Assessment of pain during labor with pupillometry: a prospective observational study.

Background: Pain intensity is usually self-rated by patients with a numeric rating scale (NRS) but this scale cannot be used for noncommunicating patients. In anesthetized patients, experimental noxious stimulus increases pupillary diameter (PD) and pupillary light reflex amplitude (PLRA), the difference between PD before and after light stimulation. Labor pain is an intense acute nonexperimental stimulus, effectively relieved by epidural analgesia. In this prospective observational study, we therefore describe the effects of labor pain and pain relief with epidural analgesia on PD and PLRA, determine their association with pain intensity and determine the ability of a single measurement of PD or PLRA to assess pain.

Methods: In the first stage, pain (11-point NRS), PD, and PLRA were measured in 4 conditions in 26 laboring women: before and after epidural analgesia and in the presence and absence of a uterine contraction. Pupillometry values among the 4 conditions were compared, and the strength of the association between absolute values of pain and PD or PLRA and between pain and changes in PD or PLRA brought about by uterine contraction was assessed with r(2). In the second stage, 1 measurement was performed in 104 laboring women. The strength of the association between pain and PD or PLRA was assessed with r(2). The ability of PD or PLRA to discriminate pain (NRS > 4) was also assessed.

Results: In the first stage, a statistically significant increase in pain, PD, and PLRA was observed during a contraction, and this change was abolished after epidural analgesia. The r(2) for the association between pain and changes in PD (r(2) = 0.25 [95% confidence interval, 0.07-0.46] or PLRA (r(2) = 0.34 [0.14-0.56]) brought about by a uterine contraction was higher than the r(2) for the association between pain and absolute values of PD (r(2) = 0.14 [0.04-0.28]) or PLRA (r(2) = 0.22 [0.10-0.37]) suggesting a stronger association for changes than for absolute values. In the second stage, r(2) was 0.23 [0.10-0.38] for PD and 0.26 [0.11-0.40] for PLRA and the area under the receiver operating characteristics curve was 0.82 [0.73-0.91] and 0.80 [0.71-0.89], respectively.

Conclusions: Changes in PD and PLRA brought about by a uterine contraction may be used as a tool to assess analgesia in noncommunicating patients.