Of ce hysteroscopy and compliance: mini-hysteroscopy versus traditional hysteroscopy in a randomized trial

Transcription

1 Human Reproduction Vol.18, No.11 pp. 2441±2445, 2003 DOI: /humrep/deg463 Of ce hysteroscopy and compliance: mini-hysteroscopy versus traditional hysteroscopy in a randomized trial Carlo De Angelis 1,3, Giuseppina Santoro 1, Maria Elisa Re 1 and Italo Nofroni 2 1 Centre for Minimally Invasive Surgery, Department of Gynecological Sciences and Perinatology, Policlinico Umberto I `La Sapienza' UniversitaÁ di Roma and 2 First Clinic of Medical Statistics and Biometry, Department of Experimental Medicine and Pathology, UniversitaÁ di Roma `La Sapienza', Rome, Italy 3 To whom correspondence should be addressed: Via G. Del Monte, 13 int. 13, Rome, Italy. cdeangelis@ tiscalinet.it BACKGROUND: Diagnostic hysteroscopy has not yet been generally accepted as a well-tolerated of ce procedure. The aim of our study was to verify compliance, side-effects and haemodynamic variations when a mini-hysteroscope is used. METHODS: A prospective randomized trial on of ce hysteroscopy was performed by comparing the use of a traditional 5 mm hysteroscope (group A) and of a 3.3 mm mini-hysteroscope (group B). Two patient groups (A and B), each comprising 100 cases, were formed on the basis of a randomized computer-generated list. RESULTS: A marked reduction in the mean (6 SD) pelvic pain score during of ce hysteroscopy was seen in group B ( ) as compared with group A ( ) (P < , Mann±Whitney test). This result was also con rmed when using an alternative approach: four classes of pelvic pain at the visual analogue score (VAS). A signi cant reduction was observed in the incidence of moderate and severe pelvic pain in group B at the end of the examination (P = 0.001) and 5±10 min later (P < 0.05). CONCLUSIONS: The use of mini-hysteroscopes (3.3 mm with diagnostic sheath) lowers considerably the level of pelvic pain the patients feel: it is halved in comparison with traditional calibre hysteroscopes ( , on a 0±10 VAS). Furthermore the outpatient hysteroscopy failure rate is less than half (2%) with the mini-hysteroscope compared with the traditional 5 mm hysteroscope (5%). As for side-effects and haemodynamic parameters, no differences were observed except for an increase (P < 0.05) in bradycardia in group B. The advantage of this technique is self-evident, if the patients' compliance is taken into account: in many cases the introduction or withdrawal of the vaginal speculum was reported as the greatest discomfort. Key words: compliance/mini-hysteroscopy/of ce hysteroscopy/pelvic pain/side-effects Introduction The technological advances made in recent yearsðfor example the introduction of small-diameter hysteroscopesðhave brought about remarkable progresses in the eld of of ce hysteroscopy, even though large-diameter instruments are still widely used (De Iaco et al., 2000; Giorda et al., 2000). However, the main limiting factor to a large-scale use of of ce hysteroscopy is the level of pain or discomfort a patient feels during or soon after the procedure, which is often caused by the instrument diameter (Bettocchi, 1996). The level of pelvic pain experienced by the patients during of ce hysteroscopy has been studied by several authors: Munro et al. (1994) asserted that in spite of local anaesthesia many patients still experienced pain and discomfort; Nagele et al. (1997) reported that an increasing number of diagnostic hysteroscopies were being performed in an outpatient setting, but pain was the most common cause for failure to complete the investigation (Nagele et al., 1997; Paschopoulos et al., 1997). In a study on the acceptability and pain of of ce hysteroscopy, De Iaco (2000) observed that this procedure is painful even when it is performed by an experienced surgeon using a non-traumatic technique. For these reasons, numerous studies have been carried out in the last decade on the use of local anaesthesia such as paracervical (Finikiotis et al., 1992; Vercellini et al., 1994; Cicinelli et al., 1998), intracervical (Broadbent et al., 1992) or transcervical (Zupi et al., 1995; Cicinelli et al., 1996; Cicinelli et al., 1997; Lau et al., 2000) anaesthesia; lignocaine spray (Davies et al., 1997; Zullo et al., 1999; Soriano et al., 2000); eutectic mixture of local anaesthetics (EMLA) cream (Stigliano et al., 1997) before of ce hysteroscopy; however, its ef cacy as a pain-relieving method has not been de nitely proven (Wieser et al., 1998). The main purpose of our Gynecological Endoscopy Unit has always been to diminish the level of pelvic pain or discomfort felt by the patient during of ce hysteroscopy in order to make this procedure acceptable and well tolerated; our aim was to make it `pain-free' and therefore widespread as against its presently limited application in Italy. Human Reproduction 18(11) ã European Society of Human Reproduction and Embryology 2003; all rights reserved 2441

2 C.DeAngelis et al. Figure 1. Patient ow through the phases of the randomized controlled trial. An Italian multicentre study (Tantini et al., 2000) evaluated the use of diagnostic and operative hysteroscopy as well as the number of hysteroscopies performed in Italian Gynecological Units every year. A total of 394 Operative Units in public hospitals (50.5% of the departments in our country) and 50 Operative Units in private clinics (27.4%) were surveyed. The results demonstrated that diagnostic and operative hysteroscopies had never been performed in 21.1 and 40.8% of the samples respectively. Within the last study year only 5.7% of the Units had carried out >500 examinations, whereas <300 and <100 procedures per year had been performed by 82.4 and 48% of the Units respectively. In our previous study, we had investigated the use of an electrical nerve stimulation device as a pain-relieving method during hysteroscopy. The data were extremely positive with 2442 reference to the reduction of the level of pelvic pain during of ce hysteroscopy and greater acceptability of the procedure by the patients (De Angelis et al., 2003). The aim of the present study was to compare two endoscopes of different calibresða traditional 5 mm Hamou I hysteroscope and a smaller 3.3 mm hysteroscopeðin order to evaluate the level of pelvic pain, the incidence of side-effects, haemodynamic parameters (heart rate, systolic and diastolic pressure) and the image quality. Materials and methods This prospective randomized study, performed between from January 2 to June 30, 2000, included 215 patients who had been referred to our Center for Minimally Invasive Therapy, Department of Obstetrics and

3 Of ce mini-hysteroscopy and compliance Table I. Patient characteristics Group A Group B P Mean age (years) NS Vaginal delivery (%) NS Caesarean section (%) NS Nulliparous (%) NS Menopause (%) NS Ages from menopause (years) < HRT (%) NS Cervical diathermy (%) NS Values are means 6 SD. NS = not signi cant (Mann±Whitney test and c 2 -test). Gynaecology, with different indications for treatment with of ce hysteroscopy. Departmental Institutional Approval and the patient's informed consent to the study were obtained. A randomized computer-generated list was used to allocate the subjects into two groups, A and B, of 105 and 102 patients respectively (Figure 1). In group A, diagnostic hysteroscopy was performed by means of a 4 mm traditional optic Hamou I Storz (Tuttlingen, Germany) with a 5 mm thick outer diagnostic sheath; the patients in group B were treated by means of a 2.7 mini-hysteroscope (Circon, USA) with a 3.3 mm diagnostic sheath. Eight patients out of 215 were excluded from the study before randomization for the following reasons: history of cardiovascular disease in ve cases (ischaemic myocardial disease, atrio-ventricular conduction disease) and refusal to undergo of ce hysteroscopy after they had read the informed consent form in the remaining three cases. With the patient lying in a lithotomic position, a