Hysteroscopy in daily practice Dongen, H. van

Hysteroscopy in daily practice

Dongen, H. van

Citation

Dongen, H. van. (2009, February 26). Hysteroscopy in daily practice.

Retrieved from https://hdl.handle.net/1887/13533

Version: Corrected Publisher’s Version

License:

Downloaded from:

Note: To cite this publication please use the final published version (if

applicable).

Chapter 8

Follow-up after incomplete hysteroscopic removal of uterine myomas

Heleen van Dongen Mark Hans Emanuel Maddy J.G.H. Smeets J. Baptist M.Z. Trimbos Frank Willem Jansen Adapted from Acta Obstet Gynecol Scand 2006;85:1463-1467

Chapter 8 76

Introduction

Submucous myomas often lead to abnormal uterine bleeding, which may be accom- panied by anaemia or reproductive dysfunction [Stewart 2001]. Hysteroscopic removal is the standard surgical approach for myomas within the uterine cavity [Brooks 1989;

Hallez 1987; Neuwirth 1978; Valle 1990]. The ability to remove these myomas depends on the degree to which myomas are embedded in the myometrium [Wamsteker 1993b].

Type 0 myomas are usually easy to remove. Type I lesions can often be completely removed as the uterus contracts and tends to expel the intramural component into the uterine cavity during surgery [Loffer 1990]. However, removal of type II myomas can be problematic, as fluid overload or fear of perforation may lead to incomplete resec- tion (see for classification of submucous myomas chapter 2). Wamsteker et al. found that the likelihood of achieving a complete removal of type II myomas was only 50%

[Wamsteker 1993b]. Therefore the hysteroscopic removal of type II myomas is ques- tioned by several authors [Loffer 1990; Wamsteker 1993b].

Although incomplete removal is associated with failure of treatment, there are some reports suggesting that residual tissue may undergo spontaneous regression, without the need for subsequent surgery [Corson 1991; Donnez 1990; Loffer 1990]. However, it remains unclear how these results should be interpreted on the long term. The objec- tive of this study was to assess the surgery-free survival following incomplete removal of submucous myomas in patients with abnormal uterine bleeding and to assess risk factors predicting recurrence of clinical symptoms.

Methods

The study was conducted in three affiliated hospitals; the Leiden University Medical Center (Leiden, The Netherlands), the Spaarne Hospital (Heemstede/Hoofddorp, The Netherlands) and the Bronovo Hospital (The Hague, The Netherlands). Patients who had undergone incomplete removal of submucous myomas after hysteroscopic myomec- tomy for abnormal uterine bleeding between January 1997 and January 2005 were eligi- ble for the study. Abnormal uterine bleeding was defined as menorrhagia, metrorrhagia and/or intermenstrual bleeding. Patients diagnosed with multiple myomas, and having one or more non-treated myomas in situ after hysteroscopic surgery were excluded from the study.

Preoperatively, the size of the myomas was assessed by saline infusion sonography or transvaginal ultrasound. The location of myomas was assessed by saline infusion sono- graphy or diagnostic hysteroscopy. The degree of intramural extension was determined during therapeutic hysteroscopy by observing the angle of the myoma to the uterine wall and classified as proposed by Wamsteker et al. [Wamsteker 1993a]. If multiple myomas were removed, the location and intramural extension were determined by the myoma that was incompletely removed. The diameter of all removed myomas was derived from the total volume of all submucous myomas combined that presented in one patient. The

Incomplete removal of submucous myomas 77

volume of each single myoma was calculated as follows: 4/3πr³ (formula of volume of sphere, r=radius).

In all centres, transcervical resection was obtained with a rigid 8 or 9mm resectoscope (Olympus Winter&Ibe, Hamburg, Germany) equipped with a 12º optic. Sorbitol (5%) was used for distension and irrigation of the uterine cavity. In a few cases hysteroscopic myomectomy was performed with the hysteroscopic morcellator (Smith&Nephew, Andover MASS, USA), which uses normal saline as distension medium. Fluid control was maintained, and if an excessive deficit of fluid developed the procedure was termi- nated. Fluid administration was considered excessive if the deficit exceeded a limit of 1500ml when sorbitol was used and 2500ml when using normal saline. Further details of both techniques have been described previously [Corson 1991; Emanuel 2005;

Wamsteker 1993b]. All procedures were performed by experienced hysteroscopists (MHE, MS or FWJ).

Recurrence was defined as abnormal uterine bleeding requiring repeat or further surgi- cal treatment. Follow-up was assessed from the time primary hysteroscopic myomec- tomy was performed until surgical intervention or the end of the study period.

Medical records of all patients were reviewed in order to evaluate the clinical course after surgery. If insufficient information was available, a questionnaire was sent to the patient, inquiring whether a surgical intervention had been necessary and what their menopausal state was. Patients were considered postmenopausal if they experienced amenorrhoea for at least twelve months. In case of non-response a second request was sent followed by a phone call.

