Association between pre-pregnancy uterine volume and preterm birth in women with adenomyosis: a retrospective cohort study

This retrospective cohort study was conducted at the Department of Obstetrics, Shanghai First Maternity and Infant Hospital. To investigate the correlation between pre-pregnancy uterine volume and pregnancy outcomes in patients with adenomyosis, pregnant women with a singleton pregnancy between January 2014 and December 2022 who received a pre-pregnancy diagnosis of adenomyosis based on transvaginal ultrasound (TVUS) scan were enrolled. Patient data were obtained from the obstetrics department’s electronic medical records. Data were retrieved for women who were prenatally managed and delivered after 24 weeks of gestation. The exclusion criteria were as follows: chronic hypertension, multiple pregnancies, uterine malformation [ 18 ] (congenital Müllerian tract anomalies: unicornuate, bicornuate septate, or arcuate uterus, or uterine septa), uterine surgery (myomectomy or resection of adenomyosis lesions), uterine myoma (multiple uterine fibroids ≥ 3 or affecting the endometrium), ovarian endometriomas which caused infertility, fetal anomalies, cervical incompetency and pregnancies with missing or incomplete follow up data. Women with a history of preterm birth and second-trimester miscarriage were also excluded. The sonographic diagnostic criteria for women with adenomyosis were consistent with those of previous reports [ 17 , 19 ]. The direct sonographic appearance of adenomyosis included: myometrial cysts, hyperechogenic islands, echogenic sub-endometrial lines, and buds. The indirect sonographic appearance of adenomyosis included: a globular uterus, asymmetrical myometrial thickening, fan-shaped shadowing, translesional vascularity, and an irregular or interrupted junctional zone. A diagnosis of adenomyosis was confirmed by at least two experienced sonographers. Medical records were collected, including maternal age, pre-pregnancy body mass index (BMI), gravidity, parity (at time of delivery), mode of conception, adenomyosis type, and pregnancy complications. Ultrasonographic examinations were performed at any stage of the menstrual cycle, regardless of the use of hormonal therapy. The recommended window for ultrasound evaluation is less than 6 months prior to conception. The morphological features of adenomyosis were recorded with a TVUS scan to determine the classification and size of the uterus, including the uterine body length (L), from the fundus to the internal os ; and the transverse (T) and anteroposterior (AP) diameters (cm). The uterine volume was calculated using a geometric formula for a prolate ellipsoid: uterine volume (cm 3 ) = L ×T × AP × π/6 [ 20 ]. The type of adenomyosis was classified as focal or diffuse according to the adenomyotic lesion [ 21 ]. Focal adenomyosis (including adenomyoma) was subdivided into grossly circumscribed adenomyotic masses within the myometrium and cystic adenomyosis. Diffuse adenomyosis was defined as typical adenomyotic changes that spread throughout the uterine musculature. The analysis of obstetric complications included PTB (delivery at < 37 gestational weeks), gestational diabetes mellitus (GDM), pregnancy induced-hypertension (including gestational hypertension (GH), pre-eclampsia, and eclampsia), placental malposition (placenta previa or low-lying placenta that required caesarean delivery), PPROM, and placental abruption. GDM was defined as fasting blood glucose level ≥ 92 mg/dL, 1-h post 75 g-glucose load ≥ 180 mg/dL, or 2-h post 75 g-glucose load ≥ 153 mg/dL using 75-g oral glucose. Data were analyzed using the R statistical packages (The R