Laparoscopic myomectomy for fibroids penetrating the uterine cavity: is it a safe procedure?


* Dr R. Seracchioli, Center for Reconstructive Pelvic Endosurgery, Reproductive Medicine Unit, S. Orsola Hospital, University of Bologna, Massarenti 13, 40138 Bologna, Italy.


Objective The purpose of the study was to evaluate the post-operative course and follow up of women who had undergone laparoscopic removal of intramural fibroids penetrating the uterine cavity.

Design Retrospective study.

Setting Center for Reconstructive Pelvic Endosurgery, Italy.

Population Thirty-four women with fibroids penetrating the uterine cavity.

Methods Laparoscopic myomectomy.

Main outcome measures Feasibility and safety of surgical technique, length of operation, blood loss, intra- or post-operative complications, length of hospital stay, resolution of symptoms and future obstetric outcome.

Results The mean operative time was 79 (SD 30) minutes; the mean reduction in haemoglobin was 1.1 ± 0.9 g/dL. No intra- or post-operative complications were observed. The average post-operative stay in hospital was 54 (SD 22) hours. Nineteen (73%) out of 26 patients who had experienced symptoms prior to surgery reported resolution of these symptoms post-operatively. All patients resumed work within a mean time of 20 (SD 8) days. Among 23 of the 32 patients attempting pregnancy during the follow up period, nine (39%) conceived within one year. Seven pregnancies went to term without complications.

Conclusion The clinical results of this study suggest that laparoscopic myomectomy for intramural fibroids penetrating the uterine cavity is a safe procedure, providing well known advantages of minimal access surgery.


Laparoscopic myomectomy is a procedure whose feasibility has been definitely demonstrated1–3 but is indicated for selected patients only4. For example, the choice between laparotomy and operative laparoscopy is based on a number of restrictions such as the number, size and location of the fibroids4. These parameters can limit the use of the laparoscopic approach, but no definitive criteria have been established5. It is well known that intramural fibroids, reaching the uterine cavity, might represent a contraindication to laparoscopic myomectomy, due to intra- and post-operative bleeding and inadequate closure of the endometrium/myometrium6,7.

The purpose of this study was to evaluate the post-operative course and follow up of a series of patients who underwent laparoscopic removal of intramural fibroids penetrating the uterine cavity.

There are a limited number of published studies evaluating laparoscopic myomectomy where fibroids penetrate the endometrial cavity. One group8 reported a series of 32 patients who had laparoscopic suturing of the endometrial cavity, with good reconstructive outcome. There are no data about obstetric outcome and long term follow up. In another study9, four women had entry and repair of the uterine cavity but no specific clinical details such as type of fibroids, size, location, subsequent pregnancy and delivery history were given. Recently, a series of seven pregnancies after laparoscopic removal of fibroids from the endometrial cavity followed by laparoscopic repair of the uterine defect has been reported10. Four patients were delivered at or near term by caesarean section, one delivered preterm at 28 weeks (vaginally), without dehiscence and two patients had a miscarriage at eight weeks.


This study concerns 34 non-consecutive patients who underwent laparoscopic myomectomy at the Center for Reconstructive Pelvic Endosurgery, Reproductive Medicine Unit, S. Orsola Hospital, University of Bologna, Italy, between January 1996 and January 2000. All patients had at least one intramural fibroid penetrating the uterine cavity, defined as a tumour located intramurally extending through the entire thickness of the myometrium and reaching the endometrial cavity. The presence of the fibroid in the endometrial cavity was confirmed in all cases by hysteroscopy. Patients with intramural fibroids not reaching the uterine cavity were excluded.

Patient details are presented in Table 1. The mean age was 34.9 (SD 4.2) [range 28–45] years; 27 were nulliparous and 7 were primiparous. At diagnosis, 26 of the 34 patients were symptomatic with menorrhagia (n= 15), pelvic pain (n= 6), lower abdominal discomfort (n= 3), dysmenorrhoea (n= 2). The remaining eight patients were asymptomatic, but the increase in fibroid size measured by ultrasound represented an indication for surgery.

