BackgroundAdults aged 80 years or older frequently require peripherally inserted central catheter (PICC) placement for prolonged intravenous therapy, nutritional support, and difficult vascular access. In this oldest-old population, PICC-related bloodstream infection (PICC-CRBSI) risk may be shaped by catheter factors, frailty, nutritional depletion, immune-inflammatory imbalance, and competing events such as death or non-infectious catheter removal. This study developed and internally validated a competing-risk model for 60-day definite PICC-CRBSI in patients aged ≥80 years and separately explored the antibiofilm and anti-inflammatory activity of berberine in vitro.MethodsConsecutive patients aged ≥80 years who underwent PICC placement at a single tertiary teaching hospital between January 2019 and December 2025 were included in a retrospective cohort. The prediction landmark was 24 h after PICC insertion. Definite PICC-CRBSI within 60 days was analyzed with death and non-infectious PICC removal treated as competing events. Clinical, biomarker, and combined models were developed using L2-penalized Fine–Gray regression and internally validated with 1,000 patient-level bootstrap resamples. After model locking, a separate exploratory substudy evaluated berberine effects on Staphylococcus epidermidis biofilms formed on PICC material and on macrophage inflammatory responses induced by risk-stratified donor plasma and sterile biofilm-conditioned medium.ResultsA total of 386 patients were analyzed, and 46 developed definite PICC-CRBSI over 19,528 catheter-days, corresponding to an incidence density of 2.36 per 1,000 catheter-days. Patients with definite PICC-CRBSI had greater frailty, longer catheter dwell time, lower lymphocyte and albumin levels, and higher systemic immune-inflammation index (SII) and C-reactive protein-to-albumin ratio (CAR) than those without definite infection. The combined model showed an apparent 60-day time-dependent area under the curve (AUC) of 0.758 and a bootstrap-corrected AUC of 0.731, indicating moderate internally validated discrimination. The Aalen–Johansen 60-day cumulative incidence of PICC-CRBSI was 3.2%, 8.5%, and 28.1% in the low-, intermediate-, and high-risk groups, respectively. In vitro, berberine reduced viable S. epidermidis biofilm burden from 7.656 log10 CFU/coupon under vehicle exposure to 6.919 log10 CFU/coupon at one-half of the strain-specific minimum inhibitory concentration. In macrophage experiments, high-risk donor plasma induced higher interleukin-6 release than low-risk donor plasma under vehicle conditions, while 10 μM berberine reduced interleukin-6 by approximately 23% without substantial cytotoxicity.ConclusionA single-center competing-risk model combining catheter-related variables with routine immune-inflammatory biomarkers provided preliminary internal risk stratification for 60-day PICC-CRBSI in patients aged ≥80 years. The berberine substudy suggested antibiofilm and anti-inflammatory activity in vitro, but it did not directly validate the clinical prediction model or establish clinical efficacy. Before the locked prediction equation can be used clinically, independent multicenter validation is required across settings with different PICC products, patient case-mix, catheter-maintenance protocols, blood-culture practices, and local microbiological epidemiology. Further mechanistic studies are also required before berberine-based catheter-infection prevention can be considered.