Physiologic and Psychologic Risk Factors in

Central Serous Chorioretinopathy

Ahmad M. Mansour, MD,1,2 Mona Koaik, MD,1 Luiz H. Lima, MD,3 Antonio Marcelo B. Casella, MD,4
Sami H. Uwaydat, MD,5 Maha Shahin, MD,6 Hani Tamim, PhD,7 Maria-Jose Sanchez-Ruiz, PhD,8
Hana A. Mansour, BS,1 David Dodwell, MD9

Purpose: Central serous chorioretinopathy (CSCR) is characterized by macular detachment due to thickened
choroid, mostly affecting young men under perceived stress. Although most previous studies on CSCR have
been retrospective and have focused on a single facet of the patient’s personality, we conducted a prospective,
intercontinental, controlled study to analyze the multifaceted personality profile in CSCR.
Design: Prospective, cross-sectional, case-control study.
Participants: Subjects with CSCR from 6 university-based eye clinics consented to participate in a ques-
tionnaire. Controls without retinal disease were recruited from the same clinics.
Methods: The interview consisted of a 60-item questionnaire. Recruitment of participants was from January
2015 to February 2016. Controls were matched for age, gender, and race. Statistical analyses were performed
using univariate and multivariate analyses.
Main Outcome Measures: The main parameters registered were presence of stress, daily number of cups
caffeine intake, and personality traits (Type A; obsessiveecompulsive; aggressive).
Results: A total of 83 consecutive patients with CSCR (mean age, 45.9 years; male, 80.7%) and 83 controls
(mean age, 46.0 years; male, 80.7%) were analyzed for 60 variables. Multivariate analysis revealed a strong
association with obsessiveecompulsive behavior (P ¼ 0.001), caffeine intake (P ¼ 0.002), Type A personality
(P ¼ 0.002), continuous stress (P ¼ 0.001), and premature ejaculation (P ¼ 0.001).
Conclusions: This study sheds light on the unique psychologic functioning of patients with CSCR: preoc-
cupied, inflexible, perfectionist (obsessiveecompulsive tendency), competitive, ambitious, impatient, high
achiever (Type A personality), and under continuous stress. In addition, caffeine abuse and premature ejaculation
were linked to CSCR. Ophthalmology Retina 2017;-:1e11 ª 2017 Published by Elsevier Inc. on behalf of the
American Academy of Ophthalmology

Central serous chorioretinopathy (CSCR) characterized by selection of controls, questionnaire vs. interview by physi-
serous macular detachment is the fourth most common cian assistant vs. interview by physician) and focus on
retinal disease after age-related macular degeneration, dia- single risk factors in the literature,7,9,10,13 we designed a
betic retinopathy, and retinal vein occlusion.1e3 In a study that targets a vast number of known or potential risk
population-based retrospective cohort and case-control factors in various collaborative centers and compared it with
study in Olmsted County, Minnesota, Kitzmann et al4 the results reported in the literature.
reported a mean age-adjusted incidence of 9.9 (95% confi-
dence interval [CI], 7.4e12.4) per 100 000 in men and 1.7
Methods
(95% CI, 0.7e2.7) per 100 000 in women. The mean age of
CSCR onset is usually between 41 and 45 years.5 The In this prospective study, each collaborating senior ophthalmolo-
disease affects various racial groups5e7 and occasionally gist from 6 different medical centers and 4 different countries
can be familial. The pathogenesis remains poorly under- (Lebanon, Brazil, United States, Egypt) personally administered a
stood, and currently the exudation is thought to result from standard open-ended interview to their consecutive patients with
hyperpermeable choroid secondary to venous stasis, CSCR and to an equal number of consecutive age-, gender-, race-,
ischemia, or inflammation.3 Because of the lack of proper and center-matched controls (patients without CSCR), after
animal models1,8 and definite cure, clinical research has informed consent. The period of this cross-sectional, case-control
study was from January 2015 to February 2016. Institutional Re-
focused on risk factors, such as stress, corticosteroid intake,
view Board approval was obtained from the collaborating centers,
and Type A personality.9,10 and the described research adhered to the tenets of the Declaration
Different hospital-based studies yielded different risk of Helsinki. The study received clinical trial registration
factors,8,10e12 and epidemiologic studies failed to ascertain NCT02819622.
many of these risk factors.4,6 Because of different method- The interview lasted a minimum of 8 minutes and included 60
ology (retrospective vs. prospective case-control studies, potential risk factors. The interview was carried out during the

