Purpose
To investigate if shift work or sleep disturbances are risk factors for central serous chorioretinopathy (CSCR).
Design
Prospective case-control study.
Methods
Forty patients with active CSCR and 40 controls (age- and sex-matched) were prospectively recruited from the Ophthalmology Department of Hôtel Dieu Hospital, Paris, between November 2013 and December 2014. All patients were asked to complete a questionnaire addressing previously described risk factors and working hours, as well as the Insomnia Severity Index (ISI), a validated instrument for assessing sleep disturbances.
Results
The mean age of the CSCR group was 44 ± 9 years, whereas the mean age of the control group was 43 ± 10 years. By use of multivariate analysis, shift work (odds ratio [OR] [95% confidence interval]: 5 [1.2–20.4]; P = .02), steroid use (OR: 5.5 [1.1–26.2]; P = .03), and recent psychological stress (OR: 15.3 [4.1–54.5]; P < .001) were found to be independently associated with CSCR.
Conclusion
The outcomes of this study suggest that shift work is an independent risk factor of CSCR. Further studies are required to confirm these results and to examine if work reconversion would be beneficial in the treatment of patients with chronic/recurrent CSCR.
Central serous chorioretinopathy (CSCR) is a chorioretinal disease that most often affects middle-aged men. It is characterized by posterior and focal serous detachments of the neurosensory retina and retinal pigment epithelium alterations. The role of choroid hyperpermeability in the pathogenesis of CSCR has been well documented recently with multimodal imaging modalities. In CSCR patients, a thick choroid has been reported not only in the affected eye but also in the fellow eye, which is consistent with bilateral choroidal hyperpermeability.
To date, several risk factors for CSCR have been identified, and the most consistent is corticosteroid exposure from therapeutic administration or from endogenous overproduction, as in Cushing syndrome. Corticosteroids were recently shown to induce choroidal vasodilation through mineralocorticoid receptor activation in animal models. CSCR has also been associated with increased sympathetic activity and sympathomimetic medication. Psychological stress and type A personality with narcissistic traits have also been identified as contributing factors. Additional associations have been reported with hypertension, coronary heart disease, peptic ulcer disease, antihistamines, antibiotics, and psychopharmacologic medications.
“Shift work” includes work hours that fall outside the standard daylight hours (7 AM to 6/7 PM), including evening, night, morning, rotating, and irregular shifts. In industrialized countries, 15%–20% of the work force have regular shift work. It leads to circadian misalignment and sleep disturbances that have been associated with increased risk of obesity, diabetes, cardiovascular disease, depression, and cancer. Mechanisms underlying these health problems are not fully elucidated but are thought to be related to sleep and circadian disruption. Indeed, sleep disturbances have been associated with increased activities of the hypothalamic-pituitary adrenal axis and the autonomic sympathoadrenal system, characterized by altered secretion of cortisol and catecholamine hormones.
Since CSCR patients have higher levels of urine and plasma cortisol compared with the control group and increased levels of plasma catecholamines, we questioned whether shift work and/or sleep disturbance evaluated by the Insomnia Severity Index (ISI) could be risk factors for CSCR.
Methods
Study Patients
Patients were prospectively included between November 4, 2013 and December 31, 2014 at the Department of Ophthalmology in the Hôtel Dieu Hospital, Paris, France. The Ethics Committee of the French Society of Ophthalmology approved this study. Informed signed consent was obtained from all subjects in compliance with the tenets of the Helsinki agreement.
Patients of working age between 20 and 60 years and presenting symptoms of CSCR were enrolled. Active CSCR was defined as a localized serous retinal detachment on optical coherence tomography associated with angiographic leakage on fluorescein angiography without signs of choroidal neovascularization, polypoidal choroidal vasculopathy, or inflammation. Chronic CSCR was defined by the persistence of serous retinal detachment (SRD) for more than 6 months and/or with recurrent SRD associated with widespread decompensation of the retinal pigment epithelium.
Control patients were matched for age and sex at a ratio of 1:1 and recruited at the emergency consultation with ocular complaint but with no previous or ongoing retinal pathology. The frequency of diseases including conjunctivitis, blepharitis, chalazion, dry eye, keratitis, anterior uveitis, refractive disorders, vitreous floaters, eye contusion, endophthalmitis, and migraine could not exceed 10% of the whole sample.
Study Protocol
All patients were asked to complete study questionnaires, which included an ISI questionnaire in addition to questions addressing previously identified risk factors such as history of systemic diseases (hypertension, depression, allergic disease), current medication (current or recent [<3 months] corticosteroid intake [oral, intranasal, inhalational, topical skin application, intravenous, intramuscular, eye drop]), and degree of alcohol or tobacco use. Recent psychological stresses including life changes (death, divorce, familial strife, layoff) and stress at work were also assessed. Participants provided information on present and past (<1 year) working hours. Shift work was defined as work starting before 7:00 AM or finishing after 7:00 PM, and thus included evening, night, or early morning working within this broader schema.
Insomnia Severity Index
Sleep disturbances were evaluated using the ISI, which is a widely used instrument for evaluating insomnia symptoms and severity. The ISI is a 7-item self-report questionnaire assessing the severity of sleep onset, sleep maintenance, early-morning awakening problems, sleep dissatisfaction, interference of sleep difficulties with daytime functioning, noticeability of sleep problems by others, and distress caused by the sleep difficulties, with a total score ranging from 0 to 28 ( Supplementary Material , available at AJO.com ). A cutoff score of 10 has been previously selected to detect insomnia (86.1% sensitivity and 87.7% specificity).
