Early View
ORIGINAL RESEARCH ARTICLE
Open Access

Impact of patient-reported salpingitis on the outcome of hysterectomy and adnexal surgery: A national register-based cohort study in Sweden

Josefin Jännebring

Josefin Jännebring

Department of Clinical Sciences, Obstetrics and Gynecology, Umeå University, Umeå, Sweden

Search for more papers by this author
Per Liv

Per Liv

Department of Public Health and Clinical Medicine, Umeå University, Umeå, Sweden

Search for more papers by this author
Malin Knuts

Malin Knuts

Department of Clinical Sciences, Obstetrics and Gynecology, Umeå University, Umeå, Sweden

Search for more papers by this author
Annika Idahl

Corresponding Author

Annika Idahl

Department of Clinical Sciences, Obstetrics and Gynecology, Umeå University, Umeå, Sweden

Correspondence

Annika Idahl, Department of Clinical Sciences, Obstetrics and Gynecology, Umeå University, 901 87 Umeå, Sweden.

Email: [email protected]

Search for more papers by this author
First published: 15 February 2024

Abstract

Introduction

Salpingitis is caused by ascending microbes from the lower reproductive tract and contributes to tubal factor infertility, ectopic pregnancy, and chronic pelvic pain. The aim of this study was to analyze if the risk for complications and dissatisfaction after hysterectomy and adnexal surgery was increased in women reporting previous salpingitis.

Material and methods

This is an observational cohort study including women undergoing gynecologic surgery from 1997 to 2020, registered in the Swedish National Quality Register of Gynecologic Surgery (GynOp). Patient-reported previous salpingitis was the exposure. Complications up to 8 weeks and satisfaction at 1 year postoperatively were the outcomes. Multivariable logistic regression and ordinal regression were performed. Results were adjusted for potential confounders including age, body mass index, smoking and year of procedure as well as endometriosis and previous abdominal surgery. Multiple imputation was used to handle missing data.

Results

In this study, 61 222 women were included, of whom 5636 (9.2%) women reported a previous salpingitis. There was an increased risk for women reporting previous salpingitis in both the unadjusted and fully adjusted models to have complications within 8 weeks of surgery (adjusted odds ratio [aOR] 1.22, 95% confidence interval [CI] 1.14–1.32). The highest odds ratios were found for bowel injury (aOR 1.62, 95% CI 1.29–2.03), bladder injury (aOR 1.52, 95% CI 1.23–1.58), and postoperative pain (aOR 1.37, 95% CI 1.22–1.54). Women exposed to salpingitis were also more likely to report a lower level of satisfaction 1 year after surgery compared with unexposed women (aOR 0.87, 95% CI 0.81–0.92).

Conclusions

Self-reported salpingitis appears to be a risk factor for complications and dissatisfaction after gynecologic surgery. This implies that known previous salpingitis should be included in the risk assessment before gynecologic procedures.

Abbreviations

  • BMI
  • body mass index
  • C. trachomatis
  • Chlamydia trachomatis
  • CI
  • confidence interval
  • GynOp
  • The Swedish National Quality Register of Gynecologic Surgery
  • OR
  • odds ratio
  • Key message

    The impact of previous salpingitis on subsequent gynecologic surgery has not been investigated before. Women reporting previous salpingitis have an increased risk for complications up to 8 weeks and dissatisfaction 1 year after benign hysterectomy or adnexal surgery.

    1 INTRODUCTION

    Salpingitis is defined as an inflammation of the fallopian tubes and is caused by ascending microbes from the vagina and cervix. Together with inflammation of the endometrium, the ovaries and the pelvic peritoneum, salpingitis is part of the concept of pelvic inflammatory disease, an inflammation of the female upper reproductive tract.1-3 The life time incidence of pelvic inflammatory disease in Swedish women is approximately 15%.3 Salpingitis is a clinical diagnosis, and the symptoms are lower abdominal pain, fever, abnormal vaginal discharge, vomiting, menstrual irregularities, and urinary symptoms.1, 2 Symptoms are, however, frequently subtle as well as unspecific, which constitutes a risk for misclassification.4

    More than 85% of the infections of the upper reproductive tract are caused by sexually transmitted pathogens or microbes associated with bacterial vaginosis.1 The most common pathogen Chlamydia trachomatis (C. trachomatis) is an intracellular bacterium with a tendency to cause chronic or persistent infections. It is detected in 60% of women with confirmed salpingitis or endometritis.5 Results from prospective studies indicate that 15% of untreated C. trachomatis infections develop into clinically diagnosed pelvic inflammatory disease.3, 6, 7 Infection causes inflammation along the fallopian tubes and ovaries leading to scarring, adhesions, and possibly obstruction of the fallopian tubes.1, 4 As a consequence, salpingitis increases the risk of tubal factor infertility which is the primary cause of female infertility, ectopic pregnancy, and chronic pelvic pain.1, 3, 8 Women with chronic pelvic pain are in turn at higher risk of developing central sensitization and hence generalized pain conditions, with or without confirmed endometriosis.9 Moreover, pelvic inflammatory disease increases the risk of recurrent infection4 and has been associated with an increased risk of ovarian cancer.10

    Hysterectomy and adnexal surgical procedures are important for female health. However, the procedures still entail some risks. Finding indicators that imply increased risk for complications and dissatisfaction, and that might be potential targets for personalized treatment, is vital to improve outcomes. Salpingitis is known to have many negative long-term consequences. Nevertheless, the impact of previous salpingitis on the outcome of subsequent gynecologic surgery for other indications has not been studied thoroughly. We hypothesized that women reporting previous salpingitis would more frequently suffer from complications related to gynecologic surgery and experience a lower level of satisfaction.