 2017 Published by Elsevier Inc. on behalf of the American Academy of http://dx.doi.org/10.1016/j.oret.2017.02.009 1

Ophthalmology ISSN 2468-6530/17
 Ophthalmology Retina Volume -, Number -, Month 2017

setting of clinical examination after the diagnosis of CSCR was aggressive behavior (P < 0.0001), presence of continuous
fully established (best-corrected visual acuity using Snellen charts; stressful conditions (P < 0.0001), work devotion (P < 0.0001),
slit-lamp examination of the anterior chamber, vitreous, and retina; history of premature ejaculation (in men) (P < 0.0001), irritable
indirect ophthalmoscopy; intravenous fluorescein angiography; bowel syndrome (P ¼ 0.001), inability to sleep (P ¼ 0.001),
spectral domain OCT; and clinical history). Cases of chronic,
use of erection disorder medication (in men) (P ¼ 0.001),
nonresolving CSCR were excluded (to not include possible cases
of polypoidal choroidal vasculopathy, especially when indocyanine history of sinusitis (P ¼ 0.004), and sleep apnea (P ¼ 0.005).
green angiography was not performed). Controls included subjects Strong associations were found with migraine (P ¼ 0.01), cold
who attended the same eye clinic during the same period with no extremities in winter (P ¼ 0.02), rheumatic disease (P ¼ 0.02),
retinal disease (more or less equally distributed into 5 categories: use of psychopharmacologic drugs (P ¼ 0.02), gastroesophageal
dry eyes, conjunctivitis, chalazion, refractive error, and conver- reflux (P ¼ 0.03), tendonitis (P ¼ 0.03), financial burden
gence insufficiency). The interviewees were given the option to (P ¼ 0.05), and history of panic attack (P ¼ 0.05).
skip questions deemed personal. Multivariate analysis revealed a strong association with
The core questions were derived, but in a truncated form, from obsessiveecompulsive behavior (P ¼ 0.001), caffeine intake
several standard tests. The Pittsburgh Sleep Quality Index was trun- (P ¼ 0.002), Type A personality (P ¼ 0.002), continuous stress
cated from 20 to 2 questions14; the Type A personality test was
(P ¼ 0.001), and premature ejaculation (P ¼ 0.001) (Table 3).
truncated from 14 items to 1 long sentence15; the Imperial Stress
assessment test was truncated to 3 short questions16; and 2 Literature review of English-written prospective controlled
questions were used instead of the 5-item version of International studies yielded 9 studies summarized in Table 4.10,18e23
Index of Erectile Function.17 Type A personality was defined to the
interviewee as competitive, outgoing, ambitious, impatient, and high
achiever. Obsessiveecompulsive behavior was defined as excessive Discussion
preoccupation with details, inflexibility, perfectionism, and extreme
devotion to work. Caffeine intake included coffee, tea, and cola. Central serous chorioretinopathy is idiopathic in nature, and
We converted tea and cola servings into equivalent cups of coffee. the literature includes contradicting reports on the risk fac-
tors. One of the first risk factors identified was having a
Statistical Analyses Type A personality.10 The onset of CSCR has been
associated with severe psychosocial stressors, such as
The data obtained were analyzed with independent t test or
ManneWhitney test for continuous variables. Categoric variables
divorce, bankruptcy, or critical illness often in patients
were compared using the chi-square or Fisher exact test. Risk with poor coping mechanisms.5 Others found an
factors initially were analyzed using univariate linear regression association with systemic hypertension,11 erectile
analysis. Those with statistical significance on univariate analysis dysfunction (in men),24 sleep apnea,7 gastroesophageal
were included in multivariate analysis using logistic regression reflux,13 and alcohol use.11 The occurrence of CSCR
with forward stepwise selection. Risk factors with small numbers seems potentiated by the administration of corticosteroids
were not included in the regression analysis. Statistical analyses (intralesional, oral, intravenous, topical eye drops, nasal
were performed using SPSS version 23.0 software (IBM, Armonk, drops),8,11,12 and to a lesser extent sympathomimetic
NY). A P value of less than 0.05 was considered statistically agents, phosphodiesterase type 5 inhibitors, and inhibitors
significant. of mitogen-activated protein kinases or extracellular signal-
Moreover, we reviewed prospective case-control studies on risk
factors for CSCR by searching PubMed from 1965 to June 2016
regulated kinases. Haimovici et al11 reported an odds ratio
for English-written articles. The review included publications with (OR) of 10.3 (95% CI, 6.2e221.8) for corticosteroid use
a good level of evidence following the same methodology outlined in their retrospective case-control study of 312 patients
in a previous meta-analysis study by Liu et al.2 with CSCR, whereas Tittl et al12 reported an OR of 3.17
(95% CI, 1.30e7.70) in their retrospective case-control
study of 230 patients with CSCR. Overactivation of the
Results hypothalamic-pituitary-adrenal (HPA) axis and sympathetic
nervous system (Cushing’s syndrome, pregnancy, obstruc-
The demographics are listed in Table 1. A total of 83 patients with tive sleep apnea) may contribute to the development of
CSCR (Table 2) were enrolled from Lebanon (33), Brazil (28), the CSCR.11,24 This is consistent with the presence of higher
United States (11), and Egypt (11). Controls included 83 subjects. levels of urine and plasma cortisol and higher plasma
The 2 groups were well balanced with regard to age, gender, and catecholamines in those with CSCR compared with con-
race. A total of 83 consecutive patients with CSCR (mean age, trols.11,24 In a retrospective case-control series of 74 patients
45.9 years; male, 80.7%) and 83 controls (mean age, 46.0 years; with CSCR, Kitzmann et al4 failed to confirm any of these
male, 80.7%) were analyzed for 60 variables. Central serous associations with CSCR, including corticosteroid use. In a
chorioretinopathy affected women at a later age than men, systematic review and meta-analysis of CSCR series pub-
usually after menopause. The mean age (standard deviation) of lished before March 2015,2 a total of 9839 patients with
men with CSCR was 44.8 years (11.3), and the mean age CSCR (391 in prospective studies or 3.9%) in 17 studies
(standard deviation) of women with CSR was 50.8 years (14.6) (7 prospective studies or 40%) were included. In that
(P ¼ 0.076) (Fig 1). Caffeine intake is depicted in Figure 2. analysis, the risk factors with statistically significant
The results are divided into 2 categories according to strength of differences found between CSCR and control groups were
the association by univariate analysis. Strong associations listed as follows: hypertension (OR, 1.7; 95% CI,
included Type A personality (P < 0.0001), 1.28e2.25), Helicobacter pylori infection (OR, 3.12; 95%
obsessiveecompulsive personality disorder (P < 0.0001), CI, 1.81e5.40), corticosteroid use (OR, 4.29; 95% CI,

