Abstract Study question What is the prevalence of antinuclear antibodies (ANAs) in women with recurrent early pregnancy loss (REPL) and do they affect subsequent pregnancy outcome? Summary answer The prevalence of ANAs in women with REPL was 13.0%, and their presence did not negatively impact subsequent pregnancy outcomes, including EPL and livebirth rates. What is known already An etiological factor cannot be identified in more than half of the couples experiencing REPL. Immunologic mechanisms could be involved, with the presence of autoimmune disease (AD) associated to lower reproductive outcomes. For the mere presence of ANAs, in the absence of AD, the evidence on the association with REPL is controversial, with most studies having a small sample size with important clinical and methodological heterogeneity. Study design, size, duration This is a single-centre, retrospective cohort study from a tertiary referral hospital including 960 women referred for REPL between January 2014 and December 2019. During the entire study period, maternal serum ANA detection was part of the routine diagnostic REPL work-up, defined as two or more pregnancy losses before the 10th week of gestation, including previous spontaneous conceptions and pregnancies following autologous IVF/ICSI treatment. Participants/materials, setting, methods Women under the age of 40 at ANA analysis were eligible for inclusion. Confirmed AD, uterine anomalies, parental chromosomal abnormalities and uncontrolled endocrine disorders were exclusion criteria. The prevalence of positive ANA titers and the impact of ANA positivity on the outcome of a subsequent pregnancy were investigated with EPL rate as the main outcome parameter and biochemical pregnancy rate, clinical miscarriage rate and live birth rate as secondary outcome measures. Main results and the role of chance Positive ANA screening at a titer of ≥ 1/80 was observed in 125/940 women (13.0%), a prevalence in line with previously published data in the general population not experiencing REPL. There were no statistically significant differences in maternal age, body mass index, ethnicity, or conception method between ANA-positive and -negative women. No higher number of previous EPLs was observed in ANA-positive compared to ANA-negative women (respectively 2.7 versus 2.8, p = 0.06). Women with ANAs had significantly more thyroid antibodies (21.6% versus 10.2%, p < 0.001). For a subsequent pregnancy, no association was observed between ANA positivity and reproductive outcome with an unadjusted EPL rate of 30/86 (34.9%, ANA-positive) versus 206/509 (40.5%, ANA-negative, p = 0.32). Logistic regression analysis did not identify ANAs as an independent predictive risk factor for additional EPL (OR 0.73, 95%CI 0.45-1.20, p = 0.22). Similarly, there was no significant difference in live birth rates between both groups (p = 0.5), with 54/86 (62.8%) in ANA-positive and 300/509 (58.9%) in ANA-negative patients. Limitations, reasons for caution The current data are limited by the absence of a formal control group without REPL, the retrospective nature of the study and potential bias due to unmeasured confounders. Wider implications of the findings These findings further questions systematic ANA measurement in the diagnostic work-up of REPL without AD. Furthermore, when ANAs are detected, patient counselling and treatment adaptations should be considered with caution, given our inability to eventually detect any significant differences in the pregnancy outcomes between ANA-positive and -negative REPL patients. Trial registration number not applicable