Hysteroscopic findings after laparoscopic and open myomectomy with or without uterine cavity breach: historical cohort study

Abstract

Introduction

The aim of this study was to evaluate hysteroscopic findings after laparoscopic and laparotomic myomectomy with a focus on the presence of postoperative intrauterine adhesions in groups of patients with and without perioperative uterine cavity breach (UCB).

Material and methods

This is a historical cohort study. Our database was searched to identify patients with UCB during myomectomy and matched the same number of patients after myomectomy without UCB to create a control group. All relevant data were retrieved from our medical records. In both groups, the results of follow-up hysteroscopy were analyzed.

Results

Low prevalence of intrauterine adhesions after myomectomy was observed in only 3.5% of the 170 patients in our samples. No significant difference in the occurrence of synechiae between the patients with and without UCB was found (2 vs. 4, RR 0.5, 95% CI 0.1–2.7, p = .341), nor was the difference in other hysteroscopic findings. Follow-up hysteroscopy was performed with slender optics and expandable casing system without need of any anesthesia in 87.1% of cases.

Conclusions

According to our findings, the prevalence of post-myomectomy intrauterine adhesions after myomectomy is low. Our study did not demonstrate that UCB during myomectomy is a risk factor for the formation of intrauterine synechiae.

Keywords:

• Fertility
• hysteroscopy
• intrauterine adhesions
• leiomyoma
• myomectomy

Introduction

Uterine leiomyomas are the most common benign pelvic tumors [1]. They occur in 40–60% of women of reproductive age. The exact pathogenesis of these tumors remains unknown; however, it includes a heterogeneous assemblage of hormones, growth factors, stem cells, genetic, and epigenetic abnormalities [2,3]. Myomectomy still represents irreplaceable management of uterine leiomyomas, especially in patients with undetermined reproductive plans and a deformed uterine cavity or symptomatology related to fibroids [1,4,5]. The harm to the uterus, either the myometrium or endometrium, can be diverse and depends on the size and location of the myoma. Surgical procedures with uterine cavity breach (UCB) may represent a special category while speaking of saving fertility. Regardless of the reason for surgery, whether it is clinical symptomatology, size of uterine pathology or infertility, the main goal of these procedures remains the same – to create a spacious uterine cavity and compact myometrium suitable for receiving and carrying pregnancy. For submucosal leiomyomas (FIGO type 0–2) [6], hysteroscopic resection is the method of choice [7,8]. However, for other types of myomas (FIGO type 3–8) or multiple fibroids, the abdominal approach is the only option.

The opening of the uterine cavity during abdominal surgery is usually unavoidable in transmural or large submucosal leiomyomas [6], nevertheless, the approach is open or laparoscopic. In these cases, the surgeon harms the uterine cavity, which is a crucial space for pregnancy. There are few papers which examine the presence of post-myomectomy synechiae. Those that do exist usually observe heterogeneous groups and high percentages of intrauterine adhesions presented in follow-up hysteroscopy [9,10]. According to our pre-study experience, the frequency of post-myomectomy synechiae is not high, and it may vary depending on periprocedural insult to the endometrium.

Material and methods

Study design and setting

This was a historical cohort study performed in the Department of Minimally Invasive Gynecological Surgery of General Faculty Hospital. All patients who underwent laparoscopic or open myomectomy with UCB between January 2013 and August 2018 were searched in our database (UCB group). For patients with UCB during myomectomy, follow-up hysteroscopy has always been part of the standard management in our department. In the history of our department, second-look laparoscopy and hysteroscopy used to be performed on all pregnancy-planning patients after myomectomy, regardless of uterine cavity breach. As there was no benefit seen from those follow-up procedures, starting in 2010, the second-look laparoscopy or hysteroscopy was no longer obligatory follow-up for all patients. From that historical time period (2007–2009), the same number of patients with open or laparoscopic myomectomy was picked, but without UCB. They were matched on the dominant fibroid size to create a control group because the postoperative hysteroscopic findings of those patients were available.

Ethical approval

All patients gave written consent to the use of their clinical data for scientific purposes before the surgical procedure. The study was approved by the hospital Ethical Committee (1444/19 S-IV).

Patients

The inclusion criteria for both groups were as follows: laparoscopic or open myomectomy in our department; dominant leiomyoma with mainly intramural location – FIGO 3, 4, 5 or 2–5 [6]; passing all the mandatory postoperative procedures (ultrasound, hysteroscopy).

