One of the fundamental maxims of surgery is to work with excellent visualisation of the operative field. The dark, inaccessible nature of the uterine cavity has led to procedures in this environment being conducted ‘blindly’, flying in the face of the surgical axiom of good visualisation. Dilatation and curettage, or ‘D&C’, was the defining procedure of gynaecology for the diagnosis and treatment of abnormal uterine bleeding and miscarriage.
One of the fundamental maxims of surgery is to work with excellent visualisation of the operative field. The dark, inaccessible nature of the uterine cavity has led to procedures in this environment being conducted ‘blindly’, flying in the face of the surgical axiom of good visualisation. Dilatation and curettage, or ‘D&C’, was the defining procedure of gynaecology for the diagnosis and treatment of abnormal uterine bleeding and miscarriage. The advent of hysteroscopy, endoscopic visualisation within the uterus, has threatened the existence of blind intrauterine procedures. Technological advances, particularly in optics and ancillary equipment, have led to hysteroscopic surgery becoming increasingly adopted. Moreover, the miniaturisation and portability of hysteroscopic equipment has allowed many procedures to be done in the outpatient setting without general anaesthesia.
In comparison to other health technologies, hysteroscopic surgery has been well evaluated, especially transcervical resection of the endometrium, second-generation endometrial ablation and hysteroscopic removal of uterine polyps. Randomised controlled trials (RCTs) have been conducted evaluating these interventions. However, many questions remain unanswered regarding the role of hysteroscopic intervention. Examples include the place of hysteroscopy in the diagnostic workup of women with abnormal uterine bleeding, miscarriage and subfertility, the effectiveness of hysteroscopic removal of intrauterine pathologies, such as submucosal fibroids, in treating heavy menstrual bleeding and reproductive problems, and optimisation of the patient experience of hysteroscopic surgery in the outpatient setting.
This chapter discusses current uncertainties regarding the role of hysteroscopy in the diagnosis and treatment of uterine disorders, and proposes areas where future research should be targeted. In addition, recent technological innovations are introduced.
The role of cervical preparation in the patient experience in the office setting continues to be the subject of much research. A systematic review and meta-analysis of six studies showed there may be a benefit of using prostaglandins for post-menopausal women and there is some evidence that prostaglandins reduce the force and requirement for dilating the cervix beyond 5 mm . They concluded that cervical priming with vaginal prostaglandins may be considered in post-menopausal women if using hysteroscopic systems more than 5 mm in diameter.
Since the completion of this systematic review in 2011, further trials comparing prostaglandins with placebos or analgesics have been completed, although the results are conflicting. One RCT looked at pain during office hysteroscopy in three comparative groups, all a mixture of pre- and post-menopausal women: placebo vs intravenous ketoprofen, placebo vs intravaginal misoprostol, intravenous ketoprofen vs intravaginal misoprostol . When compared with placebo, the mean visual analogue scale (VAS) score during the procedure was significantly lower in the misoprostol group (p = 0.04) and the ketoprofen group (p = 0.05). In addition, the median VAS score after the procedure was significantly lower in the misoprostol group than in the ketoprofen group (p = 0.02). Therefore, the authors concluded that 400 μg of vaginal misoprostol significantly reduced pain during and immediately after hysteroscopy. A further RCT in pre-menopausal women also supported the effectiveness of vaginal prostaglandin. This was a double-blind RCT comparing 200 μg misoprostol given vaginally three hours before office hysteroscopy to placebo . Results showed that vaginal misoprostol significantly facilitated the procedure of office hysteroscopy. Ease of cervical entry (assessed on a five-point Likert scale) was greater with misoprostol (3.47 ± 1.05) than with placebo (2.93 ± 1.02); p = 0.002. Procedural time (minutes) from introduction of the hysteroscope through the external cervical os to completion of visualisation of the uterine cavity was shorter with misoprostol (1.95 ± 0.47) than with placebo (2.48 ± 0.56); p <0.001. Patient acceptability (also assessed on a five-point Likert scale) was higher with misoprostol (2.87 ± 0.96) than with placebo (2.51 ± 0.89); p = 0.019. Pain scoring (measured using VAS) was lower with misoprostol (3.26 ± 1.56) than with placebo (4.87 ± 1.79); p <0.001.
