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: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/13533
Note: To cite this publication please use the final published version (if
applicable).
Chapter 10
Hysteroscopic polypectomy in premenopausal women with abnormal uterine bleeding:
a long-term follow-up
Dacia D.C.A. Henriquez Heleen van Dongen Ron Wolterbeek Frank Willem jansen Adapted from J Minim Invasive Gynecol 2007;14:59-63
Chapter 10 90
Introduction
As discussed in chapter 9, there is a paucity of literature on the effectiveness of the removal of endometrial polyps in cases of abnormal uterine bleeding in premenopausal women [Nathani 2006]. In the past, several authors [Cravello 1995; Cravello 1996; Cravello 2000;
Nagele 1996a; Preutthipan 2005; Shushan 2002; Tjarks 2000] have studied the effective- ness of hysteroscopic polypectomy. Nevertheless, few studies differentiate between pre- and postmenopausal women and little is known about predictive factors for persistence or recurrence of symptoms.
The objective of this study was to estimate the effectiveness of hysteroscopic polypectomy in premenopausal women with abnormal uterine bleeding on the long term. In addition, we evaluated prognostic factors for persistence or recurrence of symptoms after surgery in these women.
Methods
All premenopausal women who had undergone hysteroscopic polypectomy because of abnormal uterine bleeding between December 1993 and July 2005, whether or not com- bined with endometrial ablation or insertion of a levonorgestrel-intrauterine device (IUD), at the department of Gynaecology of the Leiden University Medical Center, (Leiden, The Netherlands) were included in this study. Endometrial polyps were diagnosed by saline infusion sonography or diagnostic hysteroscopy and abnormal uterine bleeding was defined as menorrhagia, metrorrhagia and/or intermenstrual bleeding. Patients receiv- ing dilatation and curettage as primary treatment for endometrial polyps were excluded from this study. If histological examination revealed no endometrial polyp, patients were excluded as well.
Hysteroscopic polypectomy was performed using a continuous flow operative hystero- scope (Olympus Winter&Ibe, Hamburg, Germany) equipped with a 7 or 9mm operative sheath and a 12º optic. Endometrial polyps were removed mechanically (e.g. with scissors or forceps) or electrosurgically. Sorbitol (5%) or normal saline were used for distension and irrigation of the uterine cavity and fluid balance was carefully monitored in order to prevent excessive intravasation. All procedures were performed under general or spinal anaesthesia.
Failure of treatment was defined as the persistence or recurrence of abnormal uterine bleeding after polypectomy requiring medical therapy or surgical re-intervention. Medical therapy included oral contraceptives, levonorgestrel-IUD’s and iron supplements because of anaemia caused by menstrual bleedings. Surgical re-intervention included repeated hysteroscopic polypectomy, endometrial ablation or hysterectomy. The follow-up period was defined as the period of time between the primary hysteroscopic polypectomy and repeat or further medical or surgical intervention. In case no recurrence occurred, the
Effectiveness of polypectomy on the long term 91
follow-up period was defined as the time period between the hysteroscopic removal and the end of the study period. Data were retrieved from the medical records of all patients.
Collected data included patient characteristics (i.e. age, obstetric and gynaecological his- tory, medication use and preoperative symptoms), information concerning the procedure, the removed lesions and follow-up after polypectomy. If no additional information was available, a questionnaire was sent to the patients, inquiring whether a further interven- tion 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 telephone call.
Statistical analyses were performed in the statistical SPSS program (SPSS, version 11.0, SPSS Inc., Chicago, IL). Kaplan-Meier survival curves were used to calculate the cumu- lative proportion of patients requiring further treatment over time. Differences between patients who underwent only hysteroscopic polypectomy and patients who underwent polypectomy combined with endometrial ablation or insertion of a levonorgestrel-IUD were analysed using the log-rank test.
Prognostic factors for persistence or recurrence of symptoms after polypectomy were analysed by means of univariable and multivariable backward Cox regression analyses.