The exclusion criteria for both groups were as follows: presence of any fibroid FIGO type 0, 1 or 2 on the preoperative ultrasound; dominant myoma FIGO type 6, 7 or 8 [6]; unsigned informed consent form with an agreement to the use of patient’s data for scientific work; and hysteroscopic resection of leiomyomas. The coexistence of smaller fibroids FIGO types 6–8 was not considered an exclusion criterion.

Our medical records were used to evaluate the perioperative data. After the main surgery, all patients had standard follow-up care with vaginal ultrasound (USG) examination of the uterus 1 month postoperatively and with hysteroscopy performed between 2 and 3 months postoperatively for evaluation of the uterine cavity.

Study objectives

Our primary objective was the assessment of the intrauterine findings focusing on the presence of synechiae or cavity deformation and the need for intervention during postoperative hysteroscopy while comparing the UCB group with controls.

The secondary goal was to analyze the parameters of the follow-up hysteroscopy in the group with UCB: the need for general anesthesia, the type of hysteroscope used, and the length of the procedure. Follow-up hysteroscopy in the control group was part of second-look laparoscopy meaning that it was performed under general anesthesia; therefore, we did not closely analyze the parameters of the procedure itself for this group.

Surgical techniques

All surgical procedures took place in the Department of Minimally Invasive Gynecological Surgery and were performed by one surgical team – four surgeons who alternated as the main surgeon and the assistants. Patients did not have any pharmacological pre-treatment prior to myomectomy (i.e., selective progesterone modulators or GnRH analogs).

Myomectomy

The myomectomy was performed by laparoscopy or laparotomy according to preoperative USG findings or first laparoscopic perspective. For dominant leiomyoma ≥10 cm on preoperative USG, the open approach was chosen prior to the surgery. Elective conversion to laparotomy occurred after the first laparoscopic trial due to the size or/and number of fibroids in some cases. There was no surgical tool vaginally inserted into the uterine cavity for guidance or need to manipulate the uterus. All leiomyomas were enucleated with the same technique (adopted in our department since 2005), with no difference between the open or laparoscopic approach: after vertical incision on the most protruding surface (sometimes with the need for intraoperative ultrasound guidance for deeper localized leiomyomas) of fibroid, the serosa and musculature were cut with unipolar scissors to reach the fibroid. Additionally, the leiomyoma was enucleated with traction and countertraction while using bipolar coagulation to minimize blood loss and to preserve pseudocapsula [11]. Description of eventual breaching of the uterine cavity (identified visually as the presence of endometrium – shown in Figure 1) has always been an obligatory part of the surgical protocol. In the UCB group, there is also histological verification of the presence of endometrial tissue in specimens. When the removal of leiomyoma was completed, the multilayer running suture of myometrium with atraumatic 1–0 absorbable polyglactin stitches (Coated Vicryl 1–0 by Johnson & Johnson, Ethicon US, LLC) was performed, usually with two or three layers according to the leiomyoma bed size, including the serosa in the last layer. The suture was performed by saving the endometrium, which means the endometrium was excluded from the suturing and the breached space was overlaid by myometrial tissue.

Figure 1. On the left – intrauterine breach during myomectomy with endometrial tissue visible on the bottom of the fibroid bed. On the right – myomectomy without opening of the uterine cavity during the enucleation of leiomyoma.

Ultrasound follow-up

Control USG was performed approximately 1 month after myomectomy, with a focus on echo structure of the myometrium to assess the compactness of the uterine wall as well as the healing scar after myomectomy, presence of residual leiomyoma/s and symmetry of the uterine cavity.

Hysteroscopy

Hysteroscopy was scheduled 2–3 months after primary surgery during the first one-third of the menstrual cycle. The uterine cavity (spaciousness and symmetry), the ostia of the tubes, the endometrium, and the presence of intrauterine pathology were assessed during the procedure. All these characteristics were evaluated subjectively by a surgeon. In cases with intrauterine pathology, there was also histological confirmation. All hysteroscopic procedures took place in the operating room, given the possibility of immediate conversion to general anesthesia and/or operative hysteroscopy.

