Niche development after caesarean section: Impact and preventive strategies

Chapter 1 provides an introduction on niches after caesarean section (CS). A niche can be the result of impaired healing of the uterus and may have clinical consequences for subsequent pregnancies, but could also result in menstrual disorders or fertility-related problems. Chapter 2 presents a qualitative study with patients who were treated for niche-related problems. Fifteen themes were reported in four domains (physical, psychological, social and environment). 3/5 prioritized themes were studied in literature very often: abnormal uterine bleeding, subfertility and abdominal pain. However, 2/5 of the prioritized themes were hardly studied in literature: sexuality and (lack of) self-esteem. In chapter 3 we present the results of a study in which we assessed the changes in QoL after a laparoscopic niche resection. After one year, the subdomains bodily pain, mental health and social functioning improved statistically significantly. Women with ‘symptom relief’ as main goal of the intervention benefit more from laparoscopic niche resection in terms of physical health than women with ‘restoring anatomy’ as main goal of the intervention. Chapter 4 is a comment on a systematic review. Chapter 5 presents the results of a systematic review and meta-analysis. We included 20 studies on three variations in uterine closure technique. Double-layer closure and unlocked suturing resulted in a thicker residual myometrium compared to single-layer closure and locked suturing. Inclusion of the decidua resulted in a lower prevalence of large niches. We concluded that double-layer unlocked uterine closure including the decidua could possibly result in better sonographic outcomes. Long-term clinical outcomes were reported very infrequently. Chapter 6 describes the study protocol for the evaluation of the (cost-)effectiveness of double- versus single-layer uterine closure after a first CS: the 2Close study, a multicenter, randomised controlled superiority trial. Women aged over 18 years who underwent a first planned or unplanned CS were assessed for eligibility. We calculated that 2290 women had to be included to show superiority. We randomised women to single-layer unlocked closure with the decidua in- or excluded in the suture, or double-layer unlocked closure including the decidua. Chapter 7 presents the results of the 2Close study. 2292 women were included and randomised to single-layer (n=1144) or double-layer closure (n=1148). The mean number of postmenstrual spotting days was not different at 9 months follow-up. Operative time was slightly shorter, and niche prevalence was slightly lower after single-layer closure. Quality of life and sexual functioning did not differ between groups. Since we observed no difference in prevalence of large niches and residual myometrium thickness, we do not expect long-term outcomes to be different. However, this has to be confirmed in our long-term follow-up after three years. Based on our results, we concluded that none of the techniques should be advocated in clinical guidelines. In chapter 8 the results of the cost-effectiveness analysis are described. Total societal costs and total healthcare costs were not different between groups. Additionally, the cost utility analysis showed that double-layer closure did not improve quality-adjusted life-years (QALY). The probability of the intervention being cost-effective compared with standard care was low. We concluded that there is no reason to advocate double-layer closure in any clinical guideline. Chapter 9 describes the development and internal validation of a prediction model for the development of a niche. We describe univariate analysis of these factors on the outcome ‘niche’ (indentation ≥2 mm). The strongest predictors were gestational age, twin pregnancy, double-layer closure, less surgical experience, smoking (risk factors) and multiparity and vicryl suture material (protective factors). It remains very difficult to predict who will develop a niche and who will not and we have probably not taken into account all possible predictors