Table 1.  Main characteristics of 34 patients. Values are expressed as n (%) or mean [SD].
Age (years)34.9 [4.2]
Nullipara27 (79.4)
Primipara7 (20.6)
Symptomatic26 (76.4)
Bleeding15 (44.1)
Pelvic pain6 (17.6)
Lower abdominal discomfort3 (8.8)
Dysmenorrhoea2 (5.8)

Twenty-three out of 34 patients had been trying to conceive for a mean of 32 (SD 26) months. In 10 patients, the presence of fibroid was the only known cause of their subfertility. In the others, 11 had tubal pathology (hydrosalpinx, proximal anomalies) and in two cases, there was a male infertility factor.

All patients underwent complete physical examination, vaginal and abdominal ultrasound and hysteroscopy. We recommended all patients to use a reliable method of contraception from the first day of the menstrual cycle before surgery. All patients with irregular menstrual cycles had a pregnancy test the day before surgery.

No pre-operative treatment with gonadotrophin-releasing hormone was used to reduce the size of the uterine fibroids.

Prophylactic antibiotic therapy (ampicillin 2 g im) was given per-operatively, and continued only if there was a persistent post-operative pyrexia.

All surgical procedures were performed by the same investigators (RS and SV) who are experienced in laparoscopic surgery.

For each patient, the total operating time from skin incision to closure was recorded. Change in haemoglobin concentration, measured the day before surgery and 24 hours after, and the incidence of febrile morbidity (indicated by a temperature of 38°C or higher on two consecutive measurements at least 6 hours apart, excluding the first 24 hours) were reported. Blood loss at the time of surgery was recorded.

The length of hospital stay was also noted. Patients were discharged when they were able to tolerate a normal diet, dress themselves and be fully mobile around the ward, be analgesic-free and be satisfied that they could manage at home.

The mean follow up was 32 (SD 19) months, and ultrasound assessments were performed at 3, 6 and 12 months in the first year and every 6 months thereafter. Patients wishing to become pregnant agreed to use contraceptives until they had undergone hysteroscopy six months after surgery.

Under general anaesthesia, the pneumoperitoneum and video laparoscopy were established. Two suprapubic access routes (5 mm) were inserted lateral to the inferior epigastric arteries but slightly higher than usual. A third trocar (10 mm) was inserted in the midline. Uterine cannulation was used in order to obtain optimal exposure of the fibroid, in particular, when it was in a posterior location. A methylene blue test was carried out to confirm tubal patency and to stain the endometrium (this is a useful technique for fibroids embedded deeply in the myometrium). An incision was made in the serosa overlying the fibroid using a monopolar pointed knife. The myometrium retracted as the incision was made, exposing the fibroid which was then enucleated by entering into the cleavage plane using claw forceps and scissors. Traction on the fibroid with counter-traction on the uterus facilitated dissection. Vessels in the connective tissue bridges between the fibroid and the uterus were electrocoagulated with bipolar current before being cut.

Once the fibroid had been completely enucleated, the opening of the uterine cavity was confirmed by the spreading of the methylene blue.

Haemostasis was completed by bipolar coagulation.

The myometrial defect was sutured in two layers. The first layer was approximated with separate stitches with Biosyn, 1/0–2/0 (Glycomer 631, monofilament synthetic absorbable, Auto Suture, Milan, Italy) using intracorporeal knots. The second myometrial layer was closed using the same suture material either interrupted or continuous.

Removal of the fibroids from the abdominal cavity was achieved using an electrical laparoscopic morcellator (Wisap, Sauerlach, Germany), and a trocar, 15 or 20 mm in diameter inserted through a suprapubic incision. The pelvis was irrigated with saline solution, but no absorbable adhesion barrier or macromolecular solutions were left in the peritoneal cavity.


All 34 patients underwent laparoscopic myomectomy. A total of 96 fibroids were removed [mean 2.8 (SD 2.2) for each patient]. The number reaching the uterine cavity was 34 (one for each patient), with a mean diameter of 6.5 (SD 2.9) [range 4–15] cm (Table 2).

Table 2.  Characteristics of fibroids removed. Values are expressed as n or mean (SD).
No. of fibroids removed96
No. of fibroids per patient2.8 (2.2)
No. of fibroids reaching uterine cavity34
Mean diameter of fibroids reaching uterine cavity (cm)6.5 (2.9)

In 12 patients, only one fibroid was removed; multiple myomectomy was achieved in the remaining 22 patients.