bimanual pelvic examination was perfomed; the cervix was then visualized through a small-size vaginal speculum. At this point the hysteroscope was introduced into the uterine cavity without dilating the cervix or using a tenaculum. No pharmacological preparations or local anaesthesia were administered before the examination. In both groups, hysteroscopy was performed using a 270 W metal halide light source and a 3-CCD Microdigital IIIe enhanced camera (Circon). Once the instrument entered into the uterine cavity, CO 2 was delivered by means of an automatic, constant- ow and variablepressure hystero ator (Storz, Germany) with a mean ow rate of 25±30 ml/min and a 100 mmhg intrauterine pressure limit. The hysteroscopic procedure was not brought to an end in ve cases in the Hamou hysteroscope group and in two cases treated with the miniendoscope (Figure 1). The indications for hysteroscopy were the same for both groups: abnormal uterine bleeding (AUB) in pre- and post-menopausal age, AUB following HRT; ultrasound indications (endometrial thickening pattern, endometrial polyps); infertility; abnormal cytology; monitoring of the endometrium (tamoxifen therapy, HRT, previous hyperplasia); cervical polyps; post-surgery assessment; and others. The level of pelvic pain was rated according to a 0±10 cm visual analogue scale (VAS: 0 = no pain, 1±3 mild pain, 4±7 moderate pain, 8±10 severe pain). Before performing of ce hysteroscopy, the patient's obstetric and gynaecological history was recorded, and both basal blood pressure and heart rate were measured. At the very end of the procedure, the patient was asked to measure the level of pain she had experienced on a VAS. Also the presence of side-effects like shoulder pain, nausea, vomiting, bradycardia (bpm <60), dizziness was recorded, and blood pressure and heart rate were measured again at the end of the examination. The doctor interviewing the patients was different from the examiner and blinded to the study: as a pain analogue scale she used a plastic ruler with a mobile cursor that the patient would move along a non-graduated line from `no pain' to `extreme pain'. The back of the ruler was graduated from 0 to 10 and divided into cm and mm, so that when the patient moved the cursor it would read a precise objective numerical value. The duration of the procedure(s) and the CO 2 ow used during hysteroscopy were evaluated in each group. All examinations were performed by the same operator (C.D.). Statistical analysis A computer-generated randomization list was applied without any restriction, using a Basic program which applies the DOS random function connected to the computer clock. Initially a descriptive analysis of all the variables was made in order to single out any possible abnormal or incorrect data. The mean, being a variable that does not require the normality of the data, was compared using the non-parametric Mann±Whitney test. The Z-test was applied to compare the proportions. The c 2 -test was used to evaluate the association of categorical or nominal variables. The BMDP software (release 7, 1993) was used for the statistical analysis. P < 0.05 was considered signi cant. Results No statistically signi cant differences were observed between the two groups as to mean age, vaginal deliveries, Caesarean section, nulliparity, menopausal status, use of HRT and previous diathermocoagulation of the cervix (DTC) (Table I). Nevertheless the post-menopause time interval was longer in group B (group A, versus group B, ; P < , Mann±Whitney test). The indications for hysteroscopy in both groups were: AUB in pre- and post-menopausal age, abnormal uterine bleeding with HRT (25%); ultrasound ndings (endometrial thickening pattern, endometrial polyps) (37%); infertility (6.5%); abnormal cytology (1.5% = three cases with cytological nding of macrophages and hystiocytes with a pathologist's indication for an endometrial examination); monitoring of the endometrium (tamoxifen therapy, HRT, previous