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 12

General discussion

Chapter 12 104

Implementation of hysteroscopy

Hysteroscopy has become an important component of gynaecological practice. In the Netherlands, the introduction of hysteroscopic surgery came relatively late compared to other endoscopic fields. This delay was due to technical difficulties typical of the uterus, e.g. narrowness of the cervix and problems of distending the uterine cavity. In addition, serious complications occurred during developmental stages, and are still happening in unskilled hands. This caused gynaecologists to be sceptical towards hysteroscopy in the early 1990s [de Wit 1993]. As described in this thesis, at the turn of the last century the number of hospitals that had incorporated hysteroscopic surgery, had changed promis- ingly (chapter 2). This applied predominantly to diagnostic hysteroscopy and hystero- scopic surgery related to the treatment of abnormal uterine bleeding. In addition, the residents and gynaecologists surveyed in this thesis were found to be interested as well as competent in the aforementioned procedures (chapter 3). On the contrary, the imple- mentation of more advanced procedures, e.g. synechiolysis and sterilisation, seemed to lag behind (chapter 2). Though, today’s gynaecologist is interested in these procedures and considers himself, surprisingly, quite competent as well. This contradiction raises the question whether implementation of advanced hysteroscopic procedures in the gen- eral practice should be pursued. Difficult procedures are prone to a higher complica- tion rate, underlining the need for experience [Jansen 2000], whilst the caseload is rather small. For this reason, the establishment of special centres (particularly teaching hospitals) in combination with an adequate referral system for infrequently performed procedures, would probably be a good solution.

Since surgeons are still dealing with learning curve issues, improvement on this point could be of importance to implementation as well. Recently, a new technique, the so-called hysteroscopic morcellator, was introduced for the removal of endometrial polyps and submucous myomas [Emanuel 2005]. Like the bipolar resection technique [Kung 1999], the morcellator uses normal saline as distension medium, which relatively decreases the risk of complications due to fluid overload. Another advantage of the morcellator technique is that tissue fragments can be removed easily by means of suc- tion through the instrument. In addition, although no difference in learning curve could be demonstrated, we proved in this thesis that operating time was reduced significantly when residents in training used this technique instead of the conventional resectoscope.

Furthermore, residents and their trainers felt more comfortable using the morcellator than using the conventional resectoscope (chapter 4). The hysteroscopic morcella- tor has shown to be most convenient in the removal of endometrial polyps and small submucous myomas. Recently it was advocated, with reference to a randomised con- trolled trial, to treat polyps smaller than 2cm with a bipolar electrode passed through an operating sheath of a small-calibre hysteroscope in an outpatient setting, and to treat polyps bigger than 2cm in an inpatient setting [Muzii2007]. To date, the use of the hysteroscopic morcellator requires cervical dilatation and therefore general or regional anaesthesia. The indication area of the hysteroscopic morcellator is thus primarily the

General discussion 105

removal of large endometrial polyps (>2cm). Inexperienced surgeons (like residents in training) can use it for the removal of polyps and small submucous myomas. Once this technique is mastered, they should preferably proceed to remove small endometrial polyps (<2cm) in an outpatient setting. Submucous myomas bigger than 3cm or with an intramural extension of more than 50% are not suitable for removal by the hystero- scopic morcellator. Removal of these lesions is considered an advanced hysteroscopic technique, and in combination with its small caseload it will not be possible to provide this care in every hospital. Therefore, it should be confined to clinics with experienced surgeons as discussed before.

Diagnostic hysteroscopy

Despite the lack of adequate information regarding its accuracy, diagnostic hysteros- copy has been suggested the method of choice in the evaluation of the uterine cavity, ever since 1988 when Gimpelson and Rappold [Gimpelson 1988] reported that hystero- scopy was more accurate than dilatation and curettage. In 2002, the first systematic review was published on the accuracy of hysteroscopy in the evaluation of (pre-)malig- nant disorders [Clark 2002b]. Farquhar et al. [Farquhar 2003] conducted a meta-analysis of premenopausal women with benign intracavitary abnormalities in 2003. However, since in that review only studies written in English were included and -due to hetero- geneity- the positive likelihood ratio was not calculated, we have repeated a systematic review and meta-analysis. In this review it was proved that diagnostic hysteroscopy is an accurate and feasible tool in the diagnosis of intrauterine pathology in women with abnormal uterine bleeding (chapter 5).

