Spontaneous Fracture of Copper Intrauterine Devices: A Decade-Long Retrospective Analysis From a Single Tertiary Center.

The intrauterine device (IUD) is an effective method of contraception that has been used reliably for many years. Commonly used IUDs are generally of 2 types: hormonal and copper-containing IUDs. They stay in place for up to10 years, depending on the device. In addition to contraceptive treatment, hormonal IUDs have been used in the treatment of excessive menstrual bleeding, and pelvic pain associated with adenomyosis [ 1 – 3 ]. Some studies suggest that IUD use can reduce the risk of gynecological cancers and aid in clearing HPV infection [ 4 – 6 ]. Given these benefits, the use of IUDs has been increasing worldwide. Between 2015 and 2017 in the USA, 64.5% of women aged 15–49 years had used some form of contraception, and 10.3% of these methods were long-acting reversible contraceptives (LARC) such as an IUD or hormonal implant [ 7 ]. IUD use was also shown to increase between 2008 and 2014 from 6% to 14% [ 8 ]. Although IUDs provide reliable, safe, and appropriate contraception, they can cause complications due to device failure, anatomical variations, and practitioner errors. These complications include unplanned pregnancy (<1%), uterine perforation (0.1–0.3%), device displacement, breakage, ectopic pregnancy, cramping, abnormal bleeding, pelvic infection, infertility, device embedment, fragmentation, or expulsion [ 1 , 9 ]. Spontaneous IUD fractures, although rare, can have important clinical and public health implications. Fractured fragments may remain undetected within the uterus, leading to complications such as abnormal uterine bleeding, pelvic pain, infection, infertility, or even injury to adjacent organs if migration occurs. From a clinical standpoint, unexpected fracture can complicate removal procedures, increase the need for invasive interventions, and cause significant patient anxiety. Despite the widespread use of IUDs worldwide, the true prevalence and mechanisms of spontaneous fracture remain poorly understood. Addressing this knowledge gap is crucial for guiding follow-up protocols, informing device design improvements, and safeguarding patient outcomes [ 10 – 12 ]. Fracture of an IUD refers to the structural breakage of its components, typically the horizontal arms. A ‘spontaneous fracture’ is defined as breakage that occurs while the IUD is in situ, without any external manipulation, such as insertion or removal. This is distinct from iatrogenic fractures, which result from mechanical stress during device placement or extraction [ 13 ]. Case reports of spontaneous IUD breakage in the uterus have been published, but they are limited in number and primarily date back to the 1970s and 1980s, with only a few scattered reports in recent decades [ 14 ]. More recently, pharmacovigilance analyses have suggested that copper IUDs may be more frequently associated with fracture events than hormonal devices. For example, Latack et al evaluated the FDA Adverse Event Reporting System (FAERS) database and reported that copper IUDs were implicated in 9.6% of device-related adverse events, compared with 1.7% for hormonal IUDs, despite their lower overall usage [ 15 ]. In addition, a recent mechanical study demonstrated that structural stress tends to concentrate at the T-junction, predisposing copper IUDs to fatigue-related breakage [ 16 ]. These findings underscore the need for updated prevalence data and provide further rationale for examining spontaneous IUD fracture in contemporary practice. To our knowledge, there is no prevalence study on spontaneous IUD breakage in the uterus, especially in recent years. This lack of data poses a challenge for clinicians, as unrecognized fractures may lead to complications such as bleeding, pain, infection, infertility, or even migration into adjacent organs. Without robust prevalence data, routine screening protocols and device safety assessments remain inadequately informed. Therefore, the aim of this study was to investigate the prevalence and clinical implications of spontaneous IUD fractures in patients followed by our clinic.

