Effect of different long-acting reversible contraceptive methods on female sexual function: a retrospective cohort study.

An observational retrospective cohort study was carried out. All medical information in the cohort study was obtained from the medical record system and patient health records. Married women aged 21–45 years who visited the Gynecology Department of Hangzhou Women’s Hospital for contraceptive services between September 2022 and September 2023 were included in the study. They were considered if they met the following criteria: (i) had been using Cu-IUD, LNG-IUS or Implanon for at least six months. (ii) had regular menstrual cycles and periods without dysmenorrhoea and dyspareunia prior to the use of LARCs. (iii) had a regular heterosexual partner and an active sexual life. Exclusion criteria included pregnancy or suspected pregnancy, vascular or metabolic disease, known or suspected malignancy, severe liver disease and undiagnosed vaginal bleeding, use of psychotropic medication, and use of hormonal contraceptives in the past year. This study was approved by the Ethics Review Committee of Hangzhou Women’s Hospital (No. 2024-025). This research was conducted in accordance with the principles of the Declaration of Helsinki. Participants’ privacy was protected during data analysis. Each patient chose the type of LARC available according to their wishes. (i) Cu-IUD (Shanghai Medical Suture Factory Ltd); (ii) LNG-IUS (Mirena ® . Bayer SA, Bayer Oy, Turku, Finland); (iii) Implanon (Schering-Plough, Organon, The Netherlands). The main indications for LARC are as follows [ 2 ]: Cu-IUD is indicated for long-acting, highly effective reversible contraception. And it can also be used for emergency contraception. LNG-IUS is indicated for contraception, treatment of menorrhagia, and management of symptoms related to adenomyosis or endometriosis. Implanon is indicated for long-acting reversible contraception or for women whose uterine cavity conditions are unsuitable for an IUD. The population was divided into three groups according to the type of LARC: the Cu-IUD group (30 samples), the LNG-IUS group (30 samples), and the Implant group (30 samples). Female Sexual Function Index (FSFI) is one of the key tools used to assess female sexual function. For outpatient use, we have adopted the faster and more manageable FSFI-6 [ 8 ].The questionnaire consists of six domains (desire, arousal, lubrication, orgasm, satisfaction and pain) covering different aspects of female sexual function. Each question is scored from 0 to 5, with a total score of 2–30, and a total score of ≤ 19 can be diagnosed as female sexual dysfunction (FSD) [ 8 ]. Demographic data (age, occupation, education level), previous contraceptive methods, history of abortion, and health status were collected at baseline. Sexual activity was assessed using the FSFI-6 questionnaire at one, three, and six months after placement of LARCs, corresponding to experiences during each respective month. And the changes in sexual frequency and changes in partner sexual satisfaction were recorded. Quantitative variables were analyzed using the Shapiro-Wilk normality test. Normally distributed quantitative data were expressed as N ± s and comparisons among groups were performed with ANOVA. Non-normally distributed quantitative data were expressed as M (P25, P75) and comparisons among groups were performed using the Kruskal-Walli’s rank sum test. Qualitative variables were expressed as frequencies and comparisons between groups were made using the chi-square test. Rank-sum test was used for hierarchical data. Generalized estimated equation and Wald Chi-Square tests recorded differences in categorical variables at different stages and trends over time. General linear models documented the differences and trends over time for continuous numerical variables at different stages. Statistical analyses were performed using SPSS 25 and p  < 0.05 was considered statistically significant.