    The purpose of this study was to investigate the impact of patient-reported previous salpingitis on hysterectomy and adnexal surgery in a national register-based cohort. The specific aim was to analyze if the risk for complications and dissatisfaction after hysterectomy and adnexal surgery was increased in women reporting previous salpingitis.

    2 MATERIAL AND METHODS

    2.1 Study design

    This is a historical cohort study based on prospectively collected data from the Swedish National Quality Register of Gynecologic Surgery (GynOp). GynOp was founded 1997 and had a coverage of 88.6% of gynecologic procedures in Sweden 2020. Data are collected from 60 gynecologic clinics in Sweden from women undergoing gynecologic surgery. Regulated by law, all women going through a gynecologic procedure are included in the register, albeit withdrawal of consent is possible at any time. The register is certified at the highest level and approved by the Swedish Data Protection Authority and approved for use by healthcare systems.11 Patients answer preoperative questionnaires, and postoperative questionnaires 8 weeks and 1 year after surgery. The questionnaires are sent automatically to the women followed by three reminders if answers are not registered. Printouts are available, although input to the register is primarily web-based. The questionnaires are assessed by a physician at the operating clinic. In addition to the questionnaires answered by the women, a physician fills out pre-, per- and postoperative questionnaires and performs a postoperative assessment. Copies of the collection forms and questionnaires can be requested via www.gynop.se.12

    2.2 Patient selection

    Previous salpingitis has been reported since 1997 for hysterectomies and from 2004 for adnexal procedures, including not only salpingectomy, oophorectomy, salpingo-oophorectomy, sterilization, puncture of ovarian cysts, and fallopian tubes, but also ovarian and tubal biopsy, evacuation of ectopic pregnancy in the fallopian tubes, enucleation and deconstruction of alterations in the ovaries, and reconstructive surgery. Patients with a benign histopathologic diagnosis who had abdominal or laparoscopic surgery since the start of salpingitis registration until 31 December 2020 were included. Exclusion criteria were age under 15 year, if the preoperative questionnaire was missing or if there was no answer regarding previous salpingitis (Figure 1).

    Details are in the caption following the image
    Flow chart illustrating patient selection of women eligible for the study.

    2.3 Exposure

    The exposure was patient-reported salpingitis. In the preoperative questionnaire, the question regarding salpingitis is formulated: “Has a physician informed you that you have or have had any of the following diseases/conditions?”. One of the suggested diseases is “inflammation of the fallopian tubes” and the women are asked to answer either Yes or No.

    2.4 Primary outcomes

    The primary outcomes were complications up to 8 weeks and satisfaction at 1 year. In addition to the questionnaires filled out by the surgeon perioperatively and at discharge from the hospital, complications up to 8 weeks were reported by the patient and assessed by a physician. Complications included minor and major complications and were specified regarding type and location (bowel, bladder, ureter, vessel, nerve, fistula, postoperative pain). Satisfaction was stated by the women in the postoperative questionnaire 1 year after surgery. The levels of satisfaction ranged from “Very dissatisfied” to “Very satisfied” on a five-grade Likert scale.

    The analysis of complications was stratified regarding severity (major complications compared with minor and no complications) as well as type and location (see above). In the analysis of satisfaction, results were stratified on surgical procedure, primary incision, conversion to laparotomy and indication for surgery, which comprised pain, infertility, ovarian cyst, menorrhagia, dysmenorrhea, cancer prophylaxis, sterilization, infection, endometriosis and other indications (pregnancy-related, intra-abdominal bleeding, benign ovarian tumors, diagnostic laparoscopy, unspecified adnexal/uterine disorder, menstrual disorder, urinary tract problems, postmenopausal bleeding, pressure/weight symptoms, dyspareunia, other).

    2.5 Secondary outcomes

    The secondary outcomes were tubal pathology, adhesions, and conversion to laparotomy. Furthermore, perioperative measures including primary incision (abdominal, laparoscopic, and robot-assisted laparoscopic), duration of surgery, and bleeding were presented descriptively for exposed and unexposed patients.

    2.6 Potential confounders

    In addition to estimating crude odds ratios (OR), results were adjusted for potential confounders, identified a priori using directed acyclic graphs. In model 1, adjustment was made for age, body mass index (BMI), smoking and year of surgery. In model 2, adjustment was made for the factors in model 1 as well as previous abdominal surgery—including previous cesarean section, adnexal procedure, hysterectomy, sterilization, myomectomy, ectopic pregnancy, and appendectomy, and endometriosis. As an exploratory analysis, two additional models (model 3 and model 4) were created for post-hoc analyses with adjustments for tubal pathology and adhesions, respectively, as potential mediators for the primary outcomes complications (overall) and satisfaction (overall) in addition to the adjustments of model 1.

