Excerpts from the World Medical Literature

      Taylor C, Ellett L, Hiscock R, et al. Hysteroscopic management of retained products of conception: a systematic review. Aust N Z J Obstet Gynaecol 2022;62:22–32.
      Summary: A review of 19 prospective and retrospective studies evaluating hysteroscopic versus nonhysteroscopic management of retained products of conception (RPOC) that included 2314 patients. The primary outcome was reproductive outcomes, specifically, conception, time to conception, and live birth; secondary outcomes included acute surgical complications, development of intrauterine adhesions, and future pregnancy complications. Studies with either a single treatment arm or comparison arms (e.g., hysteroscopy vs. suction curettage, manual vacuum aspiration, or medical management) were included. Fourteen studies had no comparison group, and five compared hysteroscopic removal to either suction or manual evacuation. No randomized controlled trials were identified. All five comparison studies were considered to have a critical risk of bias; therefore, a formal meta-analysis was not done, although meta-analysis techniques were used to provide summary estimates for the primary and secondary outcomes. The conception rate after hysteroscopic resection for RPOC was 81.1% and 65.4% after nonhysteroscopic management. Live birth rates were 87.3% and 93.8%, respectively, for both approaches. Complications were very low (1%–2%) in both groups. The rate of intrauterine adhesions after hysteroscopic resection of RPOCs was 6.8%, but this outcome was reported in only one study.
      Comments: I have reviewed and commented on the increasing “medicalization” of miscarriage in these excerpts over the past few years. I think the combination of increasing use of high-resolution ultrasound, clinician and patient anxiety and impatience, and temptation to use complicated surgical devices when simple procedures are available have contributed to this trend. I have observed with a mixture of curiosity and scepticism colleagues booking hysteroscopic removal of RPOC soon after an early first-trimester miscarriage, wondering why an expectant approach wasn’t considered in a minimally symptomatic patient with a small focus of RPOC. This paper does little to support that practice but at least reassures us that this approach is not dangerous. It is not possible to draw definitive conclusions from the limited research so far, as the authors have alluded to at the outset. My own practice is to use suction curettage with ultrasound guidance, and I will reserve judgment until time and better data allow us to make the best and most cost-effective decision for these cases.
      Chan JK, Tian C, Kesterson JP, et al. Symptoms of women with high-risk early-stage ovarian cancer. Obstet Gynecol 2022;139:157–62.
      Summary: A retrospective chart review of 419 patients enrolled in a phase 3 clinical trial (Gynecological Oncology Group 157) of patients with early stage (1–2) ovarian cancer. The goal of the study was to assess the presentation, characteristics, and prognostic significance of symptoms in patients with high-risk, early-stage ovarian cancer. A Cox proportional hazards model was used to examine the association between presenting symptoms and recurrence-free survival and overall survival. Kaplan-Meier survival plots, based on presenting symptoms, were calculated and compared using the log-rank test. Three-hundred one of the 419 patients (72%) presented with one or more symptoms and 118 (28%) were asymptomatic. The two most common symptoms were pelvic pain and fullness. There was a significant association between primary tumour size and symptoms; 65% of patients with tumours <10 cm experienced symptoms compared with 71% of those with tumours 10 to 15 cm and 79% of those with tumours >15 cm (P < 0.001). Symptoms were not associated with age, stage, or histological type. There was no association between the presence of symptoms (single or multiple) and either the risk of recurrence or survival (log-rank test P = 0.712 and P = 0.917, respectively).
      Comments: The literature on early detection and screening for ovarian cancer remains depressing; the large, randomized screening trials using ultrasound and CA-125 measurements show zero impact on overall survival, and other novel tests (endometrial biopsy, other serum markers) are still not ready for prime time. This study looks retrospectively at the symptoms experienced by thoroughly staged patients with early-stage ovarian cancer to determine whether the presence of symptoms in women with “the disease that whispers” is associated with improved prognosis. Sadly, the answer is no. However, the data here partially refute the claim that symptoms (pain, fullness, or sensation of increasing girth, vaginal bleeding, urinary and gastrointestinal problems) are only present in advanced stage disease. In fact, the authors compared the symptoms of these 419 patients with those of a historical cohort with advanced disease and found that those with late-stage disease generally had more gastrointestinal symptoms. So, what’s the message to the generalist here? I would argue we should abandon requests for screening in asymptomatic patients but react promptly when postmenopausal patients complain of new onset pelvic pain. Early, effective detection remains elusive for the time being.
      Cusimano MC, Ferguson SE, Moineddin R, et al. Ovarian cancer incidence and death in average-risk women undergoing bilateral salpingo-oophorectomy at benign hysterectomy. Am J Obstet Gynecol 2022;226:220.e1–26.
