The TCu 380Ag (Outokumpu Oy, Pori, Finland) and the Multiload Cu375 (Multilan, Organon, Oss, The

Netherlands) were evaluated in 1477 women in a multicenter clinical trial. The intrauterine devices showed similar, low-event rates. Cumulative life-table pregnancy rates were less than 1.0, and continuation rates were approximately 90 per 100 women at 1 year after insertion. The risk of subsequent hospitalization or pelvic infection was low. PIP: This study evaluated the effectiveness, safety, and acceptability of the TCu 380Ag and the Multiload Cu375 IUDs in a multicenter international trial involving 1499 acceptors. The 2 devices were randomly assigned to women who entered the trial at 5 centers in Yugoslavis, Panama, Costa Rica, and Egypt between September 1980 and June 1982. The median age of the 737 women in the TCu 380Ag group was 26.7 years, with an average parity of 1.8; the median age of the 740 women in the Multiload Cu375 group was 27.5 years, with an average parity of 1.9. At insertion, 6.8% of women in the TCu 380Ag group and 6.2% of those in the Multiload Cu375 group complained of mild pain, 1.1% of women in both groups reported moderate pain, and there were no reports of severe pain. 1-year bleeding and pain removal rates were 3.8; removal rates for other medical reasons were less than 1.0/100 women. Significantly more women in the TCu 380Ag group reported intermenstrual pelvic pain, but this difference was confined to women under 30 years of age. The continuation rate at 1 year was 90.9% for the TCu 380Ag group and 88.7% for the Multiload Cu375 group. These rates are considerably higher than those reported for other IUDs. Of the 7 pregnancies reported in the 1st year after insertion, 2 occurred in the Tcu 380Ag group and 5 occurred in the Multiload Cu375 group. The results from this trial suggest that both the TCu 380Ag and Multiload Cu375 IUDs provide a highly effective, safe, and acceptable method of contraception. For women who wish to achieve contraceptive effectiveness for long periods, the TCu 380Ag may be the IUD of choice. It remains effective for 10-15 years, while the Multiload Cu375 has a life span of only 3-5 years. Shadow:
Objective. An intrauterine device (IUD) is a highly effective and prevalent form of birth control with a low failure rate. It is not without complications, including irregular bleeding and pelvic pain, perforation, expulsion, and pregnancy. Investigation of the symptomatic patient and routine follow-up of asymptomatic women with IUDs include transvaginal ultrasonography to rule out IUD malposition and other complications. Three-dimensional ultrasonography (3DUS) has been shown to have added value in imaging the pelvic organs, including the uterus. However, even with 3DUS, the operator may encounter difficulty in imaging the IUD in its entirety, because of variability in both echogenicity and exact location. Methods. More than 30 women undergoing IUD placement evaluation, including routine examinations and cases of suspected pregnancy, irregular bleeding, infection, and pelvic pain, were evaluated with 3DUS and volume contrast imaging in the C plane (VCI-C), which visualized acoustic shadows of the IUDs, facilitating identification of the intrauterine position and type of the device. Results. Four examples are presented from women undergoing evaluation of IUD placement in whom 2-dimensional ultrasonography failed to visualize the device satisfactorily. Volume contrast imaging in the C plane allowed visualization of the acoustic shadows of the IUDs, thereby providing more exact imaging of the type of device as well as aiding in locating the IUD. In a woman with irregular bleeding, a bicornuate

uterus with pregnancy in 1 horn and a bowed T-type IUD in the contralateral horn were diagnosed with VCI-C. Conclusions. We conclude from these shadow images that VCI-C is a useful modality in cases of difficult visualization of the IUD; VCI-C visualization of the acoustic shadow of the IUD may provide an adjunctive approach. imaging: Objective. Intrauterine contraceptive devices (IUDs) are reemerging as common methods of birth control in the United States. Imaging, especially sonography, has an important role in their evaluation. This review illustrates the normal and abnormal imaging appearances of IUDs. Methods. We describe and illustrate the appearance of different types of IUDs on different imaging modalities as well as radiologically relevant complications associated with IUDs. Results. On sonography, the IUD should be visualized as centrally located within the endometrial cavity, with the crossbar (if present) in the fundal portion of the endometrial cavity. Some older patients have IUDs in place that are no longer commonly used, such as the Lippes Loop (Ortho Pharmaceutical, Raritan, NJ) and Saf-T-Coil (Julius Schmid Laboratories, Little Falls, NJ), which have a pathognomonic appearance. Newer IUDs, such as the early version of the Mirena IUD (Leiras Oy, Turku, Finland), may be difficult to visualize on sonography. Patients from China frequently have a ring-shaped IUD. Sonography is important in assessing the complications of IUDs, including a low position, associated infection, myometrial migration, uterine perforation, intrauterine or extrauterine pregnancy associated with the IUD, and retention and fragmentation of the IUD. If an IUD is known to be present but not visualized sonographically, plain radiography is helpful in assessing the location. Computed tomography and magnetic resonance imaging are not typically used to assess IUDs, but the appearances of IUDs should be recognized with these modalities. Conclusions. Imaging, specifically sonography, has a crucial role in the evaluation and management of IUDs and associated complications.