With regard to diagnostics of the uterine cavity, it is noteworthy that a meta-analysis on the accuracy of saline infusion sonography in women with abnormal uterine bleed- ing reported a sensitivity of 0.95 and a specificity of 0.88, equalling the accuracy of diagnostic hysteroscopy described in this thesis (0.94 and 0.89 respectively) [de Kroon 2003a]. Moreover, a technology assessment of saline infusion sonography demonstrated that it could replace 84% of the diagnostic hysteroscopies in the evaluation of abnormal uterine bleeding [de Kroon 2003a; de Kroon 2003b]. In addition, saline infusion sonog- raphy was thought to reduce costs as well as discomfort for women concerned [Carlos 2001; Rogerson 2002; Widrich 1996]. However, nowadays, diagnostic hysteroscopy is performed by means of the so-called vaginoscopic approach without the use of a speculum and a tenaculum, thus reducing discomfort significantly [Bettocchi 1997].

Furthermore, studies on recent advances in endoscopic instrumentation make that out- patient therapeutic hysteroscopic procedures provide significant savings in costs and are preferred by women when compared to day case procedures [Marsh 2002; Marsh 2006]. In chapter 6 patient pain scores of saline infusion sonography and vaginoscopic hysteroscopy were compared. This randomised controlled study demonstrated that women who received saline infusion sonography experienced significantly less discom-

Chapter 12 106

fort. However, in a preference study patients with and without a history of hysteroscopy or saline infusion sonography predominantly opted for diagnostic hysteroscopy because of the possibility of treatment during the same procedure (chapter 7). Additionally, ther- apy in an outpatient setting is preferred to a day case setting by the patient. Even so one third of the women would like to be treated under regional or general anaesthesia.

Moreover, the prevalence of intrauterine abnormalities after selection by transvaginal ultrasound examination is only 50% [Towbin 1996], and we do not know yet what the success rates are of treatment by office hysteroscopy. So for clinical practice it is impor- tant to have access to both saline infusion sonography and diagnostic hysteroscopy with its treatment options. As one might expect, the removal of submucous myomas or large endometrial polyps in an ambulatory setting by hysteroscopy is inconvenient for the patient. Therefore, the likeliness of a specific intracavitary abnormality determined after a transvaginal ultrasound might be helpful to decide whether a patient should be sched- uled for saline infusion sonography or hysteroscopy with the possibility of treatment at the same time. Whether it will be possible to efficiently employ triage after a transvagi- nal ultrasound still needs clarification. Recent literature suggests to schedule women with intracavitary pathology smaller than 2cm for office hysteroscopy. Intracavitary lesions bigger than 2cm are advised to be removed under regional or general anaesthe- sia [Bettocchi 2002; Litta 2008; Muzii 2007]. These conditions are largely based on the time necessary for the removal of the specific intrauterine lesion, and therefore depend on the experience of the surgeon and available equipment as well.

Hysteroscopic surgery

In the past, hysterectomy was the preferred surgical treatment for patients suffering from abnormal uterine bleeding. Nowadays the introduction of several alternatives means that undergoing hysterectomy should only be decided upon after considering other treatment options, especially in premenopausal women. Drugs are frequently recom- mended for this indication [Lethaby 2000], although this kind of treatment, with excep- tion of the levonorgestrel intrauterine device (IUD), has proved to be less successful than hysterectomy [Learman 2004]. In two randomised controlled studies, the levonor- gestrel-IUD was compared to hysterectomy, and showed similar improvements in qual- ity of life [Hurskainen 2001; Lahteenmaki 1998].

Hysteroscopic endometrial ablation has also an established role in the treatment of abnormal uterine bleeding. Several randomised controlled trials have compared the effectiveness of endometrial ablation techniques with hysterectomy [Dwyer 1993;

Gannon 1991; O’Connor 1997; Pinion 1994]. Each of these studies demonstrated advantages for the less invasive endometrial ablation over hysterectomy, including lower morbidity, shorter treatment and recovery times, and cheaper procedures. However, on the long term a substantial number of patients after having received endometrial abla- tion, needed re-operation (25-30%). Hysterectomy revealed also a significant advan-

General discussion 107

tage of improving premenstrual symptoms and superior health related scores [Aberdeen Endometrial Ablation Trials Group 1999; Dickersin 2007; Marjoribanks 2003; Sculpher 1996].

Transcervical removal of myomas has proved to be an acceptable alternative to hys- terectomy in the treatment of submucous myomas. On the long term failure of treat- ment is estimated at 15% [Ben Baruch 1988; Derman 1991; Emanuel 1999]. The pos- sibility to remove submucous myomas depends on the degree of intramural extension [Wamsteker 1993b]. Type 0 and I myomas can often be completely removed, whereas removal of type II lesions can be problematic, as fluid overload or fear for perforation may lead to incomplete resection. Wamsteker et al. found that the chance of achieving a complete removal of type II myomas was only 50% [Wamsteker 1993b]. Interestingly, we found that approximately half of the patients who underwent incomplete removal at hysteroscopic myomectomy needed subsequent surgery within three years of follow- up. Regarding this, incomplete removal does not always seem to necessitate subsequent surgery. For this reason it may be worthwhile to wait and see if the symptoms will dimin- ish (chapter 8). Nevertheless, complete removal still has a better outcome with respect to the need of subsequent surgery [Emanuel 1999; Wamsteker 1993b]. Therefore, it remains important to weigh carefully the morbidity and cost-effectiveness of a hysterec- tomy as an alternative therapy, against the chance of incomplete removal with possibly subsequent surgery. In this context, we also have to take into account that the surgeon’s experience influences the chance of achieving a complete removal [Emanuel 1999].

