Retinal imaging with hand-held optical coherence tomography in older people with or without postoperative delirium after hip fracture surgery: A feasibility study

Abstract

Introduction: Postoperative delirium in older people may result from the interaction between intrinsic brain vulnerability (e.g. neurodegeneration) and precipitating factors (e.g. surgery induced cytokines). Intrinsic brain vulnerability may be overt (e.g. Alzheimer's disease) or preclinical. In cognitively intact older people presenting for surgery, identification of preclinical neurodegeneration using bedside tools could aid postoperative delirium risk stratification. Thinning of the circumpapillary retinal nerve fibre layer thickness is associated with neurodegenerative disorders e.g. Alzheimer's disease. We propose that thinning of the retinal nerve fibre layer may be present some older people with postoperative delirium due to preclinical neurodegeneration, albeit to a lesser extent than in overt dementia.

Objectives: The primary objective: Feasibility of acquiring usable retinal images with the hand-held optical coherence device, at the bedside of older, hip fracture surgery patients. Secondary objective: Comparison of the circumpapillary retinal nerve fibre layer thickness between people who did/did not have postoperative delirium. Proportion of exclusions due to retinal pathology.

Method: Feasibility study involving 30, cognitively intact, older people recovering from hip fracture surgery. Retinal images were obtained using the hand-held optical coherence tomography device at the participants' bedside. Imaging was deferred in participants who had postoperative delirium.

Results: Retinal images that could be assessed for circumpapillary retinal nerve fibre layer thickness were obtained in 26 participants (22 no postoperative delirium, 4 postoperative delirium). The mean circumpapillary retinal nerve fibre layer thickness was lower in the participants who had postoperative delirium compared to those who did not experience postoperative delirium (Mean (95% CI) of 76.50 (62.60-90.40) vs 89.19 (85.41-92.97) respectively).

Conclusion: Retinal imaging at the patient's bedside, using hand-held OCT is feasible. Our data suggests that the circumpapillary retinal nerve fibre layer may be thinner in older people who experience postoperative delirium compared to those who do not. Further studies are required.

Fig 1. Flowchart of participant recruitment.

POD = Postoperative Delirium, OCT = Optical Coherence Tomography.

Fig 2. HH-OCT circumpapillary retinal nerve fibre layer thickness measurement in a participant.

The cpRNFL thickness was divided into 10 radial segments (using the GDx Nerve Fiber Analyzer protocol, Carl Zeiss Meditec). The quadrants are created by amalgamating two (temporal), two (nasal), three (superior) and three (inferior) radial segments. The colours represent comparison to age-adjusted normative values (green = normal, yellow = borderline and red = out of range).

Fig 3. Box plot of Mean cpRNFL (μm) at Ring diameter 2.83mm in participants who did not have postoperative delirium and participants who had postoperative delirium with individual participant’s values.

POD = postoperative delirium, RNFL = retinal nerve fibre layer.

Fig 4. Boxplot of cpRNFL thickness (μm) by Quadrants at ring diameter 2.83mm comparing participants who had postoperative delirium to those who did not experience postoperative delirium.

POD = Postoperative delirium, RNFL = Retinal Nerve Fibre Layer.