Pupillary response abnormalities in depressive disorders

Abstract

Depressive disorders lack objective physiological measurements to characterize the affected population and facilitate study of relevant mechanisms. The melanopsin-mediated light signaling pathway may contribute to seasonal variation and can be measured non-invasively by pupillometry. We prospectively studied changes in melanopsin-mediated pupillary constriction in 19 participants with major depressive disorder (MDD) and 10 control across the summer and winter solstices. The melanopsin-mediated response, as measured by the pupil's sustained constriction six s after a high intensity blue light stimulus, was marginally attenuated in those with MDD relative to controls (p=0.071). The participants with MDD unexpectedly showed a significantly reduced transient pupillary response to low intensity red (p=0.011) and blue light (p=0.013), but not high intensity red and blue light. Sustained pupillary constriction in response to high intensity blue light was more pronounced with increasing daylight hours (p=0.037) and was more strongly related to objectively measured versus estimated light exposure. Melanopsin-mediated impairments in pupil response may serve as a biological marker for vulnerability to depression in low light conditions. Assessment of these and other responses to light stimuli, such as response to low intensity light, may be useful for the study of the neurobiology of MDD and related mood disorders.

Keywords: Bipolar disorder; Major depressive disorder; Pupillometry; Seasonality.

Fig. 1. Example pupillometry recordings

Fig. 1 shows an example of pupil light reflexes from a patient with depression recorded during the summer (top) and winter (bottom). The right and left pupil movements were averaged together in response to binocular stimuli (both eyes open and stimulated simultaneously). Four pupil light reflexes in response to varying intensity and wavelength are shown for summer and winter; low intensity red and blue light stimuli and high intensity red and blue light stimuli. A one s duration stimulus was used (square wave stimulus tracing shown at the top in black, ending at 0 s. The pupil responses were normalized to the baseline pupil size recorded during the one s before the light stimulus to give the percent of baseline size at each time point, sampled at 30 Hz. The time of year when this patient's testing occurred in the summer corresponded to 15 h of daylight and in winter to 10 h of daylight. The subject also wore a wristwatch activity monitor with light meter which recorded a light level of 2255 lx during the summer and 207 lx in the winter Note that the transient pupil responses mediated by rods and cones were greater during the summer. The sustained post-illumination pupil response (PIPR) to bright blue light provides a measure of melanopsin-mediated pupil contraction, which was also prolonged during the summer compared to the winter in this patient.(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 2. Scatterplots of stimuli responses against the number of daylight hours by group (depressed versus normal)

The panels show the responses by group to low intensity red light (Red1), low intensity blue light (Blue1), high intensity red light (Red400), high intensity blue light (Blue400), and the primary outcome of interest, sustained post-illumination pupil constriction to high intensity blue light (Blue400S7). The response, on the y-axis, can be understood as the proportion of the pupil's size following a light stimulus compared to the pupil's baseline size (e.g. −0.2 equals a 20% sustained pupil contraction measured 6 s following termination of a one s light stimulus). Regression lines (solid for control group, dashed for MDD group) model the impact of daylight hours by group, and unlike the mixed models presented, assume independent observations.(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 3. Scatterplots of the sustained response to high intensity blue light (Blue400S7) against the number of daylight hours for the two groups of participants

Observations from the same subject are connected to show how response changes with daylight hours. The y-axis represents the percent pupil contraction of the post-illumination pupil response to a one s duration bright blue light (e.g. −0.2 equals a 20% sustained pupil contraction measured 6 s following termination of a one s light stimulus). A negative slope indicates a greater sustained post-illumination pupil contraction with increasing daylight hours of light exposure during the summer compared to the winter.(For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 4. Scatterplot of the participants with MDD luxometer response against the number of daylight hours, with observations from the same subject connected

The figure shows that in the weeks surrounding the winter solstice, the variations in the luxometer reading (light intensity sensor on the actiwatch, reported on the y-axis as “LightLuxAve”) were quite small. In the weeks, surrounding the summer solstice, we observed a large variation in luxometer readings, presumably due to greater variability in outdoor activities.(For interpretation of the references to color in this figure, the reader is referred to the web version of this article.)

Fig. 5. Scatterplots of responses to sustained pupillary constriction (post-illumination pupil response) following high intensity blue light (Blue400S7) against the number of daylight hours and against the average light lux for depressed subjects

The y-axis represents the percent pupil contraction of the post-illumination pupil response to a one s duration bright blue light (e.g. −0.2 equals a 20% sustained pupil contraction measured 6 ss following termination of a one s light stimulus). Attenuated sustained responses to high intensity blue light are seen with fewer daylight hours (left panel) and lower light exposure (right panel). Unlike the mixed models presented in the primary models, the regression lines shown do not take into account subject effects and treat all observations as independent.(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)