Original research article

Experience with GyneFIX® insertions in Spain: favorable acceptanceof the intrauterine contraceptive implant with some limitations

Francisca Martineza,*, Emilli Gimenezb, Guillem Herna´ndezc, Domingo Alvarezd,Mauricio Tejadae, Paz Garcı´af, Caridad Ruizf, Carlota Abrilg, Blanca Izquierdog,

Angels Avecillah, Angels Ramirezi, Mar Cadinanosj, Nuria Dominguezj, Jose´ Luis Dovalk,Ma Jesus Rodriguezl, Vicente Rodriguezm, Vicenta Turradon, Abu Omar-Fayezo,

GESEG: Group for the Study and Follow-up of GyneFIX® in SpainaInstitut Universitari Dexeus, Barcelona, SpainbCentro de Salud de Portugalete, Bilbao, Spain

cClınica Corachan, Barcelona, SpaindHospital de Valdecilla, Santander, Spain

eC. Esp. Inocencio Jimenez, Zaragoza, SpainfCentro de Planificacion Familiar Cartagena, Caratagena, Spain

gCentro de Planificacion Familiar La Seda, Murcia, SpainhHospital Municipal, Badalona, Spain

iMunicipal Health Care Institute, Barcelona, SpainjPrograma D’Atencio a la Dona Guell, Gerona, Spain

kCentro de Orientacion Familiar, Orense, SpainlDiatros, Barcelona, Spain

mCentro de Salud Vicalvaro, Madrid, SpainnCAP II Premia de Mar, Barcelona, Spain

oConsulta, Cadiz, Spain

Received 9 May 2002; received in revised form 24 July 2002; accepted 25 July 2002

Abstract

This study evaluates the incidents associated with GyneFIX® insertion and first-year expulsion and continuity rates within the usualintrauterine contraceptive practice of a working group of Spanish professionals (GESEG), formed specifically with this aim. It is aprospective, multicenter, observational study of GyneFIX® insertion in1684 women. Data were prospectively collected on a structured formand processed centrally. Interest was focused on difficulties encountered during the insertion procedure and symptoms experienced duringinsertion. All terms were defined by consensus. Among the total, 18.6% of the women were nulliparous. GyneFIX® insertion was rated easyin 92%, with more difficulty in nulliparous women, who showed significantly more symptoms during insertion of the device. First-yearexpulsion and continuity rates were 5.6 and 88 per 100 women, respectively. The pregnancy rate was 0.3 per 100 women/years. TheGyneFIX® system is an interesting alternative to standard IUDs for intrauterine contraception with copper, particularly in women who haveexperienced expulsion of other types of IUDs. Experienced professionals in IUD insertion quickly acquire familiarity with the GyneFIX®insertion system, but proper implantation does not completely eliminate the risk of expulsion. Thus, the insertion system should be furthermodified to achieve a simpler, safer technique. © 2002 Elsevier Science Inc. All rights reserved.

Keywords: Intrauterine contraception; Intrauterine contraceptive implant; Difficulties in insertion; Nulliparous; Expulsion

1. Introduction

Intrauterine devices (IUDs) with a high copper load arean excellent contraceptive method, providing effective pro-tection against pregnancy. Nevertheless, associated side ef-fects, mainly pain, bleeding, and expulsions, have nega-

* Corresponding author. Tel.:�34-93-227-47-83; fax:�34-93-205-79-66.

E-mail address: Pacmar@iudexeus.uab.es (F. Martinez).

Contraception 66 (2002) 315–320

0010-7824/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved.PII: S0010-7824(02)00392-X

tively affected rates of continuation and acceptance. Theseadverse effects are generally attributed to a disproportionbetween the dimensions of the IUD and that of the intra-uterine cavity. To improve IUD acceptance rates and main-tain elevated efficacy, GyneFIX®, a novel intrauterine im-plant system, was designed and presented as an alternativeto conventional IUDs with rigid frames [1].

