P-440. Shortened steroid regimen in people living with HIV/AIDS with associated moderate and severe <i>P. jirovecii</i> pneumonia

Abstract Background The 21-day steroid regimen in people with P. jirovecii pneumonia (PCP) showed positive clinical outcomes reducing mortality and immune reconstitution inflammatory syndrome (IRIS) incidence. However, steroids can perpetuate an immunocompromised state, having a negative impact on morbidity. Respiratory Function test in PLWHIV with shortened vs conventional steroid regimen for PCP Methods This is an open-label, randomized non-inferiority controlled trial in adults living with HIV, to elucidate the effect of a shortened steroid regimen (sSR) of 8 days for moderate and 14 days for severe PCP compared to conventional 21-day steroid regimen (cSR) by mortality, IRIS incidence, and respiratory function. Herpesviruses and HIV viral loads, CD4 T cells (CD4), and respiratory function tests were performed at admission and at day 90. Results A total of 44 participants were randomized, 23 in sSR and 21 in cSR. Mechanical ventilation (MV) was equal between groups (13 participants each group (OR = 0.8, 95%CI = 0.24-2.67). HIV viral loads at admission and at day 90 were similar between groups (p = 0.34 and p = 0.33 respectively). Similarly, CD4 did not show differences at admission and day 90 (p = 0.82 and p = 0.73). Respiratory function tests were similar between both groups (Figure attached). The CMV, EBV, and HHV-8 viral loads did not show statistical differences. A total of 3 participants presented IRIS in sSR vs 2 in cSR (OR = 1.5, 95%CI = 0.23-9.96) and mortality was not statistically different (OR = 3, 95%CI = 0.29-31.35). The sSR group had a shorter hospital stay (12 vs 18 days; p = 0.37). Conclusion Preliminary analysis shows that reducing steroid duration may be safe, since a sSR achieved non-inferiority compared with a cSR, by presenting similarities in mortality, IRIS incidence, hospital-stay duration, MV requirements, herpesviruses and HIV viral loads, and in respiratory function tests; suggesting that at 90 days follow-up, the use of a sSR is not inferior compared to cSR in this preliminary analysis. Disclosures All Authors: No reported disclosures