P1‐263: IDENTIFYING FACTORS THAT PREDICT LUMBAR PUNCTURE COMPLICATIONS AND PROMOTE PARTICIPATION IN LONGITUDINAL STUDIES REQUIRING SERIAL CSF SAMPLING

Cerebrospinal fluid (CSF) biomarkers are increasingly used to determine the cause of dementia, define the biochemical changes that influence rates of dementia progression, and identify cognitively normal research participants at higher risk of developing dementia. There is a need to understand the factors that promote participation in longitudinal studies of memory and aging that require serial CSF sampling. Participant- (i.e., age, gender, race, CDR, APOE genotype, medical history, BMI, mean arterial pressure) and procedure-specific factors (i.e., needle type, position, operator experience) associated with lumbar puncture (LP) complications were determined in participants engaged in longitudinal studies of memory and aging at the Knight Alzheimer Disease Research Center (Washington University School of Medicine; Saint Louis, Missouri) from 2004 to 2018. The relationship between participant-specific factors and rates of completion of ≥3 LPs were evaluated using stepwise forward logistic regression in individuals enrolled for similar amounts of time, controlling for gender and years of education. 929 participants completed 1484 LPs. Complications were reported following 331 (23.1%) LPs—most commonly headache (11.3%) or self-resolving back pain (9.5%). Headache was severe in 38 (2.8%) participants, requiring blood patch. Complications were more likely in younger (65.8±8.9 vs 69.2±8.9 years, p<0.001), female (RR 1.5, 95%CI: 1.2-1.9), cognitively normal (RR 2.1, 95%CI: 1.4, 2.5) participants, and those with a prior history of LP complications (RR 1.7, 95%CI: 1.3-2.2). Lower body mass index (23.5±3.3 vs 27.9±5.5 kg/m; p<0.001) and fewer years of operator experience (3.4±3.7 vs 6.3±6.9; p=0.01) were associated with greater numbers of severe headaches. 148 of 432 eligible participants (34%) completed ≥3 serial LPs. Younger age (OR 2.1, 95%CI: 1.6−2.6 per decade), normal cognition (OR 25.1, 95%CI: 3.4-187.0), and the absence of heart disease (OR 2.4, 95%CI: 1.2-4.6) or prior LP complications (OR 2.1, 95%CI: 1.2-3.7) were associated with increased odds of completing serial LPs. LP complications requiring medical intervention were rare in our cohort. Knowledge of the factors that predict complications and influence longitudinal participation within studies requiring serial CSF sampling may be leveraged to personalize risk assessments and improve recruitment and retention.