In contrast to these trials, another double-blind RCT, this time on just nulliparous women with no history of cervical stenosis, compared 200 μg misoprostol given vaginally to 100 mg diclofenac sodium given rectally, or placebo (‘misoprostol’ given vaginally and placebo ‘NSAID’ given rectally). No significant differences were found in pain scores, patient acceptability, vasovagal reactions, procedure time or post-procedural analgesic requirements .
The contrasting results may reflect heterogeneity in the patient cohort, operative technique and equipment. A Cochrane review in 2015 concluded that there was moderate-quality evidence that use of misoprostol for pre-operative ripening of the cervix before operative hysteroscopy is more effective than no treatment . However, while its use is associated with fewer intraoperative complications, such as lacerations and false tracks, misoprostol is associated with more side effects, including pre-operative pain and vaginal bleeding. If misoprostol is to be used, the evidence suggests that it should be given three to four hours prior to the procedure. In fact, both pharmaceutical and mechanical dilatation of the cervix require time. To be compatible with modern office hysteroscopy, which aims to avoid the need for hospital admission, future developments will have to focus on faster efficacy. With no obvious method currently meeting the criteria of speed, cervical preparation will probably only be suitable for selected cases in the office setting, such as women at high risk or known to have cervical stenosis, or where cervical dilatation beyond 5–6 mm is anticipated, such as when inserting certain types of ablative devices or operative hysteroscopic systems of larger diameter.
A systematic review and meta-analysis of 15 trials looked at local anaesthesia for pain control during office hysteroscopy . It showed that intracervical and paracervical injections of local anaesthetic significantly reduced pain (standard mean difference (SMD) −0.36, 95% confidence interval (CI) −0.61 to −0.10 and SMD −1.28, 95% CI −2.22 to −0.35, respectively) in women undergoing office hysteroscopy, whereas transcervical and topical application did not. However, the use of local anaesthetic did not have a significant effect on the incidence of vasovagal episodes.
Alongside the use of local anaesthetic to improve the experience of hysteroscopy, smaller diameter hysteroscopes minimise the degree of stretching of the cervix and resultant trauma such that stimulated pain is reduced. Moreover, the miniaturisation of hysteroscopes has facilitated the use of vaginoscopy, thus avoiding uncomfortable distension of the vagina with a speculum, and potentially painful instrumentation of the cervix with forceps. These developments may make the routine use of local anaesthetic counterproductive, because the process of injecting local anaesthetic in itself can be painful. Indeed, the one trial included in the systematic quantitative review that adopted a vaginoscopic approach compared with a conventional approach did not find any benefit from the use of local anaesthetic .
Cervical anaesthesia does not address pain generated in the upper third of the uterus, which is innervated from the thoracic nerves, largely derived from the sympathetic fibres of the superior hypogastric plexus T8–T10 and L1 roots [7, 8]. Recognition of this has led to interest in targeted intrauterine hysteroscopic injection of local anaesthesia in order to make operative procedures, such as endometrial ablation, more acceptable to women. Using a 5 Fr cystoscopy needle (Cook Medical, Indiana, USA), or a specifically designed needle (Wing Needle™, Idoman, Ireland) through the operating channel of a continuous-flow hysteroscope, local anaesthetic can be injected into the uterine fundus to target these nerves, creating an intrauterine cornual block (ICOB) . To date, one small RCT has compared the efficacy of applying a hysteroscopic local anaesthetic ICOB in addition to a direct local anaesthetic cervical block with placebo on pain experienced during outpatient endometrial ablation . The mean 10 cm VAS score during the procedure was significantly lower by 1.44 (95% CI 2.65–0.21) with ICOB, but no differences were seen in post-operative pain, rescue analgesic requirement or duration of hospital stay. Larger, multi-centre RCTs of specific outpatient hysteroscopic procedures and using specific technologies are needed to confirm these findings. Such studies should also try to tease out the clinical significance of ICOB by comparing its use in isolation with its use in conjunction with other types of cervical anaesthesia and analgesics.