A total of 90 women completed each stage of the questionnaire and examination in full. Among these 90 study participants, the majority were part-time employees (68.9%), had a college or university degree (77.8%), and had a history of abortion (78.9%). The proportion of women without any prior gynecological disease was 93.3% in both the Cu-IUD and Implanon groups, compared with 63.3% in the LNG-IUS group ( p  = 0.008). Specifically, adenomyosis was present in 30.0% of the LNG-IUS group versus 6.7% in each of the other two groups. Although the prevalence of adenomyosis was higher in the LNG-IUS group, the differences between groups were not statistically significant. The Implanon group being younger than the Cu-IUD and LNG-IUS groups ( p  = 0.002). There were also differences in previously used contraceptive methods, with a lower rate of prior male condom use and a higher rate of intrauterine device use in the LNG-IUS group ( p   0.05), suggesting comparable baseline characteristics (Table  1 ). Table 1 Baseline characteristics of the study cohort Characteristics Total, n  = 90 Cu-IUD, n  = 30 Implanon, n  = 30 LNG-IUS, n  = 30 p Age(years), M(P25,P75) 37.0(34.8,40.0) 39.0(36.0,40.3) † 35.0(32.0,38.3) 39.0(35.8,42.0) † 0.002 * Occupation, N (%) 0.261  Full-time employee 5(5.6) 2(6.7) 2(6.7) 1(3.3)  Part-time employee 62(68.9) 23(76.7) 15(50.0) 24(80.0)  Liberal professional 4(4.4) 1(3.3) 2(6.7) 1(3.3)  Homemaker 1(1.1) 0(0.0) 1(3.3) 0(0.0)  Other 18(20.0) 4(13.3) 10(33.3) 4(13.3) Education Level, N (%) 0.333  High school education or below 2(2.2) 1(3.3) 1(3.3) 0(0.0)  college or university degree 70(77.8) 24(80.0) 20(66.7) 26(86.7)  Master’s degree or above 18(20.0) 5(16.7) 9(30.0) 4(13.3) Previous contraceptive method, N (%) <0.001 *  Uncontracepted 1(1.1) 0(0.0) 0(0.0) 1(3.3)  Male condom 74(82.2) 28(93.3) 28(93.3) 18(60.0)  oral contraceptive 3(3.3) 0(0.0) 0(0.0) 3(10.0)  Natural family planning or withdrawal 2(2.2) 0(0.0) 2(6.7) 0(0.0)  Other 2(2.2) 0(0.0) 0(0.0) 2(6.7)  Intrauterine device 8(8.9) 2(6.7) 0(0.0) 6(20.0) Previous gynecological diseases, N (%) 0.008 *  None 75(83.3) 28(93.3) £ 28(93.3) £ 19(63.3)  Endometriosis 0(0.0) 0(0.0) 0(0.0) 0(0.0)  Adenomyosis 13(14.4) 2(6.7) 2(6.7) 9(30.0)  Uterine diverticulum 2(6.7) 0(0.0) 0(0.0) 2(6.7)  Dysmenorrhea 0(0.0) 0(0.0) 0(0.0) 0(0.0) Abortion, N (%) 0.850  No 19(21.1) 7(23.3) 7(23.3) 5(16.7)  Yes 71(78.9) 23(76.7) 23(76.7) 25(83.3) Frequency of sexual activity, N (%) 0.812  Once to twice per week 18(20.0) 5(16.7) 7(23.3) 6(20.0)  Two to three times per month 72(80.0) 25(83.3) 23(76.7) 24(80.0)  Once or less per month 0(0.0) 0(0.0) 0(0.0) 0(0.0) Partner sexual satisfaction, N (%) 0.654  Satisfied 1(1.1) 1(3.3) 0(0.0) 0(0.0)  Somewhat satisfied 76(84.4) 26(86.7) 24(80.0) 26(86.7)  Neutral 13(14.4) 3(10.0) 6(20.0) 4(13.3) FSFI, M(P25,P75)  Desire 3.0(3.0,3.0) 3.0(3.0,3.0) 3.0(3.0,3.0) 3.0(3.0,3.0) 0.360  Arousal 3.0(3.0,3.0) 3.0(3.0,3.0) 3.0(3.0,3.0) 3.0(3.0,3.0) 0.603  Lubrication 4.0(4.0,4.0) 4.0(4.0,4.0) 4.0(4.0,4.0) 4.0(4.0,4.0) 1.000  Orgasm 4.0(4.0,4.0) 4.0(4.0,4.0) 4.0(4.0,4.0) 4.0(4.0,4.0) 0.651  Satisfaction 4.0(4.0,4.0) 4.0(3.8,4.0) 4.0(3.8,4.0) 4.0(4.0,4.0) 0.251  Pain 4.0(4.0,4.0) 4.0(4.0,4.0) 4.0(4.0,4.0) 4.0(4.0,4.0) 0.858 Total 22.0(22.0,22.0) 22.0(21.0,22.0) 22.0(21.0,22.3) 22.0(22.0,22.0) 0.424 Abbreviations Cu-IUD copper-containing intrauterine device, LNG-IUS levonorgestrel releasing intrauterine system, FSFI female sexual function index * p  < 0.05. † p  < 0.05 vs. Implant, £ p  < 0.05 vs. LNG-IUS Baseline characteristics of the study cohort Abbreviations Cu-IUD copper-containing intrauterine device, LNG-IUS levonorgestrel releasing intrauterine system, FSFI female sexual function index * p  < 0.05. † p  < 0.05 vs. Implant, £ p  < 0.05 vs. LNG-IUS Table  2 presents changes in the frequency of sexual activity at 1, 3 and 6 months following intrauterine device placement. During the 6-month follow-up period, time exerted a significant effect on changes in the frequency of sexual activity. Specifically, frequency at 6 months was markedly higher than at 1 month ( p  < 0.001) and 3 months ( p  = 0.034). In contrast, neither age nor LARCs demonstrated a significant influence. These findings suggest that sexual activity frequency increases progressively over time post-placement, independent of the LARCs type used. Table 2 Changes in the frequency of sexual activity and partner sexual satisfaction with different LARCs Variables Changes in the frequency of sexual activity Partner sexual satisfaction p OR (95%CI) p OR (95%CI) Age 0.829 0.99(0.93 ~ 1.06) 0.654 0.98(0.90 ~ 1.07) LARCs  Implanon 0.342 1.52(0.64 ~ 3.61) <0.001* 0.13(0.04 ~ 0.36)  LNG-IUS 0.067 0.48(0.22 ~ 1.05) <0.001* 0.10(0.04 ~ 0.28)  Cu-IUD - 1 - 1 Time  1 month <0.001* 0.12(0.06 ~ 0.22) 0.003* 0.40(0.22 ~ 0.73)  3 month 0.034* 0.63(0.42 ~ 0.97) 0.112 0.67(0.40 ~ 1.10)  6 month - 1 . 