S Ismail

How well can pelvic floor muscles with major defects contract? A cross‐sectional comparative study 6 weeks after delivery using transperineal 3D/4D ultrasound and manometer

To investigate ability to contract, vaginal resting pressure (VRP), pelvic floor muscle (PFM) strength and PFM endurance 6 weeks after vaginal delivery in primiparous women, with and without major defects of the levator ani (LA) muscle.

The effects of vault drainage on postoperative morbidity after vaginal hysterectomy for benign gynaecological disease: a randomised controlled trial

Sir, I read with interest the randomised controlled trial that assessed the effect of vault drainage on postoperative morbidity after vaginal hysterectomy for benign gynaecological disease. The primary outcome measure was defined as febrile morbidity, which is an indirect manifestation of vault haematoma. A more precise measure would have been symptomatic vault haematoma. More than three in four patients had vaginal hysterectomy for pelvic organ prolapse. Such patients have easily accessible pedicles, and are thus at reduced risk of vault haematomata than patients without pelvic organ prolapse. Moreover, they are also more likely to have a vaginal pack following the pelvic floor repair, which may reduce the risk of haematoma and return to theatre, a step that could have been better standardised in the protocol. Likewise, the technique of vault closure should have been standardised, as it can also influence the risk of vault haematoma. The trial protocol specified non-suction drains, and this excludes corrugated drains that are used in vault drainage after vaginal hysterectomy. It is not clear how randomisation numbers were generated. The sealed envelopes, with trial allocation, were attached to patients’ notes at the preoperative assessment visit. This means that randomisation was carried out at this visit, rather than in theatre. As a result, some randomised patients did not have hysterectomy, and these patients should have been included in the intention-totreat analysis. This could have been avoided by keeping the sealed envelopes in theatre, to be opened consecutively upon the actual performance of hysterectomy. These points and their significance should have been highlighted in the discussion. The indication for hysterectomy should have been compared between the two groups. The number of patients having additional surgery, including pelvic floor repair, and the nature of such surgery, should have been provided and compared between the two groups. Some of the numerical data were described using the mean, without the standard deviation, alongside the median, without the interquartile range, and all were compared using a Student’s t-test, without an indication of the normality of their distribution. Parity, postoperative fall in haemoglobin and hospital stay are unlikely to fit a normal distribution, and should be compared using a non-parametric test. There is little point in providing the mean postoperative temperature of patients, assuming it was normally distributed in every patient, when the mere occurrence of postoperative febrile morbidity was chosen as the primary outcome measure. The number of women returned to theatre should have been compared between the two groups. All these points need to be looked at for a better analysis of the trial findings. j

Re: Prevalence of obstetric fistula: a population-based study in rural Pakistan.

1 Miyake Y, Tanaka K, Okubo H, Sasaki S, Arakawa M. Intake of dairy products and calcium and prevalence of depressive symptoms during pregnancy in Japan: a cross-sectional study. BJOG 2014;122: 336–43. 2 Desbonnet L, Garrett L, Clarke G, Kiely B, Cryan JF, Dinan TG. Effects of the probiotic Bifidobacterium infantis in the maternal separation model of depression. Neuroscience 2010;170:1179–88. 3 Foster JA, McVey Neufeld KA. Gut–brain axis: how the microbiome influences anxiety and depression. Trends Neurosci 2013;36: 305–12.

A comparison of the long‐term consequences of vaginal delivery versus caesarean section on the prevalence, severity and bothersomeness of urinary incontinence subtypes

Sir, I read with interest the article by Gyhagen et al. comparing the long-term impact of vaginal and caesarean delivery on the prevalence, severity and bothersomeness of urinary incontinence subtypes. A number of factors that could affect pelvic floor function were not included in the analysis; such as lifestyle, including smoking, fluid intake and exercise, as well as the intake of medications. Continence and prolapse surgery after the only delivery for the included cohort of mothers not only signify pelvic floor damage but may also cause further pelvic floor dysfunction. Some factors were not included in the analysis, despite being included in the questionnaire, like hysterectomy, which can affect pelvic floor function regardless of indication. Although the symptoms of pelvic organ prolapse were assessed using a nonvalidated questionnaire, pelvic organ prolapse may not necessarily be symptomatic despite being significant. The lack of a vaginal examination means, at least in theory, that pelvic organ prolapse may have been underestimated. About half of the original cohort was included and the number of women having a vaginal delivery was five times the number of those having a caesarean delivery. No power calculation was carried out. No distinction was made between all types of vaginal deliveries as well as caesarean deliveries. However, instrumental deliveries, especially forceps, carry more risk of pelvic floor damage than straightforward vaginal deliveries. Likewise, an emergency caesarean section in labour, especially during the second stage, can be associated with more pelvic floor damage than an elective section before the onset of labour. With these points about the estimation of impact taken into account, it is notable that the distribution of different types of incontinence was found to be the same 20 years following vaginal and caesarean deliveries. In addition, there was no significant difference in the degree of bother from urinary symptoms between the two groups. Furthermore, there was no significant difference in having any treatment for stress incontinence in the 20 years that elapsed since delivery, even when more women sought medical help following vaginal than caesarean delivery. This may indicate that the effect of caesarean delivery is basically the same as that of vaginal delivery, but to a lesser degree. It would be interesting to explore the actual mechanism of pelvic floor damage with caesarean delivery, especially elective caesarean.&

