Efficacy of salpingectomy at hysterectomy to reduce the risk of epithelial ovarian cancer: a systematic review

Transcription

1 DOI: / Systematic review Efficacy of salpingectomy at hysterectomy to reduce the risk of epithelial ovarian cancer: a systematic review A Darelius, a M Lycke, a JM Kindblom, b B Kristjansdottir, a K Sundfeldt, a,c A Strandell a,b a Department of Obstetrics and Gynaecology, Sahlgrenska University Hospital, G oteborg, Sweden b HTA-centrum, Sahlgrenska University Hospital, G oteborg, Sweden c Sahlgrenska Cancer Centre, G oteborg, Sweden Correspondence: Dr A Darelius, Department of Obstetrics and Gynaecology, Sahlgrenska University Hospital, Bla straket 6, S G oteborg, Sweden. anna.darelius@vgregion.se Accepted 4 February Background It has been argued that salpingectomy would reduce the risk of epithelial ovarian cancer (EOC), based on the theory of the tube being the site of origin. Objectives To conduct a systematic review of salpingectomy associated with ovarian cancer risk and salpingectomy with concomitant hysterectomy on outcomes of complications including endocrine function. Search strategy A comprehensive search was conducted in PubMed, Embase, and the Cochrane library. Selection criteria Original studies and systematic reviews were eligible. Data collection and analysis Each article was quality assessed. Data were extracted and, when possible, pooled in meta-analyses. The certainty of evidence across studies was evaluated using GRADE. Main results Of 844 articles found, 11 were included. No study evaluated risk reduction for EOC after salpingectomy in conjunction with hysterectomy. Two retrospective studies reported a reduced ovarian cancer risk after indicated salpingectomy, compared with no surgery: adjusted hazard ratio 0.65 (95% confidence interval, 95% CI ) and adjusted odds ratio 0.58 (95% CI ). Complications did not differ between groups with or without salpingectomy, but were nonsystematically reported. Ovarian endocrine function, measured with surrogate outcomes, did not differ at short-term follow-up in randomised or observational studies. The certainty of evidence was very low or low for all outcomes. Conclusions There is currently insufficient evidence to state that opportunistic salpingectomy reduces the risk of EOC. The impact on long-term endocrine function is unknown. The heterogeneity in results and identified knowledge gaps stress the need for a prospective trial. Keywords Hysterectomy, opportunistic salpingectomy, ovarian cancer risk reduction. Tweetable abstract Insufficient evidence for prophylactic removal of the fallopian tubes for risk reduction of ovarian cancer. Linked article This article is commented on by R Machanda, p. 890 in this issue. To view this mini commentary visit / Please cite this paper as: Darelius A, Lycke M, Kindblom JM, Kristjansdottir B, Sundfeldt K, Strandell A. Efficacy of salpingectomy at hysterectomy to reduce the risk of epithelial ovarian cancer: a systematic review. BJOG 2017;124: Introduction Epithelial ovarian cancer (EOC) is the gynaecological malignancy with the highest mortality in the Western world, and has the second highest mortality globally after cervical cancer. 1 For a majority of the patients, the disease is diagnosed at a late stage and consequently has a poor prognosis. Despite progress regarding surgical techniques and chemotherapy, the overall survival for women with EOC has only improved marginally in the past 50 years, and the 5-year survival rate is still <50%. 2 Epithelial ovarian cancer includes a group of extremely heterogeneous carcinomas, all with differences regarding origin, molecular biology, morphology, gene expression, and behaviour. 3 Recent findings indicate that the fallopian tubes play a central role, and that the ovary is seldom the origin but is instead involved secondarily. Accumulated data support this theory, which has gained widespread acceptance. 4 The most fatal type of EOC, high-grade serous cancer (HGSC), is characterised by a chaotic genome with TP53 mutation, and represents 75% of all cases. For this EOC 880 ª 2017 Royal College of Obstetricians and Gynaecologists

2 Salpingectomy for risk reduction of ovarian cancer subtype, no preclinical lesions have been found in the ovaries, but a candidate precursor for HGSC, serous tubal intraepithelial carcinoma (STIC), has been detected within the fimbriae at the distal Fallopian tube. 5 The STICs are thought to implant on the ovarian and/or peritoneal surfaces, and after an occult period will develop into fastgrowing HGSCs. Endometrioid and clear-cell carcinomas are suggested to develop from the endometrium and from endometriosis as a result of retrograde menstruation, involving the tube as a transport route. 5 8 Approaches to reduce the burden of cancer include two major strategies: screening, resulting in early detection, and more recently active prevention. Early diagnosis of EOC is challenging because of unspecific symptoms until late stages. Although existing screening methods may show satisfactory sensitivity and specificity, no conclusive impact on EOC mortality has yet been demonstrated using these methods. 