GyneFIX® was commercially introduced in Spain at theend of 1997, and it has awakened considerable curiosity andsome skepticism regarding the insertion technique and theadequacy of the intramyometrial anchoring technique. Op-timum results in intrauterine contraception require properinsertion, whatever the model of IUD. The GyneFIX® in-serter was presented as the solution for IUD expulsions [2],but in two important multicenter studies, expulsion rateshave been higher than expected [3,4]. The expulsions oc-curred mainly in the first months after inserting the deviceand early in the studies, leading to the idea that a lack ofexperience in the implant technique may have been thecause. Due to its complete flexibility, GyneFIX® was alsoexpected to be significantly better tolerated and show alower incidence of removals because of bleeding [5].

Because this system seemed to be promising and severalstudies had reported conflicting results, we felt the need toevaluate our own experience with this new intrauterineimplant. For this purpose, a group of 19 Spanish researchersfrom several centers with extensive experience in intrauter-ine contraception formed the Spanish Group for the Studyand Follow-up of GyneFIX® (GESEG) in March 1998.

The aim of the present study was to evaluate the inci-dents associated with the insertion and use of the Gyne-FIX® system, and to determine first-year rates of expulsionand continuation under the usual conditions of intrauterinecontraceptive practice.

2. Materials and methods

This is a prospective, multicenter, observational study ofGyneFIX® insertion in1684 women. Data were prospec-tively collected on a structured form and processed cen-trally. Clarifications were requested from the participatingresearchers when necessary. Meetings were held with allmembers of the group to agree on the protocol beforeinitiation of the study, and to review the preliminary resultsat 12 and 18 months.

2.1. Description of the device

The GyneFIX® IUD consists in a non-biodegradable00-monofilament polypropylene suture thread, over whichsix 5-mm copper sleeves are threaded, giving a total surfaceof 330 mm2 (Fig. 1). A detailed description of the device [6]and the insertion procedure [5,7] have been published pre-viously.

2.2. Procedure for inclusion in the study and follow-up

Specific exclusion/inclusion criteria were not defined,and were left up to the individual researchers. All thewomen included in the study had requested implantation ofan IUD for contraceptive purposes and none presented con-traindications, according to the WHO eligibility criteria [8].In all the women, insertion of the GyneFIX® was carriedout by one of the GESEG researchers. During the studyperiod, GyneFIX® was the principal device used in some ofthe centers, whereas in others, different models of copperIUDs were also used.

All the women were told to return to their respectivecenters 1 to 2 months after insertion of the device andannually thereafter, or whenever there was some reason forconcern. It was left up to each researcher to decide whetheran ultrasound study would be performed at the follow-upvisit.

The following data were analyzed.

Fig. 1. The GyneFIX insertion apparatus used in this study. A, uterinesound; B, 3.8-mm wide tube provided with centimeter marks and flange; C,the plunger (1, stainless steel stylet; 2, plastic rod; 3, middle segment; 4,handle); D, the GyneFIX system.

316 F. Martinez et al. / Contraception 66 (2002) 315–320

2.2.1. Women’s characteristics at the time of inclusion inthe study

The characteristics included age, parity status, time sincelast parturition/miscarriage, amount of menstrual bleedingbefore IUD placement, dysmenorrhea, previous use of anIUD and reason for removal, hysterometry, and uterineposition.

2.2.2. Incidents during insertionSymptoms: taking into account the woman’s declarations

and the researcher’s evaluation, were classified as: absent,pain alone, pain with hypotension, and vasovagal syncope.

Difficulty in insertion was rated as: “easy” when noadditional maneuvers were required; “difficult ”when therewas some problem passing the inserter tube through theinternal cervical orifice, and dilation or local anesthesia wasrequired; and “ failed insertion.” Reasons for failed insertionwere recorded: cervical orifice was not cannulated by inser-tion tube or the device was not anchored in the uterinefundus.

All GyneFIX® devices inserted from the beginning ofthe study up to 31 March 2001 were included in the epide-miologic analysis and the study of events during insertion.

2.2.3. Subsequent controlsThe following items were evaluated: subjectively per-

ceived modification of the bleeding pattern, presence ofother symptoms, gynecologic examination, and ultrasoundexamination, when performed.