2
 Mansour et al 
CSCR Risk Factors

Table 1. Demographic and Clinical Data in Patients with Central Serous Chorioretinopathy and Control Patients

Control CSCR
Variables N [ 83 N [ 83 P Value
Age Mean (SD) 45.915.7 46.012.2 0.99
Race Caucasian 77 (92.7) 79 (95.3) 0.80
African-American 6 (7.2) 4 (4.8)
Gender Male 67 (80.7) 67 (80.7) 1.00
Female 16 (19.3) 16 (19.3)
Highest degree Illiterate 2 (3.4) 1 (1.3) 0.53
School 11 (19.0) 14 (18.4)
Technical 4 (6.9) 7 (9.2)
Bachelor 30 (51.7) 39 (51.3)
Master 2 (3.4) 8 (10.5)
Doctorate 9 (15.5) 7 (9.2)
Working situation Unemployed 6 (13.3) 6 (10.3) 0.93
Blue collar 12 (26.7) 18 (31.0)
White collar 11 (24.4) 15 (25.9)
Self-employed 16 (35.6) 19 (32.8)
Smoker No 49 (59.0) 41 (49.4) 0.21
Yes 34 (41.0) 42 (50.6)
Substance abuse No 81 (97.6) 76 (92.7) 0.17
Yes 2 (2.4) 6 (7.3)
Alcohol intake No 63 (75.9) 53 (64.6) 0.11
Yes 20 (24.1) 29 (35.4)
Oral ulcers No 62 (75.6) 59 (72.8) 0.69
Yes 20 (24.4) 22 (27.2)
No. of recurrence of mouth ulcers per year Mean (SD) 0.491.12 1.012.45 0.08
Marital status Single 16 (19.3) 11 (13.6) 0.82
Married 60 (72.3) 63 (77.8)
Divorced 6 (7.2) 6 (7.4)
Widow 1 (1.2) 1 (1.2)
Introvert vs. extrovert Introvert 50 (60.2) 41 (50.6) 0.21
Extrovert 33 (39.8) 40 (49.4)
Work devotion No 48 (57.8) 25 (30.5) <0.0001
Yes 35 (42.2) 57 (69.5)
Continuous stress No 49 (59.0) 11 (13.4) <0.0001
Yes 34 (41.0) 71 (86.6)
Emotional stress No 65 (79.3) 33 (44.0) <0.0001
Yes 17 (20.7) 42 (56.0)
Financial burden No 76 (91.6) 65 (81.2) 0.05
Yes 7 (8.4) 15 (18.8)
Panic attack No 68 (81.9) 55 (68.8) 0.05
Yes 15 (18.1) 25 (31.2)
Violent behavior No 77 (92.8) 68 (85.0) 0.11
Yes 6 (7.2) 12 (15.0)
Cups of caffeine Mean (SD) 2.32.2 4.14.9 0.003
Oral corticosteroid No 78 (94.0) 75 (93.8) 1.00
Yes 5 (6.0) 5 (6.2)
Intralesional corticosteroid No 78 (94.0) 72 (88.9) 0.24
Yes 5 (6.0) 9 (11.1)
Inhaled corticosteroid No 77 (92.8) 76 (93.8) 0.79
Yes 6 (7.2) 5 (6.2)
Immunosuppressive therapy No 82 (98.8) 78 (96.3) 0.36
Yes 1 (1.2) 3 (3.7)
Aspirin intake No 76 (91.6) 73 (89.0) 0.58
Yes 7 (8.4) 9 (11.0)
Exercise No 62 (74.7) 63 (76.8) 0.75
Yes 21 (25.3) 19 (23.2)
Obesity No 74 (89.2) 66 (80.5) 0.12
Yes 9 (10.8) 16 (19.5)
Diabetes mellitus No 74 (89.2) 71 (86.6) 0.61
Yes 9 (10.8) 11 (13.4)
Systemic hypertension No 69 (83.1) 59 (72.0) 0.08
Yes 14 (16.9) 23 (28.0)

(Continued)

3
 Ophthalmology Retina Volume -, Number -, Month 2017

Table 1. (Continued.)