Impact of mode of delivery on levator morphology

region. Using effective contraception is preferable to termination of pregnancy, but more detailed studies are required to identify the real causes behind the observed regional trends. Furthermore, additional East–West comparisons could provide further information on how best to replace pregnancy terminations with contraception and how to reduce the use of unsafe termination methods. Data and statistics on terminations of pregnancy over time should be an essential part of the European Public Health Monitoring and Reporting System. j

Comparing clean intermittent catheterisation and transurethral indwelling catheterisation for incomplete voiding after vaginal prolapse surgery.

Sir, I read with interest the reports on the randomised trial that compared clean intermittent catheterisation with transurethral indwelling catheterisation for abnormal post-void bladder volume after vaginal prolapse surgery as well as the subsequent paper on women’s views about both options for the same condition. The first study described women having clean intermittent catheterisation by nurses while they remained on the ward, whereas clean intermittent self-catheterisation on an outpatient basis is now the established practice. The study relied on first attempt to void, whereas it is common to wait for a second void before making the diagnosis of high post-void bladder volume. The study defined abnormal post-void bladder volume as 150 ml, without reference to voided volume. Several studies have recommended a postvoid bladder volume of 100 ml following a voiding volume of over 200 ml. The significance of these points should be highlighted in the discussion. The first study included women with anterior middle as well as posterior compartment prolapse, without comparing the value of clean intermittent catheterisation for each of the three compartments. It might have been better to restrict the study to women with anterior compartment prolapse surgery, because this is the form that is likely to lead to voiding dysfunction. It would have been most likely helpful to compare the results in women undergoing prolapse surgery with and without mesh, to explore the possible impact of using mesh on voiding function. The study compared post-void bladder volume below and above 300 ml, without explicitly indicating that this applied to women having clean intermittent catheterisation and this comparison was not described in the methods nor was it stated as one of the study objectives. The second study could have followed the preferences and views expressed in it into action, in the form of shaping clinical protocols and drafting patient information leaflets. It is customary to provide the mean and standard deviation for normally distributed numerical data and for numerical data that do not fit a normal distribution, the median and interquartile range are provided. These values were not provided in either study. It is highly unlikely that the distribution of points Ba, Bp and C on the pelvic organ prolapse quantification scheme (POP-Q) fitted a normal distribution in a cohort with pelvic organ prolapse to be described by providing the mean, without including the standard deviation. The background features of those preferring clean intermittent catheterisation and those preferring transurethral indwelling catheterisation should have been compared. The tests used for comparison should be stated with the P-value for each item of data provided, which was not followed in both papers. In fact, the Mann– Whitney U test was used to compare categorical data that could have been compared using the chi-square test and the Fisher’s exact test was used to compare numerical data, which is impossible, in the first paper. If this was a typing error, it deserves an erratum, otherwise it casts doubt on the analysis of the study and its results.

Prolapse surgery with or without stress incontinence for pelvic organ prolapse

this may well differ from practice in Italy. With regards to our study outcomes, we chose incorrect diagnosis of the fetal head position as the primary outcome and powered the trial accordingly. We provided robust evidence that an ultrasound scan prior to instrumental delivery reduces the incidence of incorrect diagnosis, and furthermore that this does not delay delivery. We did not provide evidence that this approach reduces maternal or neonatal morbidity. We fully acknowledge that the trial was not powered for secondary outcome measures (maternal, neonatal and procedure-related complications), and that very large studies including several thousand women may have sufficient power to determine whether ultrasound assessment of fetal head position reduces morbidity. From the clinical perspective, however, we know that there are many factors that contribute to maternal and neonatal morbidity, and that the correct assessment of fetal head position is only one factor along the causal pathway. We stand by our statement that ‘a more integrated clinical skills-based approach is likely to be required to prevent adverse outcomes at instrumental delivery’. Rather than calling for the replication of our study, we recommended that attention be focused on stategies that not only enhance the accuracy of fetal assessment prior to instrumental delivery but also enhance the obstetrician’s ability to deal with a correctly identified fetal malposition. The definition of the primary outcome required a great deal of thought. We would emphasise that in both arms, the assessment (clinical examination alone or clinical examination and ultrasound scan) was carried out immediately prior to application of the instrument. Therefore, there was no time between assessment and delivery for the fetal head to rotate spontaneouly, unless this was performed intentionally with the chosen instrument and/or by manual rotation, both of which were documented and reported. The instrument markings were very informative, but it was sometimes challenging to differentiate between suboptimal instrument placement and incorrect diagnosis, particularly at vacuum delivery. We have performed further secondary analyses on the data set that we hope to publish in the near future. We very much look forward to reading the results of the RISPOSTA trial and are delighted that other groups are working in this challenging area of clinical practice.&

