Psychological outcomes after hysterectomy for benign conditions: a systematic review and meta-analysis

European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

Contents lists available at ScienceDirect

European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb

Review

Psychological outcomes after hysterectomy for benign conditions: a systematic review and meta-analysis Maram Darwish a,b,*, Evan Atlantis c, Tamara Mohamed-Taysir d a

School of Medicine, Hashemite University, Zarqa, Jordan Medical Intern, Prince Hamzah Hospital, Amman, Jordan c School of Nursing and Midwifery, University of Western Sydney, Campbelltown Campus, New South Wales, Australia d School of Medicine, University of Jordan, Amman, Jordan b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 19 June 2013 Received in revised form 27 August 2013 Accepted 6 December 2013

Hysterectomy is one of the commonest operative procedures in the developed world, mostly occurring among premenopausal women, with contradictory results regarding post-operative psychological wellbeing. This review aims to inform practice by examining whether hysterectomy predicts depression or anxiety outcomes. We searched PubMed, EMBASE, and PsycINFO electronic databases for articles published before November 2012. Reference lists of relevant articles were hand searched, and expert opinions were sought. Refereed studies investigating an association between hysterectomy for benign (non-cancerous) conditions and post-operative symptoms of depression or anxiety were chosen for this review. Two authors independently abstracted data from original articles. Authors of relevant studies were contacted for data that could not be extracted from the published articles. Review Manager 5.1 was used throughout the meta-analysis to calculate the summary relative risks (RRs), and the weighted standardized mean difference (WstdMD), and their corresponding 95% confidence intervals (CI). A random effects model was used in data analysis and verified using a fixed effect model. Overall, hysterectomy was associated with a decreased risk of clinically relevant depression (RR = 1.69, 95% CI 1.19–2.38). Additionally, hysterectomy was associated with a decrease in standardized depression outcomes (standardized mean difference (SMD) 0.38 (95% CI 0.27–0.49)). Conversely, there was no significant association between hysterectomy and risk of clinically relevant anxiety (RR = 1.41, 95% CI 0.72–2.75). In conclusion, data from before and after studies suggest that hysterectomy for benign gynecological conditions is not adversely associated with anxiety and may be positively rather than adversely associated with depression. ß 2013 Elsevier Ireland Ltd. All rights reserved.

Keywords: Anxiety Depression Hysterectomy Meta-analysis Psychological morbidity

Contents 1. 2.

3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Data sources . . . . . . . . . . . . . . . . . . . . 2.1. Study selection . . . . . . . . . . . . . . . . . . 2.2. Data extraction and quality assessment . . . . Statistical analysis. . . . . . . . . . . . . . . . 3.1. Overview . . . . . . . . . . . . . . . 3.1.1. Pooled analyses . . . . . . . . . . 3.1.2. Meta-regression analyses . . 3.1.3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Included studies . . . . . . . . . . . . . . . . . Descriptive data synthesis. . 4.1.1. 4.1.2. Quantitative data synthesis.

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

* Corresponding author at: Dr. Maram Darwish, Medical Intern, Prince Hamzah Hospital, P.O. Box 3678, Amman 11953, Jordan. Tel.: +962 777995009. E-mail addresses: [email protected] (M. Darwish), [email protected] (E. Atlantis), [email protected] (T. Mohamed-Taysir). 0301-2115/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejogrb.2013.12.017

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . .

6 6 6 6 6 7 7 7 8 8 8 8 10

6

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

5.

6.

Discussion . . . . . . . . . . . . . . . . . . . . . . . . Main findings and interpretations 5.1. Strength and limitations. . . . . . . . 5.2. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

1. Introduction Among many gynecological procedures, hysterectomy has shown a significant decrease in the previous three decades all over the world, but it still remains one of the most frequently performed operative procedures [1–3]. Hysterectomy involves one of two surgical approaches, traditional abdominal laparotomy and minimally invasive hysterectomy: the latter could be one of two approaches, vaginal or laparoscopic hysterectomy [4]. The current literature reports that hysterectomy is more often performed during the premenopausal period [3–6]. Approximately 10 to 30% of women in the industrialized world undergo hysterectomy by the age of 65 [7,8]. And around 20% of women in the unindustrialized world undergo hysterectomy by the age of 55 [9]. Noteworthy; there has been a fall in the prevalence of hysterectomy in the later group by almost 1% every decade since 1980 [9], perhaps due to some cultural and ethnic beliefs [10]. Epidemiological studies have reported that over 90% of hysterectomies are performed for benign surgical indications [4,6]. For instance, premenopausal women undergo hysterectomy to alleviate bothersome gynecological symptoms including dysfunctional uterine bleeding and uterine fibroids, whereas postmenopausal women undergo hysterectomy to repair a prolapsed uterus or any other pelvic organ [11]. Regardless of the surgical indication, the benefits and risks of hysterectomy remain uncertain [12–15]. Growing evidence from clinical studies indicates that women with hysterectomy suffer both psychological and physical comorbidities [16] that usually overlap and interact with one another [16,17]. Hysterectomy-related psychological morbidity typically includes depression [5,18], anxiety [6,19], and stress-related symptoms [16]. Psychological co-morbidity associated with hysterectomy could be triggered by negative perceptions about body image, femininity, youth, energy and activity levels [20], as well as loss of child-bearing capacity [21]. Physical co-morbidity associated with hysterectomy typically includes increased risk of developing pelvic floor prolapse, urinary incontinence and sexual dysfunction [16,22]. Recent literature on the sexual effects of hysterectomy has added to the confusion around the psychosocial outcomes of this surgery. Interestingly, Ryan et al. [1] suggested hysterectomy-related physical and psychological co-morbidities to be preoperative, and reported that in almost all the cases women linked their psychological symptoms with their physical symptoms as their motivation to accept hysterectomy. Conversely, evidence from randomized trials suggests that hysterectomy is associated with high levels of psychological wellbeing compared to women’s pre-hysterectomy status, since they experience relief of distressing gynecological symptoms [23–25]. The controversy regarding quality of life with regard to psychological wellbeing following hysterectomy has raised a notable doubt about the appropriateness of hysterectomy for benign conditions [26,27]. An initial analysis of the available published literature indicates an absence of a high-quality review or systematic review that examined the relationship between psychosocial wellbeing and hysterectomy, which could help resolve these uncertainties. We therefore conducted a systematic review and meta-analysis of longitudinal studies to evaluate whether hysterectomy for benign

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

14 14 16 18 18 18

(non-cancerous) conditions in adult female populations predicts depression or anxiety outcomes. 2. Methods 2.1. Data sources We searched PubMed, EMBASE, and PsycINFO electronic databases for articles published prior to November 2012. Search syntaxes were developed in consultation with an experienced university research librarian taking into account a broad range of terms and phrases used in definitions of depression or anxiety and hysterectomy (Appendix S1). Reference lists of potentially eligible articles were searched by hand to identify additional studies missed by our search strategy. We also contacted four experts in the field on 26 November 2012 to review a preliminary list of studies retrieved for inclusion and comment on the existence of additional studies not yet identified, and we updated the search on 26 May 2013. 2.2. Study selection One reviewer (MD) identified potentially relevant studies for inclusion by screening titles and/or abstracts of all citations identified with our database searches. A second screening was performed on the full text of these articles. Longitudinal studies in adult female populations (aged 18 years or older) that reported on an association between depression or anxiety and hysterectomy for benign (non-cancerous) conditions were eligible. Hysterectomy had to be the predictor variable (exposed versus non-exposed controls) while the other was the outcome. Only English-language articles were eligible. Eligible study designs included: randomized controlled trials, controlled clinical trials, before and after studies, interrupted timeseries studies, case–control studies, and cohort studies. Among the eligible study designs, we only included studies that psychologically assessed participants before and after hysterectomy for benign conditions. This review aimed to determine whether hysterectomy had adverse effects on psychological wellbeing after hysterectomy in the medium to long term, hence the use of the patients’ baseline data in a form of self-controlled designs. We included studies reporting psychological outcomes for either a single type of hysterectomy or multiple types of hysterectomy. Studies were identified for exclusion if they included hysterectomy for emergency or oncologic conditions, non-English studies, nonhuman studies, and studies that did not assess the participants before hysterectomy. 3. Data extraction and quality assessment Data extraction and quality assessment of included studies were performed and/or verified independently by two reviewers (MD and TMT). Discrepancies were resolved through discussion. Authors of relevant studies were contacted, where possible, for data that could not be extracted from the published articles. After studies had been excluded because they did not meet the inclusion criteria, a checklist devised prior to the review assessed the quality

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

7

Table 1 Quality item checklist of intervention and observational studies reviewed (1 point each). Study identification

Were study participants representative of the general population from which they were recruited?

Alexander et al., 1996 Aziz et al., 2005 Bayram et al., 2008 Donoghue et al., 2003 Ellstro¨m et al., 2003 Farquhar et al., 2008 Flory et al., 2006 Gath et al., 1982 Gibson et al., 2012 Helmy et al., 2008 Hurskainen et al., 2001 Kjerulff et al., 2000 Lambden et al., 1997 Martin et al., 1980 Moraloglu et al., 2007 Okunlola et al., 2009 Persson et al., 2006 Persson et al., 2010 Rohl et al., 2008 Ryan et al., 1989 Thakar et al., 2004 Vandyk et al., 2011

0.5 0.5 1 1 1 1 1 0.5 1 1 1 1 1 0.5 1 1 1 1 1 0.5 0.5 1

Was there any control/adjustment account for important confounding and modifying variables (age, race, parity, income, BMI, smoking, hormonal status, and indication for hysterectomy)? (0.25 each)

1.25

Study reports duration of follow-up (0.5) and duration of follow-up adequate to identify expected psychological outcomes (threshold set at 6 months) (0.5)

Was the outcome (psychological wellbeing) assessed using standardized instrument(s) (psychometrics/ questionnaire)?