hyperplasia) (10.5%); cervical polyps (3.5%); post-surgery assessment (12%); and others (4%). The evaluation of the level of pain the patient had felt during of ce hysteroscopy on a 10 cm VAS gave some very signi cant results [group A, mean ; group B, mean ; (P < ) Mann±Whitney test] (Figure 1, Figure 2). The level of pelvic pain, the main limiting factor to the large-scale use of hysteroscopy, was halved by the use of mini-endoscopes as compared with traditional hysteroscopes. When the level of pain was rated according to the four classes of the VAS (severe 8±10, moderate 4±7, mild 1±3, no pain), the incidence of moderate±severe pain was found to be lower in group B than in group A (P = 0.001, c 2 -test); group A: no pain 4%, mild pain 25%, moderate pain 57%, severe pain 14%; group B: no pain 30%, mild pain 47%, moderate pain 21%, severe pain 2% (Figure 3). Furthermore, the pelvic pain the patients had experienced was evaluated also 5 min after the examination: this time a more signi cant reduction was achieved in group B compared with group A (group A: no pain±mild pain 85%, 2443

4 C.DeAngelis et al. Figure 2. Mean pain score, rated using a 10 cm visual analogue scale in group A (`traditional hysteroscope' ( ) and in group B (`mini-hysteroscope') ( ), P < , Mann±Whitney test. Figure 3. Pelvic pain divided into four classes according to the visual analogue scale (8±10 severe, 4±7 moderate, 1±3 mild, no pain). A comparison between group A (`traditional hysteroscope') and group B (`mini-hysteroscope') (P = 0.001, c 2 -test). moderate±severe pain 15%; group B: no pain±mild pain 97%, moderate±severe pain 3%; P < 0.05, c 2 -test). As to the incidence of nausea, shoulder pain and dizziness, no statistically signi cant differences were observed between the groups. No cases of bradycardia (bpm <60) were seen in group A whereas ve cases occurred in group B where the frequency rate decreased temporarily to 56 bpm (Table II). Four of these patients recovered spontaneously without the use of any medication, but the remaining case required the administration of 0.5 mg atropine i.v. The evaluation of the heart rate at the beginning and at the end of the procedure showed a reduction in both groups (group A: versus bpm; group B: versus bpm; P = 0.96, Mann±Whitney test), but it was not statistically signi cant. Systolic and diastolic pressure increased in both groups (systolic pressure: group A, versus mmhg; group B, versus mmhg; P = 0.125, Mann±Whitney test; diastolic pressure: group A, 82 versus 83.7 mmhg; group B, 80 versus 85.5 mmhg, P = 0.84, Mann±Whitney test) probably as a result of pelvic pain or emotional stress, but these data were not statistically signi cant. The small calibre hysteroscope required a higher mean CO 2 ow than the traditional hysteroscope (group A, ; group B, ml/min; P = , Mann±Whitney test). The duration of the procedure was just as long in both groups (group A, s; group B, s; P = NS, Mann±Whitney test). Discussion The use of mini-hysteroscopes (3.3 mm) has brought about a small revolution in hysteroscopy by lowering the level of pelvic pain the patients feel. In our study it has been halved compared with traditional calibre hysteroscopes, hence the invasiveness of the examination is also reduced. Even the failure rate of outpatient hysteroscopy is less than half (2%) with the mini-hysteroscope compared with the traditional 5 mm hysteroscope (5%). This means that in at least Table II. Side-effects Group A Group B P Nausea (%) 4 2 NS Shoulder pain (%) 7 9 NS Bradycardia (%) Dizziness (%) 3 4 NS NS = not signi cant (c 2 -test). three cases out of 100 patients, we do not need the operating theatre to perform diagnostic hysteroscopy under general anaesthesia. The image quality provided by the mini-endoscope with a lens system (2.7 mm) is almost as high as that of the traditional hysteroscopic optics (4 mm) in terms of luminosity, outline de nition and width of the visual eld. However, the use of mini-optics did