    2.7 Statistical analyses

    Logistic regression was used to estimate crude OR and adjusted OR (aOR) with 95% confidence intervals (CI) of complications related to surgery to compare patients with and without previous salpingitis. The analysis was stratified for grade and type of complication. Logistic regression analysis was also conducted with tubal pathology, adhesions, and conversion to laparotomy as outcome. Ordinal regression was used to estimate OR and 95% CIs in the analysis of satisfaction at 1 year, stratified on surgical procedure, primary incision, conversion to laparotomy, and indication for surgery. Multiple imputation by chained equations and predictive mean matching was carried out to decrease the risk of bias potentially introduced by missing data. Imputations were conditioned on the outcomes (complications up to 8 weeks, satisfaction after 1 year, tubal pathology, adhesions) and variables that were correlated with at least one outcome in the logistic regression model and ordinal regression model (age, BMI, smoking, year of surgery, endometriosis, previous abdominal surgery). Primary analysis was conducted in R on imputed data.

    Complete case analysis was conducted as a sensitivity analysis in SPSS, version 28.0.1.1. For the complete case analyses a cohort was created in which women missing essential data were excluded. Variables inevitable for the analysis were factors used in the adjusted models (age, BMI, smoking, year of procedure, endometriosis, and previous abdominal surgery) and factors used in the analysis including the exposure (salpingitis), primary outcome (complications 8 weeks after surgery, Yes or No), secondary outcomes (tubal pathology and adhesions), indication for surgery, primary incision, and surgical procedure. For the analysis of satisfaction 1 year after surgery, women missing data regarding satisfaction were also excluded.

    3 RESULTS

    Data regarding 97 741 women were extracted from GynOp. After application of exclusion criteria, 61 222 women were eligible for this study, of which 5636 (9.2%) reported previous salpingitis (Figure 1). Compared with women reporting no previous salpingitis, exposed women were more often smokers or ex-smokers (25.9% vs. 14.1%), had reported endometriosis (21.9% vs. 13.0%), and had a history of abdominal surgery (68.1% vs. 48.3%). Other demographic data were similar between exposed and non-exposed women (Table 1).

    TABLE 1. Demographic and clinical descriptive data for 61 222 women that went through benign gynecologic surgery, according to patient-reported previous salpingitis exposure. Data registered in the Swedish National Quality Register of Gynecologic Surgery between 1997 and 2020.
    Self-reported previous salpingitis No self-reported previous salpingitis
    n (%) n (%)
    5636 (9.2) 55 586 (90.8)
    Age at procedure, y
    Median (range) 47.00 (16–88) 46.00 (15–94)
    15–24 51 (0.9) 1803 (3.2)
    25–34 390 (6.9) 6418 (11.5)
    35–44 1599 (28.4) 15 090 (27.1)
    45–54 2450 (43.5) 20 852 (37.5)
    55–64 734 (13.0) 6122 (11.0)
    65–74 350 (6.2) 3788 (6.8)
    ≥ 75 61 (1.1) 1509 (2.7)
    Missing data 1 (0.0) 4 (0.0)
    BMI, kg/m2
    Median (range) 25.46 (15–55) 25.21 (15–60)
    <18.5 63 (1.1) 727 (1.3)
    18.5–24 2407 (42.7) 25 374 (45.6)
    25–29 1795 (31.8) 17 769 (32.0)
    30–34 831 (14.7) 7361 (13.2)
    ≥35 307 (5.4) 2866 (5.2)
    Missing data 233 (4.1) 1489 (2.7)
    Smoking
    Yes, 1–5/day 402 (7.1) 2754 (5.0)
    Yes, 6–20/day 1018 (18.1) 4918 (8.8)
    Yes, >20 /day 41 (0.7) 144 (0.3)
    No, have quit 2114 (37.5) 16 315 (29.4)
    No, never smoked 1994 (35.4) 30 875 (55.5)
    Missing data 67 (1.2) 580 (1.0)
    ASA classification
    1 3703 (65.7) 37 370 (67.2)
    2 1419 (25.2) 14 695 (26.4)
    3 127 (2.3) 1139 (2.0)
    4 or 5 0 (0.0) 20 (0.0)
    Missing data 387 (6.9) 2362 (4.2)
    Parity
    0 726 (12.9) 10 206 (18.4)
    1 1067 (18.9) 8552 (15.4)
    2 1940 (34.4) 20 707 (37.3)
    ≥3 1504 (26.8) 12 764 (22.9)
    Missing data 399 (7.1) 3357 (6.0)
    Endometriosisa
    Yes 1235 (21.9) 7240 (13.0)
    No 3807 (67.5) 47 636 (85.7)
    Missing data 594 (10.5) 710 (1.3)
    Preoperative pain
    Yes 414 (7.3) 2954 (5.3)
    No 5075 (90.0) 51 229 (92.2)
    Missing data 147 (2.6) 1403 (2.5)
    Previous abdominal surgeryb
    Yes 3837 (68.1) 26 826 (48.3)
    No 1138 (20.2) 16 579 (29.8)
    Missing data 661 (11.7) 12 181 (21.9)
    • Note: Demographic and clinical descriptive data registered in GynOp for 61 222 women that underwent gynecologic surgery between 1997 and 2020.
    • Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index; SD, standard deviation.
    • a Patient-reported.
    • b Previous cesarean section, adnexal procedure, hysterectomy, sterilization, myomectomy, ectopic pregnancy, appendectomy.