The value of transvaginal ultrasound to monitor the position of an intrauterine device after insertion. A technology assessment study
BACKGROUND: The intrauterine device (IUD) is an effective contraceptive method. The contraceptive power as well as the sideeffects of IUD are thought to relate to the position of the IUD in the uterine cavity. We assessed the accuracy of clinical evaluation of IUD position. METHODS: A prospective comparative study was performed. The clinical evaluation was compared with the TVU measurement of IUD position both immediately after insertion and 6 weeks after insertion. The primary outcome measures were the positive and negative predictive values (PPV and NPV) of the clinical evaluation of IUD position. RESULTS: 195 women were included consecutively, 181 women (92.8%) were available for followup. The PPV and NPV of clinical evaluation of IUD position immediately after insertion were respectively 0.60 (95% CI: 0.390.81) and 0.98 (95% CI: 0.961.0). The prevalence of an abnormally positioned IUD was 7.7% (95% CI: 3.911.4). The PPV and NPV of the clinical evaluation at followup were respectively 0.54 (95% CI: 0.260.81) and 1.0 (95% CI: 0.981.0). The prevalence of abnormal

position was 4.0% (95% CI: 1.77.1). CONCLUSION: Clinical evaluation is an excellent test for the evaluation of the position of an IUD and routine TVU is not indicated for this purpose.
Key words

Key words: contraception/followup/gynaecological examination/intrauterine device/transvaginal ultrasound

Previous SectionNext Section

Introduction
The intrauterine position of an intrauterine contraceptive device (IUD) is thought to be closely related to its contraceptive power (Bernaschek et al., 1981; Tadesse et al., 1985; Anteby et al., 1993). IUD located more cervically may prevent conception to a lesser degree compared to adequately localized IUD (Anteby et al., 1993). Therefore proper insertion of an IUD is of utmost importance. Moreover, followup after the first menses is recommended since the first month is suggested to be the period with the highest risk of downwards migration and spontaneous expulsion (Anteby et al., 1993). The gynaecological interview and pelvic examination, in combination with selfexamination, can be used to ensure adequate localization of the IUD (American College of Obstetricians and Gynecologists, 1992). For the above reasons, insertion and followup of IUD has been routine in general practice (HarrisonWoolrych et al., 2002). Currently, transvaginal ultrasound (TVU) is used increasingly to complete the gynaecological examination in hospital settings (Valentin, 1999). The routine use of TVU to monitor the position of the IUD after insertion has been advocated (BonillaMusoles et al., 1996). There is much evidence that shows TVU to be highly accurate in monitoring the location of any type of IUD (Petta et al., 1996; Fandes et al., 1997, 2000; Palo, 1997; Aleem et al., 1992). The IUD endometrium distance (IUDED) seems to be the most relevant measurement, especially for copper IUD (Petta et al., 1996; Fandes et al., 1997, 2000). However, the maximum IUDED to ensure adequate contraception is under debate (Petta et al., 1996; Fandes et al., 1997), especially since Tshaped IUD tend to accommodate in their position during the first 3 months after insertion (Fandes et al., 2000). Therefore removal of all abnormally located IUD at TVU may result in a high number of unnecessary removals (Petta et al., 1996; Fandes et al., 2000). The silent incorporation of routine TVU in gynaecology and especially in IUD followup needs urgent evaluation (Valentin, 1999; Rivera and Best, 2002). Before adopting the routine use of TVU for this purpose, and consequently transfer the insertion of IUD from general practice to the gynaecological outpatient department, the clinical value of this phenomenon should be evaluated and guidelines adjusted accordingly. The aim of this prospective study was to evaluate the clinical relevance of the routine use of TVU immediately after and 6 weeks after the insertion of an IUD.
Previous SectionNext Section

