Ocular Effects of Synthetic Cannabinoids

MATERIAL AND METHODS Participants This study was designed as cross-sectional, observational, case-control study. Upon referral from the judiciary-supervised probation unit of Alcohol and Substance Research, Treatment and Training Centre (AMATEM) of the relevant tertiary psychiatry hospital between March 2015 and January 2016, 60 eyes of 60 consecutive male patients with past synthetic cannabinoids (SC) use and 30 eyes of age-matched 30 male healthy control subjects were enrolled in the study in 3 groups: a control group (n=30) included eyes of subjects with no history of use of any substance including tobacco, a seronegative SC group (n=30) included eyes of patients declaring no present use of any substance, as verified by three consecutive negative urine toxicology tests performed two weeks apart, and a seropositive SC group (n=30), including eyes of patients with positive urine toxicology tests proving present use of SC.

SCs and relevant metabolites in urine samples were screened with a direct ELISA kit (K2 Enzyme Immunoassay, Immunalysis Corp., Pomona, CA, USA). This kit was able to detect relevant metabolites, such as JWH-018, JWH-073, and AM-2201, with a cut-off level of 20 ng/mL and over for positive results. Exposures to classical cannabinoids, ecstasy, amphetamines, opioids, and ethylene glycol were also screened.

Regarding SC use, the patients were asked about the duration and the last time of use. In addition, the previous use of other substances was questioned in detail.

Ophthalmic Assessments All eyes underwent ophthalmic examinations, including autorefraction, best-corrected visual acuity (BCVA) testing, biomicroscopy, non-contact IOP measurement, and dilated fundus examination.

An experienced technician performed spectral-domain optical coherence tomography (SD-OCT) imaging of all eyes following pupil dilatation (1% tropicamide) and systolic/diastolic blood pressure measurements (SBP/DBP). All SD-OCT assessments were conducted between 14:00 and 16:00 to minimize the potential effect of diurnal variation. Only high-quality SD-OCT images with a signal strength index (SSI) greater than 50 were sent for analysis. The parafoveal retinal thickness (parafoveal RT) and central foveal thickness (CFT), CT, peripapillary RNFL thickness, and macular GCC thickness measurements were performed with an RTVue-100 OCT device (Optovue Inc., Fremont, US), which had a 5 µm axial image resolution with a speed of 27,000 A-scans per second.

We preferred the ONH map protocol for examining the RNFL parameters. This protocol creates a RNFL thickness map based on measurements around a circle 3.45 mm in diameter centred on the ONH. The protocol for the GCC scan was based on examining a square grid (7 × 7 mm) on the central macula after centring 1 mm temporal to the macula. The average, superior, and inferior values displayed by the device were studied for the mean GCC and mean RNFL parameters.

Segmental divisions were recorded with the MM5 protocol of the inbuilt software, characterized by a 5 × 5 mm2 grid of 11 horizontal and 11 vertical lines with 668 A-scans and a 3 × 3 mm2 inner grid of 6 horizontal and 6 vertical lines with 400 A-scans for the retinal thickness of a 1 mm diameter CFT. The CT was measured by cross-sectioning through the foveolar centre line, characterized by an 8 mm width, with 1024 A-scans captured on the chorioretinal mode (these settings enable enhanced depth imaging along with contrast and brightness tuning). The line was rotated horizontally and vertically so that manual measurements were taken first at the central foveola and then 750 µm apart in the nasal, temporal, superior, and inferior directions with the ruler function, using a line extending from the posterior edge of the retinal pigment epithelium to the choroidoscleral junction.

The choroidal structure was analysed using the method previously described by Agrawal et al., and all collected images were processed and analysed using the open-source and publicly available ImageJ software (version 1.50a, freely available at http://imagej.nih.gov/ij/; National Institutes of Health [NIH], Bethesda, MD). According to this method, the image was first binarized using the Niblack auto-local thresholding function. This provided a clear visualization of the choroidoscleral interface while also enabling the precise selection of a region of interest, such as the total subfoveal choroidal area (TA). A fovea-centred, 1500 μm wide, nasal-to-temporal subfoveal choroidal area was manually selected using the polygon selection tool. The dark pixels representing the luminal area (LA) were selected using the colour threshold tool, and the residual light pixels were considered the stromal area (SA). The subfoveal CVI was calculated as the ratio of the LA to the TA. Manual measurements (CT and CVI) were carried out by the same experienced investigator blinded to the groups. Each measurement was repeated twice, and the mean of the two measurements was used in the analysis.

Statistical analysis Statistical analysis was conducted with software (SPSS 22.0 Version, IBM Corporation, New York, USA). Descriptive statistics were stated as the mean, standard deviation (SD), or median (Med), the interquartile range (Q1-Q3), and the frequency and ratio (%) values, where appropriate. The distribution of normality was evaluated for each parameter with the Kolmogorov-Smirnov test. Parametric (for normal distribution) or non-parametric tests were applied as appropriate. A parametric analysis of variance (ANOVA) with a post hoc Tukey test or a non-parametric Kruskal-Wallis test with a post hoc Bonferroni-corrected Mann-Whitney U test was used to analyse the independent variables. The chi-square test was used to analyse the qualitative data. Correlations of choroidal thickness measurements with parameters relevant to SC use were studied using the Spearman test.