Control CSCR
Variables N [ 83 N [ 83 P Value
Anemia No 79 (95.2) 75 (98.7) 0.37
Yes 4 (4.8) 1 (1.3)
Dyslipidemia No 63 (75.9) 58 (70.7) 0.45
Yes 20 (24.1) 24 (29.3)
Irritable bowel syndrome No 71 (85.5) 51 (62.2) 0.001
Yes 12 (14.5) 31 (37.8)
Gastroesophageal reflux No 65 (78.3) 51 (63.0) 0.03
Yes 18 (21.7) 30 (37.0)
Gastroduodenal ulcer No 76 (91.6) 74 (91.4) 0.96
Yes 7 (8.4) 7 (8.6)
Coronary artery disease No 82 (98.8) 78 (95.1) 0.21
Yes 1 (1.2) 4 (4.9)
Tachycardia No 76 (91.6) 67 (82.7) 0.09
Yes 7 (8.4) 14 (17.3)
Cerebrovascular accident or transient repeated ischemic attack No 82 (98.8) 80 (97.6) 0.62
Yes 1 (1.2) 2 (2.4)
Migraine No 78 (94.0) 66 (81.5) 0.01
Yes 5 (6.0) 15 (18.5)
Cold extremity in winter No 82 (98.8) 73 (90.1) 0.02
Yes 1 (1.2) 8 (9.9)
Sinusitis No 82 (98.8) 71 (87.7) 0.004
Yes 1 (1.2) 10 (12.3)
Sleep apnea No 73 (88.0) 57 (70.4) 0.005
Yes 10 (12.0) 24 (29.6)
Sleep disorder (Inability to sleep) No 71 (85.5) 51 (63.0) 0.001
Yes 12 (14.5) 30 (37.0)
Asthma No 75 (90.4) 77 (95.1) 0.25
Yes 8 (9.6) 4 (4.9)
Urticaria No 80 (96.4) 76 (93.8) 0.49
Yes 3 (3.6) 5 (6.2)
Acne of face No 70 (84.3) 71 (87.7) 0.54
Yes 13 (15.7) 10 (12.3)
Kidney disease No 81 (97.6) 79 (96.3) 0.68
Yes 2 (2.4) 3 (3.7)
Kidney transplant No 82 (98.8) 82 (100.0) 1.00
Yes 1 (1.2) 0 (0.0)
Tendonitis No 83 (100.0) 76 (93.8) 0.03
Yes 0 (0.0) 5 (6.2)
Rheumatic disease No 81 (97.6) 72 (87.8) 0.02
Yes 2 (2.4) 10 (12.2)
Type A personality No 67 (80.7) 24 (29.6) <0.0001
Yes 16 (19.3) 57 (70.4)
Obsessiveecompulsive behavior No 78 (94.0) 45 (56.2) <0.0001
Yes 5 (6.0) 35 (43.8)
Phobia No 72 (86.7) 63 (77.8) 0.13
Yes 11 (13.3) 18 (22.2)
Hyperthyroidism No 82 (98.8) 79 (97.5) 0.62
Yes 1 (1.2) 2 (2.5)
Varices No 77 (92.8) 77 (95.1) 0.75
Yes 6 (7.2) 4 (4.9)
Cancer No 82 (98.8) 80 (98.8) 1.00
Yes 1 (1.2) 1 (1.2)
Aggressive vs. passive personality Aggressive 13 (15.7) 42 (52.5) <0.0001
Passive 70 (84.3) 38 (47.5)
Psychopharmacologic drugs No 80 (96.4) 70 (86.4) 0.02
Yes 3 (3.6) 11 (13.6)
Anticancer medication No 83 (100.0) 81 (100.0) —
Yes 0 0
Sympathomimetic drug No 82 (98.8) 78 (96.3) 0.36
Yes 1 (1.2) 3 (3.7)
Premature ejaculation No 54 (84.4) 29 (46.0) <0.0001
Yes 10 (15.6) 34 (54.0)

4
 Mansour et al 
CSCR Risk Factors

Table 1. (Continued.)