Study reports number lost to follow-up (0.5), and the overall number lost to follow-up is low (threshold set at 20%) (0.5)?

Total/6

Total (%)

1 1 0.5 0.5 1 1 1 1

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

1 1 1 0.5 1 0.5 0.5 1 0.5 1 1 1 0.5 1 1 1 1 1 1 1 1 1

3.5 3.5 3.5 3 4 4.75 3.5 3.5 3.5 3.5 4 5 3 3.5 3.5 3.5 4 4.75 5.5 4.25 3.5 4

58 58 58 50 67 79 58 58 58 58 67 83 50 58 58 58 67 79 92 71 58 67

1

1

0.75 1.5 0.75

0.5 1 1 0.5 1 0.5 0.5 1 1 1 1 1 1

of each study according to predefined criteria for assessment of observational studies [28]. For the purpose of this review, randomized trials were treated as a one-arm study; therefore, only one quality assessment form was needed to assess the papers. A numerical point scale 0–1 was used for all items except for one item concerning adjustment of possible confounders, which was scored 0–2. The key variables assessed were: (1) study participants representative of the general population from which they were recruited (an unselected sample of consecutive patients, or a random sample where a list of all members of the relevant population exists) (worth 0.5 point), adequate response rate (threshold set as 60%) (worth 0.5 point), (2) the distribution of known confounders (age, race, parity, income, BMI, smoking, hormonal status, and indication for hysterectomy) in the different treatment groups was described, and adjusted for the analyses through matching, stratification, multivariable analysis, or other approaches, (3) study reports duration of follow-up (worth 0.5 point), and duration of follow-up adequate to identify expected psychological outcomes (threshold set at 6 months) (worth 0.5 point), (4) the outcome (psychological wellbeing) was assessed using a clinical interview or standardized instrument(s) (psychometrics/questionnaire), (5) study reports number lost to follow-up (worth 0.5 point), and the overall number lost to follow-up is low (threshold set at 20%) (worth 0.5 point). A total quality score was obtained by adding the scores of each variable, and a quality percentage was calculated. A total maximum score of 6 points was possible in our scale, with higher score indicating better quality. Papers scoring over 55% were considered of sufficient quality and this was the case for all the included studies except for 3 studies [29–31]. However, no studies were excluded based on their methodological quality score. Table 1 shows a detailed methodological quality assessment form. Where more than one type of hysterectomy was reported, the data for each type were extracted, but when combined data were reported, they were used instead.

3.1. Statistical analysis 3.1.1. Overview Review manager (RevMan) version 5.1 was used throughout the meta-analysis to calculate the summary relative risks (RRs), and the weighted standardized mean difference (WstdMD), and their corresponding 95% confidence intervals (CI). A random effects model was used in data analysis and verified using fixed effect model. Statistical analysis of heterogeneity was carried out using the x2 and y2 tests. A P value of <0.1 for the x2 test and y2 > 50% were considered a statistically significant evidence of heterogeneity. Where evidence of heterogeneity was observed, we evaluated the possible effects of dichotomous variables (such as studies that involved postmenopausal women or women with oophorectomy vs. studies that involved premenopausal women only) in separate analyses stratified on the dichotomous variable. The effects of continuous variables, such as the interval between the hysterectomy and assessment of psychological morbidity, and the proportion of participants with baseline depression were evaluated via meta-regression. 3.1.2. Pooled analyses Data were subjected to meta-analyses stratified by hormonal status (studies that involved postmenopausal women or women with oophorectomy vs. studies that involved premenopausal women only) to obtain composite estimates of RRs separately for each hormonal status and for all studies combined. Study heterogeneity was assessed through inspection of the funnel plot and using the x2 and y2 tests. We also checked data extracted from individual outlier studies, qualitatively investigated reasons for their different results, and explored the effects of study exclusion in sensitivity analyses to investigate the robustness of the metaanalyses models. Studies that involved postmenopausal women or women with oophorectomy were excluded in the meta-analyses,

8

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

where relevant. We also excluded studies of short follow up durations (threshold set at 6 months) and studies of lower quality (score < 55%). Publication bias, which reflects the tendency for studies to be published in the literature when findings are positive, was assessed visually using funnel plots [32].

available for Review manager (RevMan) version 5.1 and Comprehensive Meta-analysis (CMA) software, version 2 (Biostat Corp, Englewood NJ (2005)).

3.1.3. Meta-regression analyses Meta-regression analyses allow the application of regression techniques to the prediction of study outcomes so that the relation of other variables of interest to outcomes can be evaluated. The technique is similar to standard linear regression except that it is usually modified to include a term, t [33], that estimates the magnitude of variability among studies. Because of its explicit allowance for an estimation of variability among studies, it is similar to random-effects analysis of variance and thus may be called random-effects meta-regression. The effects of several variables in combination on the study outcome can be evaluated and corrected for, providing estimates of the size of pooled study effects after correction of the variables of interest. In this way, meta-regression analysis provides the ability to evaluate the relation of potential covariates and confounding variables to study effects. To assess the effect of baseline depression on risk, we completed a random-effects meta-regression analysis with the baseline depression predicting the risk of developing postoperative depression. Likewise, to assess the effect of the interval between the hysterectomy and assessment of psychological morbidity on risk, we completed a random-effects meta-regression analysis with the interval between the hysterectomy and assessment of psychological morbidity predicting the risk of developing postoperative anxiety. All analyses were completed using routines

Thirty-nine articles reported information on psychological wellbeing after hysterectomy for benign conditions. In five studies [34–38], there were no adequate predictors and/or outcomes [34– 38]. Also, three studies [14,39,40], reported the same data of included studies on different timelines. Additionally, Moorman et al. [41], did not use validated tools for investigations/assessment of psychological wellbeing. In three studies [42–44], the authors did not report psychological outcome preoperatively. Finally, five studies [16,31,45–47] were excluded due to insufficient data despite contacting the authors where possible. Fig. 1 shows the selection process. Seventeen studies were excluded in total. Table 2 provides an overview of these studies. The remaining 22 studies were included in the review (Table 3).

4. Results

4.1. Included studies 4.1.1. Descriptive data synthesis Study characteristics of the 22 citations included for review are depicted in Table 3. The included studies were published in the period 1976–2012, and originated from Sweden [19,24,48,49], Canada [5,50], United States [29,51–54], United Kingdom [23,55,56], Australia [1,30], New Zealand [57], Turkey [58,59], Finland [25] Egypt [60], and Nigeria [6]. It is clear that the majority of included studies originated from western societies; hence the findings may have some bias. The sample sizes ranged from 43 to

485 Citaons idenfied from literature search of PubMed, PsycINFO, and EMBASE electronic databases

449 Citaons excluded aer de-duplicaon based on screening tles and, or abstracts for inclusion and exclusion criteria

39 Potenally relevant citaons for inclusion

3 Citaons idenfied from reference lists

17 Citaons excluded 3 Psychological well being assessed postoperavely only 1 No validated psychological assessment tool 5 Inadequate predictor and/or outcome

22 Citaons included for review

5 Insufficient data

3 Duplicate studies

Fig. 1. Flow chart of study selection process.

Table 2 Characteristics of excluded studies. Country

Baldaro et al., 2003

Italy

Barglow et al., 1965

United States

Cooper et al., 2009

United Kingdom

Sample size 100

22 1448

Mean age (years)

Treatments

Follow-up period (weeks)

Comments (outcome, assessment)

Reason for exclusion

39

Not specified

48

SQ Anxiety score [(9.32  6.58), (4.42  4.33)]d Depression score [(8.87  5.54), (4.42  4.92)]d

Not specified 47–54a

AH, TL

48

Anxiety symptoms, semi-structured interviews

Does not separate the assessment of women undergoing TAH for benign conditions from women undergoing hysterectomy for non-benign reasons Insufficient data

(AH, VH)  BSO

N/A

Psychological wellbeing,

United Kingdom

60

41

Not specified

144

Ewalds-Kvist et al., 2005

Finland

65

43

32

Farquhar et al., 2002

New Zealand

323

>46

TAH  oophorectomy TAH

24

Farquhar et al., 2006

New Zealand

314

>46

TAH

144

Hurskainen et al., 2004

Finland

236

43

AH, LH, VH

48

Lalinec-Michaud et al., 1985

Canada

102

42

HYST

48

Leppert et al., 2007

United States

1140

41

Not specified

96

Moore et al., 1976

United States

47

HYST

12

Moorman et al., 2011

United States

830

Not specified 41

(AH, LH, VH)  BSO

48

Persson et al., 2010

Sweden

178

46

TAH, STH

5

Thornton et al., 1997

United Kingdom

62

44

TAH, TAH + BSO

24

Assesses women pre- and posthysterectomy in regards to those taking the HRT with the placebo group.

