Comparative trial of the force required for，and pain of，removing

作者： 日期：2012-11-14

Rachel E D'Souza， MRCOG，Specialist Registrar in Sexual and Reproductive Health， Margaret Pyke Centre， London，UK; Walli Bounds， SCM，Principal Research Fellow，Department of Obstetrics and Gynaecology， University College London and Research Co-ordinator， Margaret Pyke Centre*，London，UK; John G uillebaud， FRCSEd，FRCOG，Medical Director and Professor of Family Planning and Reproductive Health，Margaret Pyke Centre*，London，UK

 

*N ow reti red.

 

Correspondence: Dr R E D'Souza，Margaret Pyke Centre，73 Charlotte Street，London WlT 4PL，UK. Tel: +44 (0)20 7530 3631. E-ma il: R a chel. D' Souza @ ca mdenpct，nhs. Uk

 

(Accepted 22nd January 2003)

Journal of Family Planning and Reproductive Health Care 2003; 29(2): 29-31

 

A bstr act

Objective. To assess the force required for，and pain of，removal of the GyneFix^® as compared with T-framed

intrauterine devices (IUDs).

Design. A comparative trial following patient-blinded randomisation in an outpatient clinic setting.

Method. Women requesting an IUD for emergency contraception were fitted with either a GyneFix or a Gyne-T380S^®.For those requesting remova I of the IUD，visual analogue scores were used to assess their perception of the associated pain，and a Newton dynamometer was used to measure the force required to remove the device.

Results. Removal required significantly more force for GyneFix as compared with Gyne-T380S (p = 0.004)，but there was no significant difference in pain perceived by women during removal. Interestingly， anticipated pain was worse than actual pain experienced.

Conclusion. Although more force is needed to remove the GyneFix as compared with the Gyne-T380S，this does not

 

 

Key message points ·　More force is needed to remove the anchored GyneFix as compared with the framed G yne-T 380S. ·　There is，however，no difference in the amount of pain perceived in removing the GyneFix as compared with the Gyne-T 380S.

translate into more pain.

 

 

Introduction

The GyneFix^® is a flexible，frameless intrauterine device (IUD)，developed in the mid-1980s，in an attempt to reduce

the pain and bi eedi ng associated with framed IUD s. 1 I t has been marketed in the UK since early 1998，and is currently licensed for 5-year use.

Given that the GyneFix is attached to the fundal myometrium，it is likely that removal both requires more

force and is associated with more pain than for a framed IUD which ‘sits’unattached within the uterine cavity^®

Yet to date， no studies have objectively compared the removal of various types of IUDs.

We decided to assess both the force and pain involved in removal of the GyneFix as compared with the Gyne-T380S^®.We studied women requesting removal of their device in a previously described randomised controlled trial comparing insertion-linked pain and the short-term user-acceptability and safety of the above devices.²

 

Methods

Ethics a pprova I

The project was reviewed by the local research ethics committee of the Camden and Islington Community Health

Services Trust，London，UK.

 

Description of IUDs used

The GyneFix1，3 consists of six copper sleeves threaded on a length of polypropylene thread，giving a total exposed copper surface area of 330 mm2. A knot on the proximal end of the thread is anchored in the myometrium of the uterine fundus at a controlled depth of 10 mm， using a specially designed inserter.

The Gyne-T380S4 is a T-shaped IUD that has a polyethylene frame wound with 176 mg of copper wire on the stem，and collars each containing 66.5 mg of copper on its transverse arms. The total exposed surface area of copper is approximately 380 mm2. The device includes a monofilament polyethylene tail，which aids in removal of the IUD.

