{"paper_id":"312d9299-5973-4c0c-96c9-c4c4beed30e9","body_text":"1\nVol.:(0123456789)Scientific Reports |         (2023) 13:9744  | https://doi.org/10.1038/s41598-023-36838-2\nwww.nature.com/scientificreports\nCorrelation analysis \nof hysterectomy and ovarian \npreservation with depression\nYunhong Yang 1, Xiangqi Zhang 1, Yinuo Fan 1, Jiahao Zhang 1, Bingchun Chen 2, \nXiaofeng Sun 3* & Xiaofeng Zhao 3*\nThe relationship between hysterectomy and ovarian preservation and depression is controversial. This \nstudy aimed to determine the association of hysterectomy and ovarian preservation with depression \nusing National Health and Nutrition Examination Survey. To assess the association between \nhysterectomy with or without ovariectomy and depression, we used 3 methods. Method 1: propensity \nscore model (PSM) was established. Method 2 was logistics regression analysis of hysterectomy and \ndepression before and after PSM. Method 3 was a logistics regression analysis of the relationship \nbetween hysterectomy and different depressive symptoms. At the same time, in order to evaluate \nthe association between hysterectomy with or without oophorectomy and depression, we explored \nthe effect of four different surgical procedures on depression using logistic regression equations. \nWe enrolled 12,097 women, of whom 2763 underwent hysterectomy, 34.455% were positive for \ndepression. After weighting, 33.825% of the total sample had a PHQ ≥ 5. Finally, a total of 2778 \nwomen were successfully matched by propensity score, and 35.537% of them were positive for \ndepression. The OR for PHQ ≥ 5 was 1.236 after crude adjustment of covariates and 1.234 after exact \nadjustment. This suggests that Hysterectomy is strongly associated with positive depression. Positive \ndepression (PHQ ≥ 5) was associated with little interest, feeling down and trouble concentrating. \nIt was not associated with trouble sleeping, feeling tired, poor appetite, feeling bad, slow moving \nor speaking, and suicidal thoughts. Oophorectomy-alone is not associated with depression. \nHysterectomy-alone is a risk factor for depression, but Hysterectomy combined with Oophorectomy \nhas a stronger correlation with depression than Hysterectomy-alone. Women who have had a \nHysterectomy are at higher risk of depression than women who have not had a Hysterectomy, and this \nrisk may be exacerbated if the uterus and ovaries are removed. When clinically appropriate, surgeons \nshould try to preserve the patient’s ovaries.\nHysterectomy is a frequently performed gynecological  procedure1, primarily indicated for perimenopausal uter-\nine fibroids, adenomyosis, and other conditions with high recurrence rates. Uterine fibroids and adenomyosis are \nmost prevalent in women aged 45–49  years2. In view of uterine aging and patients’ fear of malignant transforma-\ntion of leiomyoma, some scholars suggest that patients with uterine fibroids over 40 years old directly undergo \n hysterectomy3. A study involving 227,489 patients with uterine fibroids revealed a 4.1% likelihood of receiving \nmyomectomy followed by  hysterectomy4.Studies have shown that the incidence of hysterectomy is 11% 5. The \ncommonly used surgical methods include Hysterectomy and salpingectomy, or Hysterectomy and monoliteral or \nbilateral adnexectomy. Currently, total hysterectomy with or without oophorectomy is a common clinical practice \nfor genital lesions such as uterine fibroids, adenomyoma, functional bleeding, and benign ovarian cysts. Studies \nin developed countries have shown that 20–40% of women undergo hysterectomy by the age of 60  years6,7, with \nbilateral ovaries removed at the same time in 10–55% of  cases8. However, while hysterectomy effectively treats \ngynecological physiological diseases, it also gives rise to psychological issues that trouble patients.\nSince the proposal of “post-hysterectomy syndrome” in the 1970s, an increasing number of clinical studies \nhave  demonstrated9–11 that women who have undergone hysterectomy are more susceptible to psychological \ncomorbidities such as insomnia, anxiety, and depression than those who have not. Depression is a prevalent \npsychological disease characterized by persistent and significant low mood, causing immense physical and \nmental distress to patients. It has emerged as the third leading cause of global disease  burden12. According to a \nOPEN\n1Guangzhou University of Chinese Medicine, Guangzhou, China. 2Taihe Town Health Center, Baiyun District, \nGuangzhou, China. 3The First Affiliated Hospital, Guangzhou University of Chinese Medicine, Guangzhou, \nChina. *email: 13622280992@126.com; zhao13711529286@163.com\n\n2\nVol:.