Insufficient data Part 1 of Faruquhar (2008) study reported same data on shorter follow up period of time Part 2 of Faruquhar (2008) study reported same data on shorter follow up period of time MBL was the predictor of psychological outcome after hysterectomy

Insufficient data (unspecified sample inclusion and exclusion criteria) The pre-op desire of (another) child was the predictor of psychological outcome

Insufficient data (unspecified sample inclusion and exclusion criteria) Uses the ‘‘MRS’’ to assess psychological wellbeing, categorizes the psychological symptoms by race and hysterectomy status Part 1 of Persson et al. (2010) study reported same data on shorter follow up period of time and assessed psychological wellbeing preoperatively only. Insufficient data (unspecified sample inclusion and exclusion criteria)

9

Coppen et al., 1981

GHQ-28 score > 4 H + BSO post-operative GHQ-28 score > 4 prevalence [25.5%] H BSO post-operative GHQ-28 score > 4 prevalence [26.9%] Global psychological wellbeing, GHQ score Depressive symptoms, BDI score Active BDI score [8.9 (1.6)], [6.5(5.0)]d Placebo BDI scores [6.1(1.1)], [1.9{0.9)]d Active GHQ-28 score [30.5(2.2)], [26.2 (9.3)]d Placebo GHQ-28 scores [28.9(2.8)], [21.1{2.5)]d Depressive symptoms, BDI Anxiety symptoms, TMAS Depressive symptoms, CES-D score  16 Hyst prevalence of depression: [43%, 27%] Hyst + BSO prevalence of depression: [67%, 48%] Depressive symptoms, CES-D score  16 HYST prevalence of depression: [43%, 27%] HYST + BSO prevalence of depression: [67%, 50%] Depressive symptoms, BDI Anxiety symptoms, STAI BDI score changes in HYST1 [ 1.2 (0.40)]c BDI score changes in HYST 2 [ 2.1 (0.51)]c STAI score changes in HYST1 [ 1.6 (0.75)]c STAI score changes in HYST2 [ 4 (0.76)]c Depressive symptoms, SDS-D scores  51 and HAM-D scores  17 HYST prevalence of depression: (16%, 13%) Depressive scores, POMS Anxiety scores, POMS [Pre, post] Depression scores Anxiety scores Yesb [24.7 (20.1), 15.3 (19.7)] [45.5 (26.3), 26.7 (24.3)] b No [18.3 (18.6), 11.3 (15.5)] [37.2 (26.5), 20.8 (22.2)] Maybeb [26.1 (17.4), 14.6 (17.1)] [48.1 (24.1), 28.0 (23.1)] Depressive symptoms, SDS-D scores  50 HYST prevalence of depression: (36%, 32%) Depressive and anxiety symptoms, MRS summed score African American prevalence of depression: [36.6%, 31%] White prevalence of depression: [46%, 36.4%] African American prevalence of anxiety: [33.5%, 22.3%] Whiten prevalence of anxiety: [42.7%, 33.2%] Depression, BDI score, TAH [6.7 (6.3)]d Depression, BDI score, STH [6.2 (6.0)]d Anxiety STAI score, TAH [34.9 (10.0)]d Anxiety STAI score, STH [36.0 (9.8)]d Psychological wellbeing score, PGWB TAH [98.7 (16.2)]d Psychological wellbeing score, PGWB STH [98.8 (15.6)]d Depressive symptoms, HADS score [7.1 (3.9), 5.4 (3.7)] Anxiety symptoms, HADS score [10.5 (4.4), 7.7 (4.9)]

Psychological wellbeing was assessed post-operatively only. Questionnaires were distributed in different timelines regarding hysterectomy status.

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

Study identification

Does not separate the assessment of women undergoing TAH for benign conditions from women undergoing hysterectomy for non-benign reasons 16 AH, LH, VH 47 Taiwan Yen et al., 2008

68

Malaysia Wong et al., 2012

279

>18

AH, LH, VH

N/A

Psychological wellbeing was assessed post-operatively only

Depressive symptoms, SDS score > 41 Anxiety symptoms, SAS score > 40 Prevalence of depression post-operatively (4.8%) Prevalence of anxiety post-operatively (1.9) Depressive and anxiety symptoms, Author’s own developed questionnaire Prevalence of depression post-operatively (33.7%) Prevalence of anxiety post-operatively (46.6%) Depressive symptoms, HAM-D Anxiety symptoms, HAM-A Depression scores [6.85 (4.22), 5.57 (4.76)] Anxiety scores [7.31 (4.29), 5.87 (4.05)] China Wang et al., 2007

105

>20

AH, LH, VH

N/A

Reason for exclusion Comments (outcome, assessment) Follow-up period (weeks) Treatments Mean age (years) Sample size Country Study identification

Table 2 (Continued )

AH, abdominal hysterectomy; LH, laparascopic hysterectomy; VH, vaginal hysterectomy; TAH, total abdominal hysterectomy; STH, sub-total hysterectomy; BSO, bilateral salpingo-oophoroctomy; FSH, follicle-stimulating hormone; PGWB: Psychological General Well-Being; CES-D, Center for Epidemiological Studies Depression Scale; STAI, Spielberger’s State-Trait Anxiety Inventory, POMS: Profile of Moods Scale; BDI: Beck’s Depression Inventory; GHQ-28, General Health Questionnaire-28; HAM-A, Hamilton Rating Scale for Anxiety; SDS, Zung Self-Rating Depression Scale; SAS, Zung self-rating anxiety scale; HAM-D, Hamilton rating scale for depression; HAM-A, Hamilton Rating Scale for Anxiety; SQ, symptom questionnaire; MBL, menstrual blood loss. a Range. b Patients were categorized by desire for a (another) child (yes, no, maybe), Hyst1, women hysterectomized without menorrhagia; Hyst2, women hysterectomized with menorrhagia. c Values are score changes and standard errors (SE). d Values are M(SD).

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

Psychological wellbeing was assessed post-operatively only

10

1162, and resulted in a total of 5978 participants across studies. The mean age of participants ranged from 40 to 51 years. Included studies used different scales to assess psychological wellbeing including; Center for Epidemiologic Studies Depression scale (CESD) using a cut-off point of 16 for symptoms scores [5,52,57]; Beck’s Depression Inventory (BDI) using different cut-off points [24,25,30,48,50,58–60], General Health Questionnaire (GHQ-28) with a cut-off point of 4 [6,55,60], Profile of Mood States (POMS) with a cut-off point of 2.6 [51,54], Spielberger’s State-Trait Anxiety Inventory (STAI) [24,25,30,48], Hamilton Rating Scale for Anxiety (HAM-A) with a cut-off score of 14 [60], the Hospital Anxiety and Depression Scale (HADS) [23], Zung Self-Rating Depression Scale (SDS) [29,53], the PSE-Index of Definition (PSE-ID) with a cut-off level of 5 [1,56] and the Psychological General Well-Being (PGWB) [19,24,48,49]. Participants were followed for 4–480 weeks. Quality percentages ranged from 50% to 92% (Table 1). 4.1.2. Quantitative data synthesis The data extracted were the proportion of patients diagnosed to have symptoms of psychological morbidity in the form of anxiety and/or depression, using a validated tool or an interview depending on the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV). Additionally, the continuous data of the scores of validated tools regarding psychological wellbeing were also extracted when reported. Binary and continuous data were entered and analyzed using Rev-Man 5.1. Forest plots were created for graphical presentation of the data (Figs. 2–4). In each plot there are a diamond and squares with horizontal lines crossing over. The diamond illustrates the overall result of the meta-analysis. The middle of the diamond sits on the value for the overall effect estimate (RR or WstdMD) and the width of the diamond depicts the width of the overall CI. The squares represent the effect estimates (RR or WstdMD) of each study individually. The horizontal lines through the squares depict the length of the CI of the corresponding study. The size of the square corresponds directly to the ‘‘weight’’ of the study in the form of the sample size in the represented study. All analyses show a test of heterogeneity, which is variation among the results of individual studies beyond that expected from chance. Additionally, x2 statistics with the degree of freedom (df) are also reported for each analysis. The y2 statistic, which is a measure of the extent of inconsistency among the results; and the Z statistic with its P value that shows a test for overall effect were also reported. This review aimed to determine whether hysterectomy had adverse effects on psychological wellbeing, hence the use of RR in the meta-analysis. Fig. 2 shows the forest plots for random-effects meta-analysis for symptoms of depression after hysterectomy from 10 studies stratified by hormonal status and for all studies combined. The RR for studies that involved postmenopausal women or women with oophorectomy (mixed) was 1.52 (95% CI 0.89–2.60; P = 0.13) and for studies that involved premenopausal women only (premenopausal), 2.04 (95% CI 1.42–2.91; P = 0.0001). The estimate for all studies combining both studies with mixed and premenopausal hormonal status was 1.69 (95% CI 1.19–2.38; P = 0.003). Overall, symptoms of depression were significantly reduced after hysterectomy (RR = 1.69, 95% CI 1.19–2.38). The heterogeneity was statistically significant (P < 0.00001, y2 = 88%), which was mostly due to a single outlier study [60] that had an RR opposite in direction to the pooled RR (showing depression was increased after hysterectomy). After exclusion of the single outlier study, the pooled RR increased to 2.03 (95% CI 1.73–2.38) and there was less heterogeneity evident (I2 = 40%, P = 0.1). Our sensitivity analyses showed that the pooled RR was increased after exclusion of studies with short follow-up durations (RR = 1.96, 95% CI 1.60–2.28). A funnel plot was produced and showed no evidence of publication bias.