All women with a GyneFIX® implant up to 31 Decem-ber 2000, and who had at least one follow-up visit wereincluded in the analysis.

2.2.4. Discontinuation of useThe reasons for discontinued use of the device and the

total time of use were recorded. Reasons for discontinuationincluded the following: bleeding, pain, expulsion, acciden-tal pregnancy, pelvic inflammatory disease, uterine perfo-ration, other medical indications, pregnancy wish, or otherpersonal reasons (wish to change contraceptive method,menopause, no need for contraception, etc.)

2.3. Statistical analysis

For the descriptive analysis we used measures of centraltendency and dispersion. The chi-square test and relativerisk with 95% confidence intervals were used to comparequalitative variables. Quantitative variables were comparedby ANOVA, Student’s t-test or the Fisher test, where ap-propriate. The analysis of continuity was performed byapplying the life tables method of Farley [9], in which thecumulative rates of events per period elapsed since theinitiation of use of the method are established. Standarderror for the proportions was calculated. Data were pro-cessed and analyzed using the SPSS/PC®� 10.0 statisticspackage for Windows® (SPSS Inc, Chicago, IL, USA).

Corrections for the different subpopulations were taken intoaccount.

3. Results

In the present study, 1684 GyneFIX® insertions wereperformed by GESEG members, 314 of which (18.6%)were in nulliparous women. The women ranged in age from19 to 55 years, and the range of parity was 1 to 10 children.The mean follow-up time was 9.97 � 10.56 months (range0–77 months). A follow-up period of at least 12 months wascompleted by 1097 women, and a period of at least 24months by 426 women. Among the total, 58% of the women(976/1684) had used an IUD previously. In 21 women(1.2%), the GyneFIX® was placed for post-coital intercep-tion and all of them continued with the implant as a con-traceptive method (Table 1).

The most frequent reason for removal of the previousdevice was IUD expiry (58.1%). Among the women whoseprevious IUD had been removed due to complications (283/976, 29%), the most frequent complications were bleedingor pain. Expulsion of the IUD had occurred in 103 women(10.6%).

Overall, insertion was easy in 92.9% of cases, and wasdifficult and required dilatation or anesthesia in only 5.6%of cases. The women noted intense pain in 10.5% of inser-tions, which was accompanied by hypotension in 3.3%.Hypotension alone or vagal syncope was observed in 3.3%.Insertion failure, due to deficient fixation of the implantoccurred in 1.4% of cases, and there were three uterineperforations with the insertion tube (0.2%) (Table 2).

There were more difficulties in GyneFIX® insertionamong the nulliparous than the parous group, althoughdifferences were not statistically significant. Nevertheless,significantly more nulliparous than parous women experi-enced pain, hypotension, or syncope during the insertionprocedure (p � 0.001). Insertion was significantly lesslikely to be easy (relative risk 0.9 [95% confidence interval0.9–1.0], p � 0.001) and more likely to be difficult or failedamong nulliparous women. Nulliparous women were alsosignificantly more likely to have symptomatology (relativerisk 0.7 [95% confidence interval 0.7–0.8], p � 0.001).

Table 1Characteristics of GyneFIX users

No. of women N � 1684

Age in years (mean � SD; range) 35 � 7 (19–55)Parity (mean � SD; range) 1.3 � 0.9 (1–10)Hysterometry in cm (mean � SD; range) 7 � 0.8 (6–12)Indication:

Interval contraception 1608 (95.6%)Postpartum contraception �3 months 55 (3.3%)Postcoital interception 21 (1.2%)

No. of users of a previous IUD 976 (58%)

317F. Martinez et al. / Contraception 66 (2002) 315–320

Post-insertion check-ups were performed in 1453women, 1198 with ultrasound (82.5%). During the periodanalyzed, there were 66 expulsions and three pregnancies(one with the GyneFIX® in situ and the other two afterunnoticed expulsion).