(1234567890)Scientific Reports |         (2023) 13:9744  | https://doi.org/10.1038/s41598-023-36838-2\nwww.nature.com/scientificreports/\nsurvey conducted by the World Health Organization in 2015, approximately 322 million individuals worldwide \nwere afflicted with  depression13. Between 2013 and 2016, an estimated 8.1% of adults experienced symptoms of \ndepression within a 2-week  period14. In recent years, epidemiological research has revealed significant gender \ndisparities in the prevalence, incidence, course, symptoms and risk factors of  depression15,16. Current research \nindicates that the prevalence of depression is twice as high in women compared to  men17,18, However, some \nstudies have suggested that hysterectomy alone does not increase the risk of  depression19, and a positive cor -\nrelation may be observed when combined with  ovariectomy20,21. Conversely, other studies suggest that ovarian \npreservation may actually elevate the likelihood of developing  depression22.\nThe relationship between hysterectomy and ovarian preservation and depression remains a topic of debate. \nTherefore, the purpose of this study was to investigate the potential association between these procedures and \ndepression using data from the National Health and Nutrition Examination Survey.\nData and methods\nStudy population. The National Health and Nutrition Examination Survey (NHANES) is a sophisticated \nmulti-stage sampling design that selects samples to assess health and dietary status of civilian and non-institu-\ntional populations in the United States every 2 years, with resulting data  publication23,24. In this study, partici-\npants were drawn from continuous, cross-sectional NHANES data from 2007 to 2020 in the United States. We \nincluded a total of 16,821 women ages 18 and older who responded to reproductive health questions about hys-\nterectomy and mental health questions pertaining to depression using Mobile Screening Center (MEC). Exclud-\ning women with incomplete information on depression or incomplete information on hysterectomy. In the end, \na total of 12,097 women participated in our study (Fig. 1). The study was approved by the NCHS Research Eth-\nics Review Committee (https:// www. cdc. gov/ nchs/ nhanes/ irba98. htm), regulations and the written informed \nconsents were obtained from all participants and all experiments were performed in accordance with relevant \nguidelines.\nHysterectomy and oophorectomy. Hysterectomy was identified by RD280 (Had a hysterectomy?) in \nthe self-report question and oophorectomy by RHQ305 (Had both ovaries removed?), both of which were part \nof the 2007–2020 National Reproductive Health Questionnaire.\nOutcome indicator. Participants were assessed for depressive symptoms using the health Questionnaire \n(PHQ-9), which has good reliability and can effectively screen for depression or depressive symptoms in the past \n2 weeks. The questionnaire consisted of nine questions, each of which was rated on a four-point scale from 0 to \n3, with a score indicating the frequency of symptoms. The score ranges from 0 to 27. The nine diagnostic items \nincluded little interest, feeling down, trouble sleeping, poor appetite, feeling tired, feeling bad, trouble concen-\ntrating, moving or speaking slowly, suicidality and behavior. Participants with an overall score of 5 and above are \nconsidered to be positive for depression, with 5 being the PHQ  threshold25.\nCovariate. Based on previous literature and the availability of NHANES data, a number of potential con-\nfounding factors were included in this analysis. (a) demographic and socioeconomic status, including age, race/\nethnicity, education level, poverty income ratio, body mass index(BMI), marital status, smoking history, alcohol \nconsumption. (b) Past history including surgical history including ovariectomy, chronic diseases about hyper -\ncholesterolemia, diabetes, hypertension, weak or failing Kidneys and trouble sleeping, drug use includes the use \nof female hormones (Table 1).\nFigure 1.  Flowchart of screening samples from NHANES.\n\n3\nVol.:(0123456789)Scientific Reports |         (2023) 13:9744  | https://doi.org/10.1038/s41598-023-36838-2\nwww.nature.com/scientificreports/\nStatistical method. Descriptive statistics. The data are reported as mean ± SD and Min–Max for con-\ntinuous variables and percentages for categorical variables. The normality of continuous variables was assessed \nusing the Shapiro–Wilk test. If the data followed a normal distribution, statistical significance was determined \nby means of Student’s t-test. In cases where non-normal distribution was observed, Kruskal–Wallis test was em-\nployed to determine statistical significance. Due to the intricate design of NHANES, it is imperative to utilize ap-\npropriate weights while estimating data that represents the deinstitutionalized civilian population of the United \nStates. For this particular study, subsample B’s weight has been employed.