The mean operative time was 79 (SD 30) [range 45–130] minutes.

The mean reduction in haemoglobin was 1.1 ± 0.9 g/dL with a range of 0.2–2.9 g/dL. No patient required blood transfusion. The average blood loss was 330 (SD 144) (Table 3).

Table 3.  Post-operative course of patients. Values are expressed as n or mean (SD).
  1. Hb = haemoglobin.

Mean operative time (minutes)79 (30)
Average blood loss (mL)154 (75)
Average drop in Hb (g/dL)1.1 (0.9)
Fever (>38°C)3
Average hospital stay (hours)54 (22)

No intra-operative complications were observed.

The rate of conversion to laparotomy was 2.9%, this being necessary in one patient with a large fibroid (12 cm) deeply embedded in the isthmic portion of uterus, which could not be removed laparoscopically.

Antibiotic therapy was given post-operatively in three cases of pyrexia more than 38°C for 24 hours but no uterine or pelvic infection was confirmed.

The post-operative hospital stay was 54 (SD 22) [range 36–138] hours.

In 10 patients, there were concomitant surgical procedures for endometriomas (5), paradenexal cysts (2) and ovarian cyst (1). Adhesiolysis was performed in two cases.

All patients except two were followed up after surgery. These two patients failed to attend for post-operative assessment.

Nineteen (73%) of the 26 patients who had experienced symptoms prior to surgery, reported resolution of the symptomatology, and 5 (19%) declared themselves satisfied and improved. Two (8%) with pelvic pain and dysmenorrhoea pre-operatively had no change in these symptoms.

Eleven (73%) out of 15 patients stated regular menses after surgery, the remaining four patients (27%) reported no bleeding but had occasional spotting. Patients recommenced work within a mean time of 20 (SD 8) days.

Hysteroscopy was performed in the patients who wished to achieve pregnancy (n= 23). In 21 cases, a normal cavity with a normal endometrium, without evidence of uterine scar and synechia, was documented. In two cases, the uterine cavity showed mild distortion as a result of diffuse fibromatosis, but without any visible defect of the uterine closure.

Recurrence of fibroids occurred in five (16%) patients during the follow up period: two patients having one fibroid (5 and 6 cm diameter, respectively). They underwent a second laparoscopic myomectomy with a good reconstructive outcome as evaluated by ultrasound at the post-operative appointments.

Twenty-three patients (71%) attempted pregnancies during the follow up period. Nine women (39%) conceived within one year: eight spontaneously, and one after in vitro fertilisation (IVF). Seven pregnancies (six spontaneous, one after IVF) went to term: vaginal delivery occurred in one patient, while six underwent elective caesarean section. No uterine ruptures occurred during gestation and/or labour.

The remaining two pregnancies ended in miscarriage (Table 4).

Table 4.  Obstetric outcome of 23 patients attempting pregnancy after laparoscopic myomectomy.
  1. * One of these patients conceived after in vitro fertilisation.

No. of pregnancies9
No. of miscarriage2
No. of preterm delivery0
No. of vaginal delivery1
No. of caesarean sections6*
No. of uterine rupture0


The benefits of laparoscopic myomectomy have been demonstrated in several studies3,11,12.

Some authors1,13,14 have tried to state criteria for selecting women for this procedure under satisfactory and safe conditions. More recently, the indications have been modified and extended because of the improvement in laparoscopic technique and surgical skills5. This allows a greater number of women with fibroids to be treated with operative laparoscopy rather than the traditional abdominal approach.

One of the most important issues for laparoscopic myomectomy involves adequacy of approach for fibroids deeply intramural which reach the endometrial cavity15–17. There are few studies about this in the literature8,10.

The removal of fibroids penetrating the uterine cavity is of concern because of the risk of intra- and post-operative bleeding. Also, inadequate closure of the endometrium and myometrium might increase the risk of uterine rupture during a future pregnancy. There might also be a longer operative time with an increased risk of uterine infection.