Including the 66 expulsions, there were 263 discontinu-ations. The most frequent complications leading to discon-tinuation were expulsion and bleeding. There were no sig-nificant differences in overall rates of discontinuationbetween nulliparous (59/314, 18.8%) and parous (204/1370,14.9%) women. Nor were there differences in rates of re-moval between women who had used an IUD previously(90/976, 9.2%) and those who used it for the first time(76/708, 10.7%). Additionally, 55.1% (145/263) of womenwho had used an IUD previously discontinued with Gyne-FIX®, as compared to 44.9% of women who used it for thefirst time.

Table 3 details the incidence of expulsions, accidentalpregnancies, and removals for bleeding or pain per 100women/years, as well as the continuation rates at 12 and 24months of GyneFIX® use. Other, later incidents do notappear in the table. There were no differences in age be-tween the women who experienced expulsion of the deviceand those who did not (35 � 7 vs. 35 � 7).

4. Discussion

GyneFIX® is a novel system for intrauterine contracep-tion introduced not long ago on the Spanish market. Thesystem requires a special insertion technique, and severalstudies carried out by teams with different levels of expe-rience have reported dissimilar results with its use. Theadvantages claimed by the inventors of the device may becounterbalanced by the difficulties involved in its properplacement. To clarify these doubts, it was important to studythe results of this new implant in our setting. The mainobjective of IUD use is to provide effective protectionagainst pregnancy. This protection is interrupted when itsuse is discontinued because of expulsion of the device(particularly when it goes unnoticed) or its removal due tobleeding or pain. When a specific model of IUD is beingevaluated, it is essential to take into account these events inaddition to the effectiveness of the device as a contraceptive[9].

It should be mentioned that the various members of thestudy group did not participate with an equal number ofGyneFIX® insertions, and that one of the researchers car-ried out more than half of those included in the series. Allthe study group members were experienced in placing IUDsand all were familiar with the method. Moreover, beforeinitiating the study, a workshop on the specific GyneFIX®insertion technique was organized. Thus, we believe that thedifficulties encountered with placement of the device areattributable to the design of the implant rather than a lack ofexperience of the person performing the technique. Overall,insertion was easy in 92.9% of cases, with a somewhathigher, though non-significant, percentage of difficulties innulliparous than parous women. Nulliparous women re-ported significantly more discomfort than those who hadalready had a child. This is the first study that prospectivelyanalyzes the influence of parity on the difficulties of Gyne-FIX® insertion and symptoms during the procedure. Be-cause this is not a comparative study with other models ofIUD, we cannot establish whether insertion of a conven-tional model would have involved the same difference indegree of difficulty encountered in the nulliparous group. In

Table 2Incidents during GyneFIX insertion in 1684 women according to parity status

Difficulty in insertion OverallN � 1684

ParousN � 1370

NulliparousN � 314

Relative risk(95% confidence interval)

Easy 1564 (92.9%) 1287 (93.9%) 277 (88.2%) 0.9 (0.9–1.0)*Difficult 94 (5.6%) 67 (4.9%) 27 (8.6%) 1.8 (1.1–2.7)*Failed 23 (1.4%) 13 (1.0%) 10 (3.2%) 3.4 (1.5–7.6)*Perforation 3 (0.2%) 3 (0.2%) 0 (0%) —Symptoms Overall Parous NulliparousAbsent 1397 (83.0%) 1194 (87.2%) 203 (64.6%) 0.7 (0.7–0.8)*Pain 177 (10.5%) 121 (8.8%) 56 (17.8%) 2.0 (1.5–2.7)*Pain � hypotension/syncope 55 (3.3%) 26 (1.9%) 29 (9.2%) 3.7 (2.2–6.2)*Hypotension/syncope 56 (3.3%) 29 (2.1%) 26 (8.3%) 3.9 (2.3–6.6)*

* p � 0.001.