A total of 183 cases using Type 1 IUD and 280 cases using Type 2 IUD were included in the analysis. The clinical characteristics of the patients are summarized in Table 1 . The mean age of patients with a non-fractured Type 1 IUD was 35.40±4.43 years, while it was 36.50±4.66 years in those with a fractured Type 1 IUD. In comparison, the mean age of Type 2 IUD users was 35.87±4.48 years. There was no statistically significant difference in age between the groups (p=0.463, independent samples t-test). The mean duration of IUD retention was 47.12±22.76 months in non-fractured Type 1 IUD users, 51.91±18.91 months in fractured Type 1 IUD users, and 52.9±23.8 months in Type 2 IUD users. No statistically significant difference was observed in IUD retention duration between fractured and non-fractured Type 1 users (Student’s t-test, p=0.386). Additionally, there was no significant association between longer retention time and the occurrence of fracture when analyzed as a continuous variable. The retention duration was significantly shorter in the non-fractured Type 1 IUD group compared to the Type 2 IUD group (p=0.030, Mann-Whitney U test). Although this result reached statistical significance, no correction for multiple comparisons was applied; therefore, this finding should be interpreted with caution. BMI comparisons between groups are summarized in Table 2 . Although Type 2 IUD users had a slightly higher average BMI, none of the differences between groups reached statistical significance (all p>0.05). Among the 183 Type 1 IUD users, 12 patients were found to have spontaneous IUD fractures. This number corresponds to an overall fracture rate of 6.56% (95% CI: 2.97–10.15%) within the Type 1 IUD group. Considering the entire study population of 463 patients, the overall prevalence of spontaneous IUD fracture was 2.59% (95% CI: 1.49–4.48%). No fractures were observed among the 280 Type 2 IUD users. The distribution of fracture locations revealed 9 cases in the uterus’ cornua and 3 in the cervical canal. Figure 2 shows examples of broken IUDs taken during hysteroscopy ( Figure 2A ), transvaginal ultrasonography ( Figure 2B ), and a picture of the pieces that were collected ( Figure 2C ). The detailed clinical features of patients with fractured IUDs are presented in Table 3 . In 3 of the 12 fractured cases, more than 1 fracture (excluding the main body) was observed. Fractures were discovered during routine control visits in 2 patients, following episodes of heavy vaginal bleeding in another 2, and after breakthrough bleeding in 8 cases. Two patients with breakthrough bleeding reported spontaneous expulsion of IUD fragments. The location of the fractured IUD arms was most commonly the cornua of the uterus (n=9), followed by the cervical canal (n=3). Fragments located in the cervical canal were removed using Novak extraction, while those in the cornua were retrieved via hysteroscopy. Overall, 10 of the 12 fracture cases (83.3%) presented with symptoms such as breakthrough or heavy vaginal bleeding, while 2 cases (16.7%) were detected incidentally during routine follow-up without any reported symptoms. The clinical implications of fracture location are noteworthy. Fragments in the cervical canal were more likely to be spontaneously expelled and were easily accessible for removal using outpatient techniques such as Novak curettage. In contrast, fragments located in the cornual regions were embedded deeper within the uterine cavity and required operative hysteroscopy for visualization and retrieval. These findings suggest that fracture location can influence symptom presentation and the complexity of removal, underscoring the need for individualized management based on fragment position.

In this retrospective study, we found that spontaneous fractures of intrauterine devices occurred in 6.56% of women using Type 1 copper IUDs with a gold core, while none were observed in Type 2 copper IUD users. Most fractured IUD arms were located in the cornua of the uterus and were successfully taken out using hysteroscopy, while pieces in the cervical canal were removed using Novak extraction. Age and duration of IUD use were not significantly different between fractured and non-fractured groups. While the fracture rates reported in relation to the IUDs in the past are 1–2%, the frequency of fracture rates in today’s IUDs is unknown [ 16 ]. Our findings revealed a 6.56% fracture rate for Type 1 IUDs, which is substantially higher than the 1–2% rates reported in previous observational studies [ 14 ]. The complete absence of fractures in the Type 2 IUD group may point to differences in material properties or structural design. While we did not perform mechanical testing, such a marked discrepancy suggests a need for further comparative analysis of device integrity. As a result of the literature study conducted by Wilson et al in 2013 to identify fractured IUDs, it was determined that published reports were either case reports or small case series with limited clinical information [ 14 ]. In the 1970s, the first studies describing the fracture IUD reported that the majority of fractured IUDs were the Lippes Loop, a plastic, multi-loop device [ 17 – 20 ]. Multi-load Cu250 accounted for most cases of fractured IUDs in the 1980s [ 14 , 21 ]. Wilson et al reported that between 2000 and 2013, 10 cases