\nModel. To assess whether there is an association between hysterectomy and depression, we used the following \nmethods.\nMethod 1: Establishment of a propensity score model (PSM).\nThe propensity score model (PSM) is a non-parsimonious multivariate logistic regression  model26, and pro-\npensity score covariates can be found in Table 1. The data of the two groups were matched with propensity score, \nnearest neighbor matching method was adopted, caliper value was set as 0.2, and the two groups were matched \naccording to 1:127. After PSM, the distribution of covariates reached equilibrium among groups (P  > 0.05). The \nPSM model was built using the statistical software IBM SPSS Statistics 25.0.\nMethod 2 was logistic regression analysis of hysterectomy and depression.\nTable 1.  Characteristics of participants including general characteristics and past medical histories (n%). \nSignificance at P < 0.05. Q1: Non-hysterectomy, Q2: Hysterectomy.\nCharacteristics\nPropensity overlap weighting Propensity 1:1 matching Propensity overlap weighting after psm\nQ1 Q2 P value Q1 Q2 P value Q1 Q2 P value\nAge (years)% < 0.00001 < 0.001 < 0.00001\n < 30 4.457 0.103 5.832 0.216 6.87 0.213\n 30–44 6.399 1.878 15.695 1.944 19.045 2.209\n 45–54 4.085 4.506 11.663 4.968 15.371 6.121\n 55–64 2.45 4.685 14.039 5.112 13.822 4.179\n ≥ 65 2.154 7.897 19.366 9.503 13.293 6.803\n Missing 80.455 80.932 33.405 78.258 31.6 80.475\nBMI 28.007 ± 7.077 29.271 ± 6.676 < 0.00001 29.558 ± 7.284 29.831 ± 6.843 0.31 29.065 ± 7.158 29.291 ± 6.638 0.38887\nRace (%) < 0.00001 0.439 0.93634\n Hispanic and others 21.464 12.292 28.15 28.078 14.632 14.854\n Non-hispanic White 66.924 75.622 46.436 44.492 70.935 70.325\n Non-hispanic Black 11.611 12.087 25.414 27.430 14.433 14.821\nEducation level (%) < 0.00001 0.002 0.00102\n < High school 14.21 17.838 34.845 28.654 23.273 20.514\n High school 19.53 26.169 23.326 24.406 24.979 25.975\n Some college 33.741 35.833 25.990 31.461 29.004 35.285\n > College graduate 32.488 20.102 15.839 15.407 22.744 18.208\n Missing 0.03 0.057 0.000 0.072 0.00 0.018\nPIR (%) < 0.00001 0.207 0.00337\n < 1.0 15.891 11.916 22.606 20.662 13.745 13.691\n 1.0–2.0 20.031 23.704 26.782 29.518 18.996 24.16\n ≥ 2.0 60.733 60.479 50.612 49.820 67.258 62.15\n Missing 3.344 3.902 0.00 0.00\nMarital status (%) 0.73867 < 0.001 < 0.00001\n Married or living with partner 39.691 40.052 40.461 50.468 30.511 40.692\n Widowed or divorced or separated 60.309 59.948 59.539 49.532 69.489 59.308\nSmoked at least 100 cigarettes in a \nlifetime 37.375 45.518 < 0.00001 39.021 43.053 0.031 41.254 45.239 0.03399\nAlcohol consumption (%) 70.823 62.816 < 0.00001 55.22 56.587 0.468 61.969 63.775 0.3247\nOvariectomy (%) 0.339 53.4 < 0.00001 2.448 7.127 < 0.001 2.466 7.207 < 0.00001\nHormone use (%) 7.121 5.695 < 0.00001 4.68 3.528 < 0.001 6.863 4.293 < 0.00001\nHypertension (%) 26.499 56.308 < 0.00001 59.467 58.459 0.503 53.664 53.173 0.9665\nHypercholesterolemia (%) 26.974 51.81 < 0.00001 39.813 49.028 < 0.001 39.354 46.668 < 0.00001\nDiabetes (%) 7.339 16.278 < 0.00001 18.143 21.598 0.029 12.748 15.331 0.14522\nWeak or failing Kidneys (%) 2.15 4.86 < 0.00001 2.880 4.752 0.01 2.452 3.361 0.15389\nTrouble sleeping (%) 28.283 43.959 < 0.00001 31.030 39.597 < 0.001 31.504 41.775 < 0.00001\nDepression PHQ ≥ 5 25.934 33.825 < 0.00001 31.102 36.645 0.002 27.861 35.537 0.00001\n\n4\nVol:.(1234567890)Scientific Reports |         (2023) 13:9744  | https://doi.org/10.1038/s41598-023-36838-2\nwww.nature.com/scientificreports/\nAfter obtaining PSM data, we used logistic regression to analyze the relationship between hysterectomy and \nthe dichotomous depression  measure25 and each depressive  symptom28, respectively, as well as to analyze the \neffect on depression with or without ovariectomy. First, regression analysis was performed for depression posi-\ntive or negative (Table  2). Second, to explore the relationship between independent variables and depressive \nsymptoms, three models were designed (Table 3). Model 1 represents unadjusted outcomes. Model 2 is a coarsely \nadjusted logistic regression after propensity score matching, adjusting for age, marital status, poverty-income \nratio, education level, smoking history, ovariectomy status, female hormone use, hypercholesterolemia and sleep \ndisturbance. Model 3 is an adjustment for all covariables.\nFinally, in order to evaluate the correlation between the preservation of the fallopian tube ovary and depres-\nsion, logistic regression was used to analyze the relationship between different surgical methods and depression \n(Table 4). Odds ratios were obtained by adjusting for covariates. Forest plots were drawn to visualize the data \n(Fig. 2).