9,10 The current screening method (transvaginal ultrasound and serum markers) often leads to unnecessary surgery with unclear benefit for the patient. 9,10 Based on the biological and genetic evidence that the fallopian tube is the origin of most EOCs, it has been suggested that bilateral salpingectomy would reduce the risk of EOC. This systematic review aims to evaluate the present scientific support for opportunistic salpingectomy as a method to reduce the incidence of EOC, and whether it affects survival, health-related quality of life, endocrine function, operative time, and hospital stay in adult women undergoing benign hysterectomy or other abdominal gynaecological surgery leaving the ovaries intact. Methods We performed a systematic review according to established routines at the Regional Health Technology Assessment (HTA) Centre in the Region V astra G otaland, Sweden. The population, intervention, comparison, and outcomes (PICO) were defined as follows. The population (P) of interest was adult women who underwent hysterectomy or other abdominal gynaecological benign surgical procedure. The intervention (I) was opportunistic salpingectomy and the comparison (C) was no salpingectomy. The outcomes (O) were ovarian cancer (including borderline tumour, malignancy of the fallopian tubes, and peritoneal cancer), survival, endocrine function (ovarian function during fertile age and induced menopause symptoms), complications, health-related quality of life, operative time, and length of hospital stay. Literature search Systematic literature searches were conducted by two librarians in PubMed, Embase (OvidSP, 1980 to present) and the Cochrane library, including HTA databases. The last search was performed on 3 September Searches were conducted using controlled vocabulary and title/abstract words. Variations of the terms salpingectomy or tubectomy or fallopian tube removal were combined with variations of the terms prophylactic or risk or opportunistic or reduction or prevention. A detailed description of the search strategies is available in Table S1. In addition, reference lists of review articles were scrutinised for relevant references. We included systematic reviews (SRs), randomised controlled trials (RCTs), nonrandomised controlled studies, and case series, the latter for the evaluation of complications, including menopausal symptoms. No case reports or review articles were accepted for inclusion. Studies regarding ovarian function in connection with fertility treatment were not included. Included publications were restricted to English or Scandinavian languages (Swedish, Norwegian, or Danish). Abstracts and unpublished studies were not searched. No limit for publication date was applied, but no article published prior to 2007 was found. No author was contacted. In order to identify continuing or completed but still not published studies, we also searched ls.gov in October Study selection A total of 844 citations were found. Two research librarians at the Medical Library, Sahlgrenska University Hospital, Gothenburg, Sweden screened all the identified abstracts and, if needed, also read full-text articles. The studies were selected and reported according to the preferred reporting items for systematic reviews and meta-analysis (PRISMA) guidelines. 11 Those that did not fulfil the PICO were excluded. The remaining 23 articles were independently assessed for eligibility by all authors, followed by a consensus discussion for final inclusion in the systematic review. Eleven articles were finally included in the review. As the aim of the present systematic review was to evaluate the effect of opportunistic salpingectomy on ovarian cancer, our primary outcome was ovarian cancer. The systematic search did not identify any studies that evaluated this outcome following opportunistic salpingectomy (i.e. salpingectomy concomitant with hysterectomy or any other abdominal surgery), but two studies in which indicated salpingectomy was compared with no surgery were found. 12,13 We therefore extended our inclusion criteria for this specific outcome, ovarian cancer risk, to include also indicated salpingectomy in comparison with no surgery. The search did not identify any publication reporting on salpingectomy concomitant with any other gynaecological benign surgery other than hysterectomy either. Thus, the present review addresses salpingectomy per se for EOC risk evaluation and salpingectomy concomitant ª 2017 Royal College of Obstetricians and Gynaecologists 881