Foundation; http://www.r-project.org ; version 3.4.3) and EmpowerStats ( www.empowerstats.com ; X&YSolutions Inc.). P  < 0.05 was considered statistically significant. Continuous variables are presented as mean ± standard deviation or median (interquartile range for skewed distribution). Categorical variables are presented as frequency and percentage (%) and were compared using the chi-square or Fisher’s exact test. The Mann–Whitney U -test was used to analyze continuous variables. Logistic regression models were used to calculate the odds ratios (ORs) between uterine volume and PTB. The associations between uterine volume and outcomes were assessed using uterine volume as a continuous variable. The model included each covariate one at a time, and covariates were included as potential confounders in the final models if they changed the estimates of clinical outcomes by more than 10% (Criteria 1) or were associated with the outcomes ( P -value for regression coefficients < 0.1, Criteria 2). Based on clinical perspective and the results of previous studies, we select causal directed acyclic graphs (DAG) (Criteria 3, in Figure S1), which can help researchers identify confounders and mediators, and verify the stability of regression equations. To investigate whether the uterine volume was correlated with the outcomes, a multivariate logistic regression model was employed. Following the recommendations of the STROBE statement [ 22 ], we obtained the results of unadjusted (Crude Model), minimally adjusted (Model I, adjusted for covariates from Criteria 1), and fully adjusted analyses (Model II, adjusted for covariates from Criteria 1 and Criteria 2) and those from clinical adjusted analyses (Model III, adjusted for covariates from Criteria 3). The non-linear relationships between each continuous variable and outcomes were compared using smooth curve fitting (penalized spline method). If relationship was nonlinear, this variable was adjusted by curve fitting. Effect modification analyses were performed for covariates where a differential effect of pre-pregnancy uterine volume was plausible. Based on clinical cut points, we converted it to categorization and then performed an interaction test effect correction for those subgroup measures, followed by a likelihood test. The relationships between uterine volume and primary outcome in different subgroups were compared using smooth curve fitting (penalized spline method).

735 pregnant women with adenomyosis were included in this study. From this population, 149 women were excluded from the study population. Finally, 586 women with ultrasonographic signs of pre-pregnancy adenomyosis were analyzed (Flow chart of the study is visualized in Fig.  1 ). The PTB rate was 15.19% (89/586). The median age of participants was 34.0 years (interquartile range, 31.0–37.0), and the median pre-pregnancy uterine volume was 84.8 cm 3 (interquartile range, 67.0–101.3). Fig. 1 Flow chart of the study Flow chart of the study Table  1 compares the maternal demographics, pregnancy complications, and factors associated with PTB. In the PTB group, the median pre-pregnancy uterine volume was 108.9 cm 3 (interquartile range, 91.9–119.2), which was higher than that in the term birth group 85.3cm 3 (interquartile range 64.1–96.1) ( P  < 0.001). Significantly more women in the PTB group received infertility treatment than those in the term birth group (34.8% vs. 24.1%, P  = 0.034). The prevalence of pre-eclampsia or eclampsia (13.5% vs. 5.2%, P  = 0.004), placental malposition (32.6% vs. 8.9%, P  < 0.001), and PPROM (28.1% vs. 14.3%, P  = 0.001) was significantly higher in the PTB group than that in the term birth group. No