For women with symptomatic uterine myomas with more than 50% intramural exten- sion (including type II myomas), uterine artery embolisation demonstrated to have advantages over hysterectomy with regard to a shorter hospital stay and a quicker return to routine activities [Gupta 2006; Hehenkamp 2006]. Results of two randomised con- trolled trials comparing uterine artery embolisation with surgery showed a faster recov- ery in favour of uterine artery embolisation, similar pain and quality of life scores, but a greater need (20% to 25%) for further surgical treatment after 12 and 24 months of follow-up [Edwards 2007; Hehenkamp 2008; Volkers 2007].

The effectiveness of hysteroscopic polypectomy in patients with abnormal uterine bleed- ing is a less clear case. Although an overwhelming majority of gynaecologists advocate removal of endometrial polyps with the aim of treating symptoms of abnormal uterine bleeding, this rationale may be misplaced for a number of reasons [Nathani 2006]. First endometrial polyps are found in up to 10% of asymptomatic patients [Clevenger-Hoeft 1999], second, smaller polyps may regress spontaneously [DeWaay 2002], and thus may not be causative of abnormal uterine bleeding. Finally, the prevalence of serious endometrial disease, such as cancer or atypical hyperplasia within a polyp, is very rare in premenopausal patients [Savelli 2003]. A systematic review on this issue showed that there was a paucity of quality literature assessing the effectiveness of this therapy [Nathani 2006]. A few studies on the success rate of polypectomy in premenopausal patients, however, showed that these rates varied greatly from 61.2% to more than 90%

Chapter 12 108

[Cravello 1995; Nagele 1996a; Preutthipan 2005]. Since design and success rates in these studies were poorly defined, these results were rather unreliable.

In chapter 9 a prospective study showed a significant improvement of monthly blood loss among premenopausal patients with abnormal uterine bleeding six months after hysteroscopic polypectomy. Since the majority of patients felt also relieved of their com- plaints, this reduction seems clinically relevant. Two third of the premenopausal patients suffering from substantial menorrhagia before surgery still had menorrhagia postopera- tively, but experienced a significant relief of their complaints. A Kaplan-Meier survival analysis of premenopausal patients with abnormal uterine bleeding showed an inter- vention-free rate of approximately 40% after four years following polypectomy (chapter 10). When polypectomy was combined with endometrial ablation or levonorgestrel- IUD, this rate increased to 55%. So, on the short term hysteroscopic polypectomy has proved to be clinically effective. On the long term these figures seem to be less opti- mistic, although still half of the patients were saved from further treatment. Prospective studies are needed to illuminate this issue further, but until then we would suggest to perform hysteroscopic polypectomy and await recurring symptoms.

All hysteroscopic therapies related to the treatment of abnormal uterine bleeding and insertion of levonorgestrel-IUD proved to be effective in a proportion of patients and may prevent hysterectomy. Remarkably, even though these alternative treatments have been implemented, the overall hysterectomy rate in the management of abnormal uter- ine bleeding in a tertiary referral centre did not decrease (chapter 11). It appeared that the introduction of alternatives might have lowered the threshold for intervention. But why are these alternatives used as an additional technology rather than as a substitutive one? There are other examples in literature in which newer medical technologies did not simply replace existing procedures. For example, after laparoscopic cholecystec- tomy was introduced at the beginning of the 1990s, rates of open cholecystectomy rose by 28% [Escarce 1995; Steiner 1994]. The shorter recovery time with less postopera- tive morbidity might be a factor in encouraging patients with less severe symptoms to undergo minimally invasive surgery [Farquhar 2002a]. Furthermore, the treatment of premenopausal patients with abnormal uterine bleeding might have been unsatisfactory in the past, so that nowadays -with a more extensive surgical palette available- surgery rates have increased in order to supply the mounting demand. A significant improve- ment in health-related quality of life has highlighted the importance of treating abnor- mal uterine bleeding [Hurskainen 2001], it also has emphasised the inferiority of medi- cal to surgical treatment [Learman 2004]. Moreover, a considerable number of patients treated with hysteroscopic surgery, levonorgestrel-IUD or uterine artery embolisation do not need subsequent surgery, and may escape from hysterectomy.

Therefore, hysteroscopic surgery is an indispensable therapeutic tool in the surgical palette of the gynaecologist, but its limitations need to be acknowledged in order to provide optimal care.