Control CSCR
Variables N [ 83 N [ 83 P Value
Penile erection disorder No 63 (98.4) 54 (85.7) 0.009
Yes 1 (1.6) 9 (14.3)
Erection disorder medication No 66 (98.5) 53 (81.5) 0.001
Yes 1 (1.5) 12 (18.5)
Irregular menses No 16 (100.0) 14 (87.5) 0.48
Yes 0 (0.0) 2 (12.5)
Polycystic ovaries No 16 (100.0) 14 (87.5) 0.48
Yes 0 (0.0) 2 (12.5)
CSCR during pregnancy No 16 (100.0) 15 (93.8) —
Yes 0 (0.0) 1 (6.2)
Oral contraceptives No 14 (87.5) 14 (87.5) 1.00
Yes 2 (12.5) 2 (12.5)

CSCR ¼ central serous chorioretinopathy.

2.01e9.15), sleeping disturbance (OR, 1.90; 95% CI, beta-endorphins.25 Type A personality involves a greater
1.28e1.83), autoimmune disease (OR, 3.44; 95% CI, sympathetic-adrenal response due to a basic personality/
1.90e6.26), consumption of psychopharmacologic behavior inconsistency. Along the same line and in a case-
medication (OR, 2.69; 95% CI, 1.63e4.45), and Type A control psychology study by Lahousen et al26 comparing
behavior (OR, 2.53; 95% CI, 1.08e5.96). 75 controls with 95 patients with CSCR, those with CSCR
Because of different findings in different studies as the reported unfavorable stress-coping strategies and critical
result of different designs, referral bias, and special popula- life events, as well as elevated tension, aggression, strain,
tion bias, we aimed at a more varied universal multicenter emotional instability, and achievement orientation.26
design emphasizing high-quality research and controlling for Type A personality significantly predisposes the indi-
age, gender, race, and country of living, including controls vidual to experience stress. Stress was significant in 3 pre-
without retinal disease. Our prospective study suggests some vious studies,19,20,27 with an OR of 15.3 noted in a single
major risk factors in CSCR: Type A personality, continuous study27 and an OR of 4.3 noted in the current study. Stress
stress, obsessiveecompulsive behavior, intake of caffeine, also has been linked to an exaggerated cardiovascular
and premature ejaculation (in men). Although the first 3 response contributing to cardiovascular disease.28,29 The
variables represent psychologic predispositions, the other 2 response to psychosocial stress is influenced by both psy-
are correlated problems with a psychologic component. chosocial factors and genetic vulnerability. The most
Type A personality is described as a constellation of investigated gene in an abnormal response to environmental
behavioral dispositions involving competitiveness, a strong stressors is the one coding for the serotonin transporter
sense of time urgency, and hostile and aggressive tendencies. (SLC6A4).28,29 Depletion of serotonin is common in dis-
Type A personality was a significant correlate of CSCR in 3 orders with anxiety as a core symptom, such as obsessivee
studies,10,19,20 with an OR of 4.2 in our study. Type A per- compulsive disorder, depression, and social phobia.
sonality predisposes subjects to a greater reactivity with Variability within the SLC6A4 gene has been associated
higher baseline and post-stress levels of norepinephrine and with reactivity to stress and risk for various psychopatho-
logic conditions, including irritable bowel syndrome28 and
some types of migraine.29
Table 2. Ophthalmic Characteristics of Patients With Central Obsessiveecompulsive personality disorder is one of the
Serous Chorioretinopathy most common personality disorders in the general popula-
CSCR
tion, with a prevalence of approximately 10%.30
Variables N [ 83 Obsessiveecompulsive personality disorder was found to
be a significant risk factor for the development of CSCR
Eye laterality Right eye 41 (49.4)
in 2 previous studies19,20 (Table 4). In the current study,
Left eye 25 (30.1)
Bilateral 16 (19.3) obsessiveecompulsive tendencies also were linked to
Undetermined 1 (1.2) CSCR.31 Blom et al31 found an association between
Duration of CSCR attack (days) Mean (SD) 28.664.1 obsessiveecompulsive personality disorder and the
No. of CSCR attacks Mean (SD) 1.51.3 SLC6A4 gene located on chromosome 17q11.1e17q12,
Initial logMAR visual acuity Mean (SD) 0.480.40 also responsible for selective serotonin reuptake. Although
Final logMAR visual acuity Mean (SD) 0.290.36 obsessiveecompulsive personality disorder and
Mean follow-up (mos) Mean (SD) 18.928.2
obsessiveecompulsive disorder are recognized as distinct
clinical syndromes, research has shown that they
CSCR ¼ central serous chorioretinopathy; logMAR ¼ logarithm of the significantly overlap.31,32 A related overlapping disorder,
minimum angle of resolution; SD ¼ standard deviation.
obsessiveecompulsive disorder, was found to be associated

5
 Ophthalmology Retina Volume -, Number -, Month 2017

Table 3. Multivariate (Stepwise) Analysis for the Predictors of

Central Serous Chorioretinopathy

Variables* OR (95% CI) P Value

Continuous stress 4.73 (1.84e12.14) 0.001
Cups of caffeine 1.26 (1.09e1.46) 0.002
Type A personality 4.41 (1.76e11.05) 0.002
Obsessiveecompulsive behavior 11.52 (2.88e46.05) 0.001
Premature ejaculation 4.75 (1.85e12.20) 0.001

CI ¼ confidence interval; OR ¼ odds ratio.