In the control group, hysteroscopy was part of second-look laparoscopy. The rigid 5.5-mm diagnostic hysteroscope (Olympus Surgical Technologies; Europe Olympus Winter § IBE GmbH, Hamburg, Germany) was used in all patients at first. An operative 8.5-mm-wide hysteroscope or resectoscope (VersaPoint, Hysteroscopic Bipolar Resectoscopic System, Gyncare; Ethicon, Inc., Menlo Park, CA, USA) was employed later if needed. Patients were discharged on the first postoperative day after the combined procedure. For the purpose of this study, there was no need to describe the results of the second-look laparoscopy.

In the UCB group, no second-look laparoscopy was planned, so our effort was to perform the procedure without the need for general anesthesia with respect to patient’s preference and postoperative ultrasound findings. There were no other types of anesthesia or analgesia used in such cases. In cases of expected presence of intrauterine pathology (submucosal leiomyoma or polyp on the postoperative USG), the execution of procedure under general anesthesia was preferred. While working without general anesthesia, the 1.8-mm-wide optics in expandable casings (total diameter 3.2 mm; Versascope, Gynecare; Ethicon Inc., Somerville, NJ, USA) was used. Vaginoscopically, the external ostium of the cervix was reached and the progression to the uterine cavity followed. In the case of intrauterine pathology, the surgical procedure was either performed with the same device or electively converted to operative hysteroscope 8.5-mm-wide or to resectoscope (VersaPoint, Hysteroscopic Bipolar Resectoscopic System, Gynecare; Ethicon, Inc., Menlo Park, CA, USA) with the assistance of general anesthesia. In cases without general anesthesia, the patient was dismissed immediately after hysteroscopy. In cases with planned general anesthesia, the 5.5-mm-wide diagnostic hysteroscope was primarily selected. We used an operative hysteroscope or resectoscope when any surgery was needed. After the procedure with utilization of general anesthesia, patients were observed for 2 h in the recovery room, and after another 2 h, they were dismissed.

Statistical analysis

To compare quantitative parameters, the nonparametric Mann–Whitney test was used. In addition, for statistical analysis of the differences between two groups in qualitative parameters, Pearson chi-square test and Fisher’s exact test were used. The relative risks (RRs) with 95% confidence intervals (CIs) were calculated when appropriate.

Results

In the given period, we recruited 85 post-myomectomy patients with UCB who met the criteria for the UCB group. We selected the same number of post-myomectomy patients with follow-up hysteroscopy (but without UCB) who met the criteria for the control group.

While comparing the UCB group and the control group in terms of basic characteristics, we found a significant difference in their average age (35.7 ± 4.5 years vs 33.3 ± 4.6 years, p = .001), the number of removed fibroids (higher in the UCB group) as well as in the diameter of the largest removed leiomyoma and estimated blood loss (both larger and higher in the control group). The groups were not significantly different in the type of surgical approach and need for concomitant surgery. The perioperative characteristics are shown in Table 1.

Table 1. Perioperative characteristics of the UCB group and control group.

Characteristic	UCB group	Control group	p-Value	RR (95% CI)
Number of leiomyomas
Mean	2.1	1.9	˂.001^a
Median	1	1
Range	1–20	1–12
Diameter of the biggest leiomyoma (mm)
Mean ± SD	53.2 ± 18.8	60.4 ± 14.6	˂.001^a
Median	50	60
Range	25–120	25–100
Estimated blood loss (ml)
Mean ± SD	371.8 ± 256.3	399.4 ± 203.2	˂.001^a
Median	300	350
Range	100–1500	100–1500
Concomitant surgery	10	7	.443^b	1.4 (0.6–3.6)
Endometriosis	3	6
LUAO	7	1
Surgical approach	85	85
Laparoscopy	73	71	.670^b	1.0 (0.9–1.2)
Laparotomy	12	14

^aMann–Whitney test, ^bPearson chi-square test.

UCB: uterine cavity breach; LUAO: laparoscopic uterine artery occlusion; CI: confidence interval; RR: relative risk; SD: standard deviation.

There was no significant difference found in hysteroscopic findings. A completely normal uterine cavity was present in 72 patients in the UCB group and 75 patients in the controls (RR for abnormal hysteroscopic findings was 1.3, 95% CI 0.6–2.8, p = .501). The distribution of pathologies was also not significantly different between the two groups (Table 2). The presence of synechiae (Figure 2), which was our primary and main objective, was not significantly different between the groups (two in the UCB group vs. four in the control group, RR 0.5, 95% CI 0.1–2.7, p = .681). All adhesions were classified as dense but mild with easy adhesiolysis and mostly located close to the tubal ostia.

