- Top of page
Objective To evaluate the effectiveness of intrauterine lignocaine as an anaesthetic during endometrial sampling.
Design Prospective, randomised, double-blind, placebo-controlled trial.
Setting Outpatient gynaecological minor operation unit in a public hospital.
Population Two hundred premenopausal women scheduled for endometrial sampling because of abnormal uterine bleeding.
Methods Transcervical intrauterine instillation of 5 mL of 2% lignocaine or 5 mL of normal saline before performing endometrial sampling with vacuum aspirator.
Main outcome measures Evaluation of pain associated with the procedure using a visual analogue scale.
Results The use of intrauterine lignocaine reduced pain during suction curettage in endometrial sampling.
Conclusions Transcervical instillation of lignocaine reduced pain during endometrial sampling.
- Top of page
The study was approved by the Ethics Committee of Princess Margaret Hospital. Women scheduled for endometrial sampling were invited to participate in the study. Exclusion criteria included postmenopausal status, uterine size larger than eight weeks of gestational size, active vaginal infection, pregnancy, known allergy to lignocaine, use of antiarrhythmic drugs, epilepsy and refusal by patients. Early in the study period, the need for other procedures such as change or removal of intrauterine contraceptive device and cervical polypectomy was added to the exclusion criteria.
Before each minor operation session, a number of 10 mL syringes containing either 5 mL of normal saline or 2% lignocaine were prepared according to a random table by a nurse not involved in the subsequent procedures and evaluation. The syringes were labelled with consecutive numbers. The investigators, the patients and the nurses involved in the procedure did not know the content of each syringe. A record of their corresponding content was kept in a drawer in the operation room. The investigators were only allowed to break the code if any adverse reaction occurred during the study. After giving their consent, the women were allocated according to the sequence of consecutive numbers.
No premedication was given. The woman was placed in a lithotomy position. A bimanual vaginal examination was performed to assess the size and position of the uterus. The cervix was exposed with a bivalve speculum. The vagina was cleansed with antiseptic solution containing chlorhexidine 0.015% and cetrimide 0.15%. The cervix was grasped with vulsellum forceps. A 4.8-cm 16-gauge angio-catheter with metal stylet removed and the plastic catheter attached to the 10-mL syringe was inserted through the cervical os and advanced to the hub. The solution was infused into the uterine cavity through the angiocatheter, which was left in place for 3 minutes before being with-drawn. Uterine sounding was then performed. Uterine sounding might have been performed before the insertion of angiocatheter if there was difficulty in the insertion. Endometrial sampling was then performed using a 3-mm Karman Vacuum Aspirator Curette (Rocket Medical, England) connected to a negative pressure of 200 mmHg. The mucosal surface of the uterine cavity was gently and systematically curetted from the fundus down to the internal os at 3, 6, 9 and 12 o’clock.
A nurse instructed the patients before the procedure how to score their pain on the visual analogue scale (a horizontal line of 10 cm without gradations, with 0 on the left side indicating no pain and 10 on the right side indicating the worst possible pain). During the operation, the woman was informed of each stage of the procedure and asked to record the pain she felt by marking on the scale. The scores were made when the cervix was grasped (T1), insertion of angio-catheter and instillation of the solution into the uterine cavity (T2), uterine sounding (T3), suction curettage (T4) and 15 minutes after the procedure (T5).
Blood pressure and pulse were recorded before and after the procedure. Vasovagal symptom was assessed with questionnaire. Vasovagal shock was defined as bradycardia (<60 bpm) and/or hypotension (reduction of blood pressure of more than 20%), associated with faintness, sweating, nausea or vomiting.
Initially, the patients were followed up in the gynaecological clinic six weeks after the procedure. In the latter part of the study period, due to the increased caseload of the clinic, the follow up interval was lengthened to three months after the procedure. Another appointment would be arranged if the patient could not attend the first appointment. Those who were not followed up within six months after the operation were treated as having defaulted. During the follow up visit, any complication such as infection or hospital admission was noted. The histological reports of the endometrial biopsies were also reviewed. The pathologists’ comments on adequacy of the specimen for histological diagnosis were noted.