1 ps: Data represent changes from baseline (increased, unchanged, or decreased). Statistical analysis used generalized estimating equations, which was adjusted for age, employing a logit link function and an auto-regressive working correlation structure to account for repeated measures Abbreviations Cu-IUD copper intrauterine device, LNG-IUS levonorgestrel releasing intrauterine system, OR odds ratio * p  < 0.05 Changes in the frequency of sexual activity and partner sexual satisfaction with different LARCs ps: Data represent changes from baseline (increased, unchanged, or decreased). Statistical analysis used generalized estimating equations, which was adjusted for age, employing a logit link function and an auto-regressive working correlation structure to account for repeated measures Abbreviations Cu-IUD copper intrauterine device, LNG-IUS levonorgestrel releasing intrauterine system, OR odds ratio * p  < 0.05 Changes in partner sexual satisfaction during the follow-up period are shown in Table  2 . Time since placement and LARCs independently influenced satisfaction levels. Satisfaction scores were significantly higher in Cu-IUD group compared to the LNG-IUS group (B=-2.267, p  < 0.001) and the Implanon group (B=-2.072, p  < 0.001). Satisfaction levels in the Implanon and LNG-IUS groups were 13% and 10% of those observed in the Cu-IUD group, respectively. In addition, partner sexual satisfaction varied at different placement time. At the first month, satisfaction scores were only 40% of those at 6 months ( p  = 0.003), while no statistically significant difference existed between 3-month and 6-month scores ( p  = 0.112). These results indicate that sexual satisfaction of partner improved significantly by 3 months but plateaued thereafter, showing no further increase between 3 and 6 months. For the FSFI total score, the general linear model indicated significant differences across LARCs (F = 6.035, p  = 0.004), but no effect of time (F = 0.537, p  = 0.586) (Table  3 ). Further analysis revealed a significant interaction effect between time and LARCs (F = 3.841, p  = 0.025), indicating that the trajectory of total FSFI scores varied by contraceptive type over time. Figure  1 illustrates these trends across LARCs. At 1-month post-placement, total scores differed significantly between the Implanon and LNG-IUS groups ( p  = 0.008), but no differences were observed between the Cu-IUD group and the other two groups. By 3 months, a significant disparity emerged between the Cu-IUD and LNG-IUS groups ( p  = 0.037), while the Implanon group showed no differences compared to either group. At 6 months, significant gaps were observed between the LNG-IUS group and both the Cu-IUD ( p  = 0.031) and Implanon groups ( p  = 0.047), though no difference existed between the Implanon and Cu-IUD groups. These results suggest that the LNG-IUS group consistently exhibited the lowest FSFI scores regardless of time. A simple effects analysis of time revealed gradual improvements in sexual function for the Implanon group over the study period. Conversely, both the Cu-IUD and LNG-IUS groups showed significant increases in scores by month 3, with no further changes between months 3 and 6. Table 3 Comparison of FSFI among LARCs at different time Cu-IUD Implanon LNG-IUS P value1 P value2 Desire <0.001 0.522  1 month 3.00 ± 0.00†£ 2.52 ± 0.51* 2.50 ± 0.51*  3 month 3.00 ± 0.00†£ 2.55 ± 0.57* 2.60 ± 0.50*  6 month 3.00 ± 0.00†£ 2.62 ± 0.56* 2.67 ± 0.48* Arousal 0.003* 0.222  1 month 2.30 ± 1.29 2.93 ± 0.88£ 1.67 ± 1.54†  3 month 3.03 ± 0.18 3.07 ± 0.53 2.73 ± 0.83  6 month 3.03 ± 0.18 3.10 ± 0.41 2.90 ± 0.66 Lubrication 0.001* 0.685  1 month 3.07 ± 1.72 4.00 ± 1.20£ 2.27 ± 2.05†  3 month 4.00 ± 0.00 3.93 ± 0.80 3.57 ± 0.90  6 month 4.00 ± 0.00 4.21 ± 0.56£ 3.77 ± 0.50† Orgasm 0.004* 0.441  1 month 3.07 ± 1.72 3.72 ± 1.16£ 2.20 ± 1.99†  3 month 4.00 ± 0.00 3.90 ± 0.90 3.57 ± 0.86  6 month 4.00 ± 0.00£ 4.00 ± 0.60£ 3.67 ± 0.48*† Satisfaction 0.003* 0.825  1 month 3.53 ± 0.86£ 3.55 ± 0.74£ 2.87 ± 1.20*†  3 month 4.00 ± 0.00†£ 3.66 ± 0.72* 3.60 ± 0.68*  6 month 4.00 ± 0.00£ 3.86 ± 0.44 3.67 ± 0.48* Pain 0.002* 0.727  1 month 3.07 ± 1.72 3.97 ± 1.18£ 2.3 ± 2.09†  3 month 4.00 ± 0.00 4.21 ± 0.94 3.7 ± 0.92  6 month 4.00 ± 0.00† 4.31 ± 0.47*£ 3.9 ± 0.55† Total 0.004* 0.586  1 month 17.93 ± 7.28 20.69 ± 5.18£ 13.80 ± 9.13†  3 month 22.03 ± 0.18£ 21.31 ± 3.85 19.77 ± 4.36*  6 month 22.03 ± 0.18£ 22.1 ± 2.44£ 20.57 ± 2.76*† ps: Differences and temporal trends across time points for desire, arousal, lubrication, orgasm, satisfaction, and pain were analyzed using Generalized Estimating Equations (GEE); differences and temporal trends for the FSFI total score were analyzed using General Linear Models (GLM). Both models were adjusted for age Abbreviations FSFI female sexual function index, LARC long-acting reversible contraception, Cu-IUD copper intrauterine device, LNG-IUS levonorgestrel releasing intrauterine system P value1 refers to comparison between the LARC types; p value2 refers to comparison between stages. *, †, £ indicates pairs with significative differences in post-hoc test.* p  < 0.05 vs. CU-IUD, † p  < 0.05 vs. Implant, £ p  < 0.05 vs. LNG-IUS Comparison of FSFI among LARCs at different time ps: Differences and temporal trends across time points for desire, arousal, lubrication, orgasm, satisfaction, and pain were analyzed using Generalized Estimating Equations (GEE); differences and temporal trends for the FSFI total score were analyzed using General Linear Models (GLM). Both models were adjusted for age Abbreviations FSFI female sexual function index, LARC long-acting reversible contraception, Cu-IUD copper intrauterine device, LNG-IUS levonorgestrel releasing intrauterine system P value1 refers to comparison between the LARC types; p value2 refers to comparison between stages. *, †, £ indicates pairs with significative differences in post-hoc test.* p  < 0.05 vs. CU-IUD, † p  < 0.05 vs. Implant, £ p  < 0.05 vs. LNG-IUS Fig. 1 Changes in the total FSFI score of LARCs at different stages Changes in the total FSFI score of LARCs at different stages Table  3 highlights the distinct advantages of LARCs across various dimensions of sexual function. In desire, the Cu-IUD group demonstrated significantly higher scores than the Implanon and LNG-IUS groups at all time points ( p  < 0.001). For arousal, a transient difference emerged between the LNG-IUS and Implanon groups at 1 month ( p  < 0.05), but no significant disparities were observed at 3 or 6 months, suggesting only a short-term effect of LNG-IUS on arousal. In lubrication, the LNG-IUS group consistently scored lower than both the Cu-IUD and Implanon groups, with statistically significant differences from the Implanon group at 1 and 6 months ( p  < 0.05). While the Implanon group outperformed the Cu-IUD group at these time points, differences were not statistically significant. For orgasm, the LNG-IUS group scored lower than both the Cu-IUD and Implanon groups, whereas the Cu-IUD and Implanon groups performed comparably. In satisfaction, the Cu-IUD group exhibited the highest scores, significantly surpassing the LNG-IUS group at all stages and the Implanon group at 3 months ( p  < 0.05). Interestingly, this is the same result as male sexual satisfaction. The LNG-IUS group consistently ranked lowest, with the Implanon group intermediate. In pain, the Implanon group performed best overall, while the LNG-IUS group scored lowest. By 6 months, the Implanon group significantly outperformed both the LNG-IUS ( p  < 0.05) and Cu-IUD groups ( p  < 0.05). In summary, the LNG-IUS group consistently underperformed across all dimensions. The Implanon group excelled in pain reduction, while the Cu-IUD group led in desire and satisfaction. Cu-IUD and Implanon performed similarly in lubrication and orgasm. Arousal scores did not differ significantly between groups long-term, though LNG-IUS showed a temporary decline at 1 month. Generalized estimated equations analysis identified LARCs, abortion history, and changes in the frequency of sexual activity as significant predictors of FSD (Table  4 ). Whereas age, time since placement and time of placemen showed no association with FSD ( p  > 0.05). Further comparative analysis highlighted distinct FSD risks among LARCs. Specifically, LNG-IUS users exhibited the higher probability of experiencing FSD compared to those who chose Implanon ( p  = 0.023). However, no significant differences were found in the probability of FSD between users of Cu-IUD and LNG-IUS, or between users of Implanon and Cu-IUD. Furthermore, the history of abortions ( p  = 0.005) and the decreased frequency of sexual activity ( p < 0.001) were independent risk factors for FSD. Table 4 Generalized Estimating Equation Model for LARCs on FSD B Adjusted OR (95% CI) P Age 0.071 1.07(0.97 ~ 1.19) 0.189 Different LARCs 0.073  Cu-IUD vs. LNG-IUS -1.094 0.33(0.04 ~ 2.95) 0.325  Implanon vs. LNG-IUS -1.636 