SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References

An increasing number of diagnostic hysteroscopies are being performed in an outpatient setting. Most women tolerate the examination well, but die single commonest reason for failure is pain. We assessed the efficacy of a nonsteroidal, anti-inflammatory analgesic as premedication before hysteroscopy in a double-blind, placebo controlled trial. Our results showed that 500 mg mefenamic acid given one hour before hysteroscopy had no significant benefit in the discomfort experienced during the procedure but did signficantly reduce pain after hysteroscopy. A larger dose or a longer interval between premedication and hysteroscopy may possibly be associated with greater benefits.

Diagnostic hysteroscopy is now regarded as the optimum investigation for abnormal uterine bleeding. It allows complete visualisation of the uterine cavity, directed or guided endometrial biopsies, and more appropriate therapeutic management. Increasingly, diagnostic hysteroscopy is being performed as an outpatient procedure. Despite the use of local anaesthesia, the commonest reason for failure to complete the investigation is pain. For instance, in our series of 2500 outpatient hysteroscopies, 29women were not satisfied with their surgical management due to the length of time between diagnosis and operation. There were no women dissatisfied with their conservative management. One woman from each group was diagnosed and treated for pelvic infection.8% of women required local anaesthesia, 3–6% of the hysteroscopies failed, and in 84% of cases this was because of excessive discomfort1.

The injection of local anaesthetic agents into or around the cervix is not the complete solution to this problem. We have already shown that intracervical injection of lignocaine solution can be more painful than the hysteroscopy itself2. In another randomised study Vercellini et al.3 concluded that paracervical anaesthesia is ineffective in reducing the discomfort of hysteroscopy and noted that it is the endometrial biopsy which is the most painful part of the procedure.

As the above local anaesthetic regimens do not effect the sensitivity of the uterine fundus, and the discomfort of hysteroscopy may in part be related to prostaglandin induced uterine contractions, it seems logical to use prostaglandin synthesis inhibitors prophylactically for the procedure. Mefenamic acid is widely used in gynaecology to treat dysmenorrhoea and would be expected to reduce the discomfort of the procedure. We therefore studied the effects of mefenamic acid during outpatient diagnostic hysteroscopy under double-blind placebo controlled conditions.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References

Ninty-five consecutive women who were referred to our outpatient hysteroscopy clinic and gave their informed consent were included in this study, which was approved by the local medical ethics committee. Patients with a known sensitivity to nonsteroidal anti-inflammatory drugs and prostaglandin synthesis inhibitors, medication with drugs interacting with nonsteroidal anti-inflammatory drugs, inflammatory bowel disease, porphyria, and those unable to provide informed consent were excluded from the study.

Active (mefenamic acid 500 mg) and placebo tablets, which were identical in appearance, were packaged in coded bottles, randomisation being provided by the manufacturer (Parke-Davis Medical, Eastleigh, UK). The women were instructed to take their tablet orally one hour before the hysteroscopy, and were told not to use any other analgesia.

Immediately before the hysteroscopy the level of anxiety, mood, lower abdominal pain/backache, and shoulder tip pain experienced by the women were assessed using a 10 cm visual analogue scale. The hysteroscopy was then performed by a small group of experienced operators using a standard 4 mm telescope with a 30° fore-oblique lens and a 5 mm diagnostic sheath. The uterine cavity was distended with carbon dioxide via an electronic HAMOU hysteroflator set to a flow rate of 45 mL/min and a pressure of up to 100 mmHg. Illumination was provided by a 250 W high intensity cold light source, and the images were viewed on a high resolution colour monitor using a chip camera (all equipment manufactured by Karl Storz GmbH, Tuttlingen, Germany). The cervix was not dilated routinely. If the procedure was uncomfortable at any stage, local anaesthesia was given by injecting 10 mL of 1% lignocaine containing 1:200,000 adrenaline intracervically up to the level of the internal os. An 8 mm operating sheath was used for minor procedures, such as polypectomy or target biopsy. Endometrial biopsies were generally taken, if indicated, using a narrow metal curette.

The women were asked to score the worst pain experienced during hysteroscopy, and their degree of discomfort immediately after, at 30 minutes and 60 minutes after the procedure using a 4-point ordinal scale (0 = none, 1 = mild, 2 = moderate, 3 = severe). Lower abdominal pam/backache and shoulder tip pain were specifically assessed using 10 cm visual analogue scales. The women were also questioned about the relative discomfort of the hysteroscopy compared with endometrial biopsy if performed.

The data were analysed using the χ2 and Mann Whitney U tests, a result of P < 0.05 being considered statistically significant. Odds ratios were calculated using the program EPI-INFO (Centers for Disease Control, Atlanta, Georgia, USA), with exact confidence limits.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References

The mean age of the women was 43–9 years (range 26–73). The most common indication for hysteroscopy was abnormal uterine bleeding (84-2%), followed by subfertility (7–4%); 83–2% of the women were parous, and 11.5% postmenopausal. Forty-nine women were randomised to the active drug, and 46 to placebo. The two treatment groups were well matched with respect to all the above characteristics, and there were no differences in the reported anxiety, mood, or occurrence of lower abdominal and shoulder tip pain before hysteroscopy.