Table 3 Characteristics of included studies. Study identification

Country

Study design

Sample size

United Kingdom

Randomized controlled pre-/post-trial

204

Aziz et al., 2005

Sweden

Before and after study design

323

Bayram et al., 2008

Turkey

Before and after study with no control group

93

Donoghue et al., 2003

Australia

Before and after study with no control group

60

Ellstro¨m et al., 2003

Sweden

72

Farquhar et al., 2008

New Zealand

Randomized controlled pre-/post-trial Time series study design

Flory et al., 2006

Canada

Randomized controlled pre-/post-trial

133

Gath et al., 1982 Gibson et al., 2012

United Kingdom United States

Time series study design Three-arm prospective cohort study with beforeafter design

54 815

Helmy et al., 2008

Egypt

Before and after study with no control group

Hurskainen et al., 2001

Finland

Randomized controlled pre-/post-multicentre trial

Kjerulff et al., 2000

United States

Lambden et al., 1997

Martin et al., 1980

331

96

Major inclusion criteria

Major exclusion criteria

<50 years of age, women weighed <100 kg, had a clinical diagnosis of dysfunctional uterine bleeding (uterus less than size of a pregnancy of 10 weeks, and normal endometrial histology). 45–55 years, perimenopausal women, with last menstruation less than 12 months ago, scheduled for elective hysterectomy on benign indications, sexually active (at least one episode of intercourse/month for the past 6 months), being part of a partner relationship, have not previously sought medical help for sexual problems <65 years, married women, undergoing hysterectomy for benign conditions, had not had previous surgery for ostomy, pelvic procedures or mastectomy, and who did not have a mental illness. 30–60 years, women, undergoing hysterectomy for benign conditions. Speaking English as their main language.

Surgery for conditions other that DUB, malignant histology of the endometrium Hysterectomy for malignant conditions

Women scheduled for abdominal hysterectomy due to benign disorders and assessed as not suitable for vaginal hysterectomy <46 years, women without gynecological malignancy and had an early follicular phase (FSH) of <40 IU/L prior to surgery and retained one or both ovaries at the time of surgery. 18–55 years, premenopausal women, fluent in French language.

Women awaiting hysterectomy for menorrhagia of benign origin At baseline, all participants were expected to have an intact uterus and at least one ovary, aged 42–52 years; not pregnant; not using reproductive hormones; and having one or more menstrual cycles in the 3 months before the interview. 18 years, women undergoing hysterectomy for benign conditions, with no personal or family history of psychiatric illness

236

35–49 years, menstruating women, had completed their family, and were eligible for hysterectomy.

Time series study design

1162

United States

Time series study design

178

United States

Before and after study with no control group

18 years, women undergoing hysterectomy for benign indications including cervical dysplasia or endometrial hyperplasia. 18 years, women, undergoing abdominal or vaginal hysterectomy for nononcologic reasons with and without anterioposterior repair. Not specified

44

Not specified

Non-return of questionnaire or noncontactable on follow ups, non-English competence Not specified Gynecological malignancy, FSH levels of 40 IU/L prior to surgery Prior oophorectomy, uterine prolapse, chemotherapy, or neoplasia in the uterus/cervix. Not specified Not specified

Hysterectomy was performed as an emergency procedure or to manage premalignant or malignant conditions. Hysterectomy for: submucous fibroids, endometrial polyps, ovarian tumors or cysts (diameter > 5 cm), cervical disease, urinary and bowel symptoms or pain due to large fibroids, lack of indication for hysterectomy, history of cancer, menopause, severe depression, metrorrhagia as a main complaint, previous treatment failure with levonorgestrelreleasing IUS, severe acne, and uterine malformation. Participants who were found to have cancer at hysterectomy

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

Alexander et al., 1996

Population

Not specified

Not specified

11

12

Table 3 (Continued ) Country

Study design

Sample size

Moraloglu et al., 2007

Turkey

Time series study design

66

Okunlola et al., 2009

Nigeria

Before and after study with no control group

43

Persson et al., 2006

Sweden

Randomized controlled pre-/post-multicentre trial

119

18 years, women undergoing hysterectomy for benign conditions in whom laparoscopic hysterectomy was considered to be possible, fluently spoke and understood the Swedish language, and at least one ovary was to be preserved at the operation.

Persson et al., 2010

Sweden

Randomized controlled pre-/post-multicentre trial

179

18 years, women undergoing abdominal hysterectomy for benign conditions. Fluently spoke and understood the Swedish language, and at least one ovary was to be preserved at the operation.

Rohl et al., 2008

United States

Ryan et al., 1989

Australia

Two-arm prospective cohort study with beforeafter design Time series study design

Thakar et al., 2004

United Kingdom

Randomized controlled pre-/post-multicentre trial

279

Vandyk et al., 2011

Canada

Single-arm prospective cohort study with beforeafter design

384

1047

60

Population Major inclusion criteria

Major exclusion criteria

Women aged 40–77 years, with a non-malignant condition, women hysterectomy was planned in gynecology department of Zekai Tahir Burak women’s health education and research hospital, and were able to be reached by telephone for follow up interviews. 18 years, women undergoing hysterectomy for benign conditions.

Not specified

18 years, premenopausal women, undergoing hysterectomy for benign indications including cervical dysplasia or endometrial hyperplasia. Women aged 30–55 years, speaking English as their main language, having an ongoing sexual relationship, and undergoing hysterectomy for benign conditions. >18 years, women, undergoing abdominal hysterectomy for a benign indication

18 years, English speaking women, undergoing hysterectomy for benign conditions

Malignant conditions, previous mental ill health or major surgeries defined as any surgery that necessitated general anesthesia, sub-arachnoid or epidural block Genital tract malignancy, preoperative treatment with gonadotrophinreleasing hormone analogs, postmenopausal women without (HRT) and severe psychiatric disorders. Genital malignancy, previous or present cervical dysplasia, rapidly growing myomas where malignancy could not be ruled out preoperatively, preoperative treatment with gonadotrophin releasing hormone analogs, postmenopausal women without (HRT), and severe psychiatric disorders. Post-menopausal women, hysterectomy indicated for malignant conditions Not specified

>60 years, suspected malignancy, weight over 100 kg, previous pelvic surgery, known endometriosis, abnormal cervical smears, bothersome urinary incontinence and uterine prolapse. Enrollment in another study or diagnosis of Alzheimer’s disease or other forms of cognitive impairment (as indicated in the medical record) affecting the ability to comprehend study questionnaires

Mean age (years)

Treatments

Hormonal status

Control conditions

Follow-up period (weeks)

40 48 45 46 50 >46 42 42 51 35–60a 43

HYST, conservative surgery HYST  BSO AH, VH AH, VH, TAH + BSO, TAH LH, TAH AH, VH, LAVH TAH, STH, non-hysterectomy surgery (AH, VH)  oophorectomy HYST  BSO AH, VH (AH, LH, VH)  BSO

Premenopausal Premenopausal Mixed Mixed Mixed Premenopausal Premenopausal Premenopausal Premenopausal Mixed Premenopausal

Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative

48 48 12 12 48 240 24–28 weeks 72 Up to 480 weeks 4 48

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

Study identification

Table 3 (Continued ) Mean age (years)

Treatments

Hormonal status

Control conditions

Follow-up period (weeks)

(AH, VH, LAVH)  oophorectomy AH, VH HYST  BSO (TAH, VH)  oophorectomy HYST TAH, LH TAH, STH (TAH, STH, TVH, LAVH)  BSO (TAH, VH)  oophorectomy TAH, STH HYST  BSO

Mixed Mixed Not specified Mixed Mixed Mixed Mixed Premenopausal Not specified Mixed Mixed

Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative Pre-operative

96 16 48 12 4–6 weeks 24 48 48 56 48 24

Outcome, assessment

Outcome [M(SD) unless otherwise specified] (pre-operative, post-operative)

Depressive symptoms, HADS score Anxiety symptoms, HADS score Depressive symptoms, PGWB depression score Anxiety symptoms, PGWB anxiety score

Depression [5.8 (5–0 to 6–7), 1.6 (1–3 to 1–9)]b Anxiety [9.5 (4.2), 5.5 (3.8)] HYST depression [15.01 (2.29), 15.48 (2.86)] HYST + BSO depression [14.55 (2.45), 15.37 (2.53)] HYST anxiety [22.78 (4.34), 23.08 (4.82)] HYST + BSO anxiety [22.23 (4.48), 23.08 (4.50)]

Depressive symptoms, BDI score  10 Depressive symptoms, BDI score  17 Anxiety symptoms, STAI score  46 Depressive symptoms, PGWB depression score Anxiety symptoms, PGWB anxiety score

Depressive symptoms, CES-D score  16 Depressive symptoms, BDI score Neurotic depression, PSE level  5 Anxiety state, PSE level  5 Depressive symptoms, CES-D score Anxiety symptoms, CES-A score

