The use and safety of peripheral intravenous catheters (PIV) vs central venous catheters (CVC) for lymphoma patients receiving frontline anthracycline-based chemotherapy: A large multicenter real-world analysis (RWA)

Abstract Background: Given a potential risk of extravasation with vesicants and irritants, central venous catheters (CVCs), such as port-a-cath or peripherally inserted central catheters (PICC), are commonly used for the administration of anthracycline-based chemotherapy. However, CVC placement can result in treatment delay, deep vein thrombosis (DVT), catheter malfunction, and added burdens (eg, financial, travel, etc), partly due to the need for regular flushing. Peripheral intravenous (PIV) access is often utilized in the academic setting with select frontline chemotherapy regimens. However, there is limited data reporting on the safety of PIV in this setting. This large, multicenter RWA evaluated the use and associated complications of PIV and CVC for administering CHOP- or ABVD-based chemotherapy in patients (pts) with lymphoma. Methods: We conducted a retrospective RWA across 8 U.S. centers among pts aged ≥18 years with non-Hodgkin lymphoma and Hodgkin lymphoma who were consecutively treated with ABVD or CHOP-based anthracycline chemotherapy via PIV or CVC access between 2010-2024. Choice of catheter placement was at the discretion of the pt's treating physician. Detailed baseline and safety data were collected. The primary outcome was extravasation; secondary outcomes included infiltration, DVT, cellulitis, erythema, and pain. Standard definitions of extravasation and other complications were used (Jackson-Rose J. CJON 2017). Rates of complications were compared via Fischer exact testing. Results: A total of 1,414 pts were included in the analysis: 790 in each arm (ie, PIV vs CVC). Of these, 624 were unique to each group, while 166 pts received chemotherapy via both modalities (PIV and CVC during the treatment course). Overall, 78.9% of pts completed all treatment as planned via PIV. Median age was 63 years (IQR 45–75) in the PIV group and 61 years (IQR 40–73) in the CVC group. Male sex comprised 57% vs 51%, ECOG 0–1 performance status was 83% vs 76%, and advanced-stage disease (stage III/IV) was present in 51% vs 62%, respectively. Lymphoma subtypes were comparable: DLBCL/high-grade/B-cell NOS (54.5% vs 56.1%), Hodgkin lymphoma (27.2% vs 27.5%), T-cell lymphoma (12.4% vs 6.6%), and indolent/mantle cell lymphoma (5.6% vs 6.7%). Among those receiving chemotherapy via CVC, 78.5% were treated through a port-a-cath and 21.5% via PICC. CHOP-based and ABVD-based regimens were administered to 68% and 32% of PIV pts, and 69.6% and 30.4% of CVC pts, respectively. A total of 4,084 (PIV) vs 3,859 (CVC) treatment cycles were delivered, corresponding to 4,347 (PIV) vs 4,884 (CVC) anthracycline infusion days. There was 1 documented case of extravasation in the CVC group and 2 suspected cases in the PIV group (p=0.9); for PIV, this translates to a per pt risk of 0.3% (or 0.05% of infusion days). The two PIV cases involved elderly women with DLBCL receiving CHOP. Both developed localized symptoms post-infusion and were treated conservatively (DMSO/warm compresses or dexrazoxane/ice), with full symptom resolution within 1 week and 1 day, respectively. No tissue or vascular injury occurred. Additionally, there were no significant differences between the PIV and CVC groups in the rates of induration/swelling (1.8% vs 2.0%, p=0.85), phlebitis/cellulitis (2.4% vs 2.4%, p=0.9), venous thromboembolism (VTE; 1.4% vs 2.0%, p=0.32), or erythema (3.8% vs 5.1%, p=0.27). Pain or tenderness at the infusion site (8.5% vs 2.9%, p=0.001) and infiltration (1.4 vs 0.13%, p=0.006) were more frequent in the PIV group. Chronic vascular complications related to prior VTE was slightly lower in the PIV group (0.13% vs 0.9%, p=0.06), but not significant. Additional details from a propensity score-matched analysis will be presented at the meeting. Conclusions: This is the largest known analysis to date evaluating the safety of PIV access for anthracycline-based chemotherapy in >1,500 lymphoma pts over a recent 15-year period, representing >9,000 total treatments with CHOP or ABVD. PIV access was found to be safe and associated with a low rate of serious complications. Moreover, extravasation was rare and there was not a significant difference observed between PIV and CVC use. While PIV use was associated with higher rates of local pain/tenderness and infiltration, the events were uncommon and self-limited. Despite the widespread default to CVC placement in many centers, these findings support the selective use of PIV in appropriate pts as elucidated here.