3 Darelius et al. with hysterectomy for evaluation of endocrine function, complications, health-related quality of life, operative time, and length of hospital stay. Data extraction One investigator extracted the data and at least two others verified the extracted data for each outcome. The extracted data included study design, the number of individuals (intervention and control), type of intervention and comparison, and the resulting outcomes: ovarian cancer, endocrine function, complications, operative time, and hospital stay, according to sound clinical principals. Outcomes The outcomes were subdivided into critical, important, and not important for decision-making according to the GRADE handbook. 14 The critical outcomes were ovarian cancer and survival, and the important outcomes were endocrine function, complications, and health-related quality of life. The less important outcomes were operative time and hospital stay. Quality assessment The appraisal of the included articles was performed by the use of validated checklists for RCTs and CTs from The Swedish Agency for Health Technology Assessment and Assessment of Social Services (SBU), 15 and modified by HTA-centrum. 16 Case series were critically appraised using a slightly modified previously published checklist for case series. 17 The appraisal addressed directness (external validity), risk of bias (internal validity), and precision, and is presented in three levels. The certainty of evidence across studies was assessed by all authors and rated for all outcomes separately using the GRADE approach. 18,19 Disagreements were solved in consensus. Statistical analyses When possible, extracted data were pooled in meta-analyses (requiring means and SDs reported for continuous outcomes). When instead medians and ranges were reported in the original articles, data were explored for the type of distribution. If the data showed a normal distribution, means and SDs could be estimated for inclusion in the meta-analysis. Different study designs were not pooled, but analysed and presented separately. Meta-analyses were performed using REVIEW MANAGER 5.3. A random-effects model was applied. The effect estimate was expressed as mean differences with 95% confidence intervals (95% CIs). The individual studies and the pooled estimates were presented graphically in forest plots. Statistical heterogeneity was examined using the chi-square test and the I 2 characteristics. Results The systematic literature search identified 844 publications after the removal of duplicates. A total of 815 records were excluded by reading titles and abstracts. Eighteen articles were excluded after reading the full text. They are presented along with the reason for exclusion in Table S2. Eleven of the studies were eligible and were included in the final assessment. These are presented along with their study characteristics in Table S3. The selection process is summarised in Figure 1. Two of the included studies were RCTs, 20,21 seven were cohort studies, 12,22 27 one was a case control study, 13 and one was a case series. 28 Seven of the studies were prospective, 20 23,25,26,28 and were conducted between 2001 and Four studies were retrospective register studies where data were collected between 1973 and ,13,24,27 The studies came from nine different countries. No articles in Scandinavian languages were found. In all studies the reason for hysterectomy was a benign cause, but the explicit causes were generally not reported. In the majority of the studies the salpingectomies were opportunistic ,28 In the two retrospective register studies that evaluated the risk of EOC, the salpingectomy was for pathology in the fallopian tubes (indicated salpingectomy) and not concomitant with a hysterectomy. 12,13 In these studies the cases were compared with controls that had not undergone surgery. Thus, the result of the review addresses salpingectomy per se for EOC risk evaluation and salpingectomy concomitant with hysterectomy for the evaluation of all other outcomes. Idetification Screening Eligibility Included Records identified through database searching (n = 1123) Abstracts screened (n = 844) Full-text articles assessed for eligibility (n = 29) Studies included in the review (n = 11) See Table S3 Additional records identified through other sources (n = 40) Duplicates removed (n = 319) Did not fulfil PICO or other eligibility criteria (n = 815) Full-text articles excluded (n = 18) See Table S2 Figure 1. Selection process: flow diagram of articles in the literature search. 882 ª 2017 Royal College of Obstetricians and Gynaecologists

4 Salpingectomy for risk reduction of ovarian cancer The quality assessment of the included studies is reported in Table S4. No study reported the outcomes survival or healthrelated quality of life. Ovarian cancer Two observational studies reported ovarian cancer as an outcome: one cohort study and one case control study. 12,13 Neither of these studies evaluated the risk reduction for ovarian cancer in patients undergoing hysterectomy and concomitant opportunistic salpingectomy compared with only hysterectomy. Both studies compared salpingectomy as an intervention with no surgery at all. In both studies the reason for salpingectomy was pathology of one or both fallopian tubes. Of note, patients who had undergone hysterectomy and concomitant salpingectomy were excluded in the cohort study because of limited sample size (n = 2646). 