Depressive symptoms, BDI score  10 Anxiety symptoms, HAM-A score  14 Global psychiatric morbidity, GHQ-28 score  4 Depressive symptoms, BDI score Anxiety symptoms, STAI score Depressive symptoms, POMS score > 2.6 Anxiety symptoms, POMS score > 2.6 Depressive symptoms, SDS score Depressive symptoms, SDS-D scores Depressive symptoms, BDI score Depressive symptoms, GHQ-28 score  4 Anxiety symptoms, GHQ-28 score  4 Depressive symptoms, BDI score Anxiety symptoms, STAI score Psychological wellbeing, PGWB summed score

Depression [14.9(10.8), 6.4 (6.8)] Anxiety [40.8(13.1), 35.5(10.7)] LH depression [14.5 (3.2), 15.8 (2.5)] AH depression [14.4 (3.0) 14.9 (3.2)] LH anxiety [22.5 (5.4), 24.6 (4.1)] AH anxiety [21.3 (4.9), 23.1 (5.2)]

Outcome prevalence% (pre-operative, post-operative)

(38.8%, 17.2%) Depression (34%, 8%) Anxiety (29%, 22%)

(45%, 28%) [5.78 (6.28), 3.03 (3.78)] Neurotic depression (28.2%, 6.8%) Anxiety state (18%, 8.8%) Depression HYST [0.66 (1.08)]c Depression HYST + BSO [0.41 (0.84)]c Anxiety HYST [0.18 (0.27)]c Anxiety HYST + BSO [0.14 (0.21)]c Depression (16%, 58%) Anxiety (21%, 38.5%) Depression [4
2 (3
4 to 5
0), 2.5 (1.1 to 3.9)]b Anxiety [31 (30
0 to 32
0), 28 (25.8 to 30.2)]b Depression (28%, 12.4%) Anxiety (65.4%, 25.1%) [46.4 (10.9), 39.7 (10.6)] [42.8  1.6, 41.2  1.6]d Depression [15.8 (6.9), 12.3 (4.3)] Depression (6.7%, 4.4%) Anxiety (44.4%, 51.2%)

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

43 40 44 47 49 44 46 41 40 44 18

BDI AH [6.9 (6.1), 4.0 (5.2)] BDI LH [6.6 (5.8), 5.3 (6.8)] STAI AH [34.7 (10.1), 31.7 (9.2)] STAI LH [35.6 (9.1), 33.6 (10.2)] PGWB AH [96.5 (16.5), 106.1 (16.0)] PGWB LH [96.7 (17.9), 104.7 (18.5)]

13

(36%, 22%) Depression [0.6 (1.4), 0.4 (1.2)] Anxiety [2.1 (2.3), 1.1 (1.9)]

PSE ID score [4.33 (1.42), 3.10 (1.69)] POMS depression score [10.4 (11.9), 8.9 (11.0)]

Neurotic depression, PSE level  5 Anxiety State, PSE level  5 Depressive symptoms, POMS depression score Depressive symptoms, GHQ-28 depression score Anxiety symptoms, GHQ-28 anxiety score Depressive symptoms, CES-D score  16

Depressive symptoms, POMS score > 2.6

BDI TAH [6.7 (6.3), 4.5 (6.1)] BDI STH [6.2 (6.0), 4.0 (5.6)] STAI TAH [34.9 (10.0), 32.4 (10.4)] STAI STH [36.0 (9.8), 32.6 (9.1)] PGWB TAH [98.7 (16.2), 105.0 (16.0)] PGWB STH [98.8 (15.6), 105.7 (14.1)] Depressive symptoms, BDI score Anxiety symptoms, STAI score Psychological wellbeing, PGWB summed score

Outcome prevalence% (pre-operative, post-operative) Outcome [M(SD) unless otherwise specified] (pre-operative, post-operative) Outcome, assessment

Table 3 (Continued )

AH, abdominal hysterectomy; LH, laparoscopic hysterectomy; VH, vaginal hysterectomy; LAVH, laparoscopic assisted vaginal hysterectomy; TAH, total abdominal hysterectomy; STH, sub-total hysterectomy; HYST, hysterectomy; BSO, bilateral salpingo-oophorectomy; FSH, follicle-stimulating hormone; HRT, hormone replacement therapy; DUB, dysfunctional uterine bleeding; PGWB, Psychological General Well-Being; CES-D, Center for Epidemiological Studies Depression Scale; STAI, Spielberger’s State-Trait Anxiety Inventory; POMS, Profile of Moods Scale; WHQ, women health questionnaire; BDI, Beck’s Depression Inventory; GHQ-28, General Health Questionnaire-28; HAM-A, Hamilton Rating Scale for Anxiety; BSI, brief symptom inventory; TMAS, Taylor’s manifest anxiety scale; HADS, Hospital Anxiety and Depression Scale; SF-36, SF-36 health survey; SDS, Zung Self-Rating Depression Scale; HAM-D, Hamilton rating scale for depression; HAM-A, Hamilton Rating Scale for Anxiety; SQ, symptom questionnaire; MBL, menstrual blood loss; PSE, present state examination; ID, Index of Definition; PSE-ID, PSE-Index of Definition (ID). a Range. b Values are mean (interquartile range). c Values are mean change in mood score coefficient (SE). d Values are mean  SE.

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19 HYST + BSO (27%, 9%) HYST-BSO (30%, 15%) Neurotic depression (28.3%, 18.3%) Anxiety state (20%, 6.7%)

14

We evaluated the relation of preoperative depressive symptoms and observed risk through meta-regression of proportion of preoperative depressive symptoms and log RRs. This analysis was undertaken to further assess the possibility that the observed decrease in risk of developing depressive symptoms post hysterectomy might be related to study sample characteristics (i.e. preoperative depressive state) not accounted for in stratified analyses. The RR for each study on the proportion of preoperative depressive symptoms in percentage of the studied group, showed a statistically significant relation. (RR = 4.47; SE = 2.12, 95% CI 0.30– 0.63). The interpretation is therefore that the effect size, the log relative risk of postoperative depressive symptoms increases with higher preoperative depressive symptoms. There appeared no significant heterogeneity between the studies once the influence of preoperative depressive symptoms is taken into consideration (Table 4). Fig. 3 shows the meta-analysis results for symptoms of anxiety after hysterectomy from six studies. The heterogeneity was statistically significant (P < 0.00001, y2 = 92%), which was mostly due to two outlier studies [6,60], that had an RR opposite in direction to the pooled RR (showing anxiety was increased after hysterectomy). After exclusion of the outlier studies, the pooled RR significantly increased to 2.18 (95% CI 1.49–3.20; P < 0.0001) and there was less heterogeneity evident (I2 = 44%, P = 0.14). Overall, there was no significant change of symptoms of anxiety after hysterectomy (RR = 1.41, 95% CI 0.72–2.75). Our sensitivity analyses showed that the pooled RR was similar after exclusion of lower quality studies (RR = 1.43, 95% CI 0.65–3.17), but was significantly increased after exclusion of studies with short followup durations (RR = 2.62, 95% CI 2.36–2.91). A funnel plot was produced and showed no evidence of publication bias. Meta-regression of the log RR for each study on the interval between the hysterectomy and assessment of anxiety on risk showed that the interval in weeks showed a statistically significant relation (point estimate = 0.01; SE = 0.003; 95% CI 0.01–0.02). The interval between the hysterectomy and assessment of anxiety was thus positively related to increased risk for developing anxiety. There appeared no significant heterogeneity between the studies once the influence of the interval between the hysterectomy and assessment of anxiety is taken into consideration (Table 5). Finally, Fig. 4 shows a meta-analysis of changes in depression scores after hysterectomy in 10 studies stratified by type of hysterectomy and for all studies combined. Two of the studies [19,49] used the Psychological General Well-Being (PGWB) as the primary depression inventory; thus the data were reversed for these studies in order to fit in the meta-analysis. Two of the studies [23,53] were not estimable as there was no record of standard deviation in the report. Overall, there was a significant reduction in depression scores after hysterectomy, weighted standard mean difference [StdMD] 0.38 (95% CI 0.27–0.49). There was no overall significant heterogeneity among the studies (P = 0.17, y2 = 29%), and the subgroup heterogeneity was also not statistically significant (P = 0.83, y2 = 0%). A funnel plot was produced and did not show any signs of asymmetry. 5. Discussion 5.1. Main findings and interpretations The results of this systematic review of before and after studies suggest that hysterectomy for benign gynecological conditions is not adversely associated with depression or anxiety outcomes. In contrast, our meta-analysis findings suggest that hysterectomy may reduce symptoms of depression. There was significant heterogeneity in outcomes across the studies we reviewed and our analyses of moderators produced some interesting findings.