not allow a good observation of the uterine cavity in 5% of the cases. Actually the diameter of the instrument was too small for the speci c anatomic conditions of the uterus: a wide open cervical canal, a notably thickened endometrium, and the presence of blood inside the uterine cavity. In these cases, a greater optical lens (5 mm) had to be used to satisfactorily bring the examination to an end. The advantage of using mini-optics is self-evident if we consider the patient's compliance: in most examinations only a slight pain or simply a feeling of discomfort during or at the end of the examination (mean pelvic pain in the 0±10 VAS) is reported, which is comparable to the level of pain felt during transvaginal ultrasound. Sometimes the worst discomfort is linked only to the introduction or withdrawal of the speculum. It can be inferred from the above that the great diagnostic advantages of the endoscopic technique and the extremely high acceptance and tolerability by the patients may widen the use of the procedure, as is certainly desirable for the hysteroscopic technique. However, the use of mini-optics entails a suitable pre-training with traditional hysteroscopic optics, which is certainly more indicated to acquire the necessary operative skills and the correct spatial orientation. 2444

5 Of ce mini-hysteroscopy and compliance Another limitation to the use of mini-optics is its high costs: even though the initial costs do not differ very much, miniendoscopes are more delicate and wear out more easily than large-calibre instruments, thus requiring more frequent substitutions. Last, but not least, the higher incidence of bradycardia in the mini-optic group observed in our study must be considered. Such a result was very surprising: actually the smaller calibre and consequently the reduced trauma at the level of the internal uterine os should have led to a lower stimulation of Frankenauser's ganglion at the beginning of the vagal stimulus point, and therefore to a lower depression in the cardiac frequency. Other mechanisms are evidently at work. As found in this study and in a previous investigation (Yang and Vollenhoven, 2002), there is no direct correlation between the level of pain reduction and a lower incidence of bradycardia: a progressive reduction in the level of pelvic pain felt during the endoscopic examination does not correspond to a lower negative stimulus on the cardiac frequency. We have put forward various assumptions to explain this behaviour. First, the use of a 4 mm Novak cannula in ~40% of the cases might have given rise to the vaso-vagal activation. However, in the remaining 60% of the cases, only hysteroscopy with mini-optics had been performed. This means that there is a different cause for bradycardia. Second, based on the CO 2 ow data, we noted that there was a signi cant difference against the use of mini-optics versus traditional optics (a higher ow for the former). It is then possible that higher CO 2 pressure and ow velocity might be required to maintain the CO 2 ow constant. Hence it can be assumed that the uterine cavity distends more rapidly and abruptly, with a higher stimulation on the sensitive nerve bres on the uterine wall. This would somehow produce a vaso-vagal activation. Finally, as a last assumption, a chemical stimulating effect of CO 2 on the nervous ends (see phrenic re ection in laparoscopy for comparison) rather than a mechanical distention of the uterine walls can be considered. In this case, a different outcome should be reached with the use of a liquid as a means to distend the uterine cavity. Acknowledgement Special thanks to Miss Claudia Di Loreto for her kind cooperation. References Bettocchi, S. (1996) New era of of ce hysteroscopy. J. Am. Assoc. Gynecol. Laparosc., 3 (Suppl. 