There have been many alternative methods to minimise patient discomfort. Recently an RCT has been completed which assessed the role of bladder distension to facilitate ‘passive uterine straightening’ on patient comfort and ease of cervical entry . Women in the intervention group were asked to drink 500 ml of water prior to the procedure and asked not to empty their bladder. In contrast, the control group of women were asked to ensure an empty bladder prior to the procedure. The status of the bladder (full or empty) was checked by ultrasound prior to the procedure. They found bladder distension prior to outpatient hysteroscopy was associated with significantly less pain, a quicker procedure and easier cervical entry. However, there was no difference in patient acceptability of diagnostic hysteroscopy. The utility of the intervention is further questioned because of issues around the practicality of introducing bladder distension into a busy clinic where there can be delays in appointments.
Vaginoscopy, otherwise known as the ‘no-touch technique’, refers to the introduction of the hysteroscope into the uterine cavity without the need for a vaginal speculum or cervical instrumentation. A systematic review identified six small RCTs comparing traditional office hysteroscopy, using a speculum and a tenaculum, to vaginoscopy . There was no significant difference in failed procedures between the techniques (OR 1.28; 95% CI 0.74–2.24), but the vaginoscope was associated with significantly lower pain scores (SMD −0.44; 95% CI −0.65 to −0.22). Despite these data, the use of vaginoscopy remained low because concerns about infection, complications, feasibility and training were not considered.
A recent large RCT, the VAST study, has addressed these uncertainties by evaluating whether vaginoscopy or standard hysteroscopy using a vaginal speculum to visualise, clean and manipulate the cervix was more successful in the outpatient setting . Success was defined when a hysteroscopy was completed satisfactorily, with no complications, a level of pain acceptable to the patient and no signs of genitourinary tract infection within two weeks of the procedure. The trial included 1597 women undergoing an outpatient hysteroscopy. Vaginoscopy was found to be significantly more successful than standard hysteroscopy (OR 1.49; 95% CI 1.10–2.03). Vaginoscopy was significantly quicker to perform, by 1 minute, and less painful, as measured on a 100 mm VAS scale (mean pain score 42.7 versus 46.4; p = 0.02). There was no difference in rates of acceptability; only 2% of the whole cohort found outpatient hysteroscopy unacceptable. Operative complications occurred in five women receiving vaginoscopy and 19 women receiving standard hysteroscopy, and these were mainly self-limiting vasovagal reactions (OR 0.22; 95% CI 0.08–0.59). Rates of post-procedural genital tract infection were comparable, occurring in 27 (3%) women undergoing vaginoscopy and 31 (4%) undergoing standard hysteroscopy. There was no significant difference in procedure failure between groups, occurring in 40/800 (5%) of vaginoscopies compared with 59/843 (7%) of standard hysteroscopies (OR 0.66; 95% CI 0.44–1.00). The authors concluded, in light of its greater speed, reduced pain and higher overall success, that vaginoscopy should be considered the technique of choice for outpatient hysteroscopy.