12 The cohort study reported an adjusted hazard ratio (ahr) for EOC of 0.65 (95% CI ) after indicated salpingectomy, compared with no surgery at all. 12 Adjustments were made for age, calendar time, education status, and parity. Despite a large cohort (n = ) only a few cases of EOC after salpingectomy were identified (n = 81). The case control study demonstrated an adjusted odds ratio (aor) for EOC of 0.58 (95% CI ) after indicated salpingectomy, compared with no surgery at all. 13 Adjustments were made for age, parity, and tubal ligation. Seventeen cases of EOC were included. Both studies revealed a high risk of both indication and detection bias, because indicated salpingectomies rather than opportunistic salpingectomies were performed and compared with no surgery at all. The quality of evidence for the outcome ovarian cancer was rated as very low (GRADE ). Endocrine function Three studies evaluated ovarian endocrine function after hysterectomy with opportunistic salpingectomy compared with hysterectomy only (Table 1). 20,21,26 None of these studies reported any statistically significant difference in endocrine function of the ovary after short-term follow-up (1 6 months). Ovarian function was evaluated by different surrogate measures: hormones (luteinising hormone, LH; follicle-stimulating hormone, FSH; estradiol, E 2 ; and anti- M ullerian hormone, AMH) and ultrasound measurements (antral follicle count, AFC; mean ovarian volume; pulsatile index, PI; resistance index, RI; and systole/diastole ratio, S/ D). Patient-reported outcomes like menopausal symptoms have not been evaluated in any study. The change in AMH was reported in two studies: one small RCT and one cohort study. 20,26 In both studies AMH was measured before surgery and after 3 months. None of the studies found any statistically significant difference in DAMH after 3 months. Another RCT reported LH, FSH, E 2, mean ovarian volume, PI, RI, and S/D ratio as a measure of ovarian function. 21 A limitation of this study was that it compared patients who underwent total bilateral salpingectomy with partial bilateral salpingectomy, where the neighbouring para-ovarian tissue was left behind. After 6 months followup there were no statistically significant differences in any of these parameters between the two groups. All three studies had major limitations because of small sample sizes, young study populations, short follow-up, and the use of surrogate markers. One RCT used an inappropriate comparison group. 21 Another limitation for the cohort study was the use of historical controls. 26 The quality of evidence for the outcome endocrine function was rated as low (GRADE ). Surgical data Complications Five studies reported complications as an outcome after opportunistic or indicated salpingectomy compared with hysterectomy only (Table 2): four cohort studies and one case series ,27,28 None of the studies reported any significant difference between the groups concerning complications related to surgery. Two of the cohort studies evaluated laparoscopic surgery (Table 2). 23,27 Ghezzi reported no complications after total hysterectomy with bilateral salpingectomy after 1 month of follow-up compared with historical controls without bilateral salpingectomy, the data for which were found in medical records. 23 Vorwergk studied laparoscopy-assisted vaginal hysterectomy and reported complication rates of approximately 18% irrespectively of the addition of bilateral salpingectomy. 27 Both groups were retrospectively included and had different lengths of follow-up: 6 10 years for the hysterectomy group and 4 5 years for the group who had both hysterectomy and salpingectomy. The other studies did not report the method of surgery. The largest cohort study (n = ) reported only readmission and blood transfusion as outcomes, and there were no statistically significant differences between the groups. 24 No common standard for reporting complications was used and the results were therefore not possible to pool. Thus, the evidence regarding complications is based solely on observational studies, three out of four of which used historical controls as the comparison. The quality of evidence was rated as very low (GRADE ). ª 2017 Royal College of Obstetricians and Gynaecologists 883

5 Darelius et al. Table 1. Endocrine function Author Publication year Country Study design Number of patients I, interventions C, controls Drop outs Intervention Mean (SD) Results Control P Mean (SD) P-value Findley USA Seziq Turkey Morelli Italy RCT n (I) = 15 n (C) = 15 RCT n (I) = 12 n (C) = 12 Cohort n (I) = 79 n (C) = 79 6 weeks follow-up TLH + BSE TLH AMH baseline (ng/ml) 2.26 (2.72) 2.25 (2.57) 0.99 I, 3 C, 4 AMH 4 6 weeks 1.03 (1.04) 1.25 (2.09) months AMH 3 months 1.86 (1.99) 1.82 (3.12) 0.97 I, 2 C, 1 DAMH (baseline 3 months) 0.07 (0.90) 0.08 (1.45) 0.98 DFSH (IU/L) Total BSE Partial BSE 1 month 0.6 (0.2) 0.7 (0.1) NS 6 months 0.4 (0.7) 0.4 (0.4) NS DLH (IU/L) 1 month 0.1 (0.3) 0.2 (0.1) NS 6 months 0.3 (0.1) 0.2 (0.1) NS DE2 (pg/ml) 1 month 1.1 (3.3) 0.3 (1.3) NS 6 months 0.5 (3.2) 3.9 (2.1) NS DMean ovarian volume (cm 3 ) 1 month 0.1 (0.4) 0.7 (0.1) NS 6 months 0.1 (0.2) 0.3 (0.1) NS DPulsatility Index 1 month 0.15 (0.18) 0.40 (0.07) months 0.3 (0.27) 0.68 (0.03) NS DResistance Index 1 month 0.09 (0.04) 0.10 (0.04) NS 6 months 0.11 (0.09) 0.22 (0.11) DS/D ratio 1 month 0.24 (0.10) 0.5 (0.30) NS 6 months 0.13 (0.3) 1.10 (0.40) 0.07 TLH + BSE TLH DAMH (ng/ml) 0.06 (0.1) 0.08 (0.1) 0.35 DFSH (miu/ml) 1.3 (1.1) 1.0 (0.8) 0.15 DAFC (n) 0.27 (0.6) 0.14 (0.3) 0.09 Dmean ovarian diameter (mm) 0.25 (0.8) 0.19 (0.6) 0.57 AFC, antral follicle count; AMH, anti-m ullerian hormone; BSE, bilateral salpingectomy; E2, estradiol; FSH, follicle stimulating hormone; LH, luteinising hormone; NS, not significant, as reported in the article; TLH, total laparoscopic hysterectomy Operative time Six studies reported operative time as an outcome after opportunistic or indicated salpingectomy: one small RCT and five cohort studies. 