We found the mean haemoglobin loss to be low despite not using pharmacological vasoconstriction (vasopressin), mechanical vascular occlusion or pre-operative medical treatment (GnRH analogues). Earlier studies18 have recommended the use of such haemostatic techniques during myomectomy to reduce operative blood loss, but their use has not been proven to reduce blood loss or transfusion requirements19. We did not administer GnRH analogues pre-operatively because the use of GnRH analogues can result in a softening of the uterine myoma, rendering identification of the cleavage plane more difficult and lengthening the surgical dissection20,21. Furthermore, there is no significant difference in the degree of surgical difficulty in the treatment of reduced fibroids after medical therapy. On the contrary, the reduction in size makes difficult the identification of small fibroids and might increase the probability of further surgical interventions for recurrence22.

Candiani et al.23 analysed the risk of recurrence of uterine fibroids in a large series of patients who underwent abdominal myomectomy. The recurrence observed was 27% after 10 years. Laparoscopic myomectomy, according to several authors24–27, might present with a higher risk of recurrence compared with abdominal myomectomy. However, Rossetti et al.28 asserted that the recurrence rate after laparoscopic myomectomy is close to that seen after abdominal myomectomy. We found a recurrence rate of 16%. This result is comparable to those observed by several authors24–27. It is a low percentage probably because of the small number of cases analysed and the follow up duration of three years might not be sufficient to reveal all recurrences.

With penetration of the uterine cavity, the problem of uterine suturing should be highlighted because of the possibility of uterine rupture during a subsequent pregnancy. The suture should always take up the full depth of the edge of the incision and result in total contact over the whole of the myomectomy defect in order to avoid haematoma formation inside the myometrium. The risk factors for uterine rupture include failure to adequately close the endometrial cavity, by failing to approximate the tissues, using the wrong suture, electrically caused necrosis and surgeon's inexperience. In this study, as the fibroids reached the cavity, the vertical deep incision was closed in two layers (as described earlier). We avoided the extensive use of thermal energy for haemostasis because it might lead to poor vascularisation and tissue necrosis with an increased risk of uterine rupture and fistula formation29,30.

In our series, there were no documented uterine ruptures in those patients who conceived, suggesting that the uterus can be repaired satisfactorily by the laparoscopic approach even when the cavity is opened. Uterine ruptures reported after laparoscopic myomectomy15–17,31–33 could be due to technical errors connected with the surgeon's lack of experience, as suggested by Dubuisson et al.34. To date, the overall rate of uterine rupture attributable to laparoscopic myomectomy is 0.5%34.

In our opinion, if the surgeon adheres to the same closure principles performing laparoscopic myomectomy as are used in the abdominal approach, the incidence of this complication should remain low34–36. The abdominal route is still the most common approach for large intramural fibroids, but laparoscopic myomectomy, if performed by a skilled surgeon, represents a valid alternative and appears to be both safe and effective. Hysteroscopy has documented in almost all the patients a regular endometrium without evidence of scarring or shape anomalies. However, these results must be interpreted with caution because only a small number of patients who wished to, conceived.

In the presence of fibroids, deeply intramural or those several requiring extensive morcellation, Nezhat et al.37 recommend laparoscopic-assisted myomectomy to render removal and suturing easier and faster. We believe that for the experienced laparoscopic surgeon skilled in endoscopic suturing, achieving good uterine closure is possible even in difficult conditions (for site, size and number of fibroids). Laparoscopic removal of the large fibroid using the electromechanical technique38 makes the procedure less demanding and time consuming. Indeed, our data show a shorter mean operative time than those reported by other authors8,10 in the presence of fibroids penetrating the uterine cavity, and may be comparable to abdominal myomectomy39.

Opening the cavity could theoretically increase the risk of uterine and pelvic infection. In our series, we documented only three cases of transient episodes of febrile morbidity, without an increase in the inflammatory index (i.e. pyrexia, leucocytosis and raised VES C protein). In addition, no uterine fistula was observed.

This report suggests that laparoscopic myomectomy can be performed in patients with intramural fibroids penetrating the uterine cavity with acceptable operative time and risk, and if performed by skilled surgeons trained in complex laparoscopic procedures.

These preliminary data are encouraging but other studies are necessary to confirm them, and also to assess the quality of the laparoscopic uterine repair even in the presence of large fibroids penetrating the uterine cavity, to assess long term recurrence rates and pregnancy outcome.