Traction force and long-term outcome in children born after vacuum-assisted delivery

Abstract: Introduction: Vacuum-assisted delivery (VAD) is used in cases of dystocia or foetal distress in the second stage of labour. The frequency of VADs differs around the world, and in Sweden they account for six to nine percent of all deliveries. Due to several procedure-related aspects, there are short-term complications associated with this delivery mode. Traction force is one of these aspects and a high traction force is associated with worse perinatal outcome. To analyse this further, a digital handle with the capacity of measuring and registering the traction force used was developed. This thesis aimed to determine if the use of this digital handle altered the perinatal outcome, as well as to analyse the traction force profiles of children with severe neonatal outcomes. Though the short-term complications are well studied, little is known about the long-term complications. The last aim of this thesis was therefore to study the long-term outcome in children born with a VAD. Material and Methods: We included women delivered at Karolinska University Hospital’s two delivery wards, Huddinge and Solna. General inclusion criteria were women at term with a singleton pregnancy delivered with a low- or mid-station VAD. In study III, a control group with spontaneous vaginal delivery (SVD) and emergency caesarean delivery at second stage of labour (ECD) were included. The three studies had different study designs: a randomised controlled trial (RCT) (I), an observational matched case-control study (II), and a prospective observational cross-sectional cohort study (III). Severe perinatal outcome was defined as intracranial haemorrhage, hypoxic ischaemic encephalopathy (1-3), subgaleal haematoma, seizure or death. The digital extraction handle VIH-3 was compared to the conventional metal handle in study I. The traction force profiles generated from this digital handle were analysed in study II using the MATLAB® program. The two main variables were total traction force (force over time in Newton-min [Nmin]) and peak force (maximum force used in Newton [N]). The Five-to-Fifteen questionnaire was used as a screening method for neurodevelopmental problems in study III. Scoring ≥90th percentile meant an obvious difficulty in the analysed domain. All clinical data was extracted from the electronic medical records, from the National Board of Health and Welfare and from Statistics Sweden. The statistical analyses were chosen depending on the study design used. Results: The incidence of severe perinatal outcome in study I was less than expected (2.7%) and the use of a digital handle did not decrease this incidence in comparison with the conventional handle. The incidence of severe perinatal outcomes differed between the two delivery wards: 0.9% vs 4.0% (p-value <0.05). Children with severe perinatal outcomes had significantly higher total traction forces, more cup detachments, more pulls, and longer duration of the VAD procedure than the healthy children. The prevalence of severe perinatal outcome in study II was 2.3% (n =13). The cases were matched individually to three controls (n =39) based on five matching variables. Analysing the traction force profiles between the cases and controls using conditional logistic regression showed an increased odds of severe perinatal outcome in total traction force for each Nmin used (OR 1.004, 95% CI 1.001-1.007), and an increased odds in peak force for each N used (OR 1.02, 95% CI 1.004-1.04). The analysis of a possible safety limit for total traction force showed an 86% reduction in risk of severe perinatal outcome in this cohort if the total traction force was kept <343Nmin (aOR 0.14, 95% CI 0.04-0.5). In study III, the response rate in the questionnaire was 58% in the VAD group, 29% in the SVD group and 50% in the ECD group. There was an increased odds of obvious difficulties in motor skills (OR 2.2, 95% CI 1.3-3.8) and perception (OR 1.7, 95% CI 1.002-2.9) in children delivered with VAD compared to SVD. Similar results were seen in the ECD group compared to the SVD group for motor skills (OR 3.3, 95% CI 1.8-6.4) and perception (OR 2.3, 95% CI 1.2-4.4). Furthermore, there was an increased odds of having neurodevelopmental problems in two or more domains when delivered with a VAD or an ECD vs a SVD. According to a multivariate regression analysis, the problems with motor skills in the VAD and in the ECD group vs the SVD group were still significant after adjusting for predefined confounders (OR 2.5, 95% CI 1.2-5.3 and OR 3.3, 95% CI 1.4-7.9 respectively). The analysis of the ICD-10 diagnoses extracted from the National Board of Health and Welfare, showed significantly more children delivered with a VAD with an attention deficit/hyperactivity disorder compared to the other two delivery modes. Conclusion: Based on the fact that traction force impacts perinatal outcome, this thesis presents two of the few studies done where this force has been quantified. Even though the sample size calculation in study I was insufficient and a reduction in severe perinatal outcome was not seen, the use of a digital extraction handle provides objective documentation. This helps keep track of the traction force used and register the traction force profiles. In study II, these profiles showed that children with severe perinatal outcome were delivered with significantly higher forces. Furthermore, the odds of having a severe outcome increased with every increase in Nmin and Newton used during the extraction. With the help of these profiles, a total traction force limit of 343Nmin was suggested to decrease the incidence of severe perinatal outcome by 86%. To provide validity, this limit needs to be tested in prospective studies. Study III presents novel data on the long-term outcome in children delivered with VAD. It is worrying that low- and mid-station VAD compared to SVD may be associated with long-term problems in motor skills and perception. It is also of concern that similar results were seen in the ECD group. To confirm these results, further long-term studies are needed to provide the most evidence-based recommendations possible.