The collected information was analysed in the statistical SPSS program (SPSS, ver- sion 11, SPSS Inc., Chicago, IL) using the Pearson’s Chi-Square test and Kaplan-Meier method. The latter was used to calculate the cumulative proportion of women having a surgery-free survival following incomplete removal of submucous myomas. A multivari- able Cox proportional hazard regression analysis in a backward stepwise manner was used to analyse the relationships of variables with recurrence of symptoms. A level of p>0.10 was used for the removal of variables from the stepwise model. Significance was reached at p<0.05.

Results

Between 1997 and 2005, 528 consecutive patients were treated for abnormal uterine bleeding with hysteroscopic myomectomy. In 91 cases (17.2%) a complete removal was not achieved. Thirteen of these patients, diagnosed with multiple myomas, had one or more non-treated myomas left in situ after hysteroscopic surgery and were therefore excluded from the study. Because of fertility desires, 37 other patients with incomplete removals underwent a repeat hysteroscopic procedure immediately without abiding to consequences regarding symptoms and they were therefore also excluded. Except for a

Chapter 8 78

significant difference in age (mean age 36.7 years; range 25-42), patient characteristics of those excluded were similar to the study group (data not shown).

Forty-one patients (7.8%) met our inclusion criteria and were followed until January 2005. Patient demographics are detailed in table 1. The mean age of the patients was 42.5 years (standard deviation [SD] 6.9; range 27-55). The characteristics of surgery are detailed in table 2. Thirty-five patients (85.4%) were treated by resectoscopy with sorbitol used as irrigation fluid. The uterine morcellator, which uses normal saline, was

Table 1 | Demographics of patients with incomplete removal of submucous myomas after hysteroscopic myomectomy.

Patient characteristics

Number of patients 41

Mean age in years (SD; range) 42.5 (6.9; 27-55)

Parity (%) 0 14 (37.8)

1 5 (13.5)

2 13 (35.1)

3 5 (13.5)

Primary symptoms (%)

Menorrhagia 37 (90.2)

Irregular bleeding 4 (9.8)

Dysmenorrhoea 15 (36.6)

Patients with multiple myomas (%) 7 (17.1)

Table 2 | Characteristics of hysteroscopic surgery of incomplete removal of myomas.

Characteristics of surgery

Mean size (diameter) of myoma in cm (SD) 3.6 (1.2)

Mean volume of myoma in cm³ (SD) 35.5 (31.4)

Type of myoma (%) 0 4 (9.8)

I 14 (34.1)

II 23 (56.1)

Median percentage residue (range) 27.5 (10-60)

Median volume residual tissue in cm³ (range) 5.6 (0.1-80.8)

Median operating time in min (range) 45.0 (20-80)

Median fluid deficit in ml (range) 1600 (900-2500)

Sorbitol 1500 (900-2300)

Normal saline 2100 (1600-2500)

Incomplete removal of submucous myomas 79

applied to six cases (14.6%). No serious complications occurred. Although during one procedure a fluid deficit of 2300ml sorbitol was measured, no transurethral resection (TUR) syndrome was clinically observed. Termination of the procedure before complete removal was achieved, was caused by reaching the maximum fluid deficit in 90.2%

(n=37) of the cases and by technical failure in 9.8% (n=4).

Figure 1 shows the percentage of all patients subjected to hysteroscopic myomectomy during the study period stratified by intramural extension and completeness of the removal. Significantly (p<0.001) more patients with incomplete removal of submucous myomas treated with hysteroscopic myomectomy were diagnosed with type II myomas.

Furthermore, significantly (p=0.004) less patients diagnosed with type 0 myomas under- went incomplete removal.

Histological examination confirmed the diagnosis of leiomyoma of the uterus in all cases.

Six patients (14.6%) became postmenopausal during the follow-up period.

The cumulative proportion of women having recurrence is shown in figure 2. The surgery- free percentages were 70.2% (95%-confidence interval [95%-CI] 55.3-85.1) at one year, 54.8% (95%-CI 38.1-70.7) at two years, and 44.2% (95%-CI 26.9-61.5) at three years.

Nineteen patients needed subsequent surgery for recurrence of symptoms. The different procedures are listed in table 3. The median follow-up period for all patients subjected to repeat or further surgical treatment was 8.8 months (range 3-36). Patients with no subse- quent surgical treatment had a median follow-up of 41 months (range 2-77).

3 10

39

*

97 *

90

* 61

0 20 40 60 80 100

type 0 type I type II

percentage

incomplete removal complete removal

Figure 1 | Percentage of patients treated with hysteroscopic myomectomy stratified by intramural extension and completeness of removal.