differences were noted in age, pre-pregnancy BMI, gravidity, parity, GDM, GH, placentation abruption, and type of adenomyosis. Table 1 Demographic characteristics and obstetric complications of the participants Total Preterm Birth Term Birth P -value Characteristics n  = 586 n  = 89 n  = 497 Age (years) 34.0 (31.0–37.0) 34.9 (32.0–37.5) 34.0 (31.0–36.0) 0.056 Pre-pregnancy BMI (kg/m 2 ) 21.5 (20.0–23.4) 21.7 (20.0–23.0) 22.0 (19.9–23.4) 0.296 Gravidity , n (%) 0.434  1–2 377 (64.3) 54 (60.7) 323 (65.0)  ≥ 3 209 (35.7) 35 (39.3) 174 (35.0) Parity , n (%) 0.510  1–2 572 (97.6) 86 (96.6) 486 (97.8)  ≥ 3 14 (2.4) 3 (3.4) 11 (2.2) Type of adenomyosis , n (%) 0.285  Focal 508 (86.7) 74 (83.1) 434 (87.3)  Diffuse 78 (13.3) 15 (16.9) 63 (12.7)  Uterine volume (cm 3 ) 84.8 (67.0–101.3) 108.9 (91.9–119.2) 85.3 (64.1–96.1) < 0.001 *  Fertility treatment , n (%) 151 (25.8) 31 (34.8) 120 (24.1) 0.034 *  GDM , n (%) 118 (20.1) 22 (24.7) 96 (19.3) 0.242  GH , n (%) 25 (4.3) 3 (3.4) 22 (4.4) 0.650  Pre-eclampsia or eclampsia , n (%) 38 (6.5) 12 (13.5) 26 (5.2) 0.004 *  Placental malposition , n (%) 73 (12.5) 29 (32.6) 44 (8.9) < 0.001 *  PPROM , n (%) 96 (16.4) 25 (28.1) 71 (14.3) 0.001 *  Placental abruption , n (%) 5 (0.9) 1 (1.1) 4 (0.8) 0.563 Data are reported as median (Q1-Q3)/N (%) for continuous variables or n (%) for categorical variables BMI body mass index, GDM gestational diabetes mellitus, GH gestational hypertension, PPROM  preterm premature rupture of membrane * P  < 0.05 Demographic characteristics and obstetric complications of the participants Data are reported as median (Q1-Q3)/N (%) for continuous variables or n (%) for categorical variables BMI body mass index, GDM gestational diabetes mellitus, GH gestational hypertension, PPROM  preterm premature rupture of membrane * P  < 0.05 The results of the univariate analysis are summarized in Supplementary Table S1. Univariate analysis revealed an increased risk of PTB in women with large uterine volume. With each 10 cm 3 increment in uterine volume, the risk of PTB increased by 35% (OR:1.35, 95% confidence interval [CI]: 1.24–1.47). Fertility treatment, pre-eclampsia or eclampsia, placental malposition, and PPROM were correlated with PTB ( P   0.05). Smooth curve fitting plots (Supplementary Figures S2-S3) suggested that age and pre-pregnancy BMI were linearly related to PTB. Based on the above results and those of previous studies, we performed a multivariate logistic analysis to further explore the relationship between pre-pregnancy uterine volume and PTB. The non-adjusted and adjusted models in the multivariable analysis are listed in Table  2 . In the non-adjusted model, uterine volume was positively correlated with the occurrence of PTB (OR = 1.35, per 10 cm 3 increase, 95% CI: 1.24–1.47, P  < 0.001). Uterine volume was an independent risk factor for PTB in Model I (OR adj = 1.37, per 10 cm 3 increase, 95% CI: 1.25–1.51, P  < 0.001), Model II (OR adj = 1.40, per 10 cm 3 increase, 95% CI: 1.27–1.55, P  < 0.001) and Model III (OR adj = 1.36, per 10 cm 3 increase, 95% CI: 1.25–1.48, P  < 0.001). We found that the relationship between uterine volume and PTB was non-linear (after adjusting for age, pre-pregnancy BMI, gravidity, parity, fertility treatment, GDM, GH, pre-eclampsia or eclampsia, placenta malposition, PPROM, placental abruption, and type of adenomyosis) (Fig.  2 ). Based on a two-piecewise linear regression model, after adjusting for all potential covariates, uterine volume remained highly correlated with PTB, with an inflection point of 89.34 cm 3 . There may exist a clinically significant threshold effect. The Adjust ORs(95%CI) were 1.16 (1.10 to 1.24) for pre-pregnancy