Figure 2. Follow-up hysteroscopy in patients after myomectomy with periprocedural uterine cavity breach with presence of intrauterine synechia on the left and with physiological intrauterine finding on the right.

Table 2. Hysteroscopic findings and comparison of the UCB group and control group.

Hysteroscopic findings	UCB group (n = 85)		Control group (n = 85)		p-Value	RR (95% CI)
	n	%	n	%
Normal uterine cavity	72	82.35	75	88.23	.501^a	1.0 (0.9–1.1)
Submucosal leiomyoma	7	8.24	3	3.53	.328^b	2.3 (0.6–8.7)
Endometrial polyp	4	4.71	3	3.53	1.000^b	1.3 (0.3–5.8)
Synechiae	2	2.35	4	4.71	.681^b	0.5 (0.1–2.7)

^aPearson chi-square test, ^bFisher’s exact test.

UCB: uterine cavity breach; RR: relative risk; CI: confidence interval.

The main characteristics of subsequent hysteroscopy of the UCB group, which is still our standard follow-up for such patients, are shown in Table 3. Synechiolysis was needed only in two patients (2.4%). We performed hysteroscopy without any anesthesia with a 3.2-mm hysteroscope in 87.1% patients and the procedures were generally very well tolerated. There was no need of conversion into general anesthesia due to patient’s discomfort. No complications were observed.

Table 3. Characteristics of hysteroscopy in the UCB group.

Basic characteristics	Average	Range
Time between procedures	88.7 days	20–251 days
Length of hysteroscopy	9.3 min	3–32 min
NRS of pain (in performance without GA)	1.6	0–4
Hysteroscopic procedures	Number	Percentages
Without GA	74	87.1
With GA	11	12.9
No intervention	68	80.0
Synechiolysis	2	2.4
Polypectomy	4	4.7
Endometrial biopsy	4	4.7
TCRM	7	8.2

UCB: uterine cavity breach; NRS: numeric rate scale; GA: general anesthesia; TCRM: transcervical resection of myoma.

Discussion

The presence of intrauterine adhesions is a significant negative factor for women’s fertility [12] and thus, maximum effort should be devoted to preventing such a complication during any reproductive procedure. Despite great regenerative ability of endometrium, any surgical procedure on the uterus can lead to the damage of endometrium and consequently to formation of postoperative intrauterine synechiae [13]. The intrauterine procedures related to pregnancy as a leading cause of intrauterine adhesions in Asherman’s syndrome make up approximately 90% of causes [14,15], on the contrary myomectomy was considered as a potential cause only in 1.3% [15] of cases. These numbers are influenced not only by the characteristics of the pregnant uterus [16] but also by the number of procedures. In the U.S. there were approximately 30,000 to 34,000 myomectomies performed per year in the 1990s [17,18] and up to 600,000 legal abortions per year at the beginning of the new millennium [19], with unknown numbers of procedures for missed abortion or pregnancy-related residua.

In our study, a low frequency of post-myomectomy intrauterine adhesions was detected in the population studied (i.e., regardless of the intraoperative opening of the endometrial cavity). According to the literature, the prevalence of intrauterine adhesions after myomectomy ranges between 9–25% [20–23]; that is much higher than our finding of a prevalence of synechiae of <5% in both groups (Table 1). This discrepancy can be partially explained by several factors. First, in our study no intrauterine manipulation was used during myomectomy. This step of the surgery is not mentioned at all in the other studies but might not be a standard technique for many surgeons [24]. Also, no intrauterine intervention was performed perioperatively, such as hysteroscopy, in contrast with some other studies [20,21] which can markedly alter the numbers. The prevalence of postoperative intrauterine adhesions after hysteroscopic myomectomy can reach up to 45% in multiple fibroids [25] regardless of hysteroscopic technique and instruments [26]. Since hysteroscopy is the gold standard therapy for submucous leiomyomas [8] these patients were excluded to purify the study group for abdominal surgery alone. On the other hand, Tixier et al. reported an extremely low prevalence of post-myomectomy adhesions of 0% in a myomectomy subgroup with temporary surgical ligation of the uterine artery (n = 27) [27]. Our study sample of 170 women undergoing myomectomy with a prevalence of 3.53% postoperative intrauterine adhesions, however, represents one of the largest populations studied so far.