*Additional variables included in the model that were not significant
were work devotion, irritable bowel syndrome, sleep disorder, and
aggressiveepassive behavior.

Figure 1. Gender and age distribution of current central serous chorior-

etinopathy (CSCR) series. Men have a peak at 49 years, and women have a human sexual psychopharmacologic studies have found
peak at 60 years. that serotonin (5-hydroxy-tryptamine) and its receptors are
involved in ejaculation.38
Corticosteroid use was significant in 3 studies, with an OR
with the SLC6A4 gene and the catechol-O- of 5.5 and 7.3 in 2 studies,27,39 whereas no significance was
methyltransferase (COMT) gene (rs 4680), involved in detected in our study. Irritable bowel syndrome, sleep apnea,
catecholamine modulation. One meta-analysis33 showed and alcohol or tobacco use were not recognized in the current
that obsessiveecompulsive disorder was associated study to be significant risk factors. In one study, subjects with
specifically with serotonin-related polymorphisms CSCR were significantly prone to QT dispersion compared
(5-HTTLPR coded as triallelic and HTR2A rs6311/rs6313) with controls.21 QT dispersion is common in subjects with
and, in men, a polymorphism involved in catecholamine acute sleep deprivation or panic attacks, and subjects with
modulation, COMT (rs4680). The COMT gene also mod- QT dispersion are prone to ventricular arrhythmias. Other
ulates the human biologic response to stress. described possible risk factors derived from retrospective
Caffeine intake is a relatively new variable in CSCR studies included systemic hypertension, coronary artery
(Table 4). Caffeine is known to activate the HPA axis,34 disease, ischemic stroke, gastroesophageal reflux, gastric
increasing the secretion of stress hormones, such as cortisol, ulcer, and higher monthly income.1,6,13,24
in a gender-dependent manner.35 Caffeine causes most of its Most of the past information on CSCR has been based on
biological effect via antagonizing all types of adenosine hospital-based studies limited by a retrospective nature of
receptors. Moreover, caffeine may block adenosine receptors the research, potential referral bias, self-administered ques-
that may be involved in absorption of subretinal fluid.36 The tionnaires,10,19,20,27,39 controls with retinal disease18,39 or
detrimental effect of caffeine on health is set at 3 cups without retinal examination,19,20 and unavailability of
(w300 mg caffeine per day).37 Caffeine is a psychoactive spectral domain OCT imaging for diagnosis and manage-
agent that enhances anxiety and panic attack.37 ment. The present structured, standardized, prospective,
Premature ejaculation was a significant risk factor found open-ended interview administered by the trusted caring
in our study (in men), but was not observed in the other physician offers several advantages over some of the past
previous prospective case-control studies (Table 4). It similar publications: (1) The interview allows quantitative
appears that anxiety activates the sympathetic nervous research from the resulting high reliability and accuracy due
system and reduces the ejaculatory threshold as a result of to the special patientephysician relationship; (2) the inter-
an earlier emission phase of ejaculation.30 Animal and view is presented with exactly the same questions in the
same order ensuring that answers can be reliably aggregated
and compared between groups; (3) compared with a written
questionnaire, interviews allow for a significantly higher
degree of intimacy, with participants often revealing
personal information to their interviewers in a real-time,
face-to-face setting; (4) the interviewer can explain ques-
tions that the respondent has not understood; (5) when
questioned by physician assistants, weaknesses arise during
the interview from patients withholding information often
because of distrust40; (6) interviews by a sympathetic
authoritative physician allow the respondent to answer a
long list of questions versus a small number of items in
a questionnaire; (7) prospective studies offer more
reliable and complete data than retrospective studies, with
Figure 2. Caffeine intake in subjects with CSCR versus controls. Some a more even balancing of related variables; and (8) we
subjects with CSCR tend to abuse caffeine intake. used spectral domain OCT along with intravenous

6
 Table 4. Literature Review of Prospective Controlled Studies on Central Serous Chorioretinopathy Risk Factors

First Author Mansour Bousquet Dagli Conrad Roshani Misiuk-Hojło Conrad Karadimas Carvalho-Recchia Yannuzzi
City, Country Lebanon, Brazil, Paris, France Elazig, Turkey Bonn, Germany Tehran, Iran Wroclaw, Poland Bonn, Germany Athens, Greece Manhattan, New Manhattan, New
United States, York, United States York, United States
Egypt
Year 2016 2016 2015 2014 2014 2009 2007 2004 2001 1986
No. of patients with 83 40 30 57 35 55 31 38 50 110