uterine volume less than 89.34 cm 3 and 1.01 (0.99 to 1.02) for uterine volume of 89.34 cm 3 or greater (Table  3 ). Table 2 Multivariable regression of pre-pregnancy uterine volume associated with preterm birth Preterm Birth Crude Model Model I Model II Model III OR (95% CI) P -value OR (95% CI) P -value OR (95% CI) P -value OR (95% CI) P -value Uterine volume (per 10 cm 3 ) 1.35 (1.24, 1.47) < 0.001* 1.37 (1.25, 1.51) < 0.001* 1.40 (1.27, 1.55) < 0.001* 1.36 (1.25,1.48) < 0.001* Crude Model adjusted for: None Model I adjusted for: Age; Fertility treatment; Pre-eclampsia or eclampsia; Placental malposition; Preterm premature rupture of membrane Model II adjusted for: Age; Pre-pregnancy body mass index; Gravidity; Parity; Fertility treatment; Gestational diabetes mellitus; Gestational hypertension; Pre-eclampsia or eclampsia; Placental malposition; Premature rupture of membrane; Placental abruption; Type of adenomyosis Model III adjust for: Age; Gestational hypertension; Pre-pregnancy BMI. * P  < 0.05 Fig. 2 A non-linear association between pre-pregnancy uterine volume and PTB. Red band represents the smooth curve fit between variables. Blue bands represent the 95% confidence interval from the fit Table 3 Threshold effect analysis of pre-pregnancy uterine volume on preterm birth Pre-pregnancy Preterm Birth Uterine volume , cm 3 Adjust OR (95%CI) P -value < 89.34 1.16 (1.10, 1.24) < 0.001 * ≥ 89.34 1.01 (0.99, 1.02) 0.308 Model adjusted for: Age; Pre-pregnancy BMI; Gravidity; Parity; Fertility treatment; Gestational diabetes mellitus; Gestational hypertension; Pre-eclampsia or eclampsia; Placental malposition; Preterm premature rupture of membrane; Placental abruption and Type of adenomyosis. * P  < 0.05 Multivariable regression of pre-pregnancy uterine volume associated with preterm birth Model I adjusted for: Age; Fertility treatment; Pre-eclampsia or eclampsia; Placental malposition; Preterm premature rupture of membrane Model II adjusted for: Age; Pre-pregnancy body mass index; Gravidity; Parity; Fertility treatment; Gestational diabetes mellitus; Gestational hypertension; Pre-eclampsia or eclampsia; Placental malposition; Premature rupture of membrane; Placental abruption; Type of adenomyosis Model III adjust for: Age; Gestational hypertension; Pre-pregnancy BMI. * P  < 0.05 A non-linear association between pre-pregnancy uterine volume and PTB. Red band represents the smooth curve fit between variables. Blue bands represent the 95% confidence interval from the fit Threshold effect analysis of pre-pregnancy uterine volume on preterm birth Model adjusted for: Age; Pre-pregnancy BMI; Gravidity; Parity; Fertility treatment; Gestational diabetes mellitus; Gestational hypertension; Pre-eclampsia or eclampsia; Placental malposition; Preterm premature rupture of membrane; Placental abruption and Type of adenomyosis. * P  < 0.05 To evaluate the potential influences of other factors, a sub-analysis was conducted stratifying patients by type of adenomyosis, fertility treatment, pre-eclampsia or eclampsia, placental malposition, PPROM and placental abruption as presented in Fig.  3 . The increased risk was consistent across age and obstetric complications groups, among fertility treatment users and non-users, as well as type of adenomyosis, with point estimates greater than 1 (Fig.  3 ). However, subgroup analysis demonstrated that the type of adenomyosis (focal or diffuse) modified the association between uterine volume and preterm birth, pregnant women with focal adenomyosis exhibited a higher OR compared to those with diffuse adenomyosis (OR = 1.43; 95% CI: 1.29–1.58 vs. OR = 1.10; 95% CI: 0.91–1.33; p for interaction = 0.021) (Fig.  3 