 

Admission procedure

Women had previously been self-selected from among the patients attending an open-access specialist National Health Service (NHS) contraception service in central London for emergency contraception. Of those requesting the IUD method, 192 women had consented in writing to be randomly assigned to receive either the GyneFix or the Gyne-T380S or the Nova-T200~, taking into account their intentions regarding short- or long-term use. Details of the eligibility criteria, randomisation and blinding have been described el sewhere.2

 

Remova I of the IUD

We asked each woman requesting removal of the IUD at the end of the 6-week observation period to complete visual analogue scales (VAS) to document her perception of pain experienced. We showed each woman how to draw a vertical line where she felt appropriate along a horizontal line 100 mm in length, the left end representing no pain at al I, and the extreme right the worst pain she could i magi ne.

Immediately prior to IUD removal we took a VAS measurement of each woman's anticipated pain of removal. We then clamped straight Spencer Wells forceps to the string(s) of the IUD. We removed the IUD by pulling on a calibrated Newton dynamometer hooked to one handle of the forceps. By use of a peak force indicator we were able to measure the maximum force used in removal. T he force was calculated to the nearest fifth decimal figure. A stopwatch was started, and 5 minutes after removal each woman completed a VA S of the actual pain of removal, and

also of the pain at 5 minutes. The doctors used a similar VA S to record their own assessment of the woman's pain of removal.

The type of device fitted was not reveal ed to the woman until completion of removal. Because the string(s) of the IUDs are different in colour and quantity, it was not possible for the physician to be blinded as to which device he/she was removing.

 

Statistical considerations and data analyses

Outcome measures. Our primary outcome measure was the force of IUD removal. Our secondary outcome measure was pain at removal.

 

Power calculations. These were based on the pain of insertion.²

The data were transcribed into, and analysed with the aid of, the Statistical Package for the Social Sciences (SPSS) (SPSS, Chicago, IL, USA) software package,version 9. We compared the force required for removal of the GyneFix and the Gyne-T380S. We compared the VAS scores relating to anticipated and actual pain of the removal procedure, and pain 5 minutes after the removal. We also compared the investigator's VA S scores with the woman's VAS scores relating to the pain experienced during the removal. In comparisons of categorical data, Chi-squared analysis was performed, whereas continuous data were compared using two sample t-tests. For all statistical analyses p < 0.05 is considered significant, and confidence intervals (CI) are quoted at the 95% level.

 

R esu Its

Numbers recruited and followed up

In the original study, a total of 192 patients were randomised, 175 within the long-term arm of the study (90 to GyneFix and 85 to Gyne-T380S), and 17 within the short-term arm (10 to GyneFix and seven to Nova-T200). The latter arm was discontinued early due to Iow recruitment. Removal was undertaken in the research unit for 20 long-term users, and for 10 short-term users. (For a further seven long-term users removal was performed elsewhere.) Since the two groups were randomized separately, we considered it inappropriate to analyse the removal data jointly. Therefore the results below and in the Tables 1-3 relate to long-term users only. However, including in the analysis also the short-term users did not alter out findings significantly, as can be seen in Table 4. The indication for removal is shown in Table 1.

 

Demographics

T here was no significant difference between those women randomised to GyneFix and Gyne-T380S in terms of age,

parity and previous experience of vaginal examination and IUD use (Table 2).

 

I U D remova I

The force required to remove the device was significantly higher for those with GyneFix [mean 5.2 Newtons (N)] as

compared with Gyne-T380S (mean 2.2 N), p = 0.004 (Table 3).

  Yet there was no significant difference in the pain experienced during removal, either as assessed by the patient or as perceived by the doctor. Interestingly the patients anticipated the pain to be considerably more than they actually experienced. In fact, the pre-removal VAS scores were not dissimilar to the pre-insertion VAS scores at the commencement of the study.2 T he doctors perceived the removal pain to be considerably higher than actually documented by the patients.

