肝移植手术中桡动脉插管是否影响手的灌注指数（PI）？ (radial artery)

## DLiver transplantation is associated with major hemodynamic changes, and continuous invasive arterial pressure monitoring is an essential part of perioperative management. The radial artery is the most commonly used site. Repeated punctures and mechanical trauma during catheterization may cause local complications and affect peripheral perfusion. Ultrasound guidance may facilitate catheter placement with fewer punctures.

The perfusion index (Pi), derived from pulse oximetry, is the ratio of the pulsatile to the non-pulsatile optical signal and is used to monitor peripheral perfusion. Because Pi is also influenced by systemic hemodynamics, sympathetic activity, temperature, nociception and drugs, changes in Pi were not interpreted as direct evidence of radial artery vasospasm.

In this prospective, randomized, controlled, two-arm parallel-group trial, adult liver transplant recipients were randomized 1:1 to palpation-guided or ultrasound-guided radial artery catheterization. Randomization used a computer-generated sequence and sequentially numbered, opaque, sealed envelopes prepared by an independent person. The envelope was opened after induction of general anesthesia, once hemodynamic stability was achieved and immediately before catheterization. All procedures were performed by a single anesthesiologist experienced in both techniques, using the same 20-gauge non-guidewire arterial cannula in both groups, preferably on the non-dominant hand after a normal modified Allen test.

In the ultrasound group, a high-frequency linear probe was used with a short-axis out-of-plane approach and dynamic needle tip positioning. Pi was measured continuously with a pulse oximetry sensor placed on the index finger the hand. Pi, heart rate, SpO2, pleth variability index (PVI), SpHb and SpOC were recorded immediately before catheterization (baseline) and immediately after catheterization; Pi was additionally recorded 5 and 10 minutes after catheterization.

The primary outcome was the change in Pi from baseline over time. The primary analysis used a linear mixed-effects model including group, time and the group-by-time interaction to account for repeated measurements. Secondary outcomes were compared with appropriate tests (t test or Mann-Whitney U test; chi-square or Fisher's exact test). Results were reported with effect estimates and 95% confidence intervals; a two-sided p<0.05 was considered statistically significant.