and Figure S4). Fig. 3 Subgroup analysis for the effect of Pre-pregnancy uterine volume on PTB in different subgroups. A The above model adjusted for none. B The above model adjusted for confounding factors in Table  2 (Model III: Age; Gestational hypertension; Pre-pregnancy BMI). In each case, the model is not adjusted for the stratification variable Supporting Information Subgroup analysis for the effect of Pre-pregnancy uterine volume on PTB in different subgroups. A The above model adjusted for none. B The above model adjusted for confounding factors in Table  2 (Model III: Age; Gestational hypertension; Pre-pregnancy BMI). In each case, the model is not adjusted for the stratification variable Supporting Information

Adenomyosis is a challenging gynecological disorder characterized by the presence of endometrial tissue within the myometrium, often accompanied by smooth muscle cells hyperplasia and hypertrophy, leading to uterine enlargement [ 1 ]. According to the extent and location of the adenomyotic lesions and histological classification, focal or diffuse adenomyosis is diagnosed [ 2 ]. Adenomyosis can negatively impact the quality of life [ 3 ] by causing pelvic pain (chronic pelvic pain, dyspareunia, dysmenorrhea), abnormal uterine bleeding and impaired reproduction [ 4 ]. With the advancement of imaging techniques, ultrasonographic adenomyosis is increasing diagnosed in young women, affecting 20.9–34% of childbearing age women [ 5 , 6 ]. Additionally, adenomyosis is reportedly associated with some adverse pregnancy outcomes, such as small for gestational age infants, preeclampsia, preterm premature rupture of membranes (PPROM) and preterm birth (PTB) [ 7 – 9 ]. Previous studies have demonstrated the influence of uterine volume on pregnancy outcomes of women with adenomyosis. Prior to undergoing a frozen-thawed embryo transfer (FET), those with a uterine volume greater than 98.81 cm 3 had a higher rate of miscarriage and a lower rate of live births [ 10 ]. Patients with adenomyosis and a uterine volume larger than 102.2 cm 3 were associated with a lower live birth rate [ 11 ]. Ni et al. pointed that adenomyosis was associated with a greater risk of PTB and spontaneous PTB < 37 weeks of gestation, especially those with adenomyosis volume ≥ 94.7 cm 3 [ 12 , 13 ]. Uterine volume may be a risk factor for PTB. PTB is defined as a live birth at a gestational age of less than 37 weeks [ 14 ]. The global incidence of PTB is approximately 11%, and each year, approximately 15 million babies are born preterm worldwide. Statistically, 18% of child deaths under the age of 5 years and 35% of neonatal deaths are associated with PTB. Prematurity imposes a substantial economic and mental burden on families and society [ 15 ]. The normal uterine volume range is shown to be 15–56 cm 3 [ 16 ]. Uterine enlargement caused by hypertrophic and hyperplastic smooth muscle cells, results in an increase in uterine volume, is considered an important feature of adenomyosis [ 17 ]. Adenomyosis with a large uterine volume, results in reduced pregnancy and live birth rates, and an increased miscarriage rate. However, the extent of correlation between pre-pregnancy uterine volume and adverse pregnancy outcomes in adenomyosis, and whether the impact on pregnancy outcomes varies among different adenomyosis types remains uncertain. Furthermore, the association between pre-pregnancy uterine volume and PTB in women with adenomyosis is not well elucidated. This study aimed to investigate the association between pre-pregnancy uterine volume and the risk of preterm birth in women with adenomyosis, including whether this relationship varies by adenomyosis type.