 

Discussion

No previous comparison has been published of the difference in the force required to remove an implanted GyneFix as compared with framed IUDs. However, the forces required for IUD removal in our study are comparable to those quoted el sewhere. Wildemeersch et al. quote the average traction force required to remove an implanted GyneFix as 6 N, (I B~t~r and D Wildemeersch,personal communication) which is three to four times that required for framed IUDs (1.0-1.7 N).s The myometrial tissue reaction at the site of anchorage of the polypropylene knot has been shown in hysterectomy specimens to be minimal.3

Importantly, this significantly greater force required to remove an implanted GyneFix as compared with the Gyne-T380S did not translate to any significant difference in discomfort experienced by the patient, which was minimal for all devices. It should be noted that these findings are based on small numbers, with Ii mited power to detect small differences in the pain of removal experienced.

Moreover, our pre-IUD removal VAS scores suggest that patients expect the removal experience to be as painful

as the actual IUD insertion. Our findings confirm that this is not so, including for the GyneFix, and should encourage

clini cians to counsel patients reassuringly.

 

Conclusion

Although more force is needed to remove the GyneFix as compared the Gyne-T380S, this does not translate into

more pain.

Ack nowl edgements

The authors are grateful to all the women who took part. The authors are indebted to those involved in enrolling, treating and following-up the women, namely Juliet Johnson, Sophie Molloy, Tina Proctor, Suzanne Everett, Dr l-racey Masters, Dr Karin Piegsa and Dr Jayne Kavanagh, and all the Margaret Pyke C entre staff who helped i n recruitment. T he authors are grateful to Dr David Smith and Professor Douglas Altman of the Centre for Statistics in Medicine, Oxford, U K and to Alison Orr for her ever-willi ng administrative assistance.

 

Statements on funding and competing interests

Funding. The project was partially funded by the manufacturers of the GyneFix^® (Contrel Europe, K nokke-Heist, Belgium) and by the Margaret Pyke Memorial Trust.

Competing interests. Professor Guillebaud has received ad hoc consultancy payments, lecture fees and research grants from various

manufacturers of contraceptives.References

1   Van Kets H, Van der Pas H, l-hiery M, et al. The GyneFix^® implant systems for interval, postabortal and postpartum contraception: a significant advance in long-term reversible contraception.International Study Group on Intrauterine Drug Delivery. Eur J    Contracept Reprod Health Care 1997; 2: 1-13.

2   D'Souza R, Masters T, Bounds W, etal. R andomised controlled trial assessing the acceptability of GyneFix~ versus Gyne-T380S~ for emergency contraception. J Fam Plann Reprod Health Care 2003;29: 23-29.

3   Wildemeersch D, Batar I, Webb A, et al. GyneFix~. The frameless intrauterine contraceptive implant - an update. Br J Fam Plann 1999; 24:149-159.

4   l-reiman K, Liskin L, Kols A, et al. IUDs - an update. Population Reports (B6). Baltimore, MD: Johns Hopkins School of Public

    Health, Population Information Program, 1995.

5.  

 

Online Access to the Journal Free access for all to the full text of the Journal of Family Planning and Reproductive Health C are ends on 30 June 2003. After that time free access for all to abstracts will continue and free access to full text will be password protected for members of the Faculty of Family Planning and Reproductive Health Care and Journal subscribers.

Kurz KH. Role of retention in avoiding expulsion of IUDs - measuring devices for basic research. C ontr aceptive Delivery Systems 1982; 9: 107-116.

 

                          

 

 

 

Introduction of the GyneFix intra-uterine device into the UK:

Client satisfaction survey and casenotes review

 

Joanne Dennis, MRCGP, MFFP, Career Grade Trainee; Anne Webb, MRCGP, MFFP, MRCOG, Consultant; Meera K ishen, MD, DGO,Dip. Ven., MFFP, Consultant, Abacus Centres for Contraception and Reproductive Health, North Mersey Community NHS Trust,Liverpool, UK.

 

Correspondence: Dr Jo Dennis, Central Abacus, 40-46 Dale St, Liverpool, L2 5SF. Tel: 0151 284 2500, Fax: 0151 293 2005.