Materials and methods

All women who attended the outpatient Department of Gynaecology of the Leiden University Medical Center for the insertion of an IUD between July 1, 2001 and January 1, 2003 were eligible for this study, regardless of type of IUD. After informed consent, women were included and the IUD inserted according to our protocol. Pain medication was not prescribed routinely while insertion was preferably performed during or just after menses. The clinician who inserted the IUD had to fill out a standardized form with questions about the procedure: (i) whether the IUD was thought to be located properly (adequate position) or erroneously (indadequate position) inside the uterine cavity; and (ii) whether the procedure was with or without complications and/or difficulties. Clinicians were divided into groups according to their experience (respectively <25, 2575 and >75 inserted IUD). Immediately after the insertion of an IUD, an independent sonographer who was unaware of the clinicians answers to the abovementioned questions performed TVU. A 57.5 MHz multifrequency vaginal probe and a Power Vision 6000 ultrasound machine (Toshiba Medical Systems, The Netherlands) were used. With TVU, the distance between the top of the vertical arm of the IUD and the junction between the endometrium and the uterine cavity (IUDED) was measured in the midlongitudinal plane (Figure 1). Whenever this junction could not be identified clearly, the IUDED was calculated by subtracting half of the double endometrial thickness from the distance from the top of the IUD until the endomyometrial junction (Figure 2) (Petta et al., 1996; Fandes et al., 1997; 2000). The mean IUDED was calculated from three independent measurements. In 20 nonselected consecutive women, a reliability analysis was performed by means of calculation of the intraclass correlation of two independent IUDED measurements (Khan and Chien, 2001; Sackett and Haynes, 2002).

View larger version:

In this page In a new window Download as PowerPoint Slide

Figure 1. A systematic representation and transvaginal scan displaying the distance between the top of the intrauterine device (IUD) and the end of the uterine cavity: the IUDendometrium distance (IUDED).

View larger version:

In this page In a new window Download as PowerPoint Slide

Figure 2. A systematic representation and transvaginal scan displaying the intrauterine device endometrium distance (IUDED) calculated by subtracting half of the double endometrial thickness from the distance between the IUD and the endomyometrial junction. According to our protocol, all women had a followup visit 6 weeks after insertion. At this visit, a standardized interview was taken concerning complaints of the IUD with emphasis on abdominal pain and cramping, bleeding disorders and complaints possibly due to the threads of the IUD. A gynaecological examination was performed to monitor the position of the IUD by estimating the length of the threads. By this means, the position of the IUD was evaluated clinically. Thereafter TVU was performed and the IUDED measured as described above. The sonographer again was unaware of the results of the interview and the gynaecological examination. Whenever the IUDED was >5.0 mm, removal of the IUD was advised. All further followup was independent of participation in the trial. The predictive values and likelihood ratios with 95% confidence intervals (95% CI) of the clinical evaluation of IUD localization were calculated using the IUDED as the gold standard. A multiregression analysis was used to calculate the multivariate odds ratios (OR with 95% CI) of risk factors for inadequate insertion and inadequate IUD position at followup. Multivariate OR (with 95% CI) were also calculated for risk factors for improper clinical evaluation of the position of the IUD both immediately after insertion and at followup. Students ttest (continuous data), 2test and Fishers exact test (categorical data) were used to compare groups; the level of significance was set at P < 0.05.

Previous SectionNext Section

Results
A total of 195 consecutive women were eligible and included in the study. The mean age was 33.2 years (SD 7.3; range 1749 years). Seventynine women were nulliparous (40.5%) while the others had delivered at least once. Most women (73.8%) had never had an IUD before. The levonorgestrelreleasing intrauterine system (Mirena) was inserted 114 times (58.5%), all other women had a copper IUD inserted: Multiload 63 (32.3%) and NovaT and 18 (9.2%) times. Women who had a NovaT were significantly younger (28.2 versus 33.8 years, P < 0.01) and more often nulliparous (89 versus 30%, P < 0.01) compared with the other women.
Validity analysis

After insertion, the clinician was certain about adequate position of the IUD in 175 cases (90%). According to the IUDED measurements, 172 of these IUD were positioned adequately [negative predictive value (NPV) of clinical evaluation: 0.98; 95% CI: 0.961.00]. According to the clinician, the IUD was positioned inadequately in 20 cases (10%), in 12 of these the IUDED was indeed >5.0 mm [positive predictive value (PPV) of clinical evaluation: 0.6; 95% CI: 0.39 0.81]. The 22 table and likelihood ratios (with 95% CI) are shown in Table I. In a total of 15 women, the IUDED was >5.0 mm. In eight of these women, the IUD was removed because of this erroneous location.
View this table:

In this window In a new window

Table I.