1,22 25,27 Three of the studies evaluated different laparoscopic surgeries: laparoscopic-assisted supracervical hysterectomy (LASH), 22 total laparoscopic hysterectomy (TLH), 23 and laparoscopic-assisted vaginal hysterectomy (LAVH), 27 whereas the other three studies did not report the method of surgery. 20,24,25 All studies were assessed as having a moderate or high risk of bias. The RCT (n = 30) reported similar operative time comparing salpingectomy in addition to hysterectomy [115.7 (SD 33) minutes] with hysterectomy alone [115.2 (SD 44) minutes] (P = 0.97). Four of five cohort studies reported means (SDs), and their results were pooled in a meta-analysis (Figure 2A). 22,24,25,27 Data from Ghezzi were not included, 23 as they were reported as medians (ranges) and had an extreme, skewed distribution. The mean difference was 2.4 minutes (95% CI 12.5, 17.3 minutes) longer when salpingectomy was added. The heterogeneity between studies was large, mainly explained by the large cohort study (n > ) reporting 16 minutes longer operative time. 24 The quality of evidence for the outcome operative time was rated as low (GRADE ). 884 ª 2017 Royal College of Obstetricians and Gynaecologists

6 Salpingectomy for risk reduction of ovarian cancer Table 2. Complications Author Publication year Country Study design Number of patients I, intervention C, control Drop outs Type of complication Results Intervention n (%) Control n (%) P Ghezzi Italy McAlpine Canada Minig Spain Vorwergk Germany Robert Canada Cohort n (I) = 137 n (C) = 145 Cohort n (I) = 3670 n (C) = 8362 Cohort n (I) = 97 n (C) = 71 Cohort n (I) = 127 n (C) = 413 Case series n (I) = 425 TLH + BSE TLH No data Intra-op complications 0 1 (0.7) 0.49 Non-infectious post-op complications 0 1 (0.7) 0.49 Hyst+BSE Hyst Not applicable Readmission Adjusted OR (95% CI) 0.91 ( ) Blood transfusion Adjusted OR (95% CI) 0.86 ( ) Hyst + BSE Hyst No data Blood transfusion 1 (1.4) 1 (1.0) 1 Conversion to 1 (1.4) 1 (1.0) 1 laparotomy Bladder damage 2 (2.8) 2 (2.1) 1 Postoperative* 12 (12.4) 8 (11.3) Emergency visit 13 (13.4) 9 (12.7) Blood loss [mean (SD)] (100.2) (83.5) LAVH+BSE LAVH 6 week follow-up Surgical 24 (18.9) 73 (17.7) (9%) drop outs complications** Questionnaire 4 to 10 years after surgery. Vaginal hysterectomy Intraoperative 4 (0.9) Non-responders Emergency room visit 20 (4.7) LAVH+BSE 55 (43%) Readmission 24 (5.6) LAVH 190 (46%) Reoperation 13 (3.1) 15% overall complication rate, 3.8% attributable to salpingectomy BSE, bilateral salpingectomy; Hyst, hysterectomy, surgical method not stated; LAVH, laparoscopic assisted vaginal hysterectomy; TLH, total laparoscopic hysterectomy. *Urinary tract infection, blood transfusion, scar infection, intra-abdominal infection, vesicovaginal fistula, vaginal cuff dehiscence. **Urinary tract lesions, pelvic abscess, postoperative haemorrhage, vaginal vault dehiscence, increased wound pain, urinary tract infection, abdominal bloating. Length of hospital stay Five cohort studies reported on length of hospital stay after opportunistic or indicated salpingectomy ,27 All studies suffered from moderate or high risk of bias, mainly based on the use of historical controls. Hospital stay varied between 1.82 and 4.93 days in the intervention groups, 25,27 and between 2.25 and 5.11 days in the control groups. 25,27 Four of five cohort studies reported means (SDs), and their results were pooled in a meta-analysis (Figure 2B). 22,24,25,27 Data from Ghezzi were not included, as they were reported as median (range) and had an extreme, skewed distribution. 23 The mean difference was 0.18 days (95% CI 0.27, 0.10) shorter when salpingectomy was added. The quality of evidence for the outcome length of hospital stay was rated as very low (GRADE ). Results of the search for continuing studies We identified 91 continuing trials in clinicaltrials.gov, 14 of which were potentially relevant for our question at issue (Table S5). Four RCTs evaluate the effect of salpingectomy on AMH levels, and two RCTs compare salpingectomy and tubal ligation and their effect on AMH levels. Two cohort studies evaluate the effect of salpingectomy on AMH levels. Furthermore, one of the cohort studies investigates whether women would opt for risk-reducing bilateral salpingectomy or standard tubal ligation when they are enrolled for ª 2017 Royal College of Obstetricians and Gynaecologists 885