 

(Accepted 30th March 2001)

The Journal of Family Planning and Reproductive Health Care 2001: 27(3): 139-144

 

A bstr act

Objective. To assess the first year's use with the GyneFix intra-uterine device. This has been used in Liverpool since

early 1997. It is offered mainly to nulliparous women wishing to use an effective non-hormonal method and parous women who have had pain with, or expulsion of, a framed device previously. It is also used as emergency contraception.

Design. Retrospective study, by means of casenote review a nd client questi onnai re.

Participants. All GyneFix users from February 1997 to January 1998; 215 women in total.

Main outcome measures. Data were extracted from casenotes to determine reasons for choice of GyneFix, parity, whether the insertion was planned or as an emergency measure, problems reported at follow-up and reasons for removal. Users were asked by questionnaire to comment on insertion procedure, menstrual or other changes since insertion and to state their overall sa ti sfa cti on.

Results. Over half of the users (57%) were nulliparous and a quarter of insertions were for emergency contraceptive

use. A third reported that insertion was 'very painful'. Half reported that their periods had become heavier since insertion, but only 8% said that they were now unmanageable. Unacceptable bleeding was the most common rea son for removal. Some reported inter menstr ual bleeding and dysmenor r hoea. Satisfaction with the GyneF ix was high; 86% of questionnaire responders said that they would recommend it to a friend.

 

 

Key message points ·The GyneFix is an effective alternative to hormonal methods of contracepti on and i s acceptable to nulli parous women. ·Users should be at Iow risk of sexually transmitted infection and consideration should be given to screening for Chlamydia      tra chomati s prior to insertion. ·Insertion can be painful, especial l y in null i parous women. ·Periods may become heavier, although within acceptable limits for most. GyneF ix should not be offered to a woman who al ready suffers heavy periods. ·Most users are satisfied with the G y neFix. Of those that are not, the most common reason is heavy periods.

Conclusion. The GyneFix is well accepted in Liverpool in appropriately selected clients, and is an effective non-hormonal method for nulliparousas well as parous women.

 

Introduction

The GyneFix intra-uterine device (IUD) was introduced into the UK in 1997. It is a frameless copper-bearing device, designed to reduce the incidence of bleeding, pain and expulsion seen with framed devices. It is marketed as being a reliable alternative to hormonal contraception,particularly for nulliparous women in whom there may be incompati bi lity between the uteri ne cavity and the frame of a standard device, causing discomfort, bleeding and expulsion. Clinical trials have confirmed that the GyneFix is effective and acceptable to nulliparous as well as parous women,1 although to date there has not been a randomized trial involving nulliparous women comparing GyneFix to other copper IUDs.

  The Contraception and Reproductive Health Service in Liverpool was the first in the UK to offer this new device

and commence the cascade of training to other service providers. I n our service the I U D is a popular choice and is

also offered routinely as an option for emergency contraception where there are no contrai ndications, both to nulliparous and parous women. As many requests for emergency contraception are from young, nulliparous women, the availabil ity of an I U D that would be acceptable as an ongoing method to this group was particularly welcomed.

  Insertions started in Liverpool in February 1997. The GyneFix IN (interval) is the version used in our service. In order to assess the fi rst years' experience with the device, a retrospective study was carried out. T hi s was by systematic casenote review and postal questionnaire. The purpose of the study was to look at reasons for choosing the GyneFix over a framed device, events at and subsequent to insertion, and to assess c l i ent satisfaction.

 

Introduction of the GyneF ix to Liverpool

At the time of introduction, guidelines were drawn up to help clinicians decide who would benefit from using a GyneFix rather than a framed device. Based on information available?3 it was decided the device would be offered to:

·nulliparous women wishing to use an effective nonhormonal method

·parous women who had experienced a previous IUD expulsion

·parous women who had suffered pain with a framed IUD previously

·parous women who, after full counselling on all available methods, preferred to use a GyneFix rather than a framed device.

  There was some initial confusion about the effect the GyneFix might be expected to have on menses. Thus some women not wi shi ng to use a hormonal method were offered, or requested, the GyneFix instead of a framed device in the hope that it would not worsen their al ready-heavy periods.