Our hypothesis that injury to the endometrium related to breaching the uterine cavity during myomectomy is a risk factor for synechiae was not confirmed. It is well known that harm to the endometrium is not the only causative factor for intrauterine adhesions. There are other factors, such as individual response to the trauma or surrounding necrosis and fibrosis [28]. This hypothesis can be manifested in patients after uterine artery embolization, where the synechiae were present in 10.2% of patients in follow-up hysteroscopy [29] or with intrauterine adhesions after tuberculosis [15]. Other authors report a slightly higher prevalence of adhesions in the group with opening of the uterine cavity during myomectomy than in patients without periprocedural endometrium injury, but the difference was not significant [20,21]. In these studies, the suture of the endometrium was performed separately with 1–0 vicryl stiches. Our technique of post-myomectomy uterine suture purposely omitted the endometrium, thus only the running suture was performed to close the myometrial bed. There was no defect of the endometrium detected during postoperative hysteroscopy in any patient. We are aware of the possibility of missing a small breach just by visual identification as it is done during our procedures, but we do not consider such a small opening as clinically important, especially in relation to our surgical technique; therefore, we did not use any other identification tool. One study in the literature came to the conclusion that opening the endometrial cavity was a major risk factor for postoperative intrauterine adhesions (OR 6.42) [22].

In our paper it was demonstrated that follow-up hysteroscopy with slender optics and expandable 3.2-mm-wide casing system is feasible, very well tolerated and can be performed without the need of any anesthesia in almost 90% of patients. General anesthesia was used only in cases with intrauterine pathology requiring extensive hysteroscopic surgery. Diagnostic hysteroscopy with simple intervention is feasible with no anesthesia in almost all cases. The specification of follow-up hysteroscopy is not defined in other papers although it might be very important for patients. Sufficient knowledge of the feasibility of the procedure may increase the acceptability of another required surgery for the patients.

During the surgical procedure, no antiadhesive barrier was used inside the uterine cavity, which was in accordance with other studies [20,22]. In one paper, patients were given a high dose of oral contraceptive pills postoperatively for the prevention of spontaneous conception [21]. Even though there is some evidence supporting the usage of antiadhesion methods after intrauterine adhesiolysis or in an effort to prevent intraperitoneal adhesions, no evidence was found for intrauterine administration of such an agent in these cases [30].

The fact that the two groups of patients are from different time periods is definitely one of the essential limitations of our study, as the surgical team could have been more experienced during the years and it might have influenced the results. However, we did not want to burden the patients without UCP with another procedure with questionable benefits to create the time-corresponding control group. Another weak point of our study is the statistically significant differences between the groups in basic characteristics, although the groups were matched by the dominant fibroid size. On the other hand, despite these differences, the prevalence of intrauterine adhesions was very low in both groups and did not differ, so it is not likely that these factors significantly influenced the results. In fact, due to this protocol, our study samples with known hysteroscopic findings are to our best knowledge the largest and most uniform mentioned in the literature.

In our study, we observed a low prevalence of intrauterine adhesions after laparoscopic or open myomectomy, which is not consistent with other studies with smaller study populations. Our results did not demonstrate that breaching the uterine cavity during myomectomy is a risk factor for the formation of intrauterine synechiae, which agrees with the majority of other studies. These results might encourage the surgeons not to postpone or avoid the surgical procedures on the uterus. These procedures should not be associated with the fear of uterine cavity harm and should be offered to pregnancy planning patients where indicated.

Regardless of the very low incidence of post-myomectomy intrauterine adhesions, there is the possibility of follow-up hysteroscopy, which can be offered to patients with unclosed fertility plans, and especially prior to IVF. This minimally invasive procedure with slender optics and expandable casing system is easy to perform, can be completed without the need of any anesthesia, is very well tolerated, and does not postpone a patient’s effort to conceive.

Ethical aspproval

The study was approved by the hospital Ethical Committee (1444/19 S-IV).

Acknowledgments

The authors wish to thank Mrs. Alena Dohnalova, Dr. Marian Rybar Ph.D. and Dr. Martin Hynek, Ph.D. for statistical consultations and analysis of the results.

Declaration of interest

No author has any potential conflicts of interest.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

The study was financially supported by the Charles University in Prague [UNCE 204065] and Všeobecná Fakultní Nemocnice v Praze (General University Hospital in Prague) [00064165].