Table 3Life table net cumulative discontinuation rates (and standard error) byincidence per 100 women at 12 and 24 months and rate of continuationwith the method

Incident Period of use

0–12 Months 12–24 Months

Discontinuation due to: N Rate (SE) N Rate (SE)Pregnancy 3 0.3 (0.2) 0 0.0 (0.2)Expulsion 59 5.6 (0.7) 5 1.3 (0.9)Bleeding 24 2.3 (0.5) 26 6.5 (1.3)Pain 7 0.7 (0.3) 8 2.1 (0.8)Perforation 3 0.3 (0.2) 0 0.0 (0.2)Other medical reasons 19 1.8 (0.4) 8 2.1 (0.8)Other personal reasons 17 1.6 (0.4) 35 8.7 (1.4)

Total incidents 132 81Exposed women 1097 426Continuation rate 88.0 (1.0) 81.0 (1.9)

318 F. Martinez et al. / Contraception 66 (2002) 315–320

the present study, local anesthesia was not routinely offered;it was only used in cases where there were difficultiesduring the insertion.

Parity status did not have an influence in the overall rateof removals/discontinuations of GyneFIX® during the pe-riod analyzed, in keeping with the results of other authors[5,10,11].

Failed insertion occurred in 23 cases from our series, inthe majority because of deficient fixation of the implant(0.2% of insertions). Because the participants had a greatdeal of experience in IUD placement, we attribute the fix-ation failures to the design of the GyneFIX® insertionmechanism. One of the consequences of improper fixation isexpulsion of the device. In fact, most authors attribute earlyexpulsions in the first few months after placement to thismechanism [4,12]. Our results contrast with those of others[4] who reported that the greatest problem resided in pass-ing the insertion tube through the cervical canal. Perhaps adifferent type of insertion tube was used in these earlierworks.

The following have been cited as user-related risk factorsfor expulsion of an IUD: young age, amount of abnormaluterine bleeding, and dysmenorrhea before the insertion[13]. In the present study, there were no statistically signif-icant differences in age or history of dysmenorrhea betweenGyneFIX® users presenting expulsion and the remainingusers.

We found a first-year expulsion rate of 5.6, a figuresimilar to that seen in several works [4], lower than thatdescribed by Rosenberg [3], and higher than that of others[5,7,14]. Because the majority of expulsions occurred in thefirst year, the cumulative rate of expulsions at the end of thesecond year was 1.3%. Some expulsions were detectedmonths after the sonographic control had demonstrated cor-rect implantation of the device.

As was mentioned above, we suggest that some expul-sions initially attributed to a lack of familiarity with theimplantation technique on the part of the researchers, [3,4]may actually have been the result of difficulties derivedfrom the insertion tube design. This component had beenmodified previously. With the new model, a first-year ex-pulsion rate of 2.67 has been reported by another author[14], and a rate of 5.6 in our series, with the same insertionapparatus. Our experience indicates that the GyneFIX®insertion system still needs to be improved.

Some expulsions are produced even when correct im-plantation of the GyneFIX® is confirmed, although theymay be fewer than with other (rigid-frame) IUDs [14].Other authors have reported late expulsions, though themechanism leading to this event is unknown [12].

The second cause of discontinuation in this study wasbleeding, with a first-year rate of 2.9%, somewhat lowerthan that of copper IUDs having a rigid frame [14,15]. Atthe beginning, Wildemersch indicated that bleeding woulddecrease with the GyneFIX® [5], but this has not beencorroborated by our results or those of others [16]. The

exact mechanisms that produce increased bleeding in cop-per IUD users is not known, but it may be related to thepresence of a rigid structure within the cavity and thepresence of the copper itself.

Our study prospectively analyzes the incidents that oc-curred during insertion of this intrauterine implant system indaily contraceptive practice in 1684 women, and the inci-dents during the first-year of use in 1097 women, bothsubstantial numbers for analysis. On the basis of the results,we conclude that the GyneFIX® is a worthwhile alternativefor women who want intrauterine contraception with copperand who do not have contraindications to use this device,and particularly for those who have experienced expulsionwith other IUD models. Professionals who regularly prac-tice IUD placement rapidly acquire a familiarity with theGyneFIX® insertion system, which, nevertheless, should befurther developed to simplify the technique and make itsafer. The current GyneFIX® inserter design used in thisstudy makes implantation of the devices difficult and causessignificantly more discomfort in nulliparous women than inparous women, and this should be kept in mind when thewoman is being assessed and prepared for the insertion.Correct implantation of the GyneFIX® does not completelyeliminate the risk of expulsion and the professionals in-volved should be aware of this possibility. Periodic controlof the implant site is recommended.