\nAll logistic analyses were performed with R software, V 0.4.0.3 [R: a language and statistical computing \nenvironment (program). Vienna, Austria: R Foundation for Statistical Computing, 2016], and EmpowerStats \n(http:// www. empow ersta ts. com). The figures were generated using Adobe Photoshop (https:// www. adobe. com/ \nprodu cts/ photo shop. html) or Origin 2021 (https:// www. origi nlab. com/). Finally, we confirm that all methods \nwere carried out in accordance with relevant guidelines and regulations.\nTable 2.  Unadjusted, crude and adjusted odds ratios (95% confidence intervals) for positivity of depression \nafter hysterectomy. a. Crude: Adjusted for marital status, PIR, education level, smoking history, oophorectomy, \nfemale hormone use, hypercholesterolemia, and sleep disorders. b. Adjusted: Adjusted for all the covariates.\nN\nOdds ratios (95% CI)\nUnadjusted P value Crude P value Adjusted P value\nPropensity overlap weighting \nbefore psm 12,097 1.307 (1.194, 1.431) < 0.001 1.145 (1.004, 1.306) 0.04 1.128 (0.987,1.289) 0.08\nPropensity overlap weighting \nafter psm 2778 1.281 (1.095, 1.500) 0.002 1.236 (1.016, 1.505) 0.03 1.234 (1.007, 1.512) 0.04\nTable 3.  β (95% CIs) of hysterectomy associated with depressive symptoms. Model 1 represents the unadjusted \noutcome. Model 2 was adjusted for marital status, PIR, education level, smoking history, oophorectomy, female \nhormone use, hypercholesterolemia, and sleep disorders. Model 3 refers to the adjustment for all covariates. A: \nHysterectomy without Oophorectomy. B: Oophorectomy without Hysterectomy. C: Non-hysterectomy without \nOophorectomy. D: Hysterectomy with Oophorectomy.\nModel 1 β (95% CI) Model 2 β (95% CI) Model 3 β (95% CI)\nLittle interest 0.128 (0.071, 0.185) 0.076 (0.010, 0.142) 0.084 (0.018, 0.150)\nFeeling down 0.074 (0.016, 0.133) 0.062 (− 0.005, 0.129) 0.069 (0.002, 0.136)\nTrouble sleeping 0.028 (− 0.048, 0.103) 0.031 (− 0.053, 0.116) 0.034 (− 0.050, 0.119)\nFeeling tired 0.082 (0.009, 0.155) 0.018 (− 0.065, 0.100) 0.038 (− 0.044, 0.120)\nPoor appetite 0.082 (0.018, 0.146) 0.036 (− 0.039, 0.111) 0.048 (− 0.026, 0.123)\nFeeling bad − 0.005 (− 0.056, 0.047) 0.016 (− 0.044, 0.076) 0.018 (− 0.042, 0.079)\nTrouble concentrating 0.093 (0.040, 0.145) 0.080 (0.020, 0.140) 0.087 (0.026, 0.147)\nMoving/speaking slowly 0.044 (0.003, 0.086) 0.039 (− 0.009, 0.087) 0.040 (− 0.008, 0.088)\nBetter off dead 0.019 (− 0.003, 0.041) 0.009 (− 0.017, 0.035) 0.009 (− 0.018, 0.035)\nTable 4.  The distribution of patients with depression by four different surgical procedures. Subgroup A: \nHysterectomy without Oophorectomy. Subgroup B: Oophorectomy without Hysterectomy. Subgroup C: Non-\nhysterectomy without Oophorectomy. Subgroup D: Hysterectomy with Oophorectomy.\nSubgroup A Subgroup B Subgroup C Subgroup D\nY es (N = 1290) No (N = 1488) Y es (N = 34) No (N = 2744) Y es (N = 1355) No (N = 1423) Y es (N = 99) No (N = 2679)\nDepression \nPHQ ≥ 5 (N%) 464 (35.97%) 477 (32.06%) 9 (26.47%) 932 (33.97%) 423 (31.22%) 518 (36.40%) 54 (54.55%) 896 (33.45%)\nDepression \nPHQ < 5 (N%) 826 (64.03) 1011 (67.94%) 25 (73.53%) 1812 (66.03%) 932 (68.78%) 905 (63.60%) 45 45.45%) 1783 (66.55%)\nP value 0.030 0.359 0.004 0.013\n\n5\nVol.:(0123456789)Scientific Reports |         (2023) 13:9744  | https://doi.org/10.1038/s41598-023-36838-2\nwww.nature.com/scientificreports/\nResult\nDescriptive statistical analysis. Prior to propensity score matching, 2763 of 12,097 women underwent \nhysterectomy, of whom 952 were positive for depression, accounting for 34.455% of the sample (P < 0.001). The \nsample was weighted so that patients with PHQ ≥ 5 points accounted for 33.825% of the total sample (P < 0.001). \nBy adjusting for covariates for propensity matching, a total of 2778 women in the database were successfully \nmatched to the hysterectomized and non-hysterectomized population of 1389 each, with 35.537% of the sample \nin the hysterectomized group experiencing depression (P < 0.001). And 35.537% were positive for depression in \nthe hysterectomy group.\nLogistic regression. A PHQ score of 5 was used as the cut-off point for the presence or absence of depres-\nsion. Before propensity score matching, the OR for PHQ score of 5 or more was 1.145 (95%CI 1.004, 1.306) after \ncrude adjustment of covariates, and 1.128 (95%CI 0.987, 1.289) after exact adjustment. In 2778 subjects after \npropensity 1:1 matching, the OR for PHQ ≥ 5 was 1.236 (95%CI 1.016, 1.505) after crude adjustment of covari-\nates and 1.234 (95%CI 1.007, 1.512) after exact adjustment. This suggests that hysterectomy is associated with \npositivity for depression.