0.19(0.05 ~ 0.8) 0.023*  Implanon vs. Cu-IUD -0.541 0.58(0.07 ~ 5.18) 0.627 Time since placement 0.257  1 month vs. 6 month 0.638 1.89(0.84 ~ 4.25) 0.122  3 month vs. 6 month 0.347 1.41(0.82 ~ 2.43) 0.209  3 month vs. 1 month -0.291 0.75(0.36 ~ 1.54) 0.429 Time of placement (ref: Pregnancy period) 0.874  Immediately after the abortion -0.923 0.4(0.02 ~ 8.35) 0.553  One month or more after the abortion -0.356 0.7(0.04 ~ 12.03) 0.806  Menstrual period 0.025 1.03(0.09 ~ 12.23) 0.984 Abortion history  Yes (ref: NO ) 2.338 10.36(1.99 ~ 54) 0.005* Changes in the frequency of sexual activity  Decreased (ref: unchanged ) 1.919 6.81(2.44 ~ 19.07) <0.001* Ps. The model was adjusted for age, LARC type, time since placement, time of placement, abortion history, and changes in the frequency of sexual activity. The model used a logit link function and an auto‑regressive working correlation structure to account for repeated measures Abbreviations LARC long-acting reversible contraception, FSD female sexual dysfunction, Cu-IUD copper intrauterine device, LNG-IUS levonorgestrel releasing intrauterine system, OR odds ratio, CI confidence interval * p <0.05 Generalized Estimating Equation Model for LARCs on FSD Ps. The model was adjusted for age, LARC type, time since placement, time of placement, abortion history, and changes in the frequency of sexual activity. The model used a logit link function and an auto‑regressive working correlation structure to account for repeated measures Abbreviations LARC long-acting reversible contraception, FSD female sexual dysfunction, Cu-IUD copper intrauterine device, LNG-IUS levonorgestrel releasing intrauterine system, OR odds ratio, CI confidence interval * p <0.05

The sexual health of women significantly impacts both emotional well-being and overall quality of life. Moderate and healthy sexual activity not only enhances physical health, but also contributes to psychological fulfillment and self-confidence. Conversely, sexual dissatisfaction may have adverse effects on women’s self-esteem, interpersonal relationships, and even reproductive health. Research has found that sexual satisfaction plays an important role in women’s choice of contraceptive method and the duration of its use [ 1 ]. Long-acting reversible contraception (LARC), including intrauterine device and subdermal contraceptive implants, represents one of the most effective contraceptive methods currently, and has become increasingly popular due to its efficacy and convenience. Recent global data indicate that the use of LARC has been steadily increasing [ 2 ]. Among LARC methods, the intrauterine device remains the most commonly chosen contraceptive among women of reproductive age, accounting for approximately 45% of married women [ 3 ]. In recent years, the use of etonogestrel subcutaneous implant has also been rising continuously among adolescents [ 4 ]. However, preferences for specific LARC methods vary across populations and regions. A prospective cohort study conducted in a real-world setting found that history of childbirth and induced abortion are influencing factors for LARC selection, particularly among women under 25 years of age [ 5 ]. Specifically, women with a history of childbirth or induced abortion tend to prefer LARC. In recent years, most studies on LARC have focused on its contraceptive efficacy and acceptability. However, despite its significant impact on daily life, the sexual experience of women using LARC is often overlooked. There is still controversy regarding the impact of LARC on female sexual function. Most studies are cross-sectional and one-sided, lacking the assessment of partners’ sexual experiences and the evaluation of sexual experiences over time. A systematic study showed [ 6 ] that most studies on LARC users reported no change in sexual function, yet 10–16% of the participants reported negative impacts. Additionally, users in other studies believed that LARC could reduce the fear of unplanned pregnancy, facilitating a more relaxed and enjoyable sexual experience, thus improving their sexual function [ 7 ]. Therefore, further research is needed to determine the impact of LARC on female sexual function. We conducted a retrospective study to evaluate the etonogestrel subcutaneous implant (Implanon), levonorgestrel intrauterine system (LNG-IUS), and copper-containing intrauterine device (Cu-IUD) on the sexual function of women of reproductive age.