Variables included in the regression analyses were age, parity, body mass index, symp- toms of menorrhagia, metrorrhagia and intermenstrual bleeding, size of endometrial polyps, number of endometrial polyps in case of multiple polyps and method of removal (mechanically or electrosurgically). For patients with multiple endometrial polyps, the size of the largest polyp was used in the analyses, as this polyp was considered most likely to cause symptoms. The size of the endometrial polyp was measured preoperatively by using either transvaginal ultrasound or saline infusion sonography. Statistical significance of variables was evaluated by comparing models with the likelihood ratio-test [Kleinbaum 1996]. A p-value <0.05 was considered of statistical significance.
Results
Of the initial 91 patients eligible to the study, thirteen patients (14.3%) were excluded as histological analysis revealed no endometrial polyp. A total of 78 patients were included in this study. Of this group, 56 patients received a questionnaire inquiring whether further intervention had been necessary and what their menopausal state was.
Patients’ characteristics are detailed in table 1. A total of 119 endometrial polyps were removed and complete removal was achieved in all patients. Forty-eight patients (61.5%) had a single polyp and 30 patients (38.5%) had multiple polyps. The highest number of polyps observed in a patient was five. Almost 90% of the polyps were pedunculated. The mean size of the endometrial polyps was 17.5mm (standard deviation [SD] 8.3). In 75.6%
of the cases polyps were removed electrosurgically, and in the other 24.4% mechanically (e.g. scissors or forceps). No major complications occurred during surgery.
Chapter 10 92
Median length of follow-up for the study population was 13.0 months (range 0.5-103 months). Figure 1 shows the intervention-free survival over time. One year after hys- teroscopic polypectomy, intervention-free survival was 67.6% (95% confidence inter- val [95%-CI] 56.2-79.0%). Four years after polypectomy, this survival-rate decreased to 44.3% (95%-CI 30.0-58.6%).
Table 1 | Demographic characteristics of patients.
Outcome
Mean age in years (SD; 95%-CI) 44.2 (5.2; 33.9-54.4)
Median parity (range) 2 (0-8)
Median body mass index in kg/m2 (range) 24.4 (18.7-38.2) Symptoms of abnormal uterine bleeding
Menorrhagia 49 (62.8%)
Metrorrhagia 21 (26.9%)
Intermenstrual bleeding 8 (10.3%)
Median duration of symptoms in months (range) 12 (1-228)
Haemoglobin level in mmol/L (SD; 95%-CI) 8.1 (0.8; 6.5-9.6)
Follow-up (months)
100 80
60 40
20 0
Intervention-free survival
1,0
0,8
0,6
0,4
0,2
0,0
Total group-censored
Hysteroscopic polypectomy combined with endometrial ablation or insertion of levonorgestrel-IUD-censored Hysteroscopic polypectomy-censored Total group
Hysteroscopic polypectomy combined with endometrial ablation or insertion of levonorgestrel-IUD Hysteroscopic polypectomy only
Page 1 Figure 1 | Kaplan-Meier curves for the intervention-free time after hysteroscopic polypectomy of patients
with abnormal uterine bleeding receiving polypectomy solely or combined with a levonor- gestrel-IUD or endometrial ablation.
Effectiveness of polypectomy on the long term 93
Fifty-six patients (71.8%) underwent only hysteroscopic polypectomy and 22 patients (28.2%) underwent polypectomy combined with endometrial ablation (n=13) or inser- tion of a levonorgestrel-IUD (n=9). After four years the intervention-free survival rate for patients who underwent only polypectomy was 41.1% (95%-CI 24.8-57.4%). The rate for patients who underwent combined treatment was 54.7% (95%-CI 28.0-81.4%; p=0.08).
Eight patients (10.8%) reported persistence or recurrence of symptoms without further treatment. Three of these patients experienced recently a recurrence of symptoms, one patient refused further treatment and for four patients reasons remained unknown.
Eighteen patients reached a postmenopausal state in the follow-up period after polypec- tomy. Of these patients three had received additionally endometrial ablation or a levo- norgestrel-IUD. Nevertheless, further examination (e.g. blood analysis) revealed a true postmenopausal state.