A pilot study on 50 women was carried out before the study. The pain scores were not in normal distribution. Median pain score during suction curettage was 4.7 with inter-quartile range of 6.7. We believed that a reduction in pain score of 20% is of clinical significance. In order to detect such a difference with a power of 80% and type I error of 0.05, a sample size of 100 was required in each arm.
Analysis was done on the basis of intention to treat. Statistical analysis was performed using Mann–Whitney U test, Student's t test and χ2 test, where appropriate. P < 0.05 was considered statistically significant. Confidence intervals were calculated for estimates of the differences.
- Top of page
The study was carried out from May 2001 to July 2002. Ninety-eight and 102 women were randomised to the lignocaine group and the placebo group, respectively. The two groups were comparable in term of age, parity and uterine sounding measurement. The indications for endometrial sampling were also comparable between the two groups (1Table 1).
Table 1. Characteristics of women in the two groups and indications for endometrial sampling. Values are presented as n or mean (SD)
| ||Lignocaine (n= 98)||Normal saline (n= 102)|
|Age (years)||43.9 (4.4)||44.4 (4.5)|
|Uterine sound (cm)||7.9 (1.1)||7.9 (1.1)|
Other procedures were performed in five women, including one IUCD removal in lignocaine group, three IUCD removal and one polypectomy in normal saline group. All other cases had only endometrial sampling. Endometrial sampling could not be done in one case in the placebo group because of tight cervical os. In the early part of the study period, the women requiring other procedures were excluded from study (1Fig. 1).
The angiocatheter could be inserted into the uterine cavity in 92% of the cases. In 64% of the cases, angiocatheter was inserted after uterine sounding.
2Table 2 showed the pain scores at various stages of the procedure. Only mild pain was found during the insertion of the angiocatheter. Suction curettage was the most painful step. The pain score was significantly less during suction curettage in the lignocaine group. There was no significant difference in pain score in other stages of the procedure between the two groups. After excluding those cases with other procedures performed, the difference in pain score during suction curettage (2.1 vs 4.2) between the two groups remained statistically significant, P= 0.017.
Table 2. Pain scores at various stages of the procedure. Values are presented as median (inter-quartile range)
| ||Lignocaine (n= 98)||Normal saline (n= 102)||Median difference (95% CI)||P|
|At grasping of cervix (T1)||1.6 (3.3)||1.7 (4.0)||−0.2 (−0.7 to 0.2)||0.5|
|At the insertion of angiocatheter and instillation of the solution into the uterine cavity (T2)||1.3 (4.3)||1.2 (4.1)||−0.1 (−0.5 to 0.3)||0.3|
|During uterine sounding (T3)||2.2 (5.0)||3.5 (4.7)||−0.3 (−1.1 to 0.3)||0.3|
|During suction curettage (T4)||2.3 (4.3)||4.2 (4.6)||−0.9 (−1.9 to −0.2)||0.01|
|15 minutes after the procedure (T5)||0.5 (1.8)||0.4 (1.3)||0.0 (−0.1 to 0.2)||0.6|
3Table 3 showed the blood pressure and pulse before and after the procedure. There was no difference found between the two groups. There were two cases of symptomatic vaso-vagal reaction in the placebo group, but none in the lignocaine group. The difference was not statistically significant (P= 0.164). They recovered spontaneously.
Table 3. Blood pressure and pulse before and after the procedure. Values are presented as mean (SD)
| ||Lignocaine (n= 98)||Normal saline (n= 102)||P|
|Before procedure|| |
|Systolic (mmHg)||147 (20)||147 (20)||0.9|
|Diastolic (mmHg)||82 (12)||80 (13)||0.4|
|Pulse rate (bpm)||81 (12)||82 (13)||0.3|
|After procedure|| |
|Systolic (mmHg)||139 (18)||140 (21)||0.9|
|Diastolic (mmHg)||78 (12)||78 (12)||0.9|
|Pulse rate (bpm)||73 (12)||74 (12)||0.2|
Seven patients (3.5%) defaulted any follow up within six months after the procedure. One hundred and ninety-three patients were followed up. There were two cases of suspected pelvic inflammatory disease in the control group and one case in the lignocaine group. Antibiotics were given to two patients as outpatient at Accident and Emergency Department and by private doctors. One patient required admission for treatment. The overall complication rate of suspected infection was 1.6%.
