin to the intracervical stroma. In that study, the odds of having a fluid deficit of more than 500 mL was 0.15 (0.03–0.94) for vasopressin compared with placebo.15 Another placebo-controlled study on intracervical injection of vasopressin was reported by Phillips et al.16 In the 51 women with hysteroscopic myomectomy, the volume of fluid intravasation was 448.5±47 mL with vasopressin and 819.1±79.7 mL in the placebo group. However, in these two studies, some patients were also given GnRH agonists, so the degree of effect contributed by vasopressin is uncertain.
In another review, 55% of the patients were given vasopressin by deep intramural injection based on the preference of the surgeon.20 The author suggested that it was difficult to infiltrate the myomas and therefore they were not able to achieve the desired results.20 With the use of a long ovum aspiration needle through the working channel of the diagnostic hysteroscope, we were able to achieve direct contact with the submucous myomas including myomas located high at the fundus for vasopressin injection. The instruments for injection are readily available and the procedure is quick, simple, and immediately effective. Vasopressin has been used in open and laparoscopic myomectomy. Our results would also support its role in facilitating hysteroscopic myomectomy. However, adverse cardiovascular effects have been reported. To minimize the risks, patients with vascular disease, epilepsy, migraine, asthma, or heart failure should not be given vasopressin. An upper limit of 10 mL was chosen because previous studies indicated that up to 4 units of diluted vasopressin could be used without any associated side effects.15,16,21 The anesthetists should be informed to ensure close hemodynamic monitoring during the procedure.
The strengths of our study are that we have ensured blinding of the patient and the surgeon to avoid bias on outcome assessments. Only two surgeons with similar experience were involved in the resection of the myoma to minimize intersurgeon operative differences. We have also included more technically demanding grade 2 myoma, which were up to 5 cm in size. The assessment of complete resection was not only by surgeon's subjective opinion, but confirmed by postoperative ultrasound examination that was performed for all patients in the study.
In conclusion, transcervical intralesional vasopressin injection in hysteroscopic myomectomy was shown to be effective in reducing use of inflow fluid, fluid intravasation, intraoperative blood loss, and improving visual clarity.
1. Hillard PA. Benign diseases of the female reproductive tract: symptoms and signs. In: Berek JS, Adashi EY, Hillard PA, editors. Novak's gynecology. 12th ed. Baltimore (MD): Williams & Wilkins; 1996:331–97.
2. Lasmar RB, Xinmei Z, Indman PD, Celeste RK, Di Spiezio Sardo A. Feasibility of a new system of classification of submucous myomas: a multicenter study. Fertil Steril 2011;95:2073–7.
3. Polena V, Mergui JL, Perrot N, Poncelet C, Barranger E, Uzan S, et al.. Long-term results of hysteroscopic myomectomy in 235 patients. Eur J Obstet Gynecol Reprod Biol 2007;130:232–7.
4. Lower AM. Laparoscopic myomectomy. In: Studd J, editor. Progress in obstetrics and gynecology. Vol. 15. London (UK): Churchill Livingstone; 2003:381–8.
5. Di Spiezio Sardo A, Mazzon I, Bramante S, Bettocchi S, Bifulco G, Guida M, et al.. Hysteroscopic myomectomy: a comprehensive review of surgical techniques. Hum Reprod Update 2008;14:101–9.
6. Wamsteker K, Emanuel MH, de Kruif JH. Transcervical hysteroscopy resection of submucous fibroids for abnormal uterine bleeding: results regarding the degree of intramural extension. Obstet Gynecol 1993;82:736–40.
7. Propst AM, Liberman RF, Harlow BL, Ginsburg ES. Complications of hysteroscopic surgery: predicting patients at risk. Obstet Gynecol 2000;96:517–20.
8. Lethaby A, Vollenhoven B, Sowter M. Pre-operative GnRH analogue therapy before hysterectomy or myomectomy for uterine fibroids. The Cochrane Database of Systematic Reviews 2001, Issue 2. Art. No.: CD000547. DOI: 10.1002/14651858.CD000547.
9. Fedele L, Vercellini P, Bianchi S, Brioschi D, Dorta M. Treatment with GnRH agonists before myomectomy and the risk of short-term fibroid recurrence. Br J Obstet Gynecol 1990;97:393–6.
10. Campo S, Campo V, Gambadauro P. Short-term and long-term results of resectoscopic myomectomy with and without pretreatment with GnRH analogs in premenopausal women. Acta Obstet Gynecol Scand 2005;84:756–60.
11. Kamath MS, Kalampokas EE, Kalampokas TE. Use of GnRH analogues pre-operatively for hysteroscopic resection of submucous fibroids: a systematic review and meta-analysis. Eur J Obstet Gynecol Reprod Biol 2014;177:11–8.
12. Mavrelos D, Ben-Nagi J, Davies A, Lee C, Salim R, Jurkovic D. The value of pre-operative treatment with GnRH analogues in women with submucous fibroids: a double-blind, placebo-controlled randomized trial. Hum Reprod 2010;25:2264–9.
13. Muzii L, Boni T, Bellati F, Marana R, Ruggiero A, Zullo MA, et al.. GnRH analogue treatment before hysteroscopic resection of submucous myomas: a prospective, randomized, multicenter study. Fertil Steril 2010;94:1496–9.
14. Kongnyuy EJ, Van Den Broek N, Wiysonge CS. A systemic review of randomized controlled trials to reduce hemorrhage during myomectomy for uterine fibroids. Int J Gynaecol Obstet 2008;100:4–9.
15. Corson SL, Brooks PG, Serden SP, Batzer FR, Gocial B. Effects of vasopressin administration during hysteroscopic surgery. J Reprod Med 1994;39:419–23.
16. Phillips DR, Nathanson HG, Milim SJ, Haselkorn JS, Khapra A, Ross PL. The effect of dilute vasopressin solution on blood loss during operative hysteroscopy: a randomized controlled trial. Obstet Gynecol 1996;88:761–6.
17. Wong AS, Cheung EC, Leung KT, Yeung SW, Leung TY, Fung TY. Transcervical intralesional vasopressin injection in hysteroscopic myomectomy—description of a new technique. J Laparoendosc Adv Surg Tech A 2013;23:258–62.
18. Perino A, Chianchiano N, Petronio M, Cittadini E. Role of leuprolide acetate depot in hysteroscopic surgery: a controlled study. Fertil Steril 1993;59:507–10.
19. Higham JM, O'Brien PM, Shaw RW. Assessment of menstrual blood loss using a pictorial chart. BJOG 1990;97:734–9.
20. Emanuel MH, Hart A, Wamsteker K, Lammes F. An analysis of fluid loss during transcervical resection of submucous myomas. Fertil Steril 1997;68:881–6.
21. Frishman G. Vasopressin: if some is good, is more better? Obstet Gynecol 2009;113:476–7.