of fractured IUDs were published, most of which involved copper models, although one case involved a levonorgestrel IUD [ 14 ]. A recent study reported 170 215 adverse events related to IUDs in the USA from 1998 to 2022. IUD fractures accounted for 9.6% (n=4144) of adverse events for the copper IUD and 1.7% (n=2140) of adverse events for hormonal IUDs [ 15 ]. However, these figures are derived from passive surveillance systems such as the FDA’s Adverse Event Reporting System (FAERS), which are subject to several inherent limitations, including significant underreporting, selective reporting influenced by media or legal attention (reporting bias), and the absence of a defined denominator population. Therefore, such data cannot be used to estimate true incidence or prevalence and should primarily be interpreted as hypothesis-generating rather than conclusive. Nonetheless, recent FAERS pharmacovigilance data indicate that copper IUDs have been associated with higher fracture rates compared to hormonal IUDs, despite their lower overall usage [ 15 ]. Although the true prevalence cannot be determined from spontaneous adverse event reports, the available data suggest that copper IUDs are reported to break approximately 6.19 times more often than hormonal IUDs [ 15 ]. However, from 2015 to 2017, 22.9% of IUD users were using the copper IUD and 76.5% were using the hormonal IUD [ 22 ]. This discrepancy suggests that copper IUDs may have a higher relative risk of breakage compared to hormonal IUDs, despite their lower prevalence among users. Furthermore, in Latack KR et al’s study, patients reporting IUD breakage had a significantly higher median age (34 vs 30 years; p<0.001) and median weight (70.3 vs 68 kg; p<0.001) compared to those reporting non-breakage events [ 15 ]. In contrast, our findings showed a trend toward lower BMI values in patients with spontaneous IUD fracture, although this difference was not statistically significant. Differences in study design, IUD types, or anatomical characteristics of the patient populations may account for this apparent discrepancy. Although higher body weight could theoretically increase mechanical pressure on the device, this remains speculative and should be interpreted with caution. We hypothesize that lower BMI may be associated with smaller uterine dimensions or more acute uterocervical angles, potentially increasing intrauterine mechanical stress on the device. Future prospective studies incorporating 3D ultrasonography or MRI could help test this anatomical predisposition hypothesis. Although our data showed a non-significant trend toward lower BMI in fracture cases, we acknowledge that BMI alone is unlikely to directly influence intrauterine force generation. A more plausible explanation may lie in the mismatch between IUD size and the dimensions of the endometrial cavity, which determines how much mechanical stress is exerted on the device. Smaller uterine cavities – regardless of body habitus – may result in disproportion between the uterine cavity and the device, predisposing to asymmetrical uterine forces that generate bending or pressure points on the IUD [ 23 , 24 ]. Future prospective imaging studies assessing uterine cavity dimensions in relation to IUD fit may help clarify this association. Although there are reports of fractured arms during removal of the IUD, spontaneous fracture of the arms of the IUD is very rare [ 25 ]. However, there is no complete explanation of how IUDs break in the uterus [ 26 ]. A biomechanical analysis by Goldstuck et al suggests that intrauterine forces generated during the menstrual cycle can reach levels sufficient to deform or fracture certain types of IUDs, particularly at structural weak points such as the junction between the horizontal and vertical arms [ 24 ]. Their findings indicate that the uterus can generate contractile forces exceeding 20 N during menstruation, which may subject the IUD to repetitive mechanical loading. Over time, this cyclical stress could lead to structural fatigue, increasing the likelihood of material failure and device fracture [ 24 ]. Consistent with this hypothesis, a study comparing 4 copper IUDs commonly used in Europe found that the mechanical resistance of all devices progressively decreased with longer intrauterine duration [ 27 ]. How these uterine forces interact with different IUD designs and patient-specific anatomical factors remains an important area for future research, as the rigidity and shape of the device, as well as uterine size, can all influence susceptibility to fracture. Given these biomechanical considerations, attention should also be directed toward device engineering, as modifications in IUD design may help mitigate fracture risk. In addition to clinical monitoring strategies, improvements in IUD design may play a role in reducing fracture risk. For example, reinforcing structurally vulnerable junctions – such as the T-junction between the vertical and horizontal arms – with more flexible or stress-absorbing materials could help minimize fatigue-related breakage. Using materials with higher fatigue resistance or optimizing device curvature to better conform to uterine anatomy may further reduce mechanical strain. Furthermore, radiopaque markers and integrated structural stress indicators (eg, materials engineered to change