The results of the clinical evaluation of the position of the intrauterine device (IUD) immediately after insertion compared with transvaginal ultrasound (TVU) measurement of the IUD endometrial distance (>5.0 mm = inadequate position) The data from 181 women (92.8%) were available for followup 6 weeks after insertion. Fourteen women were lost to followup: in seven cases (3.6%) the sonographer was not available at the scheduled followup and seven women (3.6%) did not attend at their scheduled followup appointment. In all 14 women who were lost to followup, the adequacy of the position of the IUD had been ascertained by TVU immediately after insertion. The characteristics (age, parity and type of IUD) of these women did not differ significantly from the rest of the women. The IUD had been removed immediately after insertion in eight women (4.1%), who were excluded from further analysis. Of the remaining 173 women, the clinician was certain about the adequate position of the IUD in 160 (91.9%) women. In all these women, the IUDED was 5.0 mm (NPV of the clinical evaluation at followup: 1.0). In 13 women (8.1%) the clinician concluded, according to the

results of the gynaecological interview and pelvic examination, that the IUD was located inadequately. In six of these, the IUDED was indeed located >5.0 mm (PPV of the clinical evaluation at followup: 0.54; 95% CI: 0.260.81). In seven women, the IUD had been inserted inadequately; in two of these women, the IUD was still located inadequately at followup. In the remaining five women, the IUD had migrated upwards to an adequate intrauterine position. The 22 tables and likelihood ratios of the clinical evaluation of IUD position are shown in Table II.
View this table:

In this window In a new window

Table II.

The results of the clinical evaluation of the position of the intrauterine device (IUD) 6 weeks after insertion compared with transvaginal ultrasound (TVU) measurement of the IUD endometrium distance (>5.0 mm = inadequate position)
Multiregression analysis: prediction of IUD position

The two significant variables for the prediction of inadequate position of the IUD immediately after insertion (n = 15) were the type of IUD (NovaT versus other types respectively: 5/18 versus 10/177; OR: 5.93; 95% CI: 1.8918.6; P < 0.01) and complicated insertion (complicated versus uncomplicated respectively: 6/12 versus 13/171; OR: 5.47; 95% CI: 1.9914.9; P < 0.01). Experience of the inserting clinician was did not significantly influence adequacy of insertion (threshold <25 and <75 inserted IUD respectively; OR: 1.67; 95% CI: 0.319.2, P = 0.56; and OR: 2.27; 95% CI: 0.6112.6, P = 0.19). In contrast to women with an adequately positioned IUD, all women with an inadequately positioned IUD had complaints at followup (8.4 versus 100%, OR: 11.9, 95% CI: 7.219.6). Complaints at followup was the only significant variable for inadequate position at followup (OR: 96.61; 95% CI: 6.341471, P < 0.01). Due to the low prevalence of complaints at followup, it was not possible to differentiate which of the complaints was the single most significant predictor of inadequate position of the IUD.
Multiregression analysis: correctness of clinical evaluation

Immediately after insertion, the clinician evaluated the position of the IUD to be incorrect in 11 cases (5.6%; eight false negative evaluations and three false positive evaluations). Complicated insertion was the only significant predictor of incorrect clinical evaluation (OR: 2.89; 95% CI: 1.256.67, P = 0.01). Inexperience (<25 inserted IUD) was not of significant influence in the incorrect clinical evaluation immediately after insertion (OR: 1.81; 95% CI: 0.369.2; P = 0.47). At followup, the clinical evaluation was incorrect in six cases (3.5%; all false negative evaluations). Complaints at followup was the only significant predictive variable (OR: 11.09; 95% CI: 1.5181.39, P = 0.02). Again, inexperience was not a significant variable for improper clinical evaluation (OR: 0.71; 95% CI: 0.114.7; P = 0.71). At followup, highly experienced clinicians did not significantly more often evaluate the position of the IUD incorrectly (OR: 1.39; 95% CI: 0.228.91, P = 0.69).

Reliability analysis

The intraclass correlation (alpha) for repeatability was 0.99 (95% CI: 0.980.99). Table III shows the reliability analysis of the IUDED measurement stratified for type of IUD.
View this table:

In this window In a new window

Table III.