Acknowledgments

The authors thank Ricart Bernat for valuable data pro-cessing support, Juan Carles Suris for his enthusiasm in thestatistical treatment of results, and Dr. A. Webb for herexcellent comments. This study was made possible by par-tial financial support from Italfarmaco, which has alwaysmaintained a position of respect and neutrality toward theconclusions of GESEG. This study was conducted under theauspices of the “Catedra de Investigacion en Obstetricia yGinecologia del Institut Universitari Dexeus.” Finally, allthe researchers involved in this work wish to dedicate it tothe memory of Albert Bielsa.

References

[1] Wildemeersch D, Van der Pas H, Thiery M, Van Kets H, Parewijck,Delbarge W. The Copper-Fix (Cu-Fix): a new concept in IUD tech-nology. Adv Contracept 1988;4:197–205.

[2] Van Kets H , Wildemeersch D, Van der Pas H, et al. IUD expulsionsolved with implant technology. Contraception 1995;51:87–92.

[3] Rosenberg MJ, Foldesy P, Mishell DR, Speroff L, Waugh MS,Burkman R. Performance of the TCu380A and Cu-Fix IUDs in aninternational randomized trial. Contraception 1996;53:197–203.

[4] UNDP, UNFPA, and WHO Special Program of Research, Develop-ment, and Research Training in Human Reproduction, World Bank:IUD Research Group. The TCu380A IUD and the frameless IUD“The FlexiGard,” Interim three-year data from an international mul-ticenter trial. Contraception 1995;52:77–83.

319F. Martinez et al. / Contraception 66 (2002) 315–320

[5] Wildemeersch D, Van Kets H, Van der Pas H, et al. IUD tolerance innulligravid and parous women: optimal acceptance with the framelessCuFix implant system. Br J Fam Plann 1994;20:2–5.

[6] Wildemeersch D, Batar I, Webb A, et al. GyneFIX®. The framelessintrauterine contraceptive implant—an update. Br J Fam Plann 1999;24:149–59.

[7] Van Kets H, Vrijens M, Van Trappen Y, et al. GyneFIX intrauterinecopper releasing implant system: a major improvement in efficacy,expulsion and tolerance. Adv Contracept 1995;11:131–42.

[8] WHO: Improving Access to Quality Care in Family Planning. Med-ical Eligibility Criteria for Contraceptive Use, 2nd ed.WHO/RHR/00.02. Reproductive Health and Research. Geneva: WHO 2000.

[9] Farley TMM. Life-table methods for contraceptive research. Stat Med1986;5:475–89.

[10] International Study Group on Intrauterine Drug Delivery. The Gyne-FIX implant systems for interval, postabortal, and postpartum con-

traception: a significant advance in long-term reversible contracep-tion. Europ J Contracep Reprod H C 1997;2:1–13.

[11] Duenas JL, Albert A, Carrasco F. Intrauterine contraception in nul-ligravid vs. parous women. Contraception 1996;53:23–4.

[12] Dennis J, Webb A, Kishen M. Expulsions following 1000 GyneFIXinsertions. J Fam Plann Reprod Health 2001;27:135–8.

[13] Zhang J, Chi I, Feldlum PJ, Farr MG. Risk factors for copper T IUDexpulsion: an epidemiologic analysis. Contraception 1992;42:427–33.

[14] Wu S, Hu, J, Wildemeersch D. Performance of the frameless Gyne-FIX, and the TCu380A IUDs in a 3-year multicenter, randomized,comparative trial in parous women. Contraception 2000;61:91–8.

[15] Zhang J. Factors associated with copper T IUD removal for bleeding/pain: a multivariate analysis. Contraception 1993;48:13–21.

[16] Dennis J, Webb A, Kishen M. Introduction of the GyneFIX intra-uterine device into the UK: client satisfaction survey and case notesreview. J Fam Plann Reprod Health 2001;27:139–44.

320 F. Martinez et al. / Contraception 66 (2002) 315–320