GyneFix is only offered as emergency contraception where long-term use is intended. If removal at the next period is intended, then a less-expensive framed device is used.

  As many insertions are in nulliparous women, where dilatation of the cervix may be necessary, local anaesthetic

may be needed. An intra-cervical block techniqueis used in our service. Local anaesthetic is injected into the cervical lip where the tenaculum is to be placed, and then into the cervix at the level of the internal os through the cervical canal at 3 and 9 o'clock.

 

Chi a mydi a testing

At the time of IUD insertion, infection may be introduced into the uterus, resulting in pelvic inflammatory disease (PID). This may lead to tubal damage and subsequent subfertility. It is now known that the risk of PID is largely related to the presence or acqui siti on of sex ua ll y trans mitted infection (STI) - usually Chlamydia - and not the IUD or parity of the woman per se. This information is shared with all potential IU D users in our service, enabling each client to make an informed decision as to whether an IUD is the correct method for her. Testing for C hi a mydi a is carried out if felt necessary by clinician and/or client, for example where the relationship is new or where there are doubts about monogamy. Results are awaited, where possible, prior to insertion. If the insertion is as an emergency postcoital measure, then insertion must proceed without knowledge of the result.

 

Method

Casenotes were reviewed to determine reason for choice of GyneFix over a framed device, parity (nulliparous meaning never pregnant or no pregnancy over 24 weeks), whether the insertion was planned or an emergency measure, and probl ems reported subsequently at follow-up.

A questi onnai re designed to give a qualitative assessment of experience with the method was sent to all clients consenting to correspondenceat home. Questions related to the insertion procedure, menstrual and other changes since insertion, and to overall satisfaction. Answers were multiple-choice style, with blank spaces for comments. Questionnaires were sent out 2 months after the last of the 215 insertions had taken place. Where no reply was received within 3 weeks, a second questionnaire was sent.W here there was still no reply, a questionnai re was placed in the casenotes so that the woman could be approached opportunistically when she next attended cli nic and asked to complete the questionnaire. Women who had declined correspondence home were approached opportunistically in the same way. A short pilot was carried out amongst clinic attenders before finalising the questionnaire, and Ethical Committee approval was sought prior to distribution.

  Information was entered onto an E xcel spreadsheet. The last systematic casenote review was carried out in April 1999, although the spreadsheet continues to be updated as further information becomes available. E vents up to August 2000 are i ncI uded.

 

Results

Over the first year, from February 1997 to January 1998, a total of 215 insertions were carried out. Table 1 shows numbers of subjects according to parity, divided into emergency and planned insertions. Over half of the insertions (123; 57%) were in nulliparous women, with 92 (43%) in parous women. A quarter (56; 26%) were for emergency use. T he excess of nulli parous over parous users stems mainly from the emergency users. There was one abandoned i nserti on, due to inability to advance the uteri ne sound through the cervical canal. This client was provided with an alternative method of contraception and has not been included in the analysis.

  One hundred and eighty-nine clients were sent questionnaires (Table 2). The remaining 26, who had not consented to correspondence home, had questionnaires placed in their casenotes, along with those who had consented to correspondence home but failed to return two mailed questionnaires. Completed questionnaires were returned by nearly two-thirds (132,61%). There was no significant difference in parity between the responders and non-responders using chi-square testing, with 73 (55%) of the respondents being nulliparous and 59 (45%) parous.

  Thirty-two women (18 nulliparous, 14 parous) have neither returned a questionnaire nor attended clinic since insertion; thus follow-up information is available for 183 (85%) out of the total 215 women.

  The time interval between GyneFix insertion and completion of the questionnaire ranged from two to 20 months. The time interval between insertion and clinic follow-up ranges from a few days to 39 months.