Continuous infusion of local anaesthetic following laparoscopic hysterectomy: a randomised controlled trial.

rate was 33% (n = 981). Of the 116 women in the audit, 91 (78%) had planned their subsequent pregnancy, 80 (69%) had a previous CS as an emergency in labour, and 81 (70%) had their booking visit with a consultant. The success rate of VBAC was documented in 41 cases (35%) and the risk of uterine rupture in 64 cases (55%) (Table 1). The risks of ERCS were documented in two cases (2%). A patient information leaflet regarding VBAC was given in four of the cases (3%). At booking, 48 women (41%) wanted a VBAC, 19 (16%) wanted a CS even if spontanous onset of labour (SOL) occurred, nine (8%) wanted ERCS unless there was SOL, and ten (9%) wished to decide at a later date. There was no preference documented in 30 cases (26%). Mode of delivery was reviewed at 36–37 weeks in 65 cases (56%), of whom 30 (46%) wished for a VBAC and 35 (54%) for an ERCS. Overall, 30 women had a VBAC (26%) and 86 (74%) had a CS. Of the 48 women who hoped for a VBAC at the booking visit, 19 (40%) succeeded. This snapshot of mode of delivery following a previous CS reports outcomes similar to the national study of Knight et al. However, the data suggest that there may be important deficiencies in the information women receive, and in the documentation of counselling within the routine clinical setting. If women are to make well informed shared decisions about mode of delivery, greater attention needs to be paid to the skill of antenatal counselling and the accurate recording of discussions and management plans. This has important implications for clinical outcomes, birth experiences and the risk of litigation.&

Microwave endometrial ablation versus thermal balloon endometrial ablation (MEATBall): 5-year follow up of a randomised controlled trial

and delivery; 3 The notification of the polymerase chain reaction (PCR) result and the administration of antibiotics. These details, plus any others that would help to establish the obstacles that there were to PCR testing, and using the results to inform the timely administration of intrapartum antibiotic prophylaxis against early-onset group B streptococcus infection, would be very interesting.&

Pelvic organ prolapse after subtotal and total hysterectomy: a long‐term follow‐up of an open randomised controlled multicentre study

We were pleased to learn that our paper on the rate of subsequent surgery following endometrial ablation had been discussed at Drs Vo and Banerjee’s Journal Club, and that its results will be used to help counsel women with heavy menstrual bleeding. In our study, the Kaplan–Meier estimates for the rates of subsequent surgery at 1, 2 and 5 years for women aged under 35 years were 11.1% [95% confidence interval (CI), 10.5–11.6%], 18.5% (95% CI, 17.8–19.2%) and 26.3% (95% CI, 25.5–27.2%), respectively. For women aged over 35 years, the rates were 6.7% (95% CI, 6.6–6.9%), 10.8% (95% CI, 10.6–11.0%) and 14.9% (95% CI, 14.7–15.2%), respectively. The rates of subsequent surgery at different years for women with or without fibroids (as a secondary diagnosis only, unadjusted) were comparable with the overall population rates, whose figures were reported in the original article. As only a small proportion of women were under 35 years of age (10.7%) or had fibroids (3.7%), the subgroup of women with both risk factors was too small to estimate the rate of repeat surgery robustly. Rates of subsequent surgery at 1 and 2 years following radiofrequency ablation were 5.0% (95% CI, 4.6–5.4%) and 8.1% (95% CI, 7.6–8.6%), respectively. These estimates were lower than the rates followingmicrowave or balloon ablation. The 1-, 2and 5-year rates after microwave ablation were 8.2% (95% CI, 7.8– 8.7%), 13.1% (95% CI, 12.5–13.6%) and 17.6% (95% CI, 16.9–18.2%), respectively, which were similar to the rates after balloon ablation: 8.0% (95% CI, 7.6–8.3%), 12.4% (95% CI, 12.0–12.9%) and 16.9% (95% CI, 16.4–17.5%), respectively. We were unable to calculate the 5-year Kaplan–Meier estimates for radiofrequency ablation because of an insufficient follow-up period.&