This study compared the effects of three LARCs on female sexual function, revealing that the LNG-IUS demonstrated the poorest sexual function outcomes relative to the Implanon and Cu-IUD. Regular monitoring of sexual health post-placement is critical, with intensified follow-up recommended within the first 3 months. Clinically, sexual health should be integrated as a core component of contraceptive counseling. Given variations in effects across sexual function dimensions among LARC types, clinicians can provide tailored recommendations aligned with individual patient needs. The superior satisfaction outcomes associated with Cu-IUD further suggest that future development of non-hormonal contraceptives with reduced menstrual side effects could optimize both contraceptive efficacy and quality of life.

This study incorporates “partner sexual satisfaction” into the assessment to examine the impact of three contraceptive methods (LNG-IUS, Cu-IUD, and Implanon) on female sexual function. Behaviorally, we observed that sexual frequency increased progressively over time post-placement regardless of LARC type. Regarding partner satisfaction, women using Cu-IUD reported superior outcomes compared to those using LNG-IUS and Implanon. LNG-IUS users had the lowest total FSFI scores at all assessed time points, while Cu-IUD users scored highest in the desire and satisfaction domains. Both Cu-IUD and Implanon users reported better outcomes in lubrication and orgasm compared to LNG-IUS users. Implanon users scored highest in the pain domain. Finally, multivariate analysis indicated that FSD was associated with the type of LARC used, a history of abortion, and changes in the frequency of sexual activity. Women’s perception of sexuality can be affected by multiple factors, including biological, psychological, environmental factors and partner factors [ 9 ]. In this study, we observed that by the third month, both partner satisfaction and the FSFI total score significantly improved, while changes between the third and sixth months were less pronounced. One possible explanation is that women undergo an initial adaptation period to the contraceptive device. It has been suggested that physical and psychological foreign body sensations during the early stage after placement could contribute to intercourse discomfort and reduced sexual pleasure. Implanon is known to induce systemic progestogenic effects, which might theoretically alter the estrogen-progesterone balance and lead to side effects such as irregular bleeding and dyspareunia. These effects could influence sexual frequency and quality, as suggested by previous reports [ 10 ]. In contrast, LNG-IUS primarily exerts local intrauterine effects and may influence sexual function through local endometrial effects, bleeding patterns, pelvic discomfort, device-related symptoms, or psychological factors. It is plausible that adaptation over time contributes to the observed increase in sexual frequency and satisfaction. Our finding of improved outcomes by the third month is consistent with this interpretation. Our study suggests that the initial adaptation period may last for three months, which could be a critical observation period for assessing hormone level stability and side effects. It suggests that clinicians should emphasize follow-up during the first three months. LNG-IUS showed worse outcomes in terms of sexual function compared to Cu-IUD and Implanon. Enzlin P found that in a cross-sectional study of 402 women who had used LNG-IUS or Cu-IUD for at least 6 months, LNG-IUS was thought to have a greater negative impact on frequency of sexual activity, lubrication, and sexual arousal, but had no impact on orgasm or sexual satisfaction [ 11 ]. However, most studies have not found any negative effects of LARCs on female sexual function [ 10 , 12 ]. In addition to this, in the Implanon group, FSFI total score was found to increase with time. This is consistent with a multi-centre prospective study [ 13 ] reported by BioMed Res Int et al., which found a positive correlation between FSFI total score and Nexplanon use (the commercial name of the etonogestrel implant in the United States, which is bioequivalent to Implanon). Moreover, Nexplanon demonstrated lower metabolic and bleeding effects and also had an important positive impact on sexual function. This study also compared the advantages of different LARCs in each FSFI dimensions. In pain, Implanon showed better results. This is in line with the findings of Di Carlo [ 14 ]. As a subdermal contraceptive implants, Implanon contains 68 mg of etonogestrel and has a contraceptive efficacy rate of over 99.5% for three years. As a third-generation progestin, it significantly reduces endometrial hyperplasia and congestion, thereby decreasing pain