Global endometrial biopsy can be painful but the procedure is relatively simple and cheap. The idea of the vaginoscopic approach has also extended to global biopsy tools. The H Pipelle can be used after a hysteroscope has been introduced into the uterus by withdrawing the optic from the diagnostic sheath, then passing the H Pipelle into the uterus through the sheath before it is removed, allowing a biopsy to be taken in the usual way. An RCT comparing the H Pipelle to the standard Pipelle found that endometrial sampling with the H Pipelle was significantly quicker (median times: 39 seconds for H Pipelle versus 102 seconds for Pipelle; p <0.001) and less painful (median visual analogue scale: 1 for H Pipelle and 5 for Pipelle; p = 0.01) . However, there is mounting evidence that the accuracy of an endometrial biopsy in diagnosing endometrial hyperplasia and cancer in post-menopausal women has been overestimated because focal pathologies containing abnormal endometrium can be missed [14, 15]. Thus, new guidance has advocated the use of global biopsy devices only in conjunction with hysteroscopic visualisation . Where focal lesions are identified, then hysteroscopically directed biopsies can be taken using hysteroscopic biopsy forceps or using tissue removal systems .
The recent VAST study has also shown that outpatient hysteroscopy is acceptable to 98% of women . However, these high rates of acceptability may reflect the brevity of the procedure (median time less than 3 minutes), because 30% of women reported pain scores over 7/10 and 10% over 9/10. Thus, in addition to developing strategies to minimise pain and enhance patient experience, there is interest in being able to predict women at risk of severe pain or having an unacceptable procedure so that they can be counselled better regarding the choice of setting. One prospective observational series identified the experience of the practitioner as the most predictive factor for reducing the likelihood of unacceptable peri- and post-operative pain . Women with severe dysmenorrhoea were significantly more at risk of unacceptable cramps post-operatively but not during the procedure. Another study identified nulliparity, the presence of cervical pathology and procedures longer than 2 minutes in duration as being associated with severe or intolerable pain during outpatient hysteroscopy . Larger analyses are needed to confirm these findings, which should be designed with public and patient involvement to ensure that all potential prognostic factors and relevant outcomes are evaluated.
Outpatient hysteroscopy can be associated with significant anxiety and pain. Two RCTs evaluated the place of music during the procedure on patient response and experience. A study from Italy found music to be useful as a complementary method to control anxiety, reducing intraoperative blood pressure and heart rate as well as perception of pain. However, no such beneficial effects of music on patients’ perception of pain, anxiety or satisfaction were observed in a Dutch RCT in women undergoing hysteroscopy or colposcopy [20, 21]. Future research may need to evaluate multimodal approaches to enhancing the patient experience, which may include the option of listening to music, and more qualitative work is needed.
Patient experience of outpatient hysteroscopy, an intimate and invasive procedure, will be dictated to a large degree by non-medical factors; that is, clinic protocols (e.g. patient information, staffing) and the environment (e.g. waiting areas, changing areas, the clinic, post-operative recovery facilities). One study found that reduced waiting time was correlated with reduced pain during outpatient hysteroscopy . Another randomised study evaluated the impact on experience of patients looking at the screen during hysteroscopy . The study found that viewing the screen did not benefit the patient and may indeed be harmful by interfering with the patient–physician interaction. More work on understanding the best clinical environment and the needs of women is needed, which must include qualitative methodologies.
A recently completed double-blind, randomised, placebo-controlled trial assessed the role of antibiotic administration during hysteroscopic procedures in the outpatient setting . A total of 1046 women who had operative hysteroscopy were given either 1 g cefazolin intramuscularly or 10 ml isotonic sodium chloride solution. There was no significant difference in the rates of post-surgical infection between the two groups and these findings were consistent with an earlier RCT evaluating diagnostic hysteroscopy . These results give credence to the current practice of not routinely using antibiotics for hysteroscopic procedures. However, cases of infection post-procedure may be underestimated depending upon the definition of infection and surveillance methods. The recent VAST study shows that genital tract infection rates at two weeks post-procedure may be as high as 4%, so women should be counselled about the symptoms and signs of urinary tract infection and endometritis .
In light of the increasing use of therapeutic hysteroscopy and endometrial ablation in both outpatient and inpatient settings, and the scarcity of evidence , there is a clear need to conduct RCTs evaluating the effects of prophylactic antibiotics on infectious complications.