Mansour et al
CSCR
No. of controls 83 40 30 57 138 55 31 38 50 110
% of men in CSCR 80.7 85 66.7 78.9 91.4 65.5 80.6 73.7 72 80.9
series
Mean age CSCR 45.9 44 42.0 46.8 34.1 49.2 44.8 44.7 55 42.3
patients
Systemic hypertension NS NS NA NA NA NA NA NA NA NA


Tobacco use NS NS NA NA NA NA NA NA NA NA

CSCR Risk Factors

Alcohol use NS NS NA NA NA NA NA NA NA NA
Sleep apnea NS NS NA NA NA NA NA NA NA NA
Stress or distress OR 4.3 OR 15.3 NA SS NA NA SS NA NA NA
Premature ejaculation OR 4.2 NA NA NA NA NA NA NA NA NA
Corticosteroid use NS OR 5.5 NA NA NA NA NA OR 7.3 SS NA
Irritable bowel NS NA NA NA NA NA NA NA NA NA
syndrome
Helicobacter pylori NA NA NA NA OR 4.9 OR 2.3 NA NA NA NA
infection
Type A personality OR 4.2 NA NA SS NA NA SS NA NA SS
Shift work NA OR 5 NA NA NA NA NA NA NA NA
Obsessiveecompulsive OR 11.1 NA NA SS NA NA SS NA NA NA
behavior
Phobia NS NA NA SS NA NA NS NA NA NA
QT dispersion NA NA SS NA NA NA NA NA NA NA

CSCR ¼ central serous chorioretinopathy; NA ¼ not analyzed; NS ¼ not significant; OR ¼ odds ratio; SS ¼ statistically significant with no OR values listed.
7
 Ophthalmology Retina Volume -, Number -, Month 2017

Figure 3. Proposed pathophysiology of CSCR. Catechol-O-methyltransferase (COMT) gene regulates response to stress. ACTH ¼ adrenocorticotropic
hormone; CRF ¼ corticotropin-releasing factor; RPE ¼ retinal pigment epithelium.

fluorescein angiography to rule out mimickers, such as ulcer and CSCR; thus, the same relationship probably can be
hypertensive chorioretinopathy, retinal pigment epithelitis, applied to H. pylori and CSCR. Moreover, caution needs to
paraproteinemic maculopathy, and others.1,5 be exercised on any causality observed in cross-sectional
Obviously, our study suffers from a nonvalidated ques- studies. Our study and many previous studies based their
tionnaire, interviewer bias, racial bias (mainly Caucasian), a data on a questionnaire or an interview and, therefore, suffer
small sample size in each center, and the inclusion of a large from lack of proper psychologic assessment.
number of risk factors (comprehensive manner), yet each The mechanisms behind CSCR manifesting in the
risk factor being investigated in a concise manner lacked choroid are still vague. Several mechanisms have been
in-depth analysis. We did not include a recently described proposed, such as1e3 (1) choroidal vascular stasis and
risk factor (shift work) and H. pylori infections, which are hypoperfusion from direct modulation of the choroidal
known to cause more than 90% of gastroduodenal ulcers.38 blood flow by the autonomic nervous system or from
Indirectly, we found no correlation between gastroduodenal elevated serum cortisol levels; decreases in choroidal blood