during intercourse triggered by pelvic congestion or cervical irritation. In addition, progesterone may reduce pelvic tension by regulating prostaglandin levels, further relieving pain. While both Implanon and LNG-IUS contain progestins, we did not observe positive results with LNG-IUS. One hypothesis is that this is related to discomfort from the caudal filaments or the high local concentration of levonorgestrel. Another hypothesis is that the strong inhibitory effect of LNG-IUS on the endometrium might lead to cervical stiffness or painful uterine contractions during intercourse. However, none of these mechanisms have been directly demonstrated. Importantly, LNG-IUS is clinically effective for adenomyosis-associated dysmenorrhea and heavy menstrual bleeding. Thus, the choice of LARC in patients with pain should be individualized based on the underlying gynecological condition. There were no significant differences between the three contraceptive devices in the arousal, but LNG-IUS produced a transient negative effect during the first month. The transient negative effect could hypothetically be attributed to the initial high-dose release of levonorgestrel, which rapidly suppresses the endometrium and may cause pelvic discomfort or spotting. It has also been speculated that levonorgestrel might reduce sexual arousal motivation by decreasing dopamine activity, but evidence for this specific effect in LNG-IUS users is lacking. These possibilities require dedicated neuroendocrine studies. As for the desire and satisfaction, Cu-IUD showed a better advantage compared to LNG-IUS and Implanon, which is consistent with most studies [ 15 – 17 ]. Cu-IUD demonstrated a non-hormonal advantage. The results of a cross-sectional study in Sweden also found that women using LNG-IUS were five times more likely to subjectively report negative effects on sexual desire, especially on issues related to libido and sexual function, compared to women using Cu-IUD [ 18 ]. The underlying reasons for these findings may be both physical and psychological. Physically, estrogen plays a fundamental role in female sexual behavior and its administration can be recommended as a treatment for low libido and hypoactive sexual desire disorders [ 19 ]. Cu-IUD avoids hormonal interference with the hypothalamic-pituitary-gonadal axis and maintains physiological levels of estradiol and testosterone in the body. While testosterone is positively correlated with lubrication. In contrast, LNG-IUS and Implanon on inhibit ovulation by releasing progesterone, reduce the natural production of estrogen, and increase the level of SHBG in the body, leading to a decrease in free testosterone [ 20 , 21 ]. However, the magnitude of these endocrine changes with LNG-IUS is generally considered limited due to its predominantly local action, and their clinical impact on lubrication remains uncertain. And psychologically, both Implanon and LNG-IUS have been reported to have side effects on mood, as evidenced by deterioration of mood, and even psychiatric adverse events such as anxiety and depression [ 16 , 17 , 22 ]. One speculative explanation is that fluctuations in progesterone release after placement might enhance hypothalamic-pituitary-ovarian axis reactivity and increase cortisol levels. On the other hand, the “non-pharmacological” characteristics of the Cu-IUD may enhance a woman’s sense of control over her body [ 10 ] and increase her sexual confidence. While several biological and device-related mechanisms have been hypothesized to explain the observed differences in sexual function among LARC users, the current evidence for each mechanism is limited and largely indirect. In the lubrication and orgasm, LNG-IUS consistently scored lower than Cu-IUD and Implanon at all time points, while Cu-IUD and Implanon showed similar results. Sakinci and Malmborg et al. [ 18 , 23 ] also concluded that LNG-IUS scores were lower than Cu-IUD. In contrast, Gynecol Endocrinol et al. [ 1 ] found that Implanon had a lower lubrication score than Cu-IUD. Although Implanon and LNG-IUS are also used as hormonal contraceptive devices, both have different levels of systemic progesterone exposure throughout the body and different degrees of ovulation suppression. While it is true that intrauterine levonorgestrel concentrations with LNG-IUS are orders of magnitude higher than plasma levels, whether this leads to clinically significant ovarian suppression, vaginal mucosal atrophy, or increased risk of dyspareunia and lubrication difficulties is debated. Some authors have proposed that retrograde flow or local absorption