Overall in 8% of the patients, the specimens obtained by the curettage were inadequate for histological diagnosis. Five patients in the control group and 11 in lignocaine group had inadequate specimen (P= 0.099). In the year before the study, when endometrial samplings were done without any intrauterine instillation, the rate of insufficient specimen was 6% (15/241).
- Top of page
Endometrial sampling is a very common investigation for women with abnormal uterine bleeding and menstrual disorder. It is usually performed as day case or in outpatient setting without anaesthesia. Pain during the procedure was the main problem. In previous studies, various methods had been used to reduce the pain associated with the procedure. Broadbent et al.3 found no benefit of intracervical injection of lignocaine in term of reducing the discomfort of the procedure and actually injection itself was painful. Wong et al.4 was unable to detect significant anaesthetic effect with 2% lignocaine gel applied to cervix during hysteroscopy and Vabra endometrial sampling. Lau et al.5 found that paracervical block with lignocaine failed to reduce pain during hystreroscopy and endometrial sampling. The injection itself was also found to be painful. Moreover, increased incidence of symptomatic bradycardia and hypotension was noted and it was thought to be due to anaesthetic intravasation.
A randomised controlled trial by Zupi et al.1 suggested that transcervical instillation of local anaesthetic agent into the uterine cavity could reduce pain in hysteroscopy and endometrial biopsy but the trial was small and the difference was not statistically significant. The anaesthetic agent was thought to act on the nerve endings within the endometrial mucosa. Trolice showed that intrauterine lignocaine may reduce pain during Pipelle endometrial sampling. The study fluid was instilled with Angiocath advanced to hub and left in place for 3 minutes to minimise backflow and allow the anaesthetic to give effect.2 However, the sample size was only 41.
Endometrial sampling with vacuum aspirator requires a suction pump but covers a greater sampling area of the uterine cavity than Pipelle endometrial sampling (41.6% and 4.2%, respectively).6 The former may be a better option for patient with some suspicion of endometrial malignancy but the procedure is more painful. This study included 200 premenopausal women undergoing endometrial biopsy with vacuum aspirator, using the method of Trolice to ensure intrauterine instillation and keep the study fluid in utero for 3 minutes to reach effect. Significant reduction in pain score was found during suction curettage, the most painful step in the whole procedure. The insertion of the intrauterine catheter and the instillation of fluid agents were associated with only mild pain. It was not associated with increased risk of symptomatic vasovagal reaction. The incidence of infection mobidity was minimal. One possible undesirable effect was that the presence of fluid in utero might impair the yield of suction curettage and increase the rate of inadequate tissue sample yielded. This risk however was not detected in our study.
Hysteroscopy performed before endometrial sampling has been shown to improve the detection of focal intrauterine pathologies.7 With the use of mini-hysteroscope, the procedure can be done as outpatient procedure at a reduced cost compared with inpatient procedure done under general anaesthesia.8 An effective anaesthetic is important in outpatient setting to reduce patient discomfort and improve their satisfaction. Lau et al.5 reported the result of a trial on intrauterine instillation of 2% lignocaine on 90 women undergoing outpatient hysteroscopy and endometrial sampling. Carbon dioxide was used as distending medium during hysteroscopy. No difference was found in pain scores between the lignocaine group and the placebo group.9 However, the instillation of the study fluids was through an angiocatheter inserted only up to the internal os of cervix and it was not mentioned whether any attempt was made to keep the study fluid inside the uterine cavity to allow the local anaesthetic to act. Furthermore, hysteroscopy was performed before endometrial sampling. Uterine distension itself caused pain and the distending medium might displace the local anaesthetic agent from the uterine cavity or dilute the anaesthetic agent. Shankar et al.10 found that adding lignocaine to normal saline as distension medium for hysteroscopy and endometrial biopsy did not reduce the pain during the procedure. However, lignocaine was diluted 13.5 times compared with the 2% lignocaine used in the present study. The choice of distending medium also matters in term of patient discomfort. Carbon dioxide was associated with higher level of pain during and after the procedure than normal saline.11 Further study is required to find out the best method in reducing pain in outpatient hysteroscopy and endometrial sampling.