\nTo further explore the key to hysterectomy and depression, we performed a regression analysis for each \ndepressive symptom. The results showed that positive depression was related to little interest and trouble con -\ncentrating. It was not associated with trouble sleeping, feeling tired, poor appetite, feeling bad, slow moving or \nspeaking, and suicidal thoughts. Oophorectomy alone is not associated with depression.\nIn addition, we distinguish in detail between four different types of surgery, including Hysterectomy without \nOophorectomy, Oophorectomy without Hysterectomy, Non-hysterectomy without Oophorectomy, and Hyster-\nectomy with Oophorectomy. The number of depressed patients in each group was 464,9,423, and 54, respectively. \nVisualizations were plotted with adjusted covariates. Hysterectomy alone is a risk factor for depression, but \nhysterectomy combined with oophorectomy has a stronger correlation with depression than hysterectomy alone.\nDiscussion\nAt present, gynecological  malignancies29, endometrial hyperplasia with  dysplasia30,31, intractable postpartum \n hemorrhage32,33, or prophylactic resection with a family history of tumors are suitable diseases for hysterectomy. \nDue to the differences in individualization between patients, surgeons need to perform hysterectomy according \nto professional knowledge, indications for surgery, nature of the  disease34, patient characteristics and patient \nwillingness. However, psychosocial problems after hysterectomy should not be ignored. Post-hysterectomy syn-\ndrome makes researchers raise the concern about postoperative complications. Gupte and Nagabhirava found \nthat 9% of women had post-operative  depression35, of which 2% were post-operative new-onset depression. The \nlatest research in modern medicine also provides strong evidence for the correlation between hysterectomy and \n depression36.  Hyo37 extracted data from the Korean Health Insurance from 2002 to 2013, and they found that \nwomen who underwent hysterectomy had higher rates of depression than those who did not undergo hysterec-\ntomy. We conducted a multimodal observational study using data from the National Health and Nutrition Exami-\nnation Survey (NHANES) from 2007 to 2020 and found consistent findings across patterns. The risk of depression \nwas significantly increased after hysterectomy compared with those who did not undergo hysterectomy.\nWomen are twice as likely to be diagnosed with depression as men, because hormone levels are different in \nwomen at different times. Current studies have shown that estrogen can play an antidepressant role by regulating \nneurotransmitters through estrogen receptors, which affect the hypothalamic–pituitary–adrenal axis, and that \nthe ovaries are the organ that secretes estrogen. It follows that the risk of depression should decrease when hys-\nterectomy is performed but ovaries are preserved. However,  Wilson38 and Laughlin  Tomaso39 found that women \nwho underwent hysterectomy with preservation of the ovaries were at higher risk for depression than women \nwho underwent both hysterectomy and bilateral oophorectomy. These are two completely opposite conclusions. \nBased on the above, the current study explored the effect of four different surgical procedures on depression \nthrough regression analysis. We found an interesting result that oophorectomy was not associated with positive \ndepression, but hysterectomy was a risk factor for postoperative depression, and the risk of depression was also \nincreased when both the uterus and ovaries were removed.\nFigure 2.  Logistics regression analysis of different surgical methods on positive depression.\n\n6\nVol:.(1234567890)Scientific Reports |         (2023) 13:9744  | https://doi.org/10.1038/s41598-023-36838-2\nwww.nature.com/scientificreports/\nThe uterus is an endocrine organ. In addition to its local endocrine function, it may also regulate the hypo-\nthalamic-pituitary-ovarian (HPO) axis which refers to the complete and coordinated neuroendocrine system \ncomposing of the hypothalamus, pituitary gland, and ovary. Each of its links has unique neuroendocrine func-\ntions, and they regulate and influence each other to maintain a relatively stable dynamic  balance40. The pituitary \nsecretes follicle stimulating hormone (FSH), prolactin, and luteinizing hormone (LH) under the regulation of \nGonadotropin-releasing hormone (GnRH) secreted by the hypothalamus. FSH, prolactin, and LH can act on \nthe ovary and all three participate in the negative feedback regulation of the HPO axis. Studies have shown a \ntendency to increase FSH levels after Oophorectomy and the opposite for E2  