could reduce estradiol levels, but consistent evidence is lacking. As for orgasm, most studies reported no effect of LNG-IUS on orgasm [ 18 , 24 ]. Given the differences among LARCs in various dimensions, clinicians can offer more personalized recommendations when choosing an LARC method, taking into account both sexual function preferences and gynecological indications. For example, Implanon may be considered for women whose primary concern is painful intercourse in the absence of specific uterine conditions that would favor LNG-IUS. Cu-IUD may be recommended for women who are sensitive to hormonal effects on sexual desire and satisfaction. LNG-IUS, despite showing poorer sexual function outcomes in this study, remains an appropriate choice for women with heavy menstrual bleeding, dysmenorrhea, or adenomyosis, where its non-contraceptive benefits may outweigh potential sexual side effects. For individuals with poor short-term adaptability, LNG-IUS should be used with caution or under enhanced early follow-up, but this does not preclude its use when clinically indicated. Epidemiological studies found that 41% of women of reproductive age worldwide are affected by FSD [ 25 ]. In our study, we found that the probability of FSD was higher in LNG-IUS group. As mentioned earlier, the non-hormonal advantage of the Cu-IUD can explain part of the results. Both LNG-IUS and Implanon are progestin-based contraceptive methods, but there are significant differences between their hormone release patterns, systemic exposure levels and effects on the reproductive endocrine axis, which may be the core reason for the higher incidence of FSD in the LNG-IUS group. Pharmacological differences between levonorgestrel and etonogestrel are well documented: levonorgestrel has higher androgenic activity and greater SHBG binding affinity, which can reduce free testosterone; etonogestrel is less androgenic and induces less SHBG. However, whether these differences translate into clinically meaningful effects on sexual desire or arousal in LNG-IUS versus Implanon users is not firmly established. Similarly, the hypothesis that LNG-IUS preferentially inhibits central dopaminergic pathways while Implanon does not remains unproven. Implanon, on the other hand, has a higher molecular polarity, lower blood-brain barrier permeability, and a weaker effect on the central dopamine system, thus preserving more sexual arousal. Interestingly, we found that the results of the different contraceptive devices on partner sexual satisfaction and women’s own sexual satisfaction were completely consistent, with the Cu-IUD group being significantly higher than the LNG-IUS and Implanon groups. This finding is consistent with the theory of the “Bidirectional Sexual Feedback Model”. Sexual health is essentially relational rather than a purely individual physiological issue, and the simultaneous effect of the Cu-IUD on partners’ sexual satisfaction emphasizes the central role of partner interaction in sexual function. Widman, L et al. reported a positive association between contraceptive use and sexual communication, and that the use of an intrauterine device may improve partner satisfaction by enhancing their sexual communication [ 26 ]. It must be acknowledged that this study has several limitations.First, the study allowed participants to choose their LARC method voluntarily. Thus, selection bias and confounding by indication cannot be completely ruled out. In particular, the LNG‑IUS group had a higher proportion of women with pre‑existing gynecological conditions such as adenomyosis, which may independently affect sexual function. Although we explicitly excluded women with significant dyspareunia or dysmenorrhea at enrollment and demonstrated that baseline FSFI scores (including the pain domain) did not differ among the three groups, residual confounding may still exist. Second, we only included LARC methods and did not set a non‑LARC control group (e.g., barrier methods). Therefore, we cannot determine whether the sexual function of LARC users differs from that of women using other contraceptive approaches. Third, the follow‑up period was limited to six months. The cumulative effects of hormonal contraception, such as vaginal atrophy, may continue to influence sexual experience beyond six months. Despite these limitations, the study provides valuable real‑world evidence on the comparative effects of LARC methods on sexual function, with careful attention to baseline comparability and exclusion of symptomatic patients. Future randomized controlled trials are needed to confirm these causal effects.