4), S4. Broadbent, J.A.M., Hill, N.C.W., Molnar, B.G., Rolfe, K.J. and Magos, A.L. (1992) Randomized placebo controlled trial to assess the role of intracervical lignocaine in outpatient hysteroscopy. Br. J. Obstet. Gynecol., 99, 777±780. Cicinelli, E., Didonna, T., Fiore, G., Parisi, C., Matteo, M.G. and Castrovilli, G. (1996) Topical anesthesia for hysteroscopy in postmenopausal women. J. Am. Assoc. Gynecol. Laparosc., 4, 9±12. Cicinelli, E., Didonna, T., Ambrosi, G., Schonauer, L.M., Fiore, G. and Matteo, M.G. (1997) Topical anaesthesia for diagnostic hysteroscopy and endometrial biopsy in postmenopausal women: a randomised placebocontrolled double-blind study. Br. J. Obstet. Gynecol., 104, 316±319. Cicinelli, E., Didonna, T., Schonauer, L.M., Stragapede, S., Falco, N. and Pansini, N. (1998) Paracervical anesthesia for hysteroscopy and endometrial biopsy in postmenopausal women. A randomised, double-blind, placebocontrolled study. J. Reprod. Med., 43, 1014±1018. Davies, A., Richardson, R.E., O'Connor, H., Baskett, T.F., Nagele, F. and Magos, A.L. (1997) Lignocaine aerosol spray in outpatient hysteroscopy: a randomized double-blind placebo-controlled trial. Fertil. Steril., 67, 1019± De Angelis, C., Perrone, G., Santoro, G., Nofroni, I. and Zichella, L. (2003) Suppression of pelvic pain during hysteroscopy using a transcutaneous electrical nerve stimulation device. Fertil. Steril., 79, 1422±1427. De Iaco, P., Marabini, A., Stefanetti, M., Del Vecchio, C. and Bovicelli, L. (2000) Acceptability and pain of outpatient hysteroscopy. J. Am. Assoc. Gynecol. Laparosc., 7, 71±75. Finikiotis, G. and Tsocanos, S. (1992) Outpatient hysteroscopy: a comparison of 2 methods of local analgesia. Aust. NZ J. Obstet. Gynecol., 32, 373±374. Giorda, G., Scarabelli, C., Franceschi, S. and Campagnutta, E. (2000) Feasibility and pain control in outpatient hysteroscopy in postmenopausal women: a randomised trial. Acta Obstet. Gynecol. Scand., 79, 593±597. Lau, W., Tam, W.H., Lo, W.K. and Yuen, P.M. (2000) A randomised doubleblind placebo-controlled trial of transcervical intrauterine local anaesthesia in outpatient hysteroscopy. Br. J. Obstet. Gynecol., 107, 610±613. Munro, M.G., Jensen, J. and Purdon, T.F. (1994) Narrow caliber of ce hysteroscope. J. Am. Assoc. Gynecol. Laparosc., 1, S24. Nagele, F., Lockwood, G. and Magos, A.L. (1997) Randomized placebo controlled trial of mefenamic acid for premedication of outpatient hysteroscopy: a pilot study. Br. J. Obstet. Gynecol., 104, 842±844. Paschopoulos, M., Araskevaidis, E., Stefanidis, K., Kotinas, G. and Lolis, D. (1997) Vaginoscopic approach to outpatient hysteroscopy. J. Am. Assoc. Gynecol. Laparosc., 4, 465±467. Soriano, D., Ajaj, S., Chuong, T., Deval, B., Fauconnier, A. and Darai, E. (2000) Lidocaine spray and outpatient hysteroscopy: randomised placebocontrolled trial. Obstet. Gynecol., 96, 661±664. Stigliano, C.M., Mollo, A. and Zullo, F. (1997) Two modalities of topical anesthesia for of ce hysteroscopy. Int. J. Gynecol. Obstet., 59, 151±152. Tantini, C., Ottanelli, S. and Becorpi, A. (2000) Diagnostica isteroscopica quali limiti? Ginecologia della terza etaá V Congresso Nazionale SIGITE Isola D'Elba, CIC Edizioni Internazionali, Roma, pp. 179±182. Vercellini, P., Colombo, A., Mauro, F., Oldani, S., Bramante, T. and Crosignani, P.G. (1994) Paracervical anesthesia for outpatient hysteroscopy. Fertil. Steril., 62, 1083±1085. Wieser, F., Kurz, C., Wenzi, R., Albrecht, A., Huber, J.C. and Nagele, F. (1998) Atraumatic cervical passage at outpatient hysteroscopy. Fertil. Steril., 69, 549±551. Yang, J. and Vollenhoven, B. (2002) Pain control in outpatient hysteroscopy. Obstet. Gynecol. Surv., 57, 693±702. Zullo, F., Pellicano, M., Stigliano, C.M., Di Carlo, C., Fabrizio, A. and Nappi, C. (1999) Topical anesthesia for of ce hysteroscopy. A prospective, randomised study comparing two modalities. J. Reprod. Med., 44, 865±869. Zupi, E., Luciano, AA., Valli, E., Marconi, D., Manechi, F. and Romanini, C. (1995) The use of topical anesthesia in diagnostic hysteroscopy and endometrial biopsy. Fertil. Steril., 63, 414±416. Submitted on March 25, 2003; resubmitted on June 16, 2003; accepted on July 28,