Whether sympathetic withdrawal or endothelial dilators such as nitric oxide (NO) contributes to cold-induced vasodilation (CIVD) events is unclear. We measured blood flow and finger skin temperature (Tfinger) of the index finger in nine participants during hand immersion in a water bath at 35 °C for 30 min, then at 8 °C for 30 min. Data were binned into 10 s averages for the entire 60 min protocol for laser-Doppler flux (LDF) and Tfinger. At baseline, Tfinger was 35.3 ± 0.2 °C and LDF was 227 ± 28 PU. During hand cooling, minimum Tfinger was 10.9 ± 0.4 °C and LDF was 15 ± 4 PU. All participants exhibited at least one CIVD event (Tfinger increase ≥ 1 °C), with a mean peak Tfinger 13.2 ± 0.8 °C and a corresponding peak LDF of 116 ± 34 PU. A Morlet mother wavelet was then used to perform wavelet analysis on the LDF signal, with frequency ranges of 0.005-0.01 Hz (endothelial NO-independent), 0.01-0.02 Hz (endothelial NO-dependent), and 0.02-0.05 Hz (neurogenic). The synchronicity of wavelet fluctuations with rising LDF coincident with CIVD events was then quantified using Auto-regressive Integrated Moving Average time-series analysis. Fluctuations in neural activity were strongly synchronized in real time with increasing LDF (stationary-r2 = 0.73 and Ljung-box statistic > 0.05), while endothelial activities were only moderately synchronized (NO-independent r2 = 0.15, > 0.05; NO dependent r2 = 0.16, > 0.05). We conclude that there is a direct, real-time correlation of LDF responses with neural activity but not endothelial-mediated mechanisms. Importantly, it seems that neural activity is consistently reduced prior to CIVD, suggesting that sympathetic withdrawal directly contributes to CIVD onset.
Keywords: Auto-regressive integrated moving average; Cold injury; Finger blood flow; Sympathetic nervous system; Time-series analysis; Wavelet analysis.