    As indication for surgery, women with self-reported previous salpingitis more often had dysmenorrhea (5.1% vs. 2.9%), pain (7.3% vs. 5.3%), endometriosis (2.2% vs. 1.6%), and infection (3.4% vs. 0.3%) (Table 2). In contrast, the indication ovarian cyst was less common (24.7% vs. 29.9%). There was no marked difference regarding choice of surgical procedure between the groups, although women with previous salpingitis more often had an abdominal than a laparoscopic incision.

    TABLE 2. Indication for surgery and surgical procedure for 61 222 women that underwent benign gynecologic surgery between 1997 and 2020, according to patient-reported previous salpingitis exposure. Data registered in the Swedish National Quality Register of Gynecologic Surgery.
    Self-reported previous salpingitis No self-reported previous salpingitis
    n (%) n (%)
    5636 (9.2) 55 586 (90.8)
    Indication for surgery
    Cysts 1392 (24.7) 16 636 (29.9)
    Menorrhagia 1291 (22.9) 11 334 (20.4)
    Dysmenorrhea 290 (5.1) 1586 (2.9)
    Pain 414 (7.3) 2954 (5.3)
    Endometriosis 124 (2.2) 912 (1.6)
    Cancer prophylaxis 91 (1.6) 1147 (2.1)
    Sterilization 12 (0.2) 300 (0.5)
    Infertility 54 (1.0) 267 (0.5)
    Infection 189 (3.4) 176 (0.3)
    Other indicationsa 1632 (29.0) 18 871 (33.9)
    Missing data 147 (2.6) 1403 (2.5)
    Surgical procedure
    Hysterectomy 1297 (23.0) 14 020 (25.2)
    Adnexalb 1822 (32.3) 20 224 (36.4)
    Hysterectomy/adnexal 1221 (21.7) 13 318 (24.0)
    Missing data 1296 (23.0) 8024 (14.4)
    Primary incision
    Abdominal 3294 (58.4) 29 015 (52.2)
    Laparoscopic 2342 (41.6) 26 571 (47.8)
    Missing data 0 (0.0) 0 (0.0)
    Duration of surgery (min)
    Mean (range) 89 (12–529) 89 (0–704)
    Missing data 90 (1.6) 504 (0.9)
    Bleeding (mL)
    Mean (range) 189 (0–6200) 177 (0–6500)
    Missing data 88 (1.6) 435 (0.8)
    • a Pregnancy-related, intra-abdominal bleeding, benign ovarian tumors, diagnostic laparoscopy, unspecified adnexal/uterine disorder, menstrual disorder, urinary tract problems, postmenopausal bleeding, pressure/weight symptoms, dyspareunia, other.
    • b Adnexal procedure comprises salpingectomy, oophorectomy, salpingo-oophorectomy, sterilization, puncture of ovarian cysts and fallopian tubes, but also ovarian and tubal biopsy, evacuation of ectopic pregnancy in the fallopian tubes, enucleation and deconstruction of alterations in the ovaries, reconstructive surgery.

    Women reporting previous salpingitis vs. no previous salpingitis had increased risk of complications overall (OR 1.47, 95% CI 1.38–1.57; adjustment model 2 (aOR2) 1.22, 95% CI 1.14–1.32) and major complications (OR 1.20, 95% CI 1.04–1.37; aOR2 1.06, 95% CI 0.91–1.25). Stratified on type of complication there was an increased risk for damage to the bowel, bladder, and nerves, and for postoperative pain (Table 3 and Figure 2). For complications specified as damage to the ureter or a vessel or fistula formation, OR were generally increased, but the analysis is based on few cases and confidence intervals are wide.