7 Darelius et al. (A) Operative time (minutes) (B) Length of hospital stay (days) Figure 2. Forest plots of the meta-analyses comparing hysterectomy and bilateral salpingectomy (BSE) versus hysterectomy only. Outcomes: (A) operative time; (B) length of hospital stay. sterilisation. None of the RCTs are designed to study risk reduction for ovarian cancer as a primary outcome. Discussion Main findings The main finding in the present systematic review is that there is a lack of evidence regarding the effect of opportunistic salpingectomy on the risk reduction of EOC. In addition, the complications associated with opportunistic salpingectomy are not fully elucidated. A strong opinion has evolved suggesting that opportunistic salpingectomy should be performed when benign hysterectomy is indicated to decrease the incidence of EOC later in life. 29 We have explored the current scientific support for the benefits and risks associated with opportunistic salpingectomy. In the systematic literature search, no study was found that evaluated the risk reduction of EOC after opportunistic salpingectomy. There is, however, increasing knowledge and consensus regarding the fallopian tube as the origin of HGSC, giving rise to the hypothesis of opportunistic salpingectomy as a prophylactic measure. 8,30 Highgrade serous cancer is the most common EOC subtype, it is aggressive, carries p53 mutations in all stages, and has the worst prognosis of all EOCs. In addition, there is no effective screening method. Taken together, these factors strengthen the need to increase the knowledge regarding possible benefits and drawbacks with opportunistic salpingectomy at the time for hysterectomy and other benign procedures. Two observational studies reported a reduced risk of EOC (35 42%) after indicated salpingectomy (for tubal pathology) compared with no surgery. 12,13 Conditions involving chronic tubal inflammation (infection, hydrosalpinx, and endometriosis) constitute both an indication for salpingectomy, and are known risk factors for EOC. Indicated salpingectomy, as opposed to opportunistic salpingectomy, may therefore confound the results. Both studies were also limited by indirectness and uncertain precision because of the few cases of EOC. 886 ª 2017 Royal College of Obstetricians and Gynaecologists

8 Salpingectomy for risk reduction of ovarian cancer In two small RCTs and one cohort study only short-term surrogate outcomes (hormone and ultrasound measures) for ovarian function were investigated. 20,21,26 These studies reported no increased short-term risk for ovarian failure after salpingectomy. Hysterectomy by itself may cause procedure-induced menopause, and it is a concern that the addition of salpingectomy, because of the anatomical proximity to the ovaries, could further add to this risk. 31 Benign hysterectomies are usually performed in perimenopause, and one could argue that at this time of life, menopause will occur spontaneously within a short period; however, we have also noted suggestions of opportunistic salpingectomy at the time for sterilisation, and some health regions have already introduced this recommendation. Sterilisation is performed earlier in life, and potential ovarian failure followed by increased morbidity should be a concern. The HGSC originates in the transformation zone between tubal and peritoneal epithelium in the fimbriae of the fallopian tubes, and is often preceded by an STIC lesion. With this in mind, specific surgical precision and a very good knowledge of anatomy is needed when performing an EOC risk-reducing opportunistic salpingectomy so that the complete fallopian tube including all fimbriae will be removed with as little damage to the ovarian blood flow as possible. Even though no increased risk of surgical complications was detected in the retrospective studies in this review, their results have to be evaluated as uncertain because of the comparison with historical controls ,27 In one small RCT, operative time was the same in the control arm, whereas a large observational register study reported 16 minutes longer operative time if salpingectomy was added to hysterectomy. 20,24 The heterogeneity in results stresses the need for a prospective trial. The potential increase in operative time will have implications for costs. The difference in hospital stay is biased by the use of historical controls, and explained by the trend in recent years to minimise the length of hospital stay. Strengths This is to our knowledge the first systematic review on the topic of opportunistic salpingectomy for ovarian cancer reduction. It has been conducted according to a wellestablished routine for performing health technology assessments and was led by HTA expertise. The literature search was also conducted across HTA databases. This systematic review has included not only RCTs and other controlled studies, but also the appraisal of case series to capture complications. The confidence in the result for each outcome has been evaluated across studies and the certainty of evidence described. The search for continuing studies completes the review by identifying continuing studies and highlights the need for well-conducted prospective studies. Limitations The most important limitation of the review is the complete lack of studies evaluating opportunistic salpingectomy for the risk for EOC. The only studies addressing the critical outcome ovarian cancer were based on indicated salpingectomy and the comparison with no surgery. The reporting of complications was heterogeneous without any common standard, and the results were not possible to pool. The potential risk of procedure-induced menopause/menopausal symptoms was not reported in any study. Instead, surrogate measures for ovarian function were reported. These hormones and ultrasound measures were also heterogeneous across studies and not possible to pool. Of further importance is that the short duration of follow-up has no impact on determining the onset of menopause. 