OBJECTIVE To investigate the association between subjective pain intensity and objective parameters obtained from two autonomic function tests in a longitudinal study targeting acute pain model in otolaryngology-head and neck region: pupillary light reflex (PLR) and heart rate variability (HRV). APPROACH We enrolled 35 patients with acute otolaryngology-head and neck region inflammatory disorders at pre-treatment stage. The acute inflammatory disorders were defined as acute tonsillitis, peritonsillar abscess, acute epiglottitis, acute sinusitis, and deep neck space abscess. Patients underwent a numeric rating scale (NRS) to monitor subjective pain intensity, PLR, and HRV as objective tests at 4 time-points during the follow-up term. As main outcome variables, we used 15 analyzable PLR/HRV parameters. To improve robustness of conclusions about the association between NRS and PLR/HRV parameters, we prepared four linear mixed-effects models (LMMs) including predictor variables such as NRS, sociodemographic factors, and individual variability. And then, we selected the better-fit model based on the lowest Akaike's information criterion. MAIN RESULTS NRS significantly decreased due to treatments. In 14 out of 15 parameters, better-fit models were models including not only sociodemographic factors but also individual variability. We observed significant parameter alterations to one unit change of NRS in five PLR and four HRV parameters. SIGNIFICANCE The current study revealed that PLR/HRV parameters can be used as biomarkers reflecting pain relief effects. In addition, the findings suggest the importance of adjusting predictor variables, especially individual variability defined as random effects in LMMs, for obtaining more accurate parameter estimation in the longitudinal study.

中文翻译：

---

瞳孔光反射和心率变异性参数作为评估疼痛缓解对自主神经功能影响的生物标志物的临床潜力：一项前瞻性纵向研究

目的 在一项针对耳鼻咽喉-头颈部区域急性疼痛模型的纵向研究中，探讨主观疼痛强度与从两项自主神经功能测试中获得的客观参数之间的关联：瞳孔光反射 (PLR) 和心率变异性 (HRV)。方法 我们招募了 35 名处于治疗前阶段的急性耳鼻咽喉头颈部炎症性疾病患者。急性炎症性疾病定义为急性扁桃体炎、扁桃体周围脓肿、急性会厌炎、急性鼻窦炎和颈部深部脓肿。在随访期间，患者接受了数字评分量表 (NRS) 以监测主观疼痛强度、PLR 和 HRV 作为客观测试的 4 个时间点。作为主要结果变量，我们使用了 15 个可分析的 PLR/HRV 参数。为了提高关于 NRS 和 PLR/HRV 参数之间关联的结论的稳健性，我们准备了四个线性混合效应模型 (LMM)，包括预测变量，如 NRS、社会人口因素和个体变异性。然后，我们根据最低的 Akaike 信息准则选择了更适合的模型。主要结果 NRS 因治疗而显着降低。在 15 个参数中的 14 个参数中，拟合效果更好的模型不仅包括社会人口因素，还包括个体变异性。我们观察到五个 PLR 和四个 HRV 参数中 NRS 的一个单位变化的显着参数变化。意义 目前的研究表明，PLR/HRV 参数可用作反映疼痛缓解效果的生物标志物。此外，研究结果表明调整预测变量的重要性，