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

Study or Subgroup 1.1.1 Mixed Bayram et al, 2008 Donoghue et al, 2003 Helmy et al, 2008 Kjerulff et al, 2000 Okunlola et al, 2009 Ryan et al, 1989 Vandyk et al, 2011 Subtotal (95% CI)

Preoperatively Postoperatively Risk Ratio Events Total Events Total Weight M-H, Random, 95% CI 37 30 15 364 3 17 138

93 87 96 1299 45 60 384 2064

16 5 56 144 2 11 83

93 60 96 1162 43 60 384 1898

10.7% 7.3% 10.9% 13.5% 3.1% 9.2% 13.2% 67.9%

15

Risk Ratio M-H, Random, 95% CI

2.31 [1.39, 3.86] 4.14 [1.70, 10.05] 0.27 [0.16, 0.44] 2.26 [1.90, 2.70] 1.43 [0.25, 8.16] 1.55 [0.79, 3.02] 1.66 [1.32, 2.10] 1.52 [0.89, 2.60]

Total events 604 317 Heterogeneity: Tau² = 0.41; Chi² = 68.47, df = 6 (P < 0.00001); I² = 91% Test for overall effect: Z = 1.53 (P = 0.13) 1.1.2 Premenopausal Farquhar et al, 2008 Gath et al, 1982 Rohl et al, 2008 Subtotal (95% CI)

83 15 298

202 54 1047 1303

36 4 132

134 12.5% 54 6.2% 1047 13.5% 1235 32.1%

1.53 [1.11, 2.12] 3.75 [1.33, 10.57] 2.26 [1.87, 2.72] 2.04 [1.42, 2.91]

Total events 396 172 Heterogeneity: Tau² = 0.06; Chi² = 5.48, df = 2 (P = 0.06); I² = 63% Test for overall effect: Z = 3.88 (P = 0.0001) Total (95% CI)

3367

3133 100.0%

1.69 [1.19, 2.38]

1000 Total events 489 Heterogeneity: Tau² = 0.22; Chi² = 75.94, df = 9 (P < 0.00001); I² = 88% Test for overall effect: Z = 2.97 (P = 0.003) Test for subgroup differences: Chi² = 0.79, df = 1 (P = 0.37), I² = 0%

0.01 0.1 1 10 100 Preoperatively Postoperatively

Mixed: Studies involving premenopausal women and postmenopausal women or women with oophorectomy. Premenopausal: Studies involving

Fig. 2. Forest plot for depressive symptoms before and after hysterectomy.

Differences in baseline depressive state across studies appeared to have an effect on study outcomes. Also, the effect estimate for depression outcomes increased in size in studies among premenopausal women, in which there is no influence of the menopausal transition on the mood status, and with long follow up durations, during which most women undergoing hysterectomy return to normal physical and psychological functioning [5], remained robust after exclusion of low quality studies. Likewise, the effect estimate for anxiety outcomes remained robust after exclusion of low quality studies, and substantially increased in size for studies with long follow-up durations, during which the surgery-induced anxiety was minimal.

Overall, our meta-analysis result is in keeping with studies [30,54,56,57,59] that fail to support the current theory suggesting that hysterectomy is associated with adverse psychological wellbeing [6,43,60]. Our analysis further indicates that there is no increase in psychological morbidity such as anxiety and depression post-operatively. Concerning the type of hysterectomy, our study reported a significant improvement in the depression inventories’ scores regardless of the type of hysterectomy, with favorable outcomes for all types of hysterectomy, which is consistent with other studies [19,24,48–50,52]. Furthermore, evidence from randomized trials suggests that [55] hysterectomy, whether total or subtotal

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

16

Study or Subgroup Donoghue et al, 2003 Gath et al, 1982 Helmy et al, 2008 Kjerulff et al, 2000 Okunlola et al, 2009 Ryan et al, 1989 Total (95% CI)

Risk Ratio Preoperatively Postoperatively Total Weight M-H, Random, 95% CI Events Total Events 25 10 20 851 20 12

87 54 96 1299 45 60

17.2% 13.7% 18.1% 19.7% 18.2% 13.1%

1.33 [0.74, 2.38] 2.00 [0.73, 5.46] 0.54 [0.34, 0.86] 2.62 [2.36, 2.92] 0.87 [0.56, 1.35] 3.00 [1.03, 8.78]

1475 100.0%

1.41 [0.72, 2.75]

60 54 96 1162 43 60

13 5 37 290 22 4

1641

371 938 Total events Heterogeneity: Tau² = 0.59; Chi² = 65.73, df = 5 (P < 0.00001); I² = 92% Test for overall effect: Z = 0.99 (P = 0.32)

Risk Ratio M-H, Random, 95% CI

0.01 0.1 1 10 100 Preoperatively Postoperatively

Fig. 3. Forest plot for anxiety symptoms before and after hysterectomy.

Table 4 Random-effects model – regression results for depressive symptoms. Random effects, Z-distribution Point estimate Intercept 0.83707 Proportion of preoperative depression (%) 4.46929 Test of the model Simultaneous test that all coefficients (excluding intercept) are zero Q*model = 4.43265, df = 1, p-value = 0.03526 Goodness of fit: test that unexplained variance is zero t2 = 0.23713, Qresid = 10.98559, df = 8, p = 0.20252

Standard error

95% CI

z-score

P-value

0.67137 2.12279

2.15292 to 0.47879 0.30870 to 8.62987

1.24681 2.10539

0.21247 0.03526

Table 5 Random-effects model – regression results for anxiety symptoms. Random effects, Z-distribution

Intercept Interval between the hysterectomy and assessment of anxiety (weeks) Test of the model Simultaneous test that all coefficients (excluding intercept) are zero Q*model = 17.04681, df = 1, p-value = 0.00004 Goodness of fit: test that unexplained variance is zero t2 = 0.04787, Qresid = 3.82232, df = 4, p = 0.43059

Point estimate

Standard error

95% CI

z-score

P-value

0.26968 0.01350

0.20545 0.00327

0.67237 to 0.13300 0.00710–0.01993

1.31263 4.12878

0.18931 0.00004

[55], laparoscopic or abdominal [48], may improve quality of life and psychological outcome, and that the surgical technique has no influence on the psychological outcome. The operative indication as a potential factor to impact psychological wellbeing after hysterectomy was studied by Okunlola et al. [6]. They reported that fibroid uterus as an indication for hysterectomy seemed to be a predictor for developing postoperative depression and anxiety compared to other benign operative indications [6]. Perhaps the burden of the presenting symptoms, the myth/misconceptions about the etiology, and fear of recurrence, or simply the fact that fibroid uterus is the commonest indication for hysterectomy, are plausible justifications for such findings [6]. This review, however, shows a significant postoperative reduction in the symptoms of depression and depression scores, apart from the pre-operative indication which has been depicted in Figs. 2 and 4 respectively. This may suggest that women with non-malignant indications usually find relief with the disappearance of their distressing preoperative complaints such as menorrhagia, or chronic pelvic pain after the operation. Additionally, postoperative sexual

wellbeing after hysterectomy must be taken into account. The current literature [50,61] suggests that sexual wellbeing, in the form of sexual pleasure, arousal and desire, improves after hysterectomy regardless of the surgical technique, resulting in a better psychological status, and ultimately better quality of life [55]. It is of key importance, however, to discuss psychosexual aspects with the patients before performing hysterectomy. In particular, the severity of preoperative symptoms needs to be approached cautiously, and hysterectomy benefits must outweigh risks to assure the necessity of the procedure. It is worth mentioning that an alarming report revealed that almost 70% of performed hysterectomies were in fact unneeded, and 76% of these did not even meet the American College of Obstetricians and Gynecologists (ACOG) guidelines on hysterectomy [62]. 5.2. Strength and limitations Meta-analysis is statistically a powerful study designs, as it increases the precision of the overall results rather than the result

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

Study or Subgroup 2.1.1 LH Ellström et al, 2003 Persson et al, 2006 Subtotal (95% CI)

Preoperative Postoperative Std. Mean Difference Mean SD Total Mean SD Total Weight IV, Random, 95% CI 3.5 6.6

35 64 99

3.2 5.8

2.2 5.3

5.2 6.8

35 63 98

4.6% 7.5% 12.1%

0.30 [-0.17, 0.77] 0.20 [-0.14, 0.55] 0.24 [-0.04, 0.52]

4.8% 6.9% 9.2% 20.9%

0.16 [-0.30, 0.62] 0.51 [0.14, 0.88] 0.35 [0.05, 0.66] 0.36 [0.15, 0.57]

9.9% 9.9%

0.38 [0.09, 0.67] 0.38 [0.09, 0.67]

17

Std. Mean Difference IV, Random, 95% CI

Heterogeneity: Tau² = 0.00; Chi² = 0.10, df = 1 (P = 0.76); I² = 0% Test for overall effect: Z = 1.66 (P = 0.10) 2.1.2 TAH Ellström et al, 2003 Persson et al, 2006 Persson et al, 2010 Subtotal (95% CI)

3.6 6.9 6.7

3 6.1 6.3

37 61 85 183

3.1 4 4.5

3.2 5.2 6.1

37 56 85 178

Heterogeneity: Tau² = 0.00; Chi² = 1.35, df = 2 (P = 0.51); I² = 0% Test for overall effect: Z = 3.40 (P = 0.0007) 2.1.3 STH Persson et al, 2010 Subtotal (95% CI)

6.2

94 94

6

5.6

94 94

0 1.6 2.52 2.86 3.03 3.78 39.7 10.6 0 41.2 12.3 4.3

204 217 63 178 44 66 772

4

Heterogeneity: Not applicable Test for overall effect: Z = 2.57 (P = 0.01) 2.1.4 HYST Alexander et al, 1996 Aziz et al, 2005 Flory et al, 2006 Lambden et al, 1997 Martin et al, 1980 Moraloglu et al, 2007 Subtotal (95% CI)

0 5.8 2.99 2.29 5.78 6.28 46.4 10.9 0 42.8 15.8 6.9

204 217 77 178 44 66 786

7.5% 46.3%

Not estimable 0.18 [-0.01, 0.37] 0.52 [0.18, 0.85] 0.62 [0.41, 0.83] Not estimable 0.61 [0.26, 0.95] 0.47 [0.22, 0.71]