    TABLE 3. Complications within 8 weeks after gynecologic surgery for women reporting previous salpingitis compared with women reporting no previous salpingitis, overall and stratified on severity and type of complication. Analysis based on data from the Swedish National Quality Register of Gynecologic Surgery.
    Self-reported previous salpingitis No self-reported previous salpingitis ORa (95% CI) Model 1 Model 2
    aORa (95% CI) aORa (95% CI)
    n (%) n (%)
    5636 (9.2) 55 586 (90.8)
    Complications, surgery up to 8 weeks
    Uncomplicated 3601 (63.9) 39 916 (71.8) Reference Reference Reference
    Complication (overall)b 1603 (28.4) 12 073 (21.7) 1.47 (1.38–1.57) 1.31 (1.23–1.40) 1.22 (1.14–1.32)
    Majorc 230 (4.1) 1938 (3.5) 1.20 (1.04–1.37) 1.18 (1.02–1.36) 1.06 (0.91–1.25)
    Missing data 432 (7.7) 3597 (6.5)
    Type
    Bowel 121 (2.1) 681 (1.2) 1.97 (1.62–2.40) 1.76 (1.43–2.17) 1.62 (1.29–2.03)
    Bladder 153 (2.7) 924 (1.7) 1.84 (1.54–2.19) 1.66 (1.39–2.00) 1.51 (1.23–1.58)
    Ureter 22 (0.4) 176 (0.3) 1.39 (0.89–2.16) 1.39 (0.87–2.23) 1.12 (0.65–1.94)
    Vessel 19 (0.3) 158 (0.3) 1.33 (0.83–2.15) 1.39 (0.86–2.26) 1.15 (0.65–2.03)
    Nerve 78 (1.4) 589 (1.1) 1.47 (1.16–1.86) 1.37 (1.06–1.76) 1.21 (0.91–1.61)
    Fistula 7 (0.1) 58 (0.1) 1.34 (0.61–2.93) 1.20 (0.54–2.66) 0.89 (0.35–2.28)
    Postoperative pain 534 (9.5) 3676 (6.6) 1.61 (1.46–1.77) 1.51 (1.36–1.67) 1.37 (1.22–1.54)
    Missing data 669 (41.7) 6262 (51.9)
    • Abbreviations: aOR, adjusted odds ratio, either after Model 1 or Model 2; CI, confidence interval; OR, odds ratio; Model 1 = results adjusted for age; body mass index, smoking and year of procedure; Model 2 = Model 1 and additionally adjusted for endometriosis and previous abdominal surgery. Results presented as crude and adjusted odds ratios.
    • a Analysis conducted on imputed data.
    • b Complication (overall) comprises minor and major complications.
    • c Compared with no/minor complications.
    Details are in the caption following the image
    Complications up to 8 weeks after hysterectomy or adnexal surgery, stratified on type of complication, for women reporting previous salpingitis compared with unexposed women.

    In Table 4 the association between patient-reported previous salpingitis and satisfaction 1 year after surgery is presented as satisfaction overall and stratified on surgical procedure, primary incision, conversion to laparotomy, and indication for surgery. Here, the OR and 95% CI are estimated using ordinal regression, where an OR greater than one indicates greater odds for women reporting previous salpingitis to state a level of satisfaction at least one step higher on the scale compared with women reporting no previous salpingitis. On the other hand, an OR below one indicates lower odds for satisfaction in women reporting previous salpingitis.