3 10

39

97* *

90

61*

0 20 40 60 80 100

type 0 type I type II

percentage

incomplete removal complete removal

*p-value <0.05

Chapter 8 80

In order to analyse whether factors influenced recurrence, a multivariable Cox propor- tional hazard regression analysis was performed. The following variables were included in the model: type of myoma, size of myoma (diameter), percentage of residual tissue and age. Only the size of the myoma was found of statistically significant influence, with a hazard ratio of 1.76 (95%-CI 1.11-2.79) per cm increase in diameter (p=0.016).

Figure 2 | Surgery-free time interval after incomplete removal of submucous myomas.

Censored observations are observations with incomplete follow-up.

Table 3 | Subsequent surgical procedures after incomplete removal.

Procedure Number

Hysterectomy 5

Transcervical resection of submucous myoma 13

Abdominal myomectomy 1

Total 19

Follow-up (months)

72 60

48 36

24 12

0

Intervention-free survival

1,0

0,8

0,6

0,4

0,2

0,0

Censored

Incomplete myoma removal

Page 1

Incomplete removal of submucous myomas 81

Discussion

The results from this series indicate that after a follow-up of three years approximately half of the patients who underwent incomplete removal of submucous myomas at hys- teroscopic myomectomy had subsequent surgery. Thus, incomplete removal does not always seem to necessitate subsequent surgery. Based on these results, we consider it worthwhile to wait and see how the symptoms behave after incomplete removal, instead of subsequent surgery two to three months postoperatively as others suggest [Blok de 1995; Donnez 1990; Hallez 1987; Wamsteker 1993b]. Especially when menopause is expected to set in within a reasonable period of time, a conservative approach is desir- able. Nevertheless, if these results are compared to recurrence rates of other studies with complete resection, the latter seems to improve outcome with respect to the need for subsequent surgery [Blok de 1995; Hallez 1995; Wamsteker 1993b]. Therefore, it remains important to weigh carefully the morbidity and cost-effectiveness of hysterec- tomy as alternative therapy against the chance of incomplete removal at hysteroscopic myomectomy possibly inducing subsequent surgery. In this context, we have to consider that the surgeon’s experience probably influences the likelihood of achieving complete removal [Emanuel 1999], and therefore has to be taken into account as well.

Dueholm et al. [Dueholm 1998] systematically investigated how residual tissue behaves after incomplete removal among 45 patients. In half of their patients they observed spontaneous regression of residual tissue three months after surgery. Moreover, half of the patients with remaining residual tissue experienced normal periods three months postoperatively. Unfortunately little is known of the outcome after incomplete removal on the long term. To our knowledge only one study has described a small series of patients (n=10) with incomplete removal in a similar way with similar results [Emanuel 1999].

In accordance with other studies [Blok de 1995; O’Connor 1996; Wamsteker 1993b], we found that incomplete removal of myomas occurred more often among patients with type II myomas. In our series, most of these procedures had to be aborted because of menacing fluid overload. It has been demonstrated previously that the removal of the intramural part of submucous myomas is strongly related to an increased loss of irrigat- ing fluid [Emanuel 1997], and therefore at risk of incomplete removal. This is supported by the fact that in experienced hands the rate of incomplete removal of type II myomas decreases, suggesting a learning curve limited by the maximum fluid deficit [Emanuel 1999].

Although some authors have presented data on incomplete removal, prognostic fac- tors have never been investigated. According to our data the risk of subsequent surgery increases with an increasing size of the myoma preoperatively. The use of gonadotropin- releasing hormone analogues preoperatively causes a significant decrease of both the myoma size and fluid loss during surgery [Donnez 1990], which might improve the rate of complete removal of large myomas.

Chapter 8 82

A limitation of our study is that recurrence was defined as repeat surgery without taking persistence or regression of residual tissue into account. This could have been easily assessed with the use of an ultrasound investigation. However, the use of ultrasound would also identify asymptomatic lesions and therefore result in higher rates of recur- rence [Fauconnier 2000]. The true prevalence of asymptomatic lesions is not well known. Therefore a clinical examination with oriented ultrasound investigation would have been the appropriate method because it would correspond to the manner in which the preoperative diagnosis was made. Unfortunately the retrospective character of our study made it difficult to determine the recurrence rate based on symptoms, clinical examination and ultrasound. In order to be able to determine failure and success prop- erly, we decided to define recurrence as a repeat surgical intervention. Nevertheless, the true survival without symptoms is probably somewhat lower. Another aspect of our study that could be criticised is that half of the patients eligible for this study had to be excluded since subsequent surgery was performed without taking symptoms into account. In almost all of these cases this was done because of fertility desires, in which case complete removal is thought to be the only treatment option. Except for a differ- ence in age, these excluded patients were similar to the study group.

In conclusion, our results indicate that incomplete removal does not always necessitate subsequent surgery. Instead of this, a wait and see approach is worth considering.