      Summary: A population-based cohort study of 195 282 adult women undergoing hysterectomy, with or without bilateral salpingo-oophorectomy (BSO) for benign disease in Ontario between 1996 and 2010. The primary outcome was incidence of ovarian cancer; the secondary outcome was death from ovarian cancer. A variety of Canadian national and provincial databases and registries were used to obtain demographic and surgical data and cancer outcomes; the Ontario Cancer Registry holds the records of all incident cancers in the province from 1964 to 2019 and is 95% complete. Essentially, all patients undergoing hysterectomy with previous malignancy or for risk-reducing indications were excluded from the study. A total of 148 621 patients (76.1%) had ovarian conservation at the time of surgery and 46 661 (23.9%) had concurrent BSO. Mean age was 47.2 years; 70% of hysterectomies were by an abdominal or laparoscopic approach; patients undergoing BSO were older (50.7 vs. 46.1 y; P < 0.001). Patients undergoing BSO were also more likely to have multiple comorbidities, a premalignant condition, or endometriosis (37.2% vs. 21.2%; P < 0.001). Median follow-up was 16 years. Inverse probability of treatment weighting was used to adjust for patients undergoing BSO and ovarian conservation; propensity scores were obtained using logistic regression. Fine-Grey subdistribution hazard models were used to estimate the effect of BSO on each outcome (these are time-to-event models that account for competing risks). Absolute risk reduction was calculated at 10, 15, and 20 years of follow-up; the inverse of absolute risk reduction was used to calculate the number needed to treat (NNT) at each time period. A total of 548 women developed ovarian cancer during follow-up at a median age of 59.6 years. BSO was associated with lower incidence of ovarian cancer (hazard ratio 0.23; 95% confidence interval [CI] 0.14–0.38; P < 0.001); the NNTs at 10, 15, and 20 years were 634, 405, and 260, respectively. BSO was also associated with lower incidence ovarian cancer death (hazard ratio 0.30; 95% CI 0.16–0.57; P < 0.001); the NNTs were 1806, 939, and 569, respectively, at 10, 15, and 20 years. Sensitivity analysis showed an increased benefit in women over the age of 50 years at the time of hysterectomy compared with the overall cohort.
      Comments: I had to reread the statistical analysis portion of this paper several times, and I’m still not sure I completely understand the concepts of inverse probability of treatment weighting and Fine-Grey models (authorship by four PhDs may be a partial explanation). But the numbers are impressive, the follow-up scrupulous, and the sources are good ol’ honest Canadian databases. So, I think the information is credible. Counselling women undergoing hysterectomy about BSO is easy in women under age 45 years and those already in menopause, but the conversation is nuanced in women aged 45 to 50 years; it is often an emotional decision based on “fear of getting ovarian cancer,” and this publication enables us to base this decision on evidence. However, clinical decisions are rarely based on one publication or one outcome; I previously reviewed the paper of Tuesley et al. in the May 2021 installment of this column, which provided a more global overview of overall mortality based on age of intervention. Though having NNT may be helpful in a population sense, it is not a number that resonates easily with patients, and I’m not sure this paper will help direct the individual conversation with an undecided patient. It may, however, help us convince perimenopausal women close to age 50 to consider concurrent BSO when they’re on the fence.
      van der Vaart LR, Vollebregt A, Milani AL, et al. Pessary or surgery for a symptomatic pelvic organ prolapse: the PEssary Or surgery for symptomatic PELvic organ Prolapse study, a multicentre prospective cohort study. BJOG 2022;129:820–9.
      Summary: The PEssary Or surgery for symptomatic PELvic organ Prolapse project was initiated to compare the effectiveness of pessary and surgery in women with symptomatic pelvic organ prolapse (POP); it included a noninferiority randomized clinical trial (RCT) and the observational cohort study presented in this paper. Many women with POP had strong preferences for treatment type and refused to participate in the RCT; data from 539 women from 22 contributing Dutch hospitals are analyzed in this study. The primary outcome was subjective improvement at 24 months, according to the patient global impression of improvement (PGI-I) scale, a 7-point Likert scale validated for success in women undergoing treatment for POP. Secondary outcomes included crossover of therapy, severity of symptoms measured by the patient global impression of severity (PGI-S) scale, and pelvic floor distress inventory (PFDI-20), a 20-question scale assessing prolapse, bladder, and bowel symptoms. Three-hundred thirty-five patients (62.2%) were in the pessary group and 204 (37.8%) were in the surgery group. Surgical procedures included uterus-preserving procedures like sacrospinous hysteropexy, Manchester-Fothergill, and anterior and posterior colporrhaphy, where indicated. Patients who opted for surgery were significantly younger, heavier, and more symptomatic at baseline (higher PGI-S and PFDI-20 scores; P < 0.001). At 24 months, 83.8% of patients in the surgery group were found to have improved versus 74.4% in the pessary group, an absolute difference of 9.4% (95% CI 1.4–17.3%, P < 0.01). Between baseline and follow-up, 102 women (30.2%) stopped using the pessary, and 79 (23.6%) switched to surgery. In the surgery group, 22 women (10.8%) underwent additional surgery because of recurrent POP and/or urinary incontinence. Both groups experienced significant improvement in the PFDI-20 at 24 months, with a mean difference of 15 points in favour of surgery (mean difference –15; 95% CI 7.1–23.0, P < 0.001). Women who had surgery also had significantly greater improvement in bladder and bowel function.
      Comments: I like this paper because, as the authors point out, “ allocating intervention based on shared decision making, our study reflects real clinical practice, which enhances the external validity of the findings...” They go on to say that that although RCT trial design eliminates confounding factors, participants who agree to participate in RCTs generally don’t have a strong preference for treatment, and this limits generalizability. Interestingly, the outcomes in this study are all measured with subjective scales; there is no objective evaluation of surgical outcome, which is in keeping with the increasing tendency to look at overall quality of life rather than a single measurable parameter as outcome in contemporary research. Though there are many ways to spin the results, I come away with the impressive results of the simple pessary in alleviating symptoms of this bothersome, prevalent, nonlife-threatening condition as an alternative to surgery, with its associated risks, complications, and failures over time. I’m guessing my urogynaecology colleagues must have some interesting observations about the COVID-19 pandemic and the restrictions on access to the operating room for nononcologic procedures. I think many patients favour a conservative option and trust a physician who doesn’t jump to offer surgery as a first choice; those failed by conservative management are generally more comfortable when they decide to undergo surgery, even with its attendant risks.

      Linked Article