Abstract Study question To compare the effectiveness of levonorgestrel-releasing intrauterine system (LNG-IUS 52mg) with hysteroscopic niche resection in the reduction of niche-related postmenstrual spotting. Summary answer In women with niche-related postmenstrual spotting, LNG-IUS did not more often reduce spotting days with 50% at 6th month than hysteroscopic niche resection. What is known already Both Levonorgestrel intrauterine system (LNG-IUS) and hysteroscopic niche resection are widely implemented to reduce niche-related postmenstrual spotting. The study aimed to compare the effectiveness of LNG-IUS 52mg with hysteroscopic niche resection in the reduction of niche-related postmenstrual spotting. Study design, size, duration A randomised, open-label, controlled trial conducted at one medical center in Shanghai, China. From September 2019 to January 2022, we randomised 208 women to LNG-IUS (N = 104) or hysteroscopic niche resection (N = 104).The primary outcome was reduction of postmenstrual spotting at 6th month after randomisation, defined as percentage of women with a reduction of at least 50% in spotting days from baseline. Efficacy and safety were assessed by an intention-to-treat analyses. Participants/materials, setting, methods Women with symptoms of postmenstrual spotting after CS, with a niche depth of at least 2 mm and residual myometrium of at least 2.2mm on Magnetic Resonance Imaging (MRI), and no intention to conceive within the next year were randomly assigned to treatment with LNG-IUS 52mg or a hysteroscopic niche resection. Main results and the role of chance At 6th month follow-up, a 50% reduction of spotting had occurred in 78.4% (80/102) women in the LNG-IUS group and 73.1% (76/104) women in the hysteroscopic niche resection group (RR = 1.07, [95%CI 0.92-1.25]; P = 0.370). Spotting reduced over time (Ptrend=0.001), with a stronger reduction in the LNG-IUS group (P = 0.001), while there was also a significant interaction between time and treatment (P = 0.007). From nine months onwards, reduction of spotting occurred significantly more after LNG-IUS than hysteroscopic niche resection (9th month 89.2% vs. 72.1%, RR = 1.24, [95% CI 1.08-1.42]; 12th month 90.2% vs. 70.2%, RR = 1.29, [95% CI 1.12-1.48]). Moreover, compared with the hysteroscopic niche resection group, the LNG-IUS group had significantly fewer postmenstrual spotting days and total bleeding days from six months onwards (all P < 0.001), and less pelvic pain from three months onwards (all P < 0.010). Intervention-related complications were not reported in any of the groups. During follow-up, eleven (10.8%) women reported hormonal related side-effects and two (2.0%) women had spontaneous partial expulsion in the LNG-IUS group, while in the hysteroscopic niche resection group three unintended pregnancies were reported. Limitations, reasons for caution Due to the nature of the intervention, it was not possible to blind participants and gynaecologists for treatment allocation.The cut-off value chosen for our primary outcome and the moment of assessment for our primary endpoint can both be debated. Wider implications of the findings LNG-IUS provides an alternative treatment for women with niche-related gynaecological symptoms and no active desire to become pregnant. Trial registration number Chinese Clinical Research Center (ChiCTR1900025677)