Recent developments in hysteroscopic systems have focused on the miniaturisation of endoscopes and enhanced image quality, including data capture through better optics and the introduction of high-definition (HD) cameras that provide high-resolution pictures with excellent colour reproduction. Image and video capture of surgical procedures has become commonplace, with the ability to download data via USB ports or SD card slots. Current developments are placing an emphasis on simplifying hysteroscopy, particularly with regard to portability and disposability, in keeping with the move of contemporary hysteroscopic practice to the outpatient setting.
The EndoSee® device (CooperSurgical, Trumbull, CT, USA) takes portability a step further by incorporating a small, 3.5-inch touch screen LCD monitor, with video and control electronics and a rechargeable battery (referred to as the HandTower™; Figure 19.1a). The hysteroscopes can be used only for diagnosis at present and are single-use, semi-rigid curved cannulas with a diameter of 15 Fr (4.5 mm) and a length of 287 mm. The lens, camera and LED light source are placed at the tip and comprise a digital processing chip with a complementary metal-oxide semiconductor (CMOS) sensor, as used in mobile phone and web cameras. Other technologies are in development, looking at compatibility with mobile phones and at cheaply manufactured and entirely disposable all-in-one systems.
A recent addition to the miniature hysteroscope and screen combination is the LiNA OperåScope™ (LiNA Medical ApS, Glostrup, Denmark), which is battery operated and single-use (Figure 19.1b). It can be used for diagnostic and minor operative hysteroscopic procedures and has inflow and outflow facilities at the base of the handle to enable continuous-flow hysteroscopy. The 4.2 mm cannula contains the miniature camera and the LED light source at the distal tip. The intrauterine section of the cannula is pre-curved and is rotated by a turning knob placed under the LCD panel. In the centre of the knob is the entrance to the 1.86 mm operating channel that can be used with 5 Fr instruments of a minimum length of 310 mm. Also available is an HDMI cable that can be attached to the base of the handle and connected to an external monitor or a specific recording module (OperåScope™ Recording Module™) with USB access for transfer of still or video images. The battery life is 60 minutes.
Novel hand-activated devices have been introduced which avoid the need for an electrical generator, the blades being rotated by repeated squeezing and releasing of an integrated hand-held pump. The aim of these devices is to minimise the need for additional equipment so that when small endometrial lesions, such as polyps, are identified, ‘see-and-treat’ procedures are facilitated.
The Resectr™ (Boston Scientific, Marlborough, MA, USA) comes in two sizes (Figure 19.2). The smaller version is 5 Fr and so is compatible with many diagnostic hysteroscopes with an operating channel. The small size is due to a rotating wire that passes the tissue down the shaft of the device. It is suitable for removing endometrial polyps in an office, hospital or ambulatory setting. A simple vacuum pump is still required, but it nevertheless increases the opportunities for polyps to be removed during a single visit to the hysteroscopy clinic and using a relatively small hysteroscope. The particular advantage of this device is that when the presence of intrauterine pathology is unknown in advance, then a diagnostic procedure can be undertaken with the smallest of sheaths (≤4 mm) and the least discomfort. The sheath can then be changed to one with a wider diameter (≤5.5 mm) and an operating channel, but using the same hysteroscope, and only when something like a polyp is observed.
The larger 9 Fr diameter version is more robust and can remove denser tissue, but requires a larger hysteroscope, similar in size to the ones used with the current electronically operated morcellation devices (≥5 mm).
The MyoSure® MANUAL device fits down a bespoke hysteroscope and is also manually activated (Hologic®, Marlborough, MA, USA). However, a separate vacuum pump is not required, as the hand action creates sufficient suction for the tissue to be drawn into a novel, clear, tissue trap attached to the end of the device (Figure 19.3). This device is particularly useful when intrauterine pathology is unexpected, because there is no need for additional equipment, thus facilitating a ‘see-and-treat’ approach.