This study demonstrated that increased pre-pregnancy uterine volume in women with adenomyosis was associated with a higher risk of preterm delivery. When assessed individually, conventional risk factors for PTB, such as fertility treatment, pre-eclampsia or eclampsia, placental malposition, and PPROM, showed strong correlations with PTB in women with adenomyosis. After adjusting for potential confounding variables, logistic regression analysis revealed pre-pregnancy uterine volume was an independent risk factor for PTB. With the increasing incidence of adenomyosis, there are sufficient evidences that this disease adversely affects pregnancy outcomes. In a systematic review and meta-analysis, Razavi et al. reported the detrimental impact of adenomyosis on pregnancy outcomes, including a higher likelihood of PTB (OR = 3.05, 95% CI: 2.08–4.47, P ˂0.001), small-for-gestational age infant (OR = 3.22, 95% CI: 1.71–6.08, P ˂0.001), and pre-eclampsia (OR = 4.35, 95% CI: 1.07–17.72; P  = 0.042) [ 23 ]. In another meta-analysis, Nirgianakis et al. reported the adverse effect of adenomyosis on clinical pregnancy (OR = 0.69, 95% CI: 0.51–0.94) and miscarriage rates (OR = 2.17, 95% CI: 1.25–3.79) after treatment with assisted reproductive technology (ART) [ 24 ]. In addition, a 1:2 nested case-control study based on a cohort population of 2138 pregnant women, showed that women with adenomyosis had a 1.84-fold (95% CI: 1.32–4.31) increased risk of preterm delivery and a 1.98-fold (95% CI: 1.39–3.15) increased risk for PPROM [ 9 ]. A multicenter case-control study reported that adenomyosis group ( n  = 61) had significantly higher incidence of preterm delivery (21.3% vs. 9.4%), hypertensive disorders of pregnancy (13.1% vs. 5.3%), cesarean delivery (46.0% vs. 20.9%), and postpartum hemorrhage (57.3% vs. 36.8%) than control group ( n  = 244) [ 25 ]. Despite the evidence of adverse pregnancy outcomes, there are currently no well-established strategies for early warning, risk stratification, or prevention of these complications. Also, these studies did not focus on the association between uterine volume and preterm pregnancy outcomes. Our study focused on the pre-pregnancy assessment of adenomyosis using ultrasound and performed a sub-analysis of the risk of PTB in our cohort of women, with data stratified according to uterine volume and type of adenomyosis. Pregnancies with severe adenomyosis as reflected by volume ≥ 94.7 cm 3 had a significantly greater risk of PTB < 37 weeks of gestation [ 12 , 13 ], the adenomyosis volume was measured up to 14 weeks of gestation which was different from the preconception uterine volume we performed, but the conclusion was consistent with us, pregnancies complicated by adenomyosis, particularly those with large adenomyosis volume, had an increased risk of preterm birth. Notably, our study included a total of 586 cases, more than double the 226 adenomyosis cases reported in their study, which could decrease the heterogeneity of the study and enhance the credibility and reliability of results. Additionally, we conducted a subgroup analysis to investigate the impact of different types of adenomyosis on PTB. Severe obstetric complications such as placenta previa, placenta accreta, preeclampsia, and preterm birth were only found in women with adenomyosis located in the posterior side [ 26 ], which is similar with our findings, focal adenomyosis may be a risk factor for PTB. It is possible that a focal lesion in the uterus contributes to a significant discordance in myometrial contractions [ 27 ], which impairs placental implantation during pregnancy. But we need to increase the sample size to verify this point. Women with adenomyosis are presenting more frequently for fertility issues; ART is a risk factor for several adverse pregnancy outcomes, including pre-eclampsia, antepartum hemorrhage, placenta previa, PTB, low birthweight, and childbirth interventions [ 28 ]. A meta-analysis confirmed that pregnancies by ART were at higher odds of developing hypertensive disorders of pregnancy and pre-eclampsia than spontaneous conception, irrespective of the plurality [ 29 ]. In our study, more women required fertility treatment in the PTB group compared to the term birth group (34.8% vs. 24.1%), which aligns with the findings of the previous study. The univariate analysis also demonstrated that fertility treatment was a risk factor for PTB ( P  = 0.035). Abnormal hormonal and inflammatory mechanisms in the endometrium and myometrium may play a role in the process. Several mechanisms seem to be implicated in the link between adenomyosis and obstetric complications, including activation of local and systemic inflammatory pathways, increased myometrial prostaglandin production, altered uterine contractility and defective myometrial spiral artery remodeling on the basis of altered placentation [ 30 ]. Uterine volume reflects the severity of adenomyosis. An increase in uterine volume is