32 The less important outcomes of operative time and hospital stay were included in meta-analyses, except for one study. The authors were not contacted for these missing data, as all studies were at high or moderate risk of bias. There were too few studies to perform a quantitative assessment of publication bias. Interpretations Despite that a rationale for the removal of the fallopian tubes exists, no study has evaluated the effect of opportunistic salpingectomy in conjunction with hysterectomy on ovarian cancer risk. Furthermore, the evidence for procedure-related safety is also only observational and shortterm. This systematic review points out several major knowledge gaps that need to be addressed. The most straightforward move to elucidate a possible EOC protective effect would be to conduct an RCT with or without a salpingectomy concomitant with a benign hysterectomy or sterilisation. 33 A register-based RCT will start recruitment in Sweden this year. The risk for complications would be available within a short time frame. Also the risk for procedure-induced menopause would be possible to acquire within a few years, but to analyse the effect on EOC would require years of follow-up; however, if the results on all types of complications are reassuring, we would be confident in recommending salpingectomy to our patients and then wait for the result on EOC. The majority would agree that even a very small reduction of EOC incidence would be of benefit, because of the severity and the high mortality rate of the disease. How important is the removal of the fallopian tubes? Should we change our way of surgery to reach the goal? If it is not possible to remove the fallopian tubes vaginally, should we proceed with a laparoscopy? Should we abandon hysteroscopy sterilisation and perform laparoscopic salpingectomy only? Should we remove the fallopian tubes from anyone who asks for the procedure? Moving along this line of questions it becomes even more obvious that ª 2017 Royal College of Obstetricians and Gynaecologists 887

9 Darelius et al. we need more than circumstantial evidence on this topic to safely guide our patients. Conclusion The present systematic review demonstrates that despite a biological rationale for the removal of the fallopian tubes, no study has evaluated the effect of opportunistic salpingectomy in conjunction with hysterectomy on ovarian cancer risk. The risk of complications or adverse effects of the procedure are not yet fully elucidated. A randomised prospective trial is urgently needed to fully understand the risks and benefits of opportunistic salpingectomy. Disclosure of interests None declared. Completed disclosure of interests form available to view online as supporting Information. Contribution to authorship All authors contributed to the selection process and quality assessment of the articles, and to writing the manuscript. AD, ML, BK, JK, and KS extracted the data. AS performed the meta-analyses. All authors have read and approved the final version for publication. Details of ethics approval Not applicable. Funding None. Acknowledgements We thank the librarians Therese Svanberg and Kirsten Freadrich at the Medical Library, Sahlgrenska University Hospital, for their excellent work with the literature search and the process of selecting articles. Supporting Information Additional Supporting Information may be found in the online version of this article: Table S1. Search strategy. Table S2. Excluded studies and reason for exclusion. Table S3. Characteristics of included studies. Table S4. Quality assessment of included studies. Table S5. Search in ClinicalTrials. & References 1 Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelos M, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN Int J Cancer 2015;136: E Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer statistics, CA Cancer J Clin 2015 ;65: Prat J. New insights into ovarian cancer pathology. Ann Oncol 2012;23(Suppl 10):x Singh N, Blake Gilks C, Hirschowitz L, Kehoe S, McNeishe I-A, Miller D, et al. Adopting a uniform approach to site assignment in tuboovarian high-grade serous carcinoma: the time has come. Int J Gynecol Pathol 2016;35: Kindelberger DW, Lee Y, Miron A, Hirsch MS, Feltmate C, Medeiros F, et al. Intraepithelial carcinoma of the fimbria and pelvic serous carcinoma: Evidence for a causal relationship. Am J Surg Pathol 2007;31: Kurman RJ, Shih IM. The origin and pathogenesis of