8
 Mansour et al 
CSCR Risk Factors

flow are mediated by activation of sympathetic efferent the potential to arouse the adrenomedullary-sympathetic
nerves that release noradrenaline, activating alpha system to levels that predispose individuals to disease
1-adrenoceptors on vascular smooth muscle cells; (2) states, such as coronary artery disease and CSCR.1 Because
damage of the retinal pigment epithelium (RPE) cells by there are relatively few unproven therapy options for acute,
elevated levels of epinephrine, leading to apoptosis, RPE relapsing, and chronic CSCR, we propose to search for
decompensation, or RPE detachment; (3) RPE cells express ways to reduce sympathetic arousal by using stress-
the most common genes involved in the hypothalamic- management strategies to decrease serum cortisol levels
pituitary axis, leading to the production of cyclic- (and taper systemic corticosteroid); psychotherapeutic tech-
adenosine-monophosphate and adrenocorticotrophic niques to reduce sympathetic tone, such as autogenic relax-
hormone; cyclic-adenosine-monophosphate is depleted in ation19,20; and limited mood-altering oral therapies to help
CSCR; thus, there is less ability of RPE to absorb the with acute stressful events, as well as reducing caffeine
subretinal fluid; (4) hypercoagulability from elevated cyto- intake (and energy drinks).
kines and plasminogen activator and enhanced platelet ag-
gregation in CSCR41; and (5) dilated large choroidal vessels Conclusions
cause mechanical compartmental focal compression of the
choriocapillaris with secondary RPE decompensation.1 It The findings on the psychologic correlates of CSCR suggest
is tempting to compare choroidal swelling with colonic that it would be beneficial for health professionals to
distension in irritable bowel syndrome. incorporate assessment tools aimed at uncovering person-
The HPA axis, regulated by the corticotrophin-releasing ality processes in certain individuals that make them prone
factor, is the major centrally regulated neuroendocrine sys- to develop CSCR. Given the significant role of hostile and
tem responsible for rapid and strong responses to stress. It acts anxious personality dispositions in CSCR (Type A person-
through a cascade of brain and hormonal events that ulti- ality, stress, and obsessiveecompulsive behavior), as well
mately result in heightened release of glucocorticoids, with as maladaptive behaviors (caffeine abuse), psychologic
cortisol being an indicator of HPA axis responsivity to stress. therapy could focus on providing patients with adaptive
Current evidence supports the view that dysregulation of coping mechanisms to face life stressors. A psychosocial
brain corticotrophin-releasing factor systems and HPA axis approach would be advisable, whereby interpersonal
function is hereditary and increases with age, especially with relationships and social support would be pivotal factors in
sustained exposure to major stressors.42 Individual the therapeutic process.
differences in the physiologic response to stress recently
have emerged as a potentially important risk factor for these References
disorders. This raises the possibility that sex differences in
prevalence of disease (e.g., CSCR) could be explained at 1. Daruich A, Matet A, Dirani A, et al. Central serous chorior-
least in part by sex differences in the nature of the etinopathy: recent findings and new physiopathology hypoth-
physiologic response to stress. Between puberty and esis. Prog Retin Eye Res. 2015;48:82e118.
2. Liu B, Deng T, Zhang J. Risk factors for central serous
menopause, adult women usually show lower HPA and
chorioretinopathy: a systematic review and meta-analysis.
autonomic responses than adult men.25 Two major enzymes Retina. 2016;36:9e19.
are responsible for catecholamine catabolism in the brain: 3. Nicholson B, Noble J, Forooghian F, Meyerle C. Central
COMT and monoamine oxidase A. The transcriptional serous chorioretinopathy: update on pathophysiology and
activity of the genes coding for these enzymes is governed treatment. Surv Ophthalmol. 2013;58:103e126.
by common functional polymorphisms. Catechol-O- 4. Kitzmann AS, Pulido JS, Diehl NN, et al. The incidence of
methyltransferase and monoamine oxidase A poly- central serous chorioretinopathy in Olmsted County, Minne-
morphisms represent a basic neurobiologic mechanism that sota, 1980e2002. Ophthalmology. 2008;115:169e173.
contributes to the regulation of stress in passive-aggressive or 5. Wang M, Munch IC, Hasler PW, et al. Central serous
Type A personality. We hypothesize that the predilection of chorioretinopathy. Acta Ophthalmol. 2008;86:126e145.
6. Tsai D-C, Chen S-J, Huang C-C, et al. Epidemiology of idio-
CSCR and several stress disorders in men and post-
pathic central serous chorioretinopathy in Taiwan, 2001e2006:
menopausal women is related to COMT gene expression that a population-based study. PLoS One. 2013;8:e66858.
is altered by estrogen, with most women affected by CSCR 7. Yavaş GF, Küsbeci T, Kaşikci M, et al. Obstructive sleep
during pregnancy or more often after menopause. apnea in patients with central serous chorioretinopathy. Curr
It is hypothesized that psychologic traits that contribute to Eye Res. 2014;39:88e92.
difficulty in emotional regulation (obsessiveecompulsive, 8. Bouzas EA, Karadimas P, Pournaras CJ. Central serous cho-
Type A personality) coupled with or provoked by acute or rioretinopathy and glucocorticoids. Surv Ophthalmol. 2002;47:
chronic episodes of sympathetic arousal (tragic event, sleep 431e448.
apnea, shift work, corticosteroid intake) manifest in various 9. Spahn C, Wiek J, Burger T, Hansen L. Psychosomatic aspects
forms, such as systemic hypertension, irritable bowel syn- in patients with central serous chorioretinopathy. Br J
Ophthalmol. 2003;87:704e708.
drome, erectile dysfunction, and CSCR (Fig 3). Choroidal 10. Yannuzzi LA. Type A behavior and central serous chorior-
swelling appears to be the common pathway toward etinopathy. Trans Am Ophthalmol Soc. 1986;84:799e845.
leakage into the subretinal space. Psychologic factors are 11. Haimovici R, Koh S, Gagnon DR, et al. Central Serous
known to affect blood pressure, heart rate, serum lipids, Chorioretinopathy Case-Control Study Group. Risk factors for
and various hormones. In Type A personality trait, an central serous chorioretinopathy: a case-control study.
intense and sustained level of this behavioral pattern has Ophthalmology. 2004;111:244e249.

9
 Ophthalmology Retina Volume -, Number -, Month 2017

12. Tittl MK, Spaide RF, Wong D, et al. Systemic findings 28. Areeshi MY, Haque S, Panda AK, Mandal RK. A serotonin
associated with central serous chorioretinopathy. Am J Oph- transporter gene (SLC