configuration or visibility under mechanical strain) could assist in earlier fracture detection and safer removal. In a case study presented by Sinha et al, a 37-year-old woman reported vaginal bleeding 5 weeks after the insertion of a copper IUD following pregnancy termination [ 25 ]. A transvaginal ultrasound revealed 2 echogenic areas in the fundal region of the endometrium. During hysteroscopy, the transverse arm of the IUD was found to be fractured at the T-junction, and the fragment was removed. The authors emphasized that this was a very rare instance of spontaneous IUD breakage [ 25 ]. In our study, we aimed to determine and compare the fracture rates in 2 different IUDs with similar characteristics between January 1, 2011 and December 31, 2021. There were 280 cases of Type 2 IUDs, and no fractured IUDs were found in any of them. However, we detected a spontaneous fracture in 12 (6.56%) of the 183 Type 1 IUD cases. The age and mean interval time of patients with fractured Type 1 IUD were similar to those of patients with non-fractured Type 1 IUD and Type 2 IUD. Although the mean retention duration was slightly higher in the fracture group, the difference was not statistically significant. No demographic factors, including age, BMI, or parity, were found to be significantly associated with spontaneous IUD fracture risk in our study population. This suggests that duration of IUD use alone may not be a strong independent predictor of fracture risk in this cohort. In recent years, the number of studies about spontaneous fracture has almost been nonexistent. Except for the RUDIUS study, only case reports have been available. In the RUDIUS study, the spontaneous fracture rate was found to be 32.5% (n=39) among the cases in which a fractured fragment was detected over a 25-month period. However, the study did not specify the proportion of fractured IUDs in all cases during this period. Unlike the RUDIUS study, our cohort includes all insertions with follow-up data, allowing for a more accurate estimation of prevalence. To provide a clearer synthesis of the existing literature, Table 4 presents a comparative overview of IUD fracture rates and characteristics across major published studies. This structured comparison highlights the limited prevalence data available and situates our findings within the broader research landscape. Since the RUDIUS study only included confirmed fracture cases and did not report the total number of IUDs inserted or monitored during that timeframe, direct prevalence comparison with our study is not feasible [ 28 ]. Given the scarcity of comprehensive studies on spontaneous IUD fractures, the 6.56% fracture rate observed in Type 1 IUDs in our cohort indicates a relatively high prevalence. In our study, 2 patients with breakthrough bleeding reported the spontaneous expulsion of IUD fragments. This observation underscores the importance of considering spontaneous expulsion as a potential explanation when fractured fragments are not visualized on imaging. In clinical practice, the term “spontaneous expulsion of a fractured IUD” specifically refers to the passage of a broken fragment, rather than the entire device. Such fragments may be expelled unnoticed, especially if accompanied by minor bleeding, and can complicate the diagnostic process when residual pieces are no longer visible on imaging. Clinicians should be aware that expelled fragments may go unnoticed by patients, especially in the case of minor bleeding. Therefore, when IUD fracture is suspected but no fragment is detected radiologically or hysteroscopically, the possibility of prior spontaneous expulsion should be considered. It may be prudent to advise short-term follow-up and repeat imaging to ensure complete uterine clearance and to prevent retained fragments, which could lead to complications such as infection or infertility. The predominance of symptomatic presentations indicates that breakthrough bleeding may serve as an important clinical indicator of IUD fracture, although a minority of cases (16.7%) were entirely asymptomatic and detected incidentally during routine surveillance. A similar finding was reported by Wakrim et al and Nadgir et al, who documented spontaneously expelled fractured IUD pieces [ 12 , 29 ]. These findings indicate that, in cases of confirmed IUD fracture where no fragments are identified during hysteroscopic or radiologic evaluation, spontaneous expulsion of the broken part should be considered a plausible explanation. Even in cases where IUD fracture is suspected or some fragments are believed to have been spontaneously expelled, it is crucial to promptly evaluate the uterine cavity for any remaining fragments, as timely detection and removal of fractured IUD pieces can prevent complications such as pain, bleeding, or infertility [ 30 ]. Additionally, since IUDs can lead to infections like Actinomyces or group A streptococcal, any broken pieces need to be taken out [ 31 ]. Fractured IUD fragments can lead to serious complications, including adhesions and perforations involving adjacent organs such as the bladder and rectum. In support of this observation, Briceno et al reported a case in which a Lippes Loop that had remained in utero for an extended period was identified as the cause of a vesicouterine fistula [ 10 ]. A number of techniques, such as ultrasound