levels41. FSH levels are associated \nwith negative emotions such as perimenopausal  depression42, and women with rapidly rising FSH levels are \nmore likely to experience depressive  symptoms43, while lower FSH levels are associated with reduced depressive \n symptoms44. Disruption of LH and estrogen regulation after Hysterectomymay be the main mechanism contrib-\nuting to the increased risk of depression. Similarly, because of estrogen’s role in regulating mood and cognitive \nfunction, the sudden drop in estrogen levelsdue to Oophorectomy would presumably increase the incidence of \ndepression. But the study had found that postmenopausal oophorectomy did not affect the incidence of depres-\nsion. This is consistent with the results of the present study that Oophorectomy-alone was not associated with \ndepression. This may be due to the fact that the age of the sample with Oophorectomy-alone in this study was \nbasically close to menopause. The most significant change around menopause is the decline in ovarian func-\ntion, which is no longer able to affect hormone levels. In addition, removal of the uterus can cause them to stop \nbelieving that they are fully female. This affects their self-confidence and self-worth  level45, leading to mental \nhealth problems. Oophorectomy may exacerbate this psychological burden, which is consistent with the results \nof this study. Hysterectomy alone is a risk factor for depression, but simultaneous hysterectomy of the ovaries \nfurther increases the risk of depression.\nThe study also found that women after hysterectomy had more depressive symptoms, mainly including little \ninterest, feeling down and trouble concentrating, but not more severe symptoms such as self-denial, slow move-\nment or speech, suicidal tendencies and behaviors. That is, it is associated with depressed mood and somatic \nsymptoms. However, it was not related to slow thinking, decreased volitional activity and cognitive impairment. \nPay attention to female mood and somatic symptoms, and positive psychological intervention will improve the \nrehabilitation effect after  hysterectomy46.\nTherefore, when hysterectomy has become an established fact, but in the case of opportunistic adnexectomy, \nit is necessary to retain the patient’s adnexa as much as possible to reduce the risk of depression. Opportunistic \nadnexectomy refers to the implementation of oophorectomy and salpingectomy without known indications, \nsuch as ovarian lesions, hereditary ovarian cancer syndrome, etc.47.\nReviewing our study, there are still some limitations. First, because of its cross-sectional design, it was not \npossible to determine whether depression was present before the hysterectomy occurred. Second, although \nthis study used a control group matched for demographic factors and several medical histories, even though \npropensity score methods were used, residual and unmeasured confounding is still possible in this study. The \ndevelopment of depression may be affected by the differences in personality and mentality of each respondent, \nincluding preoperative psychosocial status, perioperative pain and postoperative infection.\nThis study has several strengths. The NHANES data provide us with a unique opportunity to examine the \nassociation between hysterectomy and depressive symptoms in this multi-ethnic, representative sample of the \npopulation in the United States. Second, to explore the association, we specifically considered the association \nof hysterectomy with each depressive symptom. Most importantly, we explored the effect of different types of \nsurgical procedures on depression positivity.\nConclusion\nWomen who have had a hysterectomy are at higher risk of depression than women who have not had a hyster-\nectomy, and this risk may be exacerbated if the uterus and ovaries are removed. When clinically appropriate, \nsurgeons should try to preserve the patient’s ovaries.\nData availability\nThe data that support the findings of this study are available, but restrictions apply to the availability of these data, \nwhich were used under license for the current study, and so are not publicly available. Data are however available \nfrom the authors via the email address ginayyh@163.com upon reasonable request and with permission of us.\nReceived: 25 August 2022; Accepted: 11 June 2023\nReferences\n 1. Hammer, A. et al. Global epidemiology of hysterectomy: Possible impact on gynecological cancer rates. Am. J. Obstet. Gynecol.  \n213(1), 23–29 (2015).\n 2. Yu, O. et al. A US population-based study of uterine fibroid diagnosis incidence, trends, and prevalence: 2005 through 2014. Am. \nJ. Obstet. Gynecol. 219(6), 591.e1-591.e8 (2018).