CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

698 WIDESPREAD PAIN AND ASSOCIATIONS WITH HIV-RELATED FACTORS IN PEOPLE WITH HIV Caroline Sabin 1 , Kennedy Nkhoma 2 , Richard Harding 2 , Adam Geressu 1 , Emmanouil Bagkeris 1 , Patrick W. Mallon 3 , Frank Post 4 , Memory Sachikonye 5 , Marta Boffito 6 , Jane Anderson 7 , Jaime Vera-Rojas 8 , Margaret Johnson 9 , Ian Williams 1 , Alan Winston 10 , for the POPPY Study 1 University College London, London, UK, 2 Kings College London, London, UK, 3 University College Dublin, Dublin, Ireland, 4 King’s College Hospital, London, UK, 5 UK Community Advisory Board, London, UK, 6 Chelsea and Westminster Hospital, London, UK, 7 Homerton University Hospital NHS Trust, London, UK, 8 Brighton and Sussex Medical School, Brighton, UK, 9 Royal Free Hospital, London, UK, 10 Imperial College London, London, UK Background: Widespread and burdensome pain is frequently reported by PWH, although associations with HIV factors, particularly in those on current antiretroviral (ART) regimens, have not been determined. We investigated the prevalence of widespread pain and its associations with HIV factors among PWH in the POPPY study. Methods: PWH on ART were included. Self-reported pain information was collected from 2013-2015 via self-completed questionnaires and through a pain mannikin identifying affected body sites (19 distinct sites). Associations between extent of pain (widespread [>6 affected sites], non-widespread [1-6 sites], none) and HIV factors (current/nadir CD4, total ART drugs received, current/cumulative exposure to each ART class, and previous exposure to stavudine, didanosine or zalcitabine (‘D’ drugs, associated with neuropathy)) were investigated using ordinal logistic regression adjusted for age and gender. Results: The 522 PWH were mainly male (86.0%), white (87.7%) with median (interquartile range [IQR]) age 53 (47-59) years. Median (IQR) exposure to NRTIs, PIs and NNRTIs was 8.5 (4.4-14.1), 1.9 (0-7.6) and 3.5 (0.1-8.9) years, respectively with 83.5%, 43.5% and 46.6% currently receiving each class; 14.4%/10.5% had ever/were currently receiving an INSTI. PWH had received 6 (4-9) ART drugs and 169 (32.4%) had received a d-drug. Median current/nadir CD4 counts were 620 (472-800) and 210 (100-300) cells/mm3. Pain was reported by 341 (65.4%), with most commonly affected sites being the lower (31.0%) and mid (22.4%) back, knees (33.1%), ankle/foot (26.3%), shoulders (23.0%) and neck (14.9%). The median (range) number of sites causing pain was 2 (0-17); 74 (14.2%) and 267 (51.2%) reported widespread and non-widespread pain, respectively. Widespread pain was more common in those with longer exposure to NRTIs, longer exposure to PIs, those currently receiving NNRTIs, those exposed to a greater number of ART drugs, those previously exposed to D-drugs and those with a higher current CD4 count (Table), with only exposure to D-drugs remaining associated with widespread pain after adjusting for other factors (aOR 2.09). Conclusion: Widespread pain reported in PWH is commoner in those with prior exposure to D-drugs, likely representing a legacy of prior ART-induced neuropathy. Although we found no other associations with any of the studied HIV-related factors in PWH on virally-suppressive ART, further analyses will investigate drug and immunosuppression associations in more depth.

699 MOOD DISORDERS & INCREASED RISK OF NONCOMMUNICABLE DISEASES IN ADULTS AGING WITH HIV Jessica L. Castilho , Peter F. Rebeiro, Bryan E. Shepherd, Cathy Jenkins, Robertson Nash, Rodney S. Adams, Megan M. Turner, Sally S. Furukawa, Todd Hulgan, John R. Koethe, Timothy R. Sterling Vanderbilt University, Nashville, TN, USA Background: Mood disorders of major depression and bipolar affective disorder have been associated with systemic inflammation and non- communicable disease (NCD) risk. They are also prevalent among persons living with HIV (PLWH), though whether they are associated with NCDs in PLWH has not been well described. Methods: PLWH attending the Vanderbilt Comprehensive Care Clinic from 1998-2015 and ≥1 year of follow-up contributed data. Mood disorder exposure was based on diagnoses one year after clinic entry date (baseline) to mitigate imprecise onset date. NCDs were: coronary artery disease, chronic kidney disease, cerebrovascular disease, diabetes, dementia, hepatic disease, hyperlipidemia, hypertension, obesity, peripheral vascular disease, and non-AIDS-defining cancers. Multimorbidity was the accumulation of ≥2 NCDs. Metabolic syndrome was ≥3 of hypertension, hyperlipidemia, diabetes, or obesity. Multivariable competing risk (death) models yielded cumulative incidences and subhazard ratios (sHR) of incident NCDs and multimorbidity. Cox regression yielded hazard ratios (HR) for mortality after multimorbidity. Adjusted models included sex, race, prevalent NCD, hepatitis C status, substance use, tobacco use, alcohol use, and time-updated CD4 cell count (CD4), CD4/CD8 ratio, and HIV RNA. Age was the time metric for all models. Results: Of 4,140 adults, 999 (24%) had a mood disorder. Mood disorder patients were significantly older at baseline (40 vs. 39 years) and more likely to be female (27 vs. 22%), white (68 vs. 48%), have a history of injection drug use (12 vs. 10%), and have any tobacco use ever (62 vs. 47%). Baseline CD4 and CD4/ CD8 ratio were similar; mood disorder patients were less likely to have HIV RNA <400 copies/mL (57 vs. 61%) and more likely to have ≥1 NCD (57 vs. 48%). A higher proportion of mood disorder patients died (15 vs. 13%). Mood disorders were associated with incident NCDs and multimorbidity in models (Figure). Mood disorders were also significantly associated with metabolic syndrome, which persisted even after adjusting for psychiatric medication use. Increased mortality risk after ≥2 NCDs by mood disorder status was not significant (adjusted HR=1.11, 95% confidence interval: 0.78-1.59). Conclusion: PLWH with mood disorders are at increased risk of incident NCDs and multimorbidity, particularly metabolic syndromes. Focused prevention and treatment of NCDs in PLWH with mood disorders may reduce the burden of multimorbidity in this high-risk group.

Poster Abstracts

CROI 2019 268