Statistical Analysis
The data obtained were analyzed with independent t test or Mann-Whitney test for continuous variables. Categorical variables were compared using χ 2 or Fisher exact test. Risk factors were initially analyzed using univariate linear regression analysis. Those with statistical significance on univariate analysis were included in multivariate analysis using logistic regression with forward stepwise selection (stepAIC, MASS package). Statistical analyses were performed with R version 3.1.3 (R foundation of Statistical Computing, Vienna, Austria). A P value of less than .05 was considered significant.
Results
Patient Characteristics
A total of 40 CSCR patients were compared with 40 sex- and age-matched control patients ( Table 1 ). The mean age (± standard deviation [SD]) of the CSCR group was 44.1 ± 8.6 years, whereas the mean age of the control group was 43 ± 10.1 years ( P = .6). Both the CSCR and the control groups were composed of 34 male (85%) and 6 female patients (15%), with a male-to-female ratio of 5.6:1.
| Exposure | Patients With CSCR (n = 40) | Control Patients (n = 40) | Odds Ratio | 95% Confidence Interval | P Value |
|---|---|---|---|---|---|
| Age, years (mean ± SD) | 44.1 ± 8.6 | 43 ± 10.1 | – | – | .6 |
| Male (%) | 85 | 85 | – | – | 1 |
| Sleep disorders (ISI >10), n (%) | 23 (57.5%) | 6 (15%) | 7.6 | 2.7–21.7 | <.001 |
| Shift work, n (%) | 17 (42.5%) | 6 (15%) | 4.6 | 1.5–11.9 | .007 |
| Steroid use, n (%) | 15 (37.5%) | 4 (10%) | 5.4 | 1.7–17.3 | .007 |
| Stress, n (%) | 27 (67.5%) | 5 (12.5%) | 14.5 | 4.8–44.1 | <.001 |
| Hypertension, n (%) | 9 (22.5%) | 3 (7.5%) | 3.6 | 0.96–13.3 | .06 |
| Depression, n (%) | 7 (17.5%) | 9 (22.5%) | 0.7 | 0.2–2.1 | .6 |
| Allergic disease, n (%) | 13 (32.5%) | 14 (35%) | 0.9 | 0.4–2.2 | .8 |
| Tobacco use, n (%) | 17 (42.5%) | 12 (30%) | 1.7 | 0.7–4.3 | .2 |
| Alcohol use, n (%) | 6 (15%) | 4 (10%) | 1.6 | 0.4–5.8 | .5 |
In the CSCR group, 26 patients (65%) had an acute form of the disease and 14 patients (35%) had a chronic CSCR, defined by the persistence of the SRD for more than 6 months and/or with recurrent SRD associated with bilateral widespread retinal pigment epithelium alterations.
Univariate Analysis
Previously reported risk factors, sleep disturbance, and shift work were first analyzed with univariate analysis ( Table 1 ).
Sleep disturbances evaluated with the ISI questionnaire occurred more frequently in patients with CSCR (57.5% had ISI score >10) than it did in the control group (15%; P < .001; odds ratio [OR] [95% confidence interval]: 7.6 [2.7–21.7]; Table 1 ). Moreover, the mean of the ISI score (± SD) was significantly higher in the CSCR group (9.6 ± 6.2) than in the control group (4.1 ± 4.5; P < .001; Figure ). We also examined the relationship between CSCR type and sleep disorders, which were more frequently observed in cases of chronic CSCR (78.6% had ISI score >10) than in acute CSCR (46.1%; P < .05).
Seventeen CSCR patients (42.5%) had performed shift work, vs 6 patients (15%) in the control group ( P = .007; OR: 4.6 [1.5–11.9], Table 1 ). Shift workers had a higher ISI score than day workers (ISI mean ± SD, shift workers: 10.9 ± 5.5 vs day workers: 5.2 ± 5.5; P < .001).
Previously reported risk factors were also analyzed. Corticosteroid use was present or reported in the 3 months before the first CSCR symptoms in 15 patients (37.5%) of the CSCR group and 4 patients (10%) in the control group ( P = .007; OR: 5.4 [1.7–17.3], Table 1 ). Among the 15 CSCR patients, use of corticosteroids was topical skin application in 5 patients (33.3%), intranasal administration in 4 patients (26.7%), oral medication in 4 patients (26.7%), intravenous injection in 1 patient (6.7%), and intramuscular injection in 1 patient (6.7%). In the control group, among the 4 patients reporting corticosteroid use, 2 reported eye drop, 1 topical skin application, and 1 oral corticosteroid administration.
Recent psychological stress was reported in 27 CSCR patients (67.5%) and in 5 control patients (12.5%; P < .001; OR: 14.5 [4.8–44.1], Table 1 ). Among the 27 CSCR patients, 14 patients reported stressful life changes (death, disease, divorce, familial strife, layoff) and 13 patients reported stress at work. The 5 control patients reported work-related stress.
Other previously reported risk factors, such as hypertension, depression, allergy, and tobacco or alcohol use, were not statistically significantly different in CSCR and control groups ( Table 1 ).
Multivariate Logistic Regression Analysis
A stepwise logistic regression confirmed the following risk factors ( Table 2 ): shift work ( P = .02; OR: 5 [1.2–20.4]), corticosteroid use ( P = .03; OR: 5.5 [1.1–26.2]), and recent psychological stress ( P < .001; OR: 15.3 [4.1–54.5]).