 

Rea son for choosing GyneF ix

Table 3 shows reasons for the choice of GyneFix over a framed device. Data were extracted from casenotes for all 215 women. Most women chose the GyneFix because of nulliparity or previous IUD related pain or expulsion.

Twenty-three percent of insertions were at the client's request, with no obvious clinical reason why the GyneFix might be a better choice than a framed device (all parous women). Five women chose the device because they had dysmenorrhoea, in the hope that the GyneFix would not worsen it. Similar reasons were documented for one woman whose natural menses were heavy and five women who had had excessive menstrual bleeding with a framed I U D.

 

The insertion

Clients were asked by questionnaire how painful the insertion was. A third responded 'very painful' and 31% 'more painful than I expected but bearable' (Table 4). More nulli parous women than parous responded to these options.

  Amongst those accepting intracervical block there was no observed reduction in those reporting that insertion was very painful. Seventy-three women had a cervical block. Forty-two of these responded to the questionnaire, 15 stating that insertion had been 'very painful'. Thus 36% of questionnaire responders having a cervical block reported insertion to be' very painful'.

 

C h la mydia testing

One hundred and fifty-two women (71%) had a C hlamydia test. Three results were positive, giving a rate of 2%. All three were emergency insertions in nulliparous women under the age of 25. All were referred to the local genitourinary medicine (GUM) clinic. One had developed upper abdominal but no pelvic pain and a presumptive diagnosis of Fitz-Hugh Curtis syndrome was made. The other two were asymptomatic, and all three opted to retain the GyneFix. There were no known cases of PID amongst those in whom C hlamydia testing was not carried out.

 

Menstr ua I changes si nce i nser tion

Clients were asked, both at clinic follow-up and by questionnaire, what, if any, problems they had experienced. The results in Table 5 are amalgamated results from both casenote and questionnai re analysis for the 183 women on whom follow-up information is available (103 nulliparous and 80 parous).

  Over half reported that their periods had become heavier si nce insertion. T he remainder said that they were the same or lighter. Those reporting that their periods were now intol erabl y heavy were mostly parous women. O ne of these women had reported heavy bleeding with a framed I U D. A number of women reported inter-menstrual bleeding and the development of dysmenorrhoea or pelvic pain since insertion. Most of these had also reported heavier periods. There were no contraceptive failures, but one pregnancy resulted from unnoticed expulsion.

 

Other events

There have been 48 known removals. Figure 1 shows removal s for nulli parous and parous users. A thi rd (16) were due to bleeding problems, either heavy menses (11) or intermenstrual bleeding (5). The remainder were for a variety of reasons, including: to conceive (11 removals); pain (10 removals; two within 30 minutes of insertion due to persistent pain on embedding the anchoring knot in the fundal myometrium, three dysmenorrhoea, four non- menstrual pelvic pain, one dyspareunia); two cases of suspected PID (swabs negative in both, one subsequently found to have an ovarian cyst), and one pregnancy, although conception was prior to insertion. Eight removals were for unrelated or non-medical reasons, such as no further need for contraception. Using chi-square testing, there is no si gnificant difference i n removal rates for bi eedi ng and pain between nul Ii parous and parous users.

 There have been 19 expulsions about which we are aware. S ix of these were within 3 months of insertion. Most of the devices expelled were inserted by clinicians undergoing insertion training. One unnoticed expulsion resulted in pregnancy, diagnosed 20 months after insertion. Six clients opted for re-insertion and continued to be included in the survey.

 

Comments from users

C li ents were asked for their comments on the method. Mostof these were encouraging:

  'The best method I've ever used'

  'It doesn't cause period pains like my old coil did'

  '1 forget it's there - it's very convenient'

'T he staff were friendly and helpful'

   B ut some negative comments were also received:

  '1 was disappointed about my heavy periods'

  '1 wasn't really prepared for just how painful insertion would be'

  '1 was devastated when it fell out'

Would you recommend the GyneF ix to a friend?