epithelial ovarian cancer: a proposed unifying theory. Am J Surg Pathol 2010;34: Kurman RJ, Shih Ie M. Molecular pathogenesis and extraovarian origin of epithelial ovarian cancer-shifting the paradigm. Hum Pathol 2011;42: Shih IM, Kurman RJ. Ovarian tumorigenesis: a proposed model based on morphological and molecular genetic analysis. Am J Pathol 2004;164: Jacobs I, Menon U. Can ovarian cancer screening save lives? The question remains unanswered Obstet Gynecol 2011;118: Menon U, Ryan A, Kalsi J, Gentry-Maharaj A, Dawnay A, Habib M, et al. Risk algorithm using serial biomarker measurements doubles the number of screen-detected cancers compared with a singlethreshold rule in the United Kingdom collaborative trial of ovarian cancer screening. J Clin Oncol 2015 ;33: Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol 2009;62: Falconer H, Yin L, Gronberg H, Altman D. Ovarian cancer risk after salpingectomy: a nationwide population-based study. J Natl Cancer Inst 2015;107:dju Madsen C, Baandrup L, Dehlendorff C, Kjaer SK. Tubal ligation and salpingectomy and the risk of epithelial ovarian cancer and borderline ovarian tumors: a nationwide case-control study. Acta obst gynecol Scand 2015;94: [ Accessed 01 November Checklist from SBU. Available from: [ metodbok/mall_randomiserade_studier.pdf] and [ ssets/ebm/metodbok/mall_observationsstudier.pdf]. Accessed 01 November Checklist from HTA-centrum. available from: [www2.sahlgrenska.se/ sv/su/forskning/hta-centrum/hogerkolumn-undersidor/hjalpmede l-under-projektet/]. Accessed 01 November Guo B, Moga C, Schopflocher D, Harstall C. Validation of a quality assessment checklist for case series studies. In: Better Knowledge for Better Health Un meilleur savoir pour une meilleure sante. Abstracts of the 21st Cochrane Colloquium. 2013; Sep; Quebec City, Canada. John Wiley & Sons; GRADE Working group. List of GRADE working group publications and grants Internet. Place unknown: GRADE Working group, c cited 2012 Mar 8. Available from: kinggroup.org/publications/index.htm 19 Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, et al. Grading quality of evidence and strength of recommendations. BMJ (Clinical research ed) 2004;328: Findley AD, Siedhoff MT, Hobbs KA, Steege JF, Carey ET, McCall CA, et al. Short-term effects of salpingectomy during laparoscopic hysterectomy on ovarian reserve: a pilot randomized controlled trial. Fertil Steril 2013;100: ª 2017 Royal College of Obstetricians and Gynaecologists

10 Salpingectomy for risk reduction of ovarian cancer 21 Sezik M, Ozkaya O, Demir F, Sezik HT, Kaya H. Total salpingectomy during abdominal hysterectomy: effects on ovarian reserve and ovarian stromal blood flow. J Obstet Gynaecol Res 2007;33: Berlit S, Tuschy B, Kehl S, Brade J, Sutterlin M, Hornemann A. Laparoscopic supracervical hysterectomy with concomitant bilateral salpingectomy why not? Anticancer Res 2013;33: Ghezzi F, Cromi A, Siesto G, Bergamini V, Zefiro F, Bolis P. Infectious morbidity after total laparoscopic hysterectomy: does concomitant salpingectomy make a difference? BJOG 2009;116: McAlpine JN, Hanley GE, Woo MM, Tone AA, Rozenberg N, Swenerton KD, et al. Opportunistic salpingectomy: uptake, risks, and complications of a regional initiative for ovarian cancer prevention. Am J Obstet Gynecol 2014;210:471.e Minig L, Chuang L, Patrono MG, Cardenas-Rebollo JM, Garcia- Donas J. Surgical outcomes and complications of prophylactic salpingectomy at the time of benign hysterectomy in premenopausal women. J Minim Invasive Gynecol 2015;22: Morelli M, Venturella R, Mocciaro R, Di Cello A, Rania E, Lico D, et al. Prophylactic salpingectomy in premenopausal low-risk women for ovarian cancer: primum non nocere. Gynecol Oncol 2013;129: Vorwergk J, Radosa MP, Nicolaus K, Baus N, Jimenez Cruz J, Rengsberger M, et al. Prophylactic bilateral salpingectomy (PBS) to reduce ovarian cancer risk incorporated in standard premenopausal hysterectomy: complications and re-operation rate. J Cancer Res Clin Oncol 2014;140: Robert M, Cenaiko D, Sepandj J, Iwanicki S. Success and complications of salpingectomy at the time of vaginal hysterectomy. J Minim Invasive Gynecol 2015;22: Tone AA, Salvador S, Finlayson SJ, Tinker AV, Kwon JS, Lee CH, et al. The role of the fallopian tube in ovarian cancer. Clin Adv Hematol Oncol 2012;10: Piek JM, Torrenga B, Hermsen B, Verheijen RH, Zweemer RP, Gille JJ, et al. Histopathological characteristics of BRCA1- and BRCA2- associated intraperitoneal cancer: a clinic-based study. Fam Cancer 2003;2: Moorman PG, Myers ER, Schildkraut JM, Iversen ES, Wang F, Warren N. Effect of hysterectomy with ovarian preservation on ovarian function. Obstet Gynecol 2011 ;118: Harlow SD, Gass M, Hall JE, Lobo R, Maki P, Rebar RW, et al. Executive summary of the Stages of Reproductive Aging Workshop + 10: addressing the unfinished agenda of staging reproductive aging. J Clin Endocrinol Metab 2012 ;97: Manchanda R, Menon U. Opportunistic bilateral salpingectomy (OBS) for the prevention of ovarian cancer should be offered in the context of a clinical trial. BJOG 2016;123:463. ª 2017 Royal College of Obstetricians and Gynaecologists 889