    TABLE 4. Level of satisfaction 1 year after benign gynecologic surgery among women reporting previous salpingitis compared with women reporting no previous salpingitis. Based on data from the Swedish National Quality Register of Gynecologic Surgery.
    Self-reported previous salpingitis No self-reported previous salpingitis ORa (95% CI) Model 1 Model 2
    aORa (95% CI) aORa (95% CI)
    n (%) n (%)
    5636 (9.2) 55 586 (90.8)
    Satisfaction (overall)b 4251 (74.4) 41 143 (74.0) 0.94 (0.89–1.00) 0.83 (0.79–0.89) 0.87 (0.81–0.92)
    Missing datac 1385 (24.6) 14 443 (26.0)
    Surgical procedure
    Hysterectomy 1297 (23.0) 14 020 (25.2) 0.94 (0.89–1.00) 0.85 (0.75–0.96) 0.88 (0.78–1.00)
    Adnexaldd 1822 (32.2) 20 224 (36.4) 0.94 (0.86–1.04) 0.88 (0.80–0.97) 0.90 (0.82–1.00)
    Hysterectomy/adnexal 1221 (21.7) 13 318 (24.0) 0.83 (0.73–0.94) 0.81 (0.71–0.92) 0.83 (0.73–0.95)
    Missingc 1296 (23.0) 8024 (14.4)
    Primary incision
    Abdominal 3294 (58.4) 29 015 (52.2) 0.91 (0.85–0.98) 0.83 (0.77–0.89) 0.86 (0.79–0.92)
    Laparoscopic 2342 (41.6) 26 571 (47.8) 0.93 (0.85–1.02) 0.85 (0.77–0.93) 0.88 (0.80–0.96)
    Conversion to laparotomye
    No 5456 (96.7) 54 006 (97.1) Reference Reference Reference
    Yes 180 (3.2) 1580 (2.8) 1.18 (0.87–1.59) 1.09 (0.80–1.48) 1.10 (0.81–1.50)
    Indication for surgery
    Cysts 1392 (24.7) 16 636 (29.9) 0.90 (0.80–1.00) 0.82 (0.74–0.92) 0.84 (0.75–0.94)
    Menorrhagia 1291 (22.9) 11 334 (20.4) 1.05 (0.81–1.35) 0.92 (0.71–1.20) 0.96 (0.73–1.25)
    Dysmenorrhea 290 (5.1) 1586 (2.9) 0.88 (0.78–1.00) 0.78 (0.68–0.89) 0.80 (0.70–0.91)
    Pain 414 (7.3) 2954 (5.3) 1.03 (0.84–1.26) 0.91 (0.74–1.13) 0.96 (0.78–1.18)
    Endometriosis 124 (2.2) 912 (1.6) 1.21 (0.82–1.78) 1.15 (0.77–1.71) 1.15 (0.77–1.71)
    Cancer prophylaxis 91 (1.6) 1147 (2.1) 1.01 (0.65–1.57) 0.89 (0.57–1.40) 0.93 (0.59–1.46)
    Infertility 54 (1.0) 267 (0.5) 0.42 (0.22–0.82) 0.38 (0.19–0.78) 0.40 (0.20–0.80)
    Otherf 1632 (29.0) 18 871 (33.9)
    Missing datac 147 (2.6) 1403 (2.5)
    • Note: Odds ratios constructed through ordinal regression are interpreted as follows: OR >1 indicates higher odds for women reporting previous salpingitis to state a higher level of satisfaction compared with women not reporting previous salpingitis (reference). OR <1 indicates on the contrary lower odds for women reporting previous salpingitis to state a higher level of satisfaction compared with the reference group.
    • Abbreviations: aOR, adjusted odds ratio, either after Model 1 or Model 2; CI, confidence interval; OR, odds ratio. Model 1 = results adjusted for age, body mass index, smoking and year of procedure. Model 2 = Model 1 and additionally adjusted for endometriosis and previous abdominal surgery.
    • a Analysis conducted on imputed data.
    • b Satisfaction: rated as Very dissatisfied, Dissatisfied, Neither satisfied nor dissatisfied, Satisfied and Very satisfied by the patient 1 year after surgery.
    • c Missing data comprises data collected from Gyn-KvalitetsRegistret (GKR) for benign hysterectomies and adnexal procedures performed in Stockholm, Karlstad and Visby, 2010–2017.
    • d Adnexal procedure comprises salpingectomy, oophorectomy, salpingo-oophorectomy, sterilization, puncture of ovarian cysts and fallopian tubes, but also ovarian and tubal biopsy, evacuation of ectopic pregnancy in the fallopian tubes, enucleation and deconstruction of alterations in the ovaries, reconstructive surgery.
    • e Conversion to laparotomy includes cases of conversion from laparoscopic to abdominal incision.
    • f Other indications: pregnancy-related, intra-abdominal bleeding, benign ovarian tumors, diagnostic laparoscopy, unspecified adnexal/uterine disorder, menstrual disorder, urinary tract problems, postmenopausal bleeding, pressure/weight symptoms, dyspareunia, other.

    A lower level of satisfaction was measured in women reporting previous salpingitis for the full cohort (OR 0.94, 95% CI 0.89–1.00; aOR2 0.87, 95% CI 0.81–0.92). Stratified on surgical procedure and primary incision, women reporting previous salpingitis still stated a lower level of satisfaction in all strata, revealing no heterogeneity based on these factors. Regarding indication for surgery, the stratified analysis illustrated heterogeneous results. For details, see Table 4 and Figure 3.

    Details are in the caption following the image
    Satisfaction 1 year after hysterectomy or adnexal surgery, stratified on indication for surgery, for women reporting previous salpingitis compared with unexposed women.

    In an exploratory analysis an increased risk for tubal pathology, adhesions, and conversion to laparotomy was found in women reporting previous salpingitis (Table 5).

    TABLE 5. Association between patient-reported previous salpingitis and occurrence of macroscopic tubal pathology, adhesions and conversion to laparotomy for women reporting previous salpingitis compared with women reporting no previous salpingitis (reference). Based on data from the Swedish National Quality Register of Gynecologic Surgery.
    Self-reported previous salpingitis No self-reported previous salpingitis ORa (95% CI) Model 1 Model 2
    aORa (95% CI) aORa (95% CI)
    n (%) n (%)
    5636 (9.2) 55 586 (90.8)
    Fallopian tubes
    Macroscopic normal 3396 (60.3) 43 033 (77.4) Reference Reference Reference
    Macroscopic pathology 1360 (24.1) 8092 (14.6) 2.12 (1.99–2.28) 2.49 (2.32–2.68) 2.19 (2.02–2.38)
    Missing data 880 (15.6) 4461 (8.0)
    Adhesions
    No 2758 (48.9) 35 578 (64.0) Reference Reference Reference
    Yes 2490 (44.2) 18 045 (32.5) 1.78 (1.68–1.88) 1.83 (1.72–1.94) 1.39 (1.30–1.49)
    Missing data 388 (6.9) 1963 (3.5)
    Conversion to laparotomyb
    No 5456 (96.7) 54 006 (97.1) Reference Reference Reference
    Yes 180 (3.2) 1580 (2.8) 1.42 (1.21–1.67) 1.18 (1.00–1.40) 1.10 (0.91–1.32)
    • Abbreviations: aOR, adjusted odds ratio, either after Model 1 or Model 2; CI, confidence interval; Model 1, results adjusted for age, body mass index, smoking and year of surgery. Model 2, Model 1 and additionally adjusted for endometriosis and previous abdominal surgery; OR, odds ratio.
    • a Analysis conducted on imputed data.
    • b Conversion to laparotomy includes cases of conversion from laparoscopic to abdominal incision.