10.8% 10.8%

0.33 [0.06, 0.60] 0.33 [0.06, 0.60]

1248 100.0%

0.38 [0.27, 0.49]

16.5% 7.8% 14.5%

Heterogeneity: Tau² = 0.05; Chi² = 11.01, df = 3 (P = 0.01); I² = 73% Test for overall effect: Z = 3.67 (P = 0.0002) 2.1.5 HYST+BSO Aziz et al, 2005 Subtotal (95% CI)

3.45 2.45

106 106

2.63 2.53

106 106

Heterogeneity: Not applicable Test for overall effect: Z = 2.37 (P = 0.02) Total (95% CI)

1268

Heterogeneity: Tau² = 0.01; Chi² = 14.06, df = 10 (P = 0.17); I² = 29% Test for overall effect: Z = 6.90 (P < 0.00001) Test for subgroup differences: Chi² = 1.49, df = 4 (P = 0.83), I² = 0%

-2

-1 0 1 Preoperative Postoperative

2

LH: Laparoscopic Hysterectomy, TAH: Total Abdominal Hysterectomy, STH: Sub-Total Hysterectomy, BSO: Bilateral Salpingo-Oophorectomy, HYST: Hysterectomy Fig. 4. Forest plot for depression scores before and after hysterectomy.

of an individual study, reducing the likelihood that the results could be affected by chance. The findings of this review, however, should be balanced against several noteworthy limitations. Firstly, only a small number of studies in specific populations were included, and some of the studies included in the meta-analysis

had a small sample size. Secondly, systematic reviews are known to be subject to publication bias, in which studies with positive results are more likely to be published, and selection bias, as the studies included in the review were chosen after meeting predefined inclusion criteria. A random effects model was used

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19

18

for adjustment. This meta-analysis showed an overall decreased risk of developing depression and anxiety states after hysterectomy for benign conditions. Yet, one should acknowledge the difference between scoring above a cut-off point on a validated tool and having a clinical diagnosis of psychological morbidity, hence the need for guidelines on the best way to assess and follow psychological wellbeing prior to and after hysterectomy, with long-term follow-up to assess the clinical reliability of our observations. Finally, the limitations in terms of sample provide us with good understanding of specific groups, but it is difficult to know the extent to which this information can be generalized more broadly to women in developing countries as most of the present studies were conducted in industrialized countries. The literature clearly points out cultural aspects to mental health. Moreover, Lalinec-Michaud et al. [10] suggested that cultural factors may have an influence on the response to hysterectomy of women from different ethnic backgrounds, thereby influencing the findings across different study populations. 6. Conclusion In summary, limited data from before and after studies suggest that hysterectomy for benign gynecological conditions is not adversely associated with anxiety and may be positively rather than adversely associated with depression. Long-term controlled studies among women in developed and developing countries need to be conducted to properly determine the psychological effects of hysterectomy, with particular emphasis on developing countries. Psychological assessment done before the decision for hysterectomy might avoid the pre-surgical stress. This may call for better standardization of psychological wellbeing assessment for measurement of similar outcomes in the future to obtain comparisons with stronger validity and generalizability. Conflict of interests We declare that we have no conflict of interest and we are responsible for the data analyses and the writing of the article. Contribution to authorship MD conception, study search and identification, inclusion/ exclusion, data extraction, quality assessment, conducted the statistical analyses, interpretation and writing of the final manuscript. EA conception, helped in designing, and provided substantial contribution in determining the inclusion criteria and editing drafts of the review. TMT helped with the selection of studies, was responsible for abstracting the data, and quality assessment. All authors have approved the final version of the article to be published. Ethical approval Not needed. Funding No funding was necessary for the undertaking of this systematic review. Acknowledgements We are grateful to Prof. Mohammad Z. Raqab and Ms. Rasmieh Al-Amer for statistical review.

Appendix A. Appendix S1 Search syntaxes PsycINFO search syntax PsycINFO (via OvidSP) was searched on 10/11/2012 for the period 1806 to November Week 1 2012 (exp Hysterectomy/or Hysterectom*.af. or Colpohysterectom*.af.) and (exp depression/or exp depressive disorder/or depressi*.af. or dysthymi*.af. or melancholi*.af. or anxiety disorders/exp or anxiety.af. or anxieties.af. or anxious*.af. or nervousness.af. or Hypervigilance.af. or psychological distress.af.) and (longitudinal studies or longitudinal stud* or longitudinal survey* or prospective studies or prospective stud* or prospective cohort* or follow up stud* or follow up stud* or follow up stud* or follow-up stud* or retrospective stud* or inciden* or cohort stud* or concurrent stud*).af PubMed search syntax PubMed was searched on 10/11/2012 for the period 1965 to November Week 1 2012 (Hysterectomy [mh] OR Hysterectom*[tw] OR Colpohysterectom*[tw]) AND (depression[mh] OR depressive disorder[mh] OR depressi*[tw] OR dysthymi*[tw] OR melancholi*[tw] OR anxiety disorders[mh] OR anxiety[tw] OR anxieties[tw] OR anxious*[tw] OR nervousness[tw] OR Hypervigilance[tw] OR psychological distress[tw]) AND (longitudinal studies[mh] OR longitudinal stud*[tw] OR longitudinal survey*[tw] OR prospective studies[mh] OR prospective stud*[tw] OR prospective cohort*[tw] OR follow up stud*[tw] OR follow up stud*[tw] OR followup stud*[tw] OR follow-up stud*[tw] OR retrospective stud*[tw] OR inciden*[tw] OR cohort stud*[tw] OR concurrent stud*[tw]) EMBASE search syntax EMBASE was searched using the Ovid interface on 10/11/2012 for the period 1947 to November Week 1 2012 (Hysterectomy/exp or Hysterectom*.mp. or Colpohysterectom*.mp.) and (depression/exp or depressive disorder/exp or depressi*.mp. or dysthymi*.mp. or melancholi*.mp. or anxiety disorders/exp or anxiety.mp. or anxieties.mp. or anxious*.mp. or nervousness.mp. or Hypervigilance.mp. or psychological distress.mp.) and (longitudinal studies or longitudinal stud* or longitudinal survey* or prospective studies or prospective stud* or prospective cohort* or follow up stud* or follow up stud* or follow up stud* or follow-up stud* or retrospective stud* or inciden* or cohort stud* or concurrent stud*).mp. References [1] Ryan MM, Dennerstein L, Pepperell R. Psychological aspects of hysterectomy: a prospective study. Br J Psychiatry 1989;154:516–22. [2] Farrell SA, Kieser K. Sexuality after hysterectomy. Obstet Gynecol 2000;95:1045–51. [3] Garry R. The future of hysterectomy. BJOG 2005;112:133–9. [4] Andryjowicz E, Wray T. Regional expansion of minimally invasive surgery for hysterectomy: implementation and methodology in a large multispecialty group. Perm J 2011;15:42–6. [5] Vandyk AD, Brenner I, Tranmer J, Van Den Kerkhof E. Depressive symptoms before and after elective hysterectomy. J Obstet Gynecol Neonatal Nurs 2011;40:566–76. [6] Okunlola MA, Umuerri C, Omigbodun OO, Morhason-Bello IO, Okonkwo SN, Ojengbede OA. Pattern of mental ill health morbidities following hysterectomy for benign gynaecological disorders among Nigerian women. Int J Ment Health Syst 2009;3:18. [7] Wilcox LS, Koonin LM, Pokras R, Strauss LT, Xia Z, Peterson HB. Hysterectomy in the United States, 1988–1990. Obstet Gynecol 1994;83:549–55.