scanning, pelvic X-ray, and hysteroscopy, can be used to identify fractured IUD parts. However, among these, hysteroscopy has the advantage of both detecting the fractured part and retrieving it. Especially for buried or fragmented devices, hysteroscopy has been quite useful [ 25 ]. It has been reported that fractured IUD fragments are visualized by hysteroscopy and successfully removed by operative hysteroscopy [ 32 ]. Canovas et al achieved 100% success with hysteroscopy for detection and removal of the IUD fragment in the uterus [ 28 ]. However, they reported that ultrasound-guided extraction with forceps did not work well. In our study, we found that 9 fractured IUD fragments were in the cornua of the uterus and 3 were in the cervical canal. Anatomical and mechanical factors may help explain why IUD fractures most commonly occur in the uterine cornua. The uterus can generate substantial contractile forces, reaching up to 50 N, particularly during peristalsis or menstruation. These forces can apply repetitive stress to the horizontal arms of the IUD, especially at the cornual curvature or the junction of the horizontal and vertical arms where structural weakness can be aggravated by narrow anatomy [ 24 ]. While the fractured IUD fragments in the cervical canal were successfully removed by Novak extraction, the fractured IUD fragments in the cornua of the uterus were successfully removed by hysteroscopy. Both methods we use – Novak extraction and hysteroscopy – are suitable for removing fractured IUD pieces. Based on our findings, we recommend that clinicians consider implementing standardized annual ultrasonographic surveillance for patients with Type 1 copper IUDs, even in the absence of symptoms. In cases of breakthrough bleeding or unexplained pelvic discomfort, prompt imaging is warranted to assess device integrity. Routine evaluation can facilitate early detection of asymptomatic fractures, enabling timely intervention to prevent complications such as migration, infection, or infertility. The strength of our study is that it addresses a topic with limited prior research in the literature. However, several limitations should be considered. The retrospective design may have introduced selection bias and restricted control over potential confounding variables, such as variations in menstrual bleeding patterns, differences in IUD insertion techniques, undetected pelvic infections, or individual patient behaviors that may influence intrauterine pressure. The evaluation was limited to only 2 types of IUDs, which may reduce the generalizability of the findings to other intrauterine devices. While ultrasound was employed for routine surveillance, it has limited sensitivity in detecting small or embedded IUD fragments. Additionally, we did not conduct any mechanical or material stress testing on the devices, which would have provided further insight into the mechanisms underlying spontaneous fracture. Given the number of statistical comparisons performed, and in the absence of formal correction for multiple testing, marginally significant findings should be interpreted with caution to avoid overestimating their clinical relevance. Further studies with larger sample sizes and prospective designs are warranted to better understand the risk factors and mechanisms underlying IUD fracture and validate our findings. Spontaneous IUD fracture is a rare complication, and its pathophysiological mechanisms remain poorly understood. Several mechanisms have been proposed, including chronic intrauterine forces during menstruation, malposition or asymmetric placement leading to uneven stress, and material fatigue at structurally weak points such as the T-junction. Device design and rigidity may further contribute, especially in patients with smaller or more angular uterine cavities. Although our study was not designed to directly evaluate these mechanisms, the marked difference in fracture rates between IUD types suggests that device-related structural characteristics may play a key role [ 15 , 24 , 33 ]. In particular, the presence of a central gold core in Type 1 IUDs, while designed to enhance durability, may inadvertently alter the flexibility of the frame. This rigidity could create a stress concentration at the T-junction between the vertical stem and the horizontal arms, predisposing the device to fatigue-related fracture under repetitive intrauterine forces. By contrast, Type 2 IUDs, without reinforcement, may allow more uniform distribution of intrauterine forces and greater long-term resilience. These device-related differences may help explain the markedly higher fracture rate observed in Type 1 IUDs compared to Type 2 devices in our study. As this was a retrospective study based on clinical records and imaging findings, direct mechanical testing of device fracture forces could not be performed . An additional dimension of analysis that could strengthen future investigations is the direct measurement of fracture force thresholds of different IUD models. Mechanical testing of IUD arms under controlled laboratory conditions could determine whether certain device designs are inherently more susceptible to fatigue or failure. Such data would be particularly valuable in elucidating the observed discrepancies in fracture rates between IUD types, as demonstrated in our study.