To this al l-i mportant questi on, the majority (86%) answered 'Yes'. Although 13% said 'No', most of these were because of heavy periods. Slightly more nulliparous than parous

users answered 'Yes' to this question (88% Vs 80%).

 

Discussion

T he introduction of a new treatment, or in this case method of contraception, is an opportunity to look at both the method itself and clinical standards within the service. If a new service is to become established, then there needs to be some evidence that there is a requirement for that service rather than providing it just because something new is

available.

  When the GyneFix was fi rst introduced into the service it was felt necessary to draw up some guidelines about who should be offered the device, or more importantly, who might benefit from using the GyneFix rather than a framed device. Nulliparous women and those who had had expulsion of a framed device appeared most likely to benefit, based on published evidence availabl e at the ti me.

  Two randomised comparative tri al s comparing prototype versions of the GyneFix (identical in all but insertion instrument, which was redesigned as a result of problems with expulsion) to the Copper T380A did not show any advantage of the GyneFix over the framed device in parous women only.2'3

  A large non-comparative study in both nulliparous and parous women reported Iow removal rates for bleeding and pain and very Iow expulsion rates in all users.1 The authors concluded that the absence of a frame explains the Iow incidence of side effects and that the GyneFix is effective and well accepted by nulliparous women. As bleeding and pain were not separated in the analysis, it was not possible to discern whether most probl ems res ulti ng i n removal were related to bleeding or pain. Thus some initial insertions were carried out in women reporting either heavy periods currently or heavy periods with a framed device in the past, in the hope that the GyneFix would not worsen the periods to the extent that a framed device might.

  As a result of over 50% reporting that their periods became heavier with the device, it is now standard practice to advise that periods are likely to become heavier. Interestingly though, none of the five women choosing the GyneFix because they had had IUD-induced bleeding problems previously has had her device removed, and only one has reported unmanageably heavy periods. Inter-menstrual bleeding occurred in almost a fifth of users -most of these had also reported heavier periods.

   It was surprising that up to 15% reported that their periods had become more painful since insertion. Most of this was reported by questionnaire - complaints of dysmenorrhoea at follow-up have been rare. The impression that the clinicians involved with GyneFix have developed is that most women do not suffer IUD-related pain. Some of these women had discontinued the combi ned pill when the GyneFix was inserted, which may account for a worsening of dysmenorrhoea. Heavier menses may also have contributed.

   It i s possi bi e that those responding to the questionnai re or attending clinic after a longer time may have had time both for problems to settle and for new ones to emerge. Due to the disparity in lengths of time involved, it is not possible to draw any further conclusions or attempt to apply significance testing. Anecdotally, some women have reported that initial ly heavy periods tended to settle over the first few months of use.

  Since this study, the 3-year interim results of a large randomised trial currently underway in China have been published4 Women are randomised to receive either the GyneFix or the TCu380A. The removal rate for both bleeding and pain is lower for the GyneFix than the framed device users, although the figures do not reach significance. Only parous women have been recruited.

   It is as anticipated that more nulliparous women than parous reported that insertion was painful. It is surprising that the use of intra-cervical block did not lead to fewer reporting that insertion was painful. Some had local anaesthetic after the procedure had already begun, due to feeling discomfort on cervical manipulation. Also, cervical

anaesthesia will not have any impact on pain felt on penetration of the fundal myometrium. It is not possible to make any further analysis of whether pain was felt on cervical manipulation or myometrial penetration using the data collected, but it shows that local anaesthetic is not always necessary even i n nulli parous women. C Ii nicians in our service have observed that the less nervous the woman, the less likely she is to suffer pain. Pain tends to resolve quickly once the device is in place. Giving oral analgesia prior to the procedure may be of benefit.

Women considered suitable for I U D use should have an inherently Iow risk of STI. The three women whose tests were positive were all under the age of 25, consistent with the trend of Chlamydia infection seen in the general populatior~ The positivity rate of 2% is lower than the reported prevalence in under-25s in the UK (upwards of 5% nationally? as high as 13% in some parts of L i verpool6). T hose undergoing planned insertion would usually have awaited negative results before proceeding. As results take 4-7 days, emergency insertions must be carried out without Chlamydia status being known. This explains why all three positive results were in emergency i nserti ons.