M. Darwish et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 174 (2014) 5–19 [8] Vessey MP, Villard-Mackintosh L, McPherson K, Coulter A, Yeates D. The epidemiology of hysterectomy: findings in a large cohort study. Br J Obstet Gynaecol 1992;99:402–7. [9] Domingo S, Pellicer A. Medscape: overview of current trends in hysterectomy; 2009, Available from: http://www.medscape.com/viewarticle/712569. [10] Lalinec-Michaud M, Engelsmann F. Cultural factors and reaction to hysterectomy. Soc Psychiatry Psychiatr Epidemiol 1989;24:165–71. [11] Al-Hendy A, Salama S. Gene therapy and uterine leiomyoma: a review. Hum Reprod Update 2006;12:385–400. [12] Ayoubi JM, Fanchin R, Monrozies X, Imbert P, Reme JM, Pons JC. Respective consequences of abdominal, vaginal, and laparoscopic hysterectomies on women’s sexuality. Eur J Obstet Gynecol Reprod Biol 2003;111:179–82. [13] Cosson M, Rajabally R, Querleu D, Crepin G. Long term complications of vaginal hysterectomy: a case control study. Eur J Obstet Gynecol Reprod Biol 2001;94:239–44. [14] Farquhar CM, Sadler L, Harvey S, McDougall J, Yazdi G, Meuli K. A prospective study of the short-term outcomes of hysterectomy with and without oophorectomy. Aust N Z J Obstet Gynaecol 2002;42:197. [15] Gu¨tl P, Greimel ER, Roth R, Winter R. Women’s sexual behavior, body image and satisfaction with surgical outcomes after hysterectomy: a comparison of vaginal and abdominal surgery. J Psychosom Obstet Gynaecol 2002;23:51–9. [16] Ewalds-Kvist SB, Hirvonen T, Kvist M, Lertola K, Niemela P. Depression, anxiety, hostility and hysterectomy. J Psychosom Obstet Gynaecol 2005;26:193–204. [17] Costantini E, Porena M, Lazzeri M, Mearini L, Bini V, Zucchi A. Changes in female sexual function after pelvic organ prolapse repair: role of hysterectomy. Int Urogynecol J 2013;24:1481–7. [18] Lowenstein L, Yarnitsky D, Gruenwald I, et al. Does hysterectomy affect genital sensation? Eur J Obstet Gynecol Reprod Biol 2005;119:242–5. [19] Aziz A, Bergquist C, Nordholm L, Moller A, Silfverstolpe G. Prophylactic oophorectomy at elective hysterectomy: effects on psychological well-being at 1-year follow-up and its correlations to sexuality. Maturitas 2005;51:349– 57. [20] Bachmann GA. Hysterectomy: a critical review. J Reprod Med 1990;35:839– 62. [21] Flory N, Bissonnette F, Binik YM. Psychosocial effects of hysterectomy: literature review. J Psychosom Res 2005;59:117–29. [22] Miao YL, Zhou R, Wang JL. Survey of long term female pelvic floor function and sexual life status after total hysterectomy. Zhonghua Fu Chan Ke Za Zhi 2012;47:496–9. [23] Alexander DA, Naji AA, Pinion SB, et al. Randomised trial comparing hysterectomy with endometrial ablation for dysfunctional uterine bleeding: psychiatric and psychosocial aspects. BMJ 1996;312:280–4. [24] Persson P, Brynhildsen J, Kjolhede P. A 1-year follow up of psychological wellbeing after subtotal and total hysterectomy – a randomised study. BJOG 2010;117:479–87. [25] Hurskainen R, Teperi J, Rissanen P, et al. Quality of life and cost-effectiveness of levonorgestrel-releasing intrauterine system versus hysterectomy for treatment of menorrhagia: a randomised trial. Lancet 2001;357:273–7. [26] Ryann Louie A, Armstrong JA, Findeiss LK, Goodwin SC. Comparison of sexual dysfunction using the Female Sexual Function Index following surgical treatments for uterine fibroids. Case Rep Obstet Gynecol 2012;2012:368136. [27] Ma¨kinen J, Johansson J, Toma´s C, et al. Morbidity of 10 110 hysterectomies by type of approach. Hum Reprod 2001;16:1473–8. [28] Centre for Reviews and Dissemination University of York. Undertaking systematic reviews of research on effectiveness: CRD’s guidance for carrying out or commissioning reviews. York, UK: NHS Centre for Reviews and Dissemination; 2001. [29] Lambden MP, Bellamy G, Ogburn-Russell L, et al. Women’s sense of well-being before and after hysterectomy. J Obstet Gynecol Neonat Nurs 1997;26:540–8. [30] Donoghue A, Jackson H, Pagano R. Understanding pre- and post-hysterectomy levels of negative affect: a stress moderation model approach. J Psychosom Obstet Gynecol 2003;24:99–109. [31] Moore JT, Tolley DH. Depression following hysterectomy. Psychosomatics 1976;17:86–9. [32] Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34. [33] Moreno SG, Sutton AJ, Ades A, et al. Assessment of regression-based methods to adjust for publication bias through a comprehensive simulation study. BMC Med Res Methodol 2009;9:2. [34] Baldaro B, Gentile G, Codispoti M, Mazzetti M, Trombini E, Flamigni C. Psychological distress of conservative and nonconservative uterine surgery: a prospective study. J Psychosom Res 2003;54:357–60. [35] Yen JY, Chen YH, Long CY, et al. Risk factors for major depressive disorder and the psychological impact of hysterectomy: a prospective investigation. Psychosomatics 2008;49:137–42.

19

[36] Coppen A, Bishop M, Beard RJ. Hysterectomy, hormones, and behaviour: a prospective study. Lancet 1981;1:126–8. [37] Leppert PC, Legro RS, Kjerulff KH. Hysterectomy and loss of fertility: implications for women’s mental health. J Psychosom Res 2007;63:269–74. [38] Hurskainen R, Teperi J, Rissanen P, et al. Clinical outcomes and costs with the levonorgestrel-releasing intrauterine system or hysterectomy for treatment of menorrhagia: randomized trial 5-year follow-up. JAMA 2004;291:1456–63. [39] Farquhar CM, Harvey SA, Yu Y, Sadler L, Stewart AW. A prospective study of 3 years of outcomes after hysterectomy with and without oophorectomy. Am J Obstet Gynecol 2006;194:711–7. [40] Persson P, Brynhildsen J, Kjolhede P. Short-term recovery after subtotal and total abdominal hysterectomy – a randomised clinical trial. BJOG 2010;117:469–78. [41] Moorman PG, Schildkraut JM, Myers ER, Wang F. Reported symptoms before and one year after hysterectomy in African American and white women. J Women’s Health 2011;20:1035–42. [42] Cooper R, Mishra G, Hardy R, Kuh D. Hysterectomy and subsequent psychological health: findings from a British birth cohort study. J Affect Disord 2009;115:122–30. [43] Wang XQ, Lambert CE, Lambert VA. Anxiety, depression and coping strategies in post-hysterectomy Chinese women prior to discharge. Int Nurs Rev 2007;54:271–9. [44] Wong LP, Arumugam K. Physical, psychological and sexual effects in multiethnic Malaysian women who have undergone hysterectomy. J Obstet Gynaecol Res 2012;38:1095–105. [45] Barglow P, Gunther MS, Johnson A, Meltzer HJ. Hysterectomy and tubal ligation; a psychiatric comparison. Obstet Gynecol 1965;25:520–7. [46] Lalinec-Michaud M, Engelsmann F. Anxiety, fears and depression related to hysterectomy. Can J Psychiatry 1985;30:44–7. [47] Thornton EW, McQueen C, Rosser R, Kneale T, Dixon K. A prospective study of changes in negative mood states of women undergoing surgical hysterectomy: the relationship to cognitive predisposition and familial support. J Psychosom Obstet Gynaecol 1997;18:22–30. [48] Persson P, Wijma K, Hammar M, Kjolhede P. Psychological wellbeing after laparoscopic and abdominal hysterectomy – a randomised controlled multicentre study. BJOG 2006;113:1023–30. [49] Ellstro¨m MA, Astro¨m M, Mo¨ller A, Olsson J-H, Hahlin M. A randomized trial comparing changes in psychological well-being and sexuality after laparoscopic and abdominal hysterectomy. Acta Obstet Gynecol Scand 2003;82:871–5. [50] Flory N, Bissonnette F, Amsel RT, Binik YM. The psychosocial outcomes of total and subtotal hysterectomy: a randomized controlled trial. J Sex Med 2006;3:483–91. [51] Rohl J, Kjerulff K, Langenberg P, Steege J. Bilateral oophorectomy and depressive symptoms 12 months after hysterectomy. Am J Obstet Gynecol 2008;199(22):e1–5. [52] Gibson CJ, Joffe H, Bromberger JT, et al. Mood symptoms after natural menopause and hysterectomy with and without bilateral oophorectomy among women in midlife. Obstet Gynecol 2012;119:935–41. [53] Martin RL, Roberts WV, Clayton PJ. Psychiatric status after hysterectomy. A one-year prospective follow-up. JAMA 1980;244:350–3. [54] Kjerulff KH, Langenberg PW, Rhodes JC, Harvey LA, Guzinski GM, Stolley PD. Effectiveness of hysterectomy. Obstet Gynecol 2000;95:319–26. [55] Thakar R, Ayers S, Georgakapolou A, Clarkson P, Stanton S, Manyonda I. Hysterectomy improves quality of life and decreases psychiatric symptoms: a prospective and randomised comparison of total versus subtotal hysterectomy. BJOG 2004;111:1115–20. [56] Gath D, Rose N, Bond A, Day A, Garrod A, Hodges S. Hysterectomy and psychiatric disorder: are the levels of psychiatric morbidity falling. Psychol Med 1995;25:277–83. [57] Farquhar CM, Sadler L, Stewart AW. A prospective study of outcomes five years after hysterectomy in premenopausal women. Aust N Z J Obstet Gynaecol 2008;48:510–6. [58] Moraloglu O, Tapisiz OL, Moroy P, et al. Functional results and quality-of-life after hysterectomy in Turkish women. Gazzetta Medica Italiana Archivio per le Scienze Mediche 2007;166:97–103. [59] Bayram GO, Sahin NH. Hysterectomy’s psychosexual effects in Turkish women. Sex Disabil 2008;26:149–58. [60] Helmy YA, Hassanin IMA, Elraheem TA, Bedaiwy AA, Peterson RS, Bedaiwy MA. Psychiatric morbidity following hysterectomy in Egypt. Int J Gynecol Obstet 2008;102:60–4. [61] Roovers J-PW, van der Bom JG, van der Vaart CH, Heintz APM. Hysterectomy and sexual wellbeing: prospective observational study of vaginal hysterectomy, subtotal abdominal hysterectomy, and total abdominal hysterectomy. BMJ 2003;327:774–8. [62] Broder MS, Kanouse DE, Mittman BS, Bernstein SJ. The appropriateness of recommendations for hysterectomy. Obstet Gynecol 2000;95:199–205.