In this retrospective cohort study of 463 women, spontaneous intrauterine device fractures were identified in 6.56% of Type 1 copper IUDs with a gold core (12/183), whereas no fractures occurred among 280 Type 2 devices. Most fractures were located in the uterine cornua, and while 83.3% of cases were symptomatic, 16.7% were detected incidentally. These findings underscore the need for routine ultrasonographic follow-up and clinician awareness of this rare but clinically relevant complication, particularly since some fractures were asymptomatic and detected only during scheduled surveillance. For clinical management, hysteroscopy is a reliable method for cornual fragments, whereas Novak extraction is a safe and effective option for fragments in the cervical canal. While our results suggest that device design contributes to fracture risk, the retrospective nature of the study and the evaluation of only 2 IUD models limit the generalizability of the findings. Future prospective studies with larger and more diverse cohorts are warranted to validate these findings and to clarify the interaction between device mechanics and patient-specific factors such as age, BMI, parity, and uterine dimensions through imaging-based assessments, to further elucidate the mechanisms underlying fracture and to guide safer device design and follow-up protocols.

Ethics approval for this retrospective study was received from the institution’s ethics committee (Approval No: 2024/13). All data were collected and analyzed in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments. All patient data were anonymized prior to analysis to protect confidentiality. As this was a retrospective study, individual informed consent was waived by the ethics committee. This was a retrospective cohort study conducted at the Department of Obstetrics and Gynecology, Memorial Şişli Hospital. Medical records of patients who underwent copper IUD insertion and follow-up between January 1, 2011, and December 31, 2021, were evaluated. Women with a history of at least 1 vaginal delivery who used an IUD were included in the study. Exclusion criteria were use of hormonal IUDs, nulliparity, uterine anomalies or pathology, prior uterine surgery (including cesarean section), removal of IUD due to menorrhagia, myomas causing menometrorrhagia, pelvic inflammatory disease, or patients with an IUD that remained less than 12 months or longer than 120 months. Patients with a prior history of cesarean section were excluded in order to avoid potential confounding effects of uterine scarring and altered cavity anatomy on fracture risk. The follow-up window was set at 12–120 months to allow adequate duration of use, and also reflecting real-world settings in which IUDs are not uncommonly retained beyond the manufacturer’s recommended duration ( Figure 1 ). In this study, 2 types of copper IUDs were evaluated: Type 1, which was a copper IUD with a gold core (Gold T, Eurogine), and Type 2, a copper T IUD (Flexi-T+ 300, Prosan ). The frame of the Type 1 IUD is made of polyethylene with barium sulfate for radiopacity and contains a central gold core for added structural strength. The Type 2 IUD consists of a polypropylene frame with barium sulfate, but without a metal core reinforcement. Neither model contains hormonal components. Type 1 IUD has 99.99% pure copper wire (375 mm2) with a 0.1 mm diameter gold core. The Type 1 IUD used in our clinic measures 31 mm in width and 33 mm in length and is Y-shaped. It provides contraceptive safety for up to 5 years. Type 2 IUD measures were 32 mm long and 28 mm in width. The IUD is wrapped with 300 mm 2 copper wire of the highest purity (99.9%) on the shaft. It offers 5 years of optimal contraceptive protection. All IUDs were inserted by the same team of physicians, and patients were followed by the same clinicians throughout the study period. A first follow-up examination was performed 1 month after insertion, during which all IUDs were confirmed to be intact. Subsequent routine evaluations were conducted annually. Over the 10-year study period, cases in which IUD integrity appeared compromised on transvaginal ultrasonography (either during scheduled follow-ups or prompted by patient complaints) were retrospectively identified and included in the analysis. The “IUD interval” was defined as the duration, in months, from the initial IUD insertion to either the last follow-up or the date of fracture detection, whichever occurred first. The primary outcome was the prevalence of spontaneous IUD fracture. Secondary measures included evaluation of the relationship between IUD fracture and variables such as Type of IUD, duration of use, and body mass index (BMI). These associations were analyzed statistically. All statistical analyses were performed using SPSS software version 20.0 (SPSS Inc., Chicago, IL). Descriptive statistics were used to summarize the clinical characteristics of the patients. Continuous variables (eg, age, BMI, IUD retention duration) were expressed as mean±standard deviation (SD), and categorical variables (eg, fracture occurrence by IUD Type) as frequencies and percentages. The Shapiro-Wilk test was used to assess the normality of continuous variables. To compare groups, the independent samples t-test was used for normally distributed continuous variables (eg, age, BMI, and IUD retention duration), and the Mann-Whitney U test for non-normally distributed variables. Categorical data were analyzed using the chi-square test or Fisher’s exact test as appropriate. A p value less than 0.05 was considered statistically significant. No missing data were encountered for the variables included in the final analysis. No formal correction for multiple comparisons was applied, as the analyses were primarily hypothesis-driven with a limited number of prespecified comparisons.