Background: Despite their elevated baseline risks for infection and thrombosis, hospitalized medical patients with active malignancy commonly receive multi-lumen peripherally inserted central catheters (PICCs). We sought to assess the relationship between number of catheter lumens and risk of PICC-related complications among patients with active malignancy.
Methods: Trained abstractors collected patient level information, including number of catheter lumens, on a sample of 5,609 PICC placements in adult patients admitted for a cancer-related condition to a general medicine ward or intensive care unit at 65 hospitals participating in the Michigan Hospital Medicine Safety Consortium (HMS). The study period was May 2014 to February 2023. Patients were followed until PICC removal, death, or 30 days from insertion, whichever came first. Outcomes of interest included image-confirmed symptomatic upper extremity venous thromboembolism (UE-VTE), laboratory-confirmed central line-associated bloodstream infection (CLABSI), catheter occlusion, and the composite of all three. For all analyses, the PICC was the level of analysis. Multivariable logistic regression, including hospital-level random effects to account for clustering of patients within hospitals, was performed to calculate the odds ratio (OR) of each outcome comparing single versus double and triple versus double lumens, adjusting for known predictors of PICC-related complications [for CLABSI: hematologic cancer, CLABSI history, ongoing chemotherapy for solid cancer, receipt of TPN and presence of another CVC at time of insertion; for UE-VTE: history of DVT, presence of another CVC at time of insertion and WBC > 12k at time of insertion; for occlusion: right arm placement, verification of tip position, flushed with saline and locked with heparin and tip malposition.
Results: There were 1,611 (28.7%) single, 3,548 (63.3%) double, and 450 (8.0%) triple lumen PICCS in the analysis. UE-VTE occurred in 191 (3.4%), CLABSI in 274 (4.9%), catheter occlusion in 751 (13.4%), and at least one complication in 1,100 (19.6%) PICCs. After adjustment, single versus double lumen PICCs had lower risks of UE-VTE (OR = 0.47; 95% confidence interval [CI] = 0.31-0.71), CLABSI (OR = 0.57; 95% CI=0.40-0.82), catheter occlusion (OR = 0.25; 95% CI = 0.20-0.33), and any of the three complications (OR = 0.35; 95% CI = 0.29-0.43). Triple versus double lumen PICCs had similar risk of UE-VTE (OR = 1.31; 95% CI = 0.80-2.15), but higher risks of CLABSI (OR = 1.93; 95% CI = 1.35-2.77), catheter occlusion (OR = 1.55; 95% CI = 1.19-2.0), and any of the three complications (OR = 1.77; 95% CI = 1.40-2.24).
Conclusions: Among PICC recipients with active malignancy, when compared to double lumen catheters, single lumen catheters are associated with lower risks of UE-VTE, CLABSI and catheter occlusion complications, while triple lumen catheters are associated with higher risks of CLABSI and catheter occlusion.