Hysteroscopy was unsuccessful in four women (two in each group): two refused because they were too anxious and two procedures had to be abandoned because of pain. These women had to be excluded from most of the analysis except when comparing discomfort during the procedure. The operative sheath was used in 11 women, and endometrial biopsy was performed in 58-2%.

Compared with placebo, mefenamic acid did not significantly affect the worst pain the women experienced during hysteroscopy or the pain at the end of the procedure (Fig. 1). As a reflection of this, a similar proportion of women required local anaesthesia during the procedure (active: 25%, placebo: 28.8%; P > 0.5). In contrast, there was a significant benefit of the active drug on overall pain 30 min and 60 min after hysteroscopy (active: median 1, lower-upper quartile 0–1; placebo: median 1, lower-upper quartile 1-1.5, P < 0.01 at 30 min; active: median 0, lower-upper quartile 0–1; placebo: median 1, lower-upper quartile 0–1, P < 0.05 at 60 min). Mefenamic acid reduced the odds of experiencing pain by 75% (95% CI34–90) at 30 min and 71% (95% CI 28–88) at 60 min following hysteroscopy.

image

Figure 1. Worst pain experienced during, immediately after (0 min), and 30 min and 60 min after outpatient hysteroscopy assessed on a 4-point scale: 0 = no pain (□), 1 = mild pain (inline image), 2 = moderate pain (inline image) and 4 = severe oain (▪).

Download figure to PowerPoint

Analysis of the visual analogue scales did not show any significant difference in lower abdominal pain/ backache or shoulder tip pain at any time during or after hysteroscopy even when allowance was made for anxiety and mood (Fig. 2). Neither were any large differences noted in pain experienced by nulliparous compared with parous or by premenopausal compared with postmenopausal women. 28% women in each group found the endometrial biopsy more painful than the hysteroscopy itself.

image

Figure 2. Lower abdominal and shoulder tip pain during (H), immediately after (0), and 30 min and 60 min after outpatient hysteroscopy assessed using a 10 cm visual analogue scale. Results are expressed as median and 75th and 25th centiles.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References

As far as we are aware, this is the first time that a prostaglandin synthesis inhibitor has been assessed during outpatient hysteroscopy. We found that mefenamic acid appears to have minimal effects on the discomfort associated with the hysteroscopy itself or endometrial biopsy but does significantly reduce the after-pains of the procedure. While disappointing, this result is comparable to the effects of a similar drug, naproxen sodium, which in a placebo controlled trial was found to significantly reduce backache and uterine cramps after outpatient suction endometrial curettage, but again had no effect on the severity of pain during the procedure4. There is one observational study which reported that prostaglandin synthesis inhibitors significantly lowered the discomfort of colposcopy, but as admitted by the authors themselves, the nonrandomised design and lack of a placebo arm makes interpretation of this result difficult5.

Why is it that prostaglandin synthesis inhibitors, such as mefenamic acid and naproxen, appear to be ineffective in reducing the discomfort of outpatient gynaecological procedures to a significant extent? Logically, not only should such drugs act directly as analgesics, but indirectly they should also reduce uterine contractility which may be another cause of pain during instrumentation of the uterus. It is possible that a better result would have been obtained had the mefenamic acid been taken at least two hours prior to the hysteroscopy as plasma levels peak between two to four hours. Another explanation is that the analgesic effect with respect to uterine manipulation is relatively minor, and hence small studies such as ours and that of Siddle et al.4 are not powerful enough to confirm the true pharmacological impact of such premedication (type II statistical error). Based on our results, over 500 women would have to be randomised to mefenamic acid or placebo to show with 95% certainty that the active agent was able to increase the percentage of women reporting no or mild pain during outpatient hysteroscopy from 10% to 20%.

Taking into account the possibility of spurious positive outcomes associated with multiple comparisons, our much smaller study was, however, able to show an advantage to mefenamic acid in terms of post-hysteroscopy pain relief. This result is not only consistent with the report of Siddle et al.4, but is also in agreement with the numerous papers documenting the post-operative analgesic benefits of prostaglandin synthesis inhibitors after surgery under general anaesthesia. In the field of gynaecology, for instance, van Ee et al.6 found in a double-blind randomised trial that naproxen significantly decreased post-operative pain and consumption of analgesics, and reduced hospital stay and post-discharge abdominal discomfort after day case diagnostic hysteroscopy/laparoscopy for infertility.

Conclusion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References

The results from this study indicate that premedication with 500 mg mefenamic acid given one hour before outpatient hysteroscopy is not superior to placebo in reducing the discomfort of the procedure but does significantly reduce post-operative pain. Better results might have been obtained with earlier premedication. An intra-operative benefit for mefenamic acid during outpatient diagnostic hysteroscopy can only be confirmed by a very large study.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusion
  7. References