associated with lesion accumulation and further deterioration of molecular expression and tissue function, potentially exerting a more pronounced adverse impact on pregnancy outcomes. Further studies could explore whether it is beneficial to treat large uterine volumes pre-pregnancy for women with adenomyosis. Pre-pregnancy medical intervention [ 31 ], surgical treatment [ 32 ], high-intensity-focused ultrasound method [ 33 ], levonorgestrel-releasing intrauterine system [ 34 ], or a combination of these treatments may reduce the uterine volume and improve the intrauterine environment. Previous research has demonstrated that gonadotropin-releasing hormone agonist (GnRH-a) pretreatment can effectively decrease uterine volume and enhance pregnancy outcomes in infertile women with adenomyosis [ 35 ].This may create more optimal conditions for embryo implantation and the subsequent process of pregnancy. However, the extent to which the uterine volume should be reduced in patients with adenomyosis is unknown. A previous study reported that women with adenomyosis with larger uterine volumes (≥ 98.81 cm 3 ) prior to FET might have a lower live birth rate due to a higher incidence of miscarriage [ 10 ]. Nevertheless, what range of pre-pregnancy uterine volume in adenomyosis is associated with PTB remains unclear. Our results showed that the inflection point of uterine volume was 89.34 cm 3 . This may be because, the larger the uterus, the harder it is for patients with adenomyosis to conceive, or possibly, more fertility intervention is required to be pregnant. Based on the current results, the risk of preterm birth may differ between type of adenomyosis when uterine volume reaches a certain threshold. Further studies are needed to investigate the pregnancy outcomes of the two groups based on the cut-off value of uterine volume. The limitations of this study were that the data were retrospectively analyzed, and selection bias was a possibility, which may affect the stability and generalizability of statistical results. Additionally, reliance on TVUS examination may result in the underdiagnosis of mild or occult adenomyosis, and could potentially lead to false-positive findings. However, the effect of this limitation is presumably modest because TVUS has high levels of sensitivity and specificity to detect adenomyosis [ 36 ]. Third, preterm births were not classified as either spontaneous or iatrogenic, which could potentially introduce bias into the results. And further stratification of PTB severity is necessary to understand its clinical implications. Given the limited sample size of diffuse adenomyosis cases in our study, future research utilizing larger cohorts will be necessary to comprehensively elucidate the association between pre-conceptional uterine volume and preterm birth within this specific patient population. Moreover, further research and prospective clinical studies are needed to find the optimal cut-off value, and confirm the value of monitoring and reducing uterine volume pre-pregnancy for women with adenomyosis to prevent PTB. Despite these limitations, our study focused on the effect of preconception uterine volume on preterm birth in adenomyosis, which may provide some reference for doctors and adenomyosis patients when planning for pregnancy.

The results demonstrated that increased uterine volume before pregnancy may be a potential risk factor for PTB in women with adenomyosis. We believe that this study will provide valuable information to clinicians, reducing the large uterine volume prior to pregnancy in women with adenomyosis may potentially improve adverse pregnancy outcomes. Additionally, these findings provide valuable evidence that gravid women with adenomyosis require more careful perinatal management than previously thought.

Supplementary Material 1: Figure S1 Directed acyclic graph (DAG) illustrating confounder selection. Supplementary Material 1: Figure S1 Directed acyclic graph (DAG) illustrating confounder selection. Supplementary Material 2: Figure S2 Smooth curve fitting plot between age and preterm birth. Supplementary Material 2: Figure S2 Smooth curve fitting plot between age and preterm birth. Supplementary Material 3: Figure S3 Smooth curve fitting plot between pre-pregnancy BMI and preterm birth. Supplementary Material 3: Figure S3 Smooth curve fitting plot between pre-pregnancy BMI and preterm birth. Supplementary Material 4: Figure S4 Interaction test for the effect of pre-pregnancy uterine volume and PTB in different type of adenomyosis subgroup. Supplementary Material 4: Figure S4 Interaction test for the effect of pre-pregnancy uterine volume and PTB in different type of adenomyosis subgroup. Supplementary Material 5: Table S1 Univariate analysis of pre-pregnancy uterine volume and preterm birth. Supplementary Material 5: Table S1 Univariate analysis of pre-pregnancy uterine volume and preterm birth. Supplementary Material 6. Supplementary Material 6.