    In a post-hoc analysis we adjusted for tubal pathology and adhesions respectively in addition to age, BMI, smoking, and year of surgery (Models 3 and 4) to analyze tubal damage and adhesions as mediators. There was still an increased risk for complications (aOR3 1.33, 95% CI 1.24–1.43; aOR4 1.28, 95% CI 1.20–1.37) and lower level of satisfaction (aOR3 0.86, 95% CI 0.80–0.91; aOR4 0.86, 95% CI 0.81–0.91). In two additional models (Models 5 and 6), analysis of satisfaction was adjusted for complications (overall) and major complications respectively together with age, BMI, smoking and year of surgery. The risk for dissatisfaction remained unchanged in both models (aOR5 0.87, 95% CI 0.82–0.89; aOR6 0.85, 95% CI 0.80–0.90).

    Sensitivity analyses were conducted using complete case data to account for possible bias introduced when adjusting for missing data using multiple imputation. The results, presented as Supporting information (Table S1–S5), show little divergence compared with analyses conducted using imputed data.

    4 DISCUSSION

    We hypothesized that women reporting previous salpingitis would more often suffer complications within 8 weeks and state a lower level of satisfaction 1 year after hysterectomy and adnexal surgery compared with women reporting no previous salpingitis. The hypothesis was confirmed, illustrating an increased risk for complications and dissatisfaction related to gynecologic surgery on various indications for women reporting previous salpingitis. The results were consistent after adjustment for age, BMI, smoking, year of procedure, patient-reported endometriosis, previous abdominal surgery, adhesions, and tubal pathology.

    There was an increased risk for exposed women to have complications overall, and particularly damage to the bowel and bladder, as well as postoperative pain. No studies investigating the impact of previous salpingitis on subsequent gynecologic surgery were encountered when conducting a general literature search. The risk for injuries to the bowel and bladder related to gynecologic surgery might be increased by adhesions, endometriosis, and previous surgery among other risk factors such as proximity to pelvic organs.13-15 Infection-induced inflammation of the fallopian tubes causes hyperplasia of fibrous tissue.16 If the condition evolves to peritonitis, which is characterized by fibrinoid exudate on serosal surfaces, a consequence is agglutination between fallopian tubes, ovaries, bowel, and omentum. Agglutination can in turn cause adhesions and pelvic pain.4 The presence of adhesions due to previous inflammation might, to some extent, explain the increased risk for bowel and bladder complications among women reporting previous salpingitis. In this study, the risk for complications was still increased after additional adjustment for adhesions and tubal pathology, indicating mechanisms beyond sequelae of salpingitis that can be detected at surgery.

    Chronic pelvic pain is a known sequela of salpingitis,1 which might explain the increased risk for postoperative pain among women reporting salpingitis in this study. Women with chronic pelvic pain are at higher risk of developing central sensitization and hence generalized pain conditions, with or without confirmed endometriosis.9 Postoperative pain in women with endometriosis going through hysterectomy has been proposed to be caused by pain sensitization in the context of chronic pelvic pain by Sandström et al.17 The same conclusion is presented in two studies examining pain thresholds in women with chronic pelvic pain and endometriosis.18, 19 In a register-based study using data from GynOp, Grundström et al. found that 22% of women with preoperative pain experienced pain 1 year after hysterectomy compared with 8% of women without preoperative pain. Postoperative complications within 8 weeks were identified as an independent risk factor for pain after 1 year.20 Women reporting previous salpingitis have, in this study, a higher risk for postoperative pain as a complication within 8 weeks of surgery. Similar to what is described regarding endometriosis, salpingitis might cause chronic pelvic pain leading to pain sensitization and hence higher risk for postoperative pain compared with women reporting no previous salpingitis. In a broader perspective, the results implicate that the anamnestic information regarding previous salpingitis is relevant in the assessment of risk for complications, including pain, after gynecologic surgery.

    In this study, women reporting previous salpingitis were less satisfied 1 year after hysterectomy or adnexal surgery. Dissatisfaction may of course be explained by many factors and the reason women are dissatisfied is not explained by the data, so no conclusion of a causal relationship can be drawn. In a register-based study using data from GynOp, Borendal Wodlin found that dissatisfaction correlates with intraoperative and postoperative complications, especially major complications.21 However, in this report, neither adjustment for complications (overall) nor major complications changed the association between patient-reported salpingitis and dissatisfaction after 1 year. This suggests that other aspects of salpingitis, not yet identified, are related to the increased risk of dissatisfaction after gynecologic surgery. Also using data from GynOp, Grundström et al. found that women reporting preoperative pain and endometriosis were more dissatisfied 1 year after hysterectomy and suffered from more severe complications postoperatively.22 In our study, lower satisfaction in women with previous salpingitis remained after adjustment for endometriosis (Model 2), indicating that there were other factors predisposing for dissatisfaction.