\n 3. Mynbaev, O. A. et al. The medical device applied to uterine fibroids morcellation: A analysis of critical biological issues and draw-\nbacks from a medical-legal prospective. Curr. Pharm. Des. 26(3), 318–325 (2020).\n 4. Wang, C. et al. Utilization of endovascular and surgical treatments for symptomatic uterine leiomyomas: A population health \nperspective. J. Vasc. Interv. Radiol. 31(10), 1552-1559.e1 (2020).\n 5. Novetsky, A. P ., Boyd, L. R. & Curtin, J. P . Trends in bilateral oophorectomy at the time of hysterectomy for benign disease. Obstet. \nGynecol. 118(6), 1280–1286. https://  doi. org/ 10. 1097/ AOG. 0b013 e3182 36fe61 (2011).\n 6. Redburn, J. C. & Murphy, M. F . Hysterectomy prevalence and adjusted cervical and uterine cancer rates in England and Wales. \nBJOG Int. J. Obstetr. Gynaecol. 108, 388–395 (2001).\n\n7\nVol.:(0123456789)Scientific Reports |         (2023) 13:9744  | https://doi.org/10.1038/s41598-023-36838-2\nwww.nature.com/scientificreports/\n 7. Rositch, A. F ., Nowak, R. G. & Gravitt, P . E. Increased age and race-specific incidence of cervical cancer after correction for hys-\nterectomy prevalence in the United States from 2000 to 2009. Cancer 120, 2032–2038 (2014).\n 8. Ramdhan, R. C., Loukas, M. & Tubbs, R. S. Anatomical complications of hysterectomy: A review. Clin. Anat. 30(7), 946–952 (2017).\n 9. Richards, D. H. A post-hysterectomy syndrome. Lancet 2(7887), 983–985 (1974).\n 10. Helmy, Y . A. et al. Psychiatric morbidity following hysterectomy in Egypt. Int. J. Gynaecol. Obstet. 102(1), 60–64 (2008).\n 11. Chou, P . H. et al. Risk of depressive disorders in women undergoing hysterectomy: A population-based follow-up study. J. Psychiatr. \nRes. 68, 186–191 (2015).\n 12. Charlson, F . J. et al. Global epidemiology and burden of schizophrenia: Findings from the global burden of disease study 2016. \nSchizophr. Bull. 44(6), 1195–1203 (2018).\n 13. World Health Organization (WHO). Depression and other common mental disorders: Global health estimates (WHO, 2017). \nhttps:// apps. who. int/ iris/ bitst ream/ handle/ 10665/ 254610/ WHO- MSD- MER- 2017.2- eng. pdf.\n 14. Brody, D. J., Pratt, L. A. & Hughes, J. P . Prevalence of depression among adults aged 20 and over: United States, 2013–2016. NCHS \nData Brief 303, 1–8 (2018).\n 15. Girgus, J. S. & Y ang, K. Gender and depression. Curr. Opin. Psychol. 4, 53–60 (2015).\n 16. Parker, G. & Brotchie, H. Gender differences in depression. Int. Rev. Psychiatry 22(5), 429–436 (2010).\n 17. Lokuge, S., Frey, B. N., Foster, J. A., Soares, C. N. & Steiner, M. Depression in women: Windows of vulnerability and new insights \ninto the link between estrogen and serotonin. J. Clin. Psychiatry 72(11), e1563–e1569 (2011).\n 18. Steiner, M., Dunn, E. & Born, L. Hormones and mood: From menarche to menopause and beyond. J. Affect. Disord. 74(1), 67–83 \n(2003).\n 19. Darwish, M., Atlantis, E. & Mohamed-Taysir, T. Psychological outcomes after hysterectomy for benign conditions: A systematic \nreview and meta-analysis. Eur. J. Obstet. Gynecol. Reprod. Biol. 174, 5–19 (2014).\n 20. Farquhar, C. M., Sadler, L. & Stewart, A. W . A prospective study of outcomes five years after hysterectomy in premenopausal \nwomen. Aust. N. Z. J. Obstet. Gynaecol. 48(5), 510–516 (2008).\n 21. Gibson, C. J. et al. Mood symptoms after natural menopause and hysterectomy with and without bilateral oophorectomy among \nwomen in midlife. Obstet. Gynecol. 119(5), 935–941 (2012).\n 22. Rocca, W . A. et al. Long-term risk of depressive and anxiety symptomsafter early bilateral oophorectomy. Menopause 15, 1050–1059 \n(2008).\n 23. Data. NHaNES. Centers for disease control and prevention (CDC) web site (accessed 24 October 2020). http:// www. cdc. gov/  \nNCHS/ nhanes. htm.\n 24. C. Centers for Disease, Prevention, National health and nutrition examination survey. Survey methods and analytic guidelines \n(accessed 24 October 2020). https:// wwwn. cdc. gov/ nchs/ nhanes/ analy ticgu ideli nes. aspx.\n 25. Kroenke, K., Spitzer, R. L. & Williams, J. B. The PHQ-9: Validity of a brief depression severity measure. J. Gen. Intern. Med. 16(9), \n606–613 (2001).\n 26. Ahmed, A. et al. Heart failure, chronic diuretic use, and increase in mortality and hospitalization: An observational study using \npropensity score methods. Eur. Heart J. 27(12), 1431–1439 (2006).\n 27. Austin, P . C. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivar. \nBehav. Res. 46(3), 399–424 (2011).\n 28. Demartini, B. et al. Depressive symptoms and major depressive disorder in patients affected by subclinical hypothyroidism: A \ncross-sectional study. J. Nerv. Ment. Dis. 202(8), 603–607 (2014).\n 29. Hamilton, C. A. et al. Endometrial cancer: A society of gynecologic oncology evidence-based review and recommendations. \nGynecol. Oncol. 