Development and validation of an updated PICC length prediction formula based on anteroposterior chest radiographs for the ultrasound-guided bedside placement.

Abstract

Bedside peripherally inserted central catheter (PICC) placement is sometimes required when the patient's intrahospital transport is restricted, and the ideal catheter length prediction is needed. This study aimed to develop an updated formula that predicts the optimal length of a PICC based on anteroposterior chest radiographs (AP-CXRs). This retrospective study collected PICC procedure data as the training and validation sets in three hospitals, including cubital crease-puncture point distance (CP), the actual PICC length (aCL), and the approach side. Horizontal and vertical measurement variables were set on the AP-CXRs. Two dependent variables were ipsilateral upper arm length (AL) and ideal truncal catheter length (iTCL). Simple and multiple regression analyses were used for formula development, and it was applied to the test set to evaluate the length prediction performance. The study included 309 patients in the training and validation sets and 91 intensive care patients in the test set. The final derived formula was: (AL + iTCL = CP + estimated PICC length, cm) = 19.831 - 0.062 × (contralateral clavicle length, cm) + 0.255 × (2nd ribs horizontal distance, cm) + 0.720 × (humero-vertebral distance, cm) + 0.761 × (thoraco-carinal distance, cm) + 1.024 × (the vertical distance of two vertebral body units, cm). (If approaching from the left, add 2.843cm, and if female, subtract 0.821cm.) In the test set, there was no case of length prediction failure. Moreover, the catheter tip position was evaluated as optimal in 82 cases (90.1%). This study's results suggest an updated formula to predict the ideal PICC length using only AP-CXRs for bedside placement.