  Some expulsions amongst the initial insertions have to be expected as the clinician is learning a new skill. As Liverpool has acted as a continuing training centre, many devices are inserted by clinicians relatively inexperienced in the technique of GyneFix insertion. This can account for ex pulsi ons within the fi rst 3 months after i nserti on, but later expulsions are more difficult to explain. As a result of the finding that the expulsion rate is higher than quoted in the literature, a further follow-up study of the first 1000 insertions was commenced and will be reported separately. All IUD users are advised to check for the presence of the thread regularly. Expulsion should not be assumed if the device is  not present on clinical and  ultrasound examination; perforation should be excluded with a plain whole abdominal X-ray once pregnancy is excluded. In the Chinese randomised comparative study, the expulsion rate with the GyneFix is significantly lower than with the TCu380A .⁴

As follow-up is opportunistic and therefore incomplete, there may be removals and expulsions about which we are unaware. Removal and expulsion rates have been similar regardless of parity and whether insertion was planned or emergency, although small numbers and i ncompl ete foil ow-up hamper statistical analysis.

   Asking for any other comments at the end of the questionnaire proved a valuable insight. On the whole comments were encouraging and complimentary, both to the device itself and the experience of using the service. Some comments related to dissatisfaction with the increased menstrual flow, which may reflect unrealistic ex pectati ons about bleeding. A s previously mentioned, this issue has been addressed.

Conclusion

The GyneFix intra-uterine device is well accepted in Liverpool in appropriately selected clients, and is an effective non-hormonal method for nulliparous as well as parous women. Most users are satisfied with the GyneFix and would recommend its use to a friend. I n our experience, periods do become heavier with GyneFix, but most women find this acceptable. Expulsions may occur.

   It is not possi bi e from this study to determi ne whether the GyneFix performs better in terms of side effects than a framed IUD. The final results of the Chinese study might give more information, but only in the population studied. A randomised comparative trial including nulliparous women is needed.

 

Statements on funding and competing interests

FundinG Funding was provided by the National Co-ordinating Unit for Clinical Audit in Family Planning.

Competing interests: In lieu of payment for training other clinicians in the technique of GyneF ix i nserti on, Abacus has received some devices free of charge from C ontrel, the manufacturer.

 

References

1  Van Kets H, Vrijens M, Van Trappen Y, et al. The frameless GyneFix~ intrauterine implant: a major improvement in efficacy, expulsion and tolerance. Advances in Contraception 1995; 11(2): 131 142.

2  UNDP, UNFPA and WHO Special Programme of Research, Development and Research Training in Human Reproduction, World Bank: IUD Research Group. The TCu 380A and the Frameless IUD "the FlexiGard": Interim three-year data from an international multicenter trial. Contraception 1995; 52:77 83.

3  Rosenberg M, Foldesy R, Mishell D, et al. Performance of theTCu 380A and Cu-Fix IUDs in an international randomized trial. Contraception 1996; 53:197 203.

4  Wu S, H u J and Wildemeersch D. Performance of the frameless GyneFix and the TCu 380A  IUDs in a three-year multicenter randomized comparative trial in parous women. Contraception 2000. 61:91 98.

5  Chief Medical Officer's Expert Advisory Group on Chlamydia Trachomatis. Summary and conclusions of the CMO's Expert Advisory Group on Chlamydia Trachomatis. Department of  Health, 1998.

6  Jones K, Darcy A, Webb A, et al. Screening, treating and contact tracing for asymptomatic Chlamydia infection in a young people's clinic (in Bootie) by Outreach Health Advisor is both acceptable and successful. European Journal of Contraception and Reproductive Health Care 2000; 5 (1): 53(a).