Results of the reliability analysis of the intrauterine device (IUD)endometrium distance measurements according to type of IUD
Previous SectionNext Section

Discussion
The routine use of transvaginal ultrasound to monitor the position of the IUD is not indicated since our results show that without clinical suspicion of an erroneous position, the chances of an inadequately positioned IUD are negligible (NPV immediately after insertion and at followup respectively: 0.98 and 1.00). In other words, clinical evaluation is a good test for evaluating the position of the IUD, especially at followup. Only in the case of clinical suspicion of inadequate positioning may additional TVU be helpful, since 60% of these IUD are still positioned adequately in the uterine cavity after all (PPV: 0.6 and 0.54 respectively). Therefore the use of TVU may prevent the unnecessary removal of adequately located IUD in women with complaints. In our opinion, the development of evidencebased guidelines for the insertion of an IUD and followup of IUD users is mandatory (Rivera and Best, 2002). Although the IUD is used infrequently, despite its reliability and costeffectiveness (Mishell, 1998), its use may increase because of three reasons. First, nulliparity is no longer a contraindication (Hubacher et al., 2001). Second, the levonorgestrelreleasing intrauterine system has been shown to be highly effective for the treatment of menorrhagia and dysfunctional uterine bleeding (Crosignani et al., 1997; Monteiro et al., 2002). Third, due to recent attention to the increased risk of venous thrombosis as a result of the use of oral contraceptives, especially third generation combined oral contraceptives, some women may switch from hormonal to nonhormonal contraceptive methods (Vandenbroucke et al., 2001). As suggested for proper diagnostic research (Sackett and Haynes, 2002), we minimized the bias of the sonographer since he was unaware of both the opinion of the clinician about the position of the IUD and of the difficulties at insertion; the sonographer was also unaware of the conclusion of the clinical evaluation at followup. However, for ethical reasons it was not possible to withhold the results of the TVU immediately after insertion to the clinicians and the

patients. This may have biased both the sonographer and the clinician in their evaluation at followup, especially in the seven women with inadequate position of the IUD after insertion. Our results indicate that clinicians with relatively little experience in inserting and monitoring IUD, like most general practitioners, are still capable of doing so. The number of incorrect clinical evaluations of IUD position did not differ significantly between inexperienced (<25 IUD inserted) and experienced (>25 IUD inserted) clinicians, either immediately after insertion or at followup (OR respectively: 1.81; 95% CI: 0.399.2; and 0.71; 95% CI: 0.114.7). Inexperienced and experienced clinicians were equally accurate in inserting IUD (OR: 1.67; 95% CI: 0.319.2). TVU has been shown to be able to monitor the position of all different types of IUD (BonillaMusoles et al., 1996; Petta et al., 1996; Fandes et al., 1997, 2000; Palo 1997); however, this is the first report on the excellent reliability of the IUDED measurement by TVU (intraclass coefficient 0.99). We choose 5.0 mm as threshold for abnormal IUD position as consensus between earlier reports. First Fandes showed that in 90% of women without complaints of their copper IUD, the IUDED was <7.0 mm (Fandes et al., 1997). Petta showed that if IUD were removed whenever the IUDED was >3.0 mm, this resulted in a significant reduction in spontaneous expulsion (Petta et al., 1996). Since there is no evidence concerning the optimal IUDED for levonorgestelreleasing intrauterine systems, we also decided to use 5.0 mm, although there are reasons to believe that this particular IUD might be as effective, both in the prevention of pregnancy as well as for the treatment of menorrhagia, if it is located more cervically (Andersson and Rybo, 1990; Andersson et al., 1994). Whether 5.0 mm is the most effective IUDED to ensure adequate contraception remains to be elucidated. However, investigation of the most effective IUDED is difficult because of the large number of women needed. In conclusion, the results of our prospective study show that TVU is not necessary on a routine basis to monitor the position of the IUD, neither immediately after insertion nor at the recommended followup after 6 weeks. Clinicians with little experience are equally capable of inserting and monitoring IUD when compared with more experienced clinicians. Therefore there is no need to refer women from primary to secondary care for insertion of an IUD. This is, especially from the costeffectiveness point of view, highly favourable. Only in the case of clinical suspicion of inadequate location of the IUD, as a result of the gynaecological interview and examination, may the use of TVU be beneficial and costeffective since this decreases the number of unnecessarily removed IUD.

European Society of Human Reproduction and Embryology

Preview or purchase options are not available More info