Abstract
Purpose :
To evaluate the association between conjunctival vessel density (VD) and common medical conditions.
Methods :
Patients who attended a preoperative office visit for cataract surgery were consecutively recruited to participate in a study to analyze the morphology of the conjunctival vasculature. All patients underwent bilateral anterior segment photographs (SL-D7, Topcon), with the same zoom and light settings, exposing most of the superior scleral quadrant (from M11 to M1). Only one high-quality image was selected per subject, defined as an image with adequate light intensity that adequately focuses on the conjunctival vessels, with no light reflections in the region of interest (ROI), and exposing at least 3 millimetres of the perilimbal sclera. All images were processed by one evaluator in ImageJ (NIH, v1.54d), and the VD was measured using the Vessel Analysis plugin as depicted in detail in Figure 1. Variables of interest such as diabetes, hypertension (HTN), smoking status, age, weight, and height were obtained from the clinical records. Haemoglobin, creatinine, glucose, and heart rate results were selected from the date closest to the date of photograph acquisition. Linear regressions were calculated to analyze the effect of variables of interest in the VD. A multivariable model was performed on variables that showed a significant or borderline effect on DV.
Results :
Ninety-six patients were recruited, from which we included high-quality photographs (one eye per patient) of eighty-seven of them for the current preliminary analysis. Mean age was 68.1±9.7, we included 35 (40.2%) males and 52 (59.8%) females. The mean VD was 9.9±5.3. The complete mean values and results are shown in Table 1. Height was significantly (P=.036) positively correlated (Spearman’s rho=0.226) with the VD, the rest of the variables had non-significative weak correlations. VD was significantly (P=.006) greater in normal-tension subjects (11.6±5.6) than in hypertense subjects (8.4±4.7). Weight (B=0.081), height (B=0.140), and HTN (B=-3.123) had a significative effect on VD at the univariable analysis. The multivariable model included gender, weight, height, HTN, and diabetes, only weight (B=0.108) and HTN (B=-3.533) remained significant.
Conclusions :
Diagnosis of hypertension was associated with a reduced VD. All patients were treated for HTN which makes it impossible to differentiate if the association is with the medications used for HTN or by the disease itself.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.