A total of fourteen patients (18.3%) required surgical re-intervention; repeated hystero- scopic polypectomy in two patients, endometrial ablation in four patients and hyster- ectomy in eight patients. Histological analysis after hysterectomy revealed adenomyo- sis in five specimens, intramural myomas in one specimen and no abnormalities in two specimens.
Individual effects of each variable included in the univariable Cox proportional haz- ard regression analyses for persistence or recurrence of symptoms after hysteroscopic polypectomy are given in table 2. None of the variables included in the analyses had a statistical significant association with persistence or recurrence of symptoms after poly- pectomy. Multivariable backward Cox regression analysis did not reveal any significant associations.
Table 2 | Hazard ratios and p-values of variables included in the univariable Cox regression analysis.
Variable Hazard ratio (95%-CI) p-value
Age 0.99 (0.93-1.07) 0.87
Parity 1.10 (0.91-1.34) 0.33
Body mass index 1.02 (0.95-1.10) 0.58
Menorrhagia 1.29 (0.61-2.73) 0.50
Metrorrhagia 1.09 (0.49-2.45) 0.83
Intermenstrual bleeding 0.42 (0.99-1.76) 0.24
Duration of symptoms 1.01 (1.00-1.01) 0.29
Number of polyps 0.90 (0.55-1.48) 0.68
Size of polyps 0.99 (0.94-1.05) 0.72
Method of removal 1.12 (0.48-2.60) 0.80
Chapter 10 94
Discussion
After hysteroscopic polypectomy almost 60% of the premenopausal women with abnormal uterine bleeding required further treatment because of persistence or recur- rence of symptoms within a 4-year follow-up. This proportion decreased to 45% when endometrial ablation or insertion of a levonorgestrel-IUD was added to polypectomy.
Probably, due to the small number of patients this difference was not statistically signifi- cant. The majority of patients who required further treatment received this within one year after polypectomy. The latter could be taken into consideration for the counselling of patients.
In the series of Nagele et al. [Nagele 1996a], 34.6% of premenopausal women with abnormal uterine bleeding failed to respond to polypectomy or experienced recurrence of symptoms. This result approximates our findings after two years of follow-up.
Remarkably, our results differ considerably from other previous studies. Cravello et al.
[Cravello 1995] reported an unfavourable result in only 6.7% of their patients. Recently, Preutthipan et al. [Preutthipan 2005] demonstrated that after hysteroscopic polypectomy more than 90% resumed normal menstruation. The different outcome of our study could be explained by differences in patient population and methodology. First, to assess and compare outcome of treatment adequately, failure and success should be well defined [Emanuel 1999]. Persistence or recurrence of symptoms defined as the need of medical therapy or surgical re-intervention are straightforward and easily reproducible. In con- trast to others, we defined survival as no further treatment instead of resumption of nor- mal menstruation. Consequently, patients receiving medical therapy in addition to sur- gery were considered failures in our study, while the same patients were not considered at all in the cohort of Preutthipan et al. [Preutthipan 2005]. Second, among the study population of Cravello et al. [Cravello 1995] nearly 20% of patients reached postmeno- pausal state during follow-up, which influenced the outcome in favour of polypectomy as effective treatment. In our study a Kaplan-Meier survival curve was used to estimate the effect of time on outcome of treatment [Dutton 2001; O’Connor 1996]. Patients who reached menopause were censored and had therefore no major influence on outcome.
A limitation of our study and the studies mentioned above is the retrospective design.
For this reason we have tried to limit undesirable effects by choosing explicit measure- ments of outcome.
Although the hysteroscopic removal of myomas and polyps share similarities in tech- nique and patient population, our results suggest that removal of myomas is a more effective intervention for symptoms of abnormal uterine bleeding than polypectomy. In a series of Emanuel et al. [Emanuel 1999], 20% of patients received further treatment within five years after myomectomy, which differs substantially from our 60% failure rate after polypectomy.
