Abstract
Purpose :
Many studies report a seasonal variation in the incidence of rhegmatogenous retinal detachment (RDD) but few address the relationship to meteorological or clinical factors. We conducted a retrospective, observational study to explore the combined influence of these factors on the incidence of RDD in the central region of Portugal.
Methods :
We included all patients over 18 years-old that underwent surgery for a de novo RRD at a tertiary referral hospital in the center of Portugal, between January 2010 and December 2014. Retinal tears or rhegmatogenous lesions treated only with laser photocoagulation were not included. Cases with mixed type detachments or previous vitreoretinal surgery were excluded. Meteorological data (temperature, humidity, precipitation, atmospheric pressure and solar radiation) from the subject’s symptom onset in his area of residence as well as demographical and clinical data on risk factors for RRD was collected. We built a forecasting regression model to account for seasonality – the chronological model. We also built a model using only weather data – the meteorological model. Finally, we built a model combining meteorological, demographical and clinical risk factors data – the biological model.
Results :
We included 914 eyes of 898 patients. Mean age was 61.84±14.00 years, 64.6% were males and 45.8% had at least one identifiable risk factor for RDD. The chronological model was significant for (1) a seasonality effect, with peaking in May (p=0.019) and September (p=0.01). This seasonality effect was lost when adjusting for temperature. The meteorological model was only significant for temperature (5 degrees increase, OR 3.1 [1.3 – 4.9], p=0.001). The biological model was significant for age from 18-39 and over 60 years-old, males, high myopia, previous trauma or ocular surgery (all p<0.001) and temperature (p=0.039), explaining 93% (adjusted R2) of RDD incidence over the study period.
Conclusions :
Our results show a seasonality effect on the incidence of RDD, associated to increased temperatures. Combining meteorological, demographical and clinical data is able to explain most of RDD incidence. These findings offer new insight on the epidemiology and lesser known predictors of RDD.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.