    As hypothesized, macroscopic damage of the fallopian tubes and adhesions were seen more frequently in women reporting previous salpingitis. Tubal damage in the context of C. trachomatis infection, the most common bacterium found in pelvic inflammatory disease, appears to be caused by both the innate and adaptive immune systems.2, 23 Not all women develop scarring following genital chlamydial infections. The severity of the sequelae is most likely influenced by host factors.24 Adjusting for tubal pathology and adhesions (Models 3 and 4) in a post-hoc analysis of complications and satisfaction did not alter the result. The risk increase for both outcomes was essentially the same after adjustment. This suggests that the risk increase for adverse events in women reporting previous salpingitis is mediated through other factors that are not yet identified.

    We believe that awareness of the increased risk for complications and dissatisfaction in women who report previous salpingitis is clinically relevant. Our data do not tell which measures could improve outcomes in this subgroup of women and there is a paucity of scientific evidence. Women with endometriosis, another disease with increased risk for chronic pain, are suggested to have extra benefit from following the Enhanced Recovery After Surgery (ERAS) approach.25 Similarly, following the ERAS protocol might be particularly important in women with previous salpingitis to improve surgical outcomes. Our study shows that 28.4% of women reporting previous salpingitis have postoperative complications compared with 21.7% of women who do not. Consequently, there is a potential to avoid nearly one out of four complications in women with self-reported previous salpingitis and increase satisfaction in this subgroup. Optimizing perioperative care, for example by ensuring that women are properly prepared and treated with multimodal analgesia, could reduce the risk for complications, postoperative pain, and dissatisfaction.26 The effectiveness of such an approach in women with previous salpingitis is a subject for future studies.

    This report is based on data from GynOp, a large national quality register collecting data from 60 gynecologic clinics in Sweden. The register is generally considered reliable with the highest certification level that can be obtained for a national quality register in Sweden.11 The wide coverage reduces the risk of selection bias, and the analyzed cohort is representative of the population undergoing gynecologic surgery in Sweden. This implies generalizability of the results. The large population also made it possible to apply desired exclusion criteria and performing stratified analyses while preserving sufficient power. As in all observational register-based studies, there is risk for bias caused by missing data. Multiple imputation was carried out to account for this type of bias, which is a strength of the study. Another aspect to bear in mind is that the present study suffers from risk of residual confounding, as do all epidemiologic studies, and that the results may be biased from true causal associations. Analysis based on a large volume of material may result in narrow confidence intervals indicating statistical significance which does not necessarily equal clinical significance. We believe, however, that the findings of the present study are relevant in a clinical context and can improve the risk assessment before gynecologic surgery. Perhaps the most central weakness of this study is that salpingitis is not confirmed objectively, which could lead to misclassification and recall bias. There is a risk that women have misinterpreted their initial diagnosis of salpingitis, as well as not recalling properly. A possible consequence of this could be large variability in reporting the clinically correct diagnosis of salpingitis. On the other hand, there is probably underreporting of previous salpingitis because many infections are asymptomatic and women may be unaware of the disease. However, we consider the misclassification as non-differential, and the risk is therefore rather an attenuation and underestimation of the association between self-reported salpingitis and complications and satisfaction.

    5 CONCLUSION

    Results of this study indicate that patient-reported previous salpingitis is associated with higher incidence and severity of complications as well as dissatisfaction after gynecologic surgery. Despite not being confirmed objectively and known to be asymptomatic in many cases, the risk for adverse events and dissatisfaction is higher in women reporting previous salpingitis when undergoing hysterectomy or an adnexal procedure. This indicates that previous salpingitis should be considered a risk factor in the preoperative risk assessment. Taking measures to lower the risk for complications and dissatisfaction, e.g. focus on pain relief for women at higher risk of postoperative pain and dissatisfaction, might alter the outcome in a positive direction.

    AUTHOR CONTRIBUTIONS

    Design of the study: Annika Idahl, Malin Knuts, Josefin Jännebring, Per Liv. Ethical approval: Annika Idahl, Malin Knuts. Main responsibility for contact with developers and programmers of the GynOp register: Malin Knuts. Statistical analyses: Josefin Jännebring, Per Liv. Drafting and editing the manuscript: Josefin Jännebring, Annika Idahl, Per Liv. All authors have contributed constructive comments and read and approved the final version of the manuscript. Annika Idahl is the principal investigator of this study.

    ACKNOWLEDGMENTS

    We would like to thank GynOp and Birgitta Renström for assistance in navigating the register.

      FUNDING INFORMATION

      This study was supported by the Medical Faculty, Umeå University, Umeå, Sweden.

      CONFLICT OF INTEREST STATEMENT

      None of the authors report any conflict of interest.

      ETHICS STATEMENT

      The study was approved by the Swedish Ethical Review Authority (Dnr. 2021–01117) on March 29, 2021. The women included in the register were informed before participating that the data might be used for quality improvement and future research after ethical approval. Also, participants were informed that partaking was voluntary, and that consent could be withdrawn at any time.