160(3), 817–826 (2021).\n 30. Kurman, R. J., Kaminski, P . F . & Norris, H. J. The behavior of endometrial hyperplasia. A long-term study of “untreated” hyperplasia \nin 170 patients. Cancer 56(2), 403–412 (1985).\n 31. Ferenczy, A. & Gelfand, M. The biologic significance of cytologic atypia in progestogen-treated endometrial hyperplasia. Am. J. \nObstet. Gynecol. 160(1), 126–131 (1989).\n 32. Mundhra, R. et al. Emergency peripartum hysterectomy during COVID-19 pandemic. Cureus 13(12), e20524 (2021).\n 33. Clark, S. L., Y eh, S. Y ., Phelan, J. P ., Bruce, S. & Paul, R. H. Emergency hysterectomy for obstetric hemorrhage. Obstet. Gynecol. \n64(3), 376–380 (1984).\n 34. Lefebvre, G., Allaire, C., Jeffrey, J. & Vilos, G. No. 109-Hysterectomy. J. Obstet. Gynaecol. Can. 40(7), e567–e579 (2018).\n 35. Gupte, S. G. & Nagabhirava, G. Prospective study of psychiatric morbidity and evaluation of quality of life in patients undergoing \nhysterectomy. MVP J. Med. Sci. 5, 204–210 (2018).\n 36. Modarres, M., Rahimikian, F . & Mehran, A. Impact of pre-hysterectomy counseling on depression among patients referred to \nTUMS Hospitals. Hayat 19(2), 40–50 (2013).\n 37. Choi, H. G. et al. Association between hysterectomy and depression: A longitudinal follow-up study using a national sample cohort. \nMenopause 27(5), 543–549 (2020).\n 38. Wilson, L. et al. Hysterectomy and incidence of depressive symptoms in midlife women: The Australian Longitudinal Study on \nWomen’s Health. Epidemiol. Psychiatr. Sci. 27(4), 381–392 (2018).\n 39. Laughlin-Tommaso, S. K. et al. Long-term risk of de novo mental health conditions after hysterectomy with ovarian conservation: \nA cohort study. Menopause 27(1), 33–42 (2020).\n 40. Cui, J. et al. Effects of ammonia on hypothalamic-pituitary-ovarian axis in female rabbits. Ecotoxicol. Environ. Saf. 227, 112922. \nhttps:// doi. org/ 10. 1016/j. ecoenv. 2021. 112922 (2021).\n 41. Adelman, M. R. & Sharp, H. T. Ovarian conservation vs removal at the time of benign hysterectomy. Am. J. Obstet. Gynecol. 218(3), \n269–279. https:// doi. org/ 10. 1016/j. ajog. 2017. 07. 037 (2018).\n 42. Berent-Spillson, A. et al. Metabolic and hormone influences on emotion processing during menopause. Psychoneuroendocrinology \n76, 218–225 (2017).\n 43. Ryan, J. et al. A prospective study of the association between endogenous hormones and depressive symptoms in postmenopausal \nwomen. Menopause 16(3), 509–517 (2009).\n 44. Gordon, J. L. et al. Ovarian hormone fluctuation, neurosteroids, and HPA axis dysregulation in perimenopausal depression: A \nnovel heuristic model. Am. J. Psychiatry 172(3), 227–236 (2015).\n 45. Goudarzi, F ., Khadivzadeh, T., Ebadi, A. & Babazadeh, R. Iranian women’s self-concept after hysterectomy: A qualitative study. \nIran J. Nurs. Midwifery Res. 26(3), 230–237 (2021).\n 46. Shehmar, M. & Gupta, J. K. The influence of psychological factors on recovery from hysterectomy. J. R. Soc. Med. 103, 56–59 (2010).\n 47. Finch, A. et al. The impact of prophylactic salpingo-oophorectomy on menopausal symptoms and sexual function in women who \ncarry a BRCA mutation. Gynecol. Oncol. 121(1), 163–168 (2011).\nAuthor contributions\nS.X. and Z.X. contributed to the experimental design. Y .Y ., Z.X. and Z.J. did the statistical analysis. Y .Y ., F .Y . and \nC.B. analyzed the data. Y .Y . wrote the manuscript. S.X. and Z.X. revised the manuscript. All authors reviewed \nthe manuscript.\n\n8\nVol:.(1234567890)Scientific Reports |         (2023) 13:9744  | https://doi.org/10.1038/s41598-023-36838-2\nwww.nature.com/scientificreports/\nCompeting interests \nThe authors declare no competing interests.\nAdditional information\nCorrespondence and requests for materials should be addressed to X.S. or X.Z.\nReprints and permissions information is available at www.nature.com/reprints.\nPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and \ninstitutional affiliations.\nOpen Access  This article is licensed under a Creative Commons Attribution 4.0 International \nLicense, which permits use, sharing, adaptation, distribution and reproduction in any medium or \nformat, as long as you give appropriate credit to the original author(s) and the source, provide a link to the \nCreative Commons licence, and indicate if changes were made. The images or other third party material in this \narticle are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the \nmaterial. If material is not included in the article’s Creative Commons licence and your intended use is not \npermitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from \nthe copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.\n© The Author(s) 2023","source_license":"CC0","license_restricted":false}