Inflammatory arthritis and pregnancy

Abstract: The aim of this thesis was to add to the existing knowledge of how inflammatory arthritis disease affects pregnancy outcomes. In particular, the aim was to describe and study the effects of a diagnosis of juvenile idiopathic arthritis (JIA) or psoriatic arthritis (PsA) on pregnancy outcomes, to further study the effect of disease severity in PsA on pregnancy outcomes and lastly to study the association of disease activity and preterm birth in pregnancies with PsA. To address these aims we used information from Swedish population based registers such as the Medical Birth Register (MBR), the National Patient Register (NPR), the Prescribed Drug Register (PDR) and the nationwide clinical quality register, Swedish Rheumatology quality register (SRQ). In study I-III, we used regression models to estimate odds ratios (ORs) with 95% confidence intervals. In study I, we assessed pregnancy outcomes in a nationwide population based cohort of singleton pregnancies between 1992 and 2011. We identified 1807 pregnancies to women with a diagnosis of JIA and 1 949 202 comparator pregnancies to women without JIA. Due to the fact that JIA is a heterogenic condition, which may or may not persist into adulthood, we stratified the pregnancies with JIA into two groups: JIA paediatric only (n=1169), where the diagnosis was confined to childhood and/or adolescence, and JIA persisting into adulthood (n=638). In conclusion, we observed increased risks of preterm birth and caesarean delivery for both JIA groups compared to comparator pregnancies. In addition, we observed increased risks of pre-eclampsia and small for gestational age (SGA) birth in pregnancies with JIA persisting into adulthood. In study II, we studied pregnancy outcomes in a cohort of 41 485 singleton pregnancies from 1997 to 2014 assembled by linkage of national registers. We identified 541 first and subsequent singleton pregnancies with PsA exposure and 40 944 pregnancies were identified as unexposed during the same time period. Pregnancies to women with PsA had increased risks (estimated with aORs) of preterm birth and caesarean delivery compared to unexposed pregnancies. In study III, the focus was to assess how disease severity of PsA (by use of a proxy, i.e. antirheumatic treatment) affected pregnancy outcomes in comparison with non-PsA pregnancies. We defined the study period from 2007 to 2017 to be able to use information from the prescribed drug register. The main study cohort consisted of 921 PsA pregnancies and 9210 matched (on parity, maternal age and year of birth) non-PsA pregnancies. In the main analysis there were increased risks of preterm birth and caesarean delivery in PsA as compared to non-PSA pregnancies. We thereafter stratified the cohort of PsA pregnancies based on presence, timing and type of antirheumatic treatment. The risks of preterm birth and caesarean delivery differed among these groups, with the most increased risks among PsA pregnancies (vs. non-PsA) with antirheumatic treatment during pregnancy. Risk of preterm birth was influenced by parity and mainly increased in first pregnancies. In study IV, we used information from SRQ and MBR to assemble a cohort of 211 pregnancies from 2007 to 2017. All identified pregnancies had a diagnosis of PsA assigned by the treating rheumatologist. We aimed to assess the association of disease activity, in the time period from one year before pregnancy until delivery, and preterm birth. Disease activity was assessed with registered values of DAS28CRP and values of health assessment questionnaire (HAQ). We dichotomised the exposure into moderate/high disease activity and low disease activity. The proportion of preterm birth was higher in pregnancies with any registered moderate/high disease activity in the year before and/or during pregnancy (vs. not). We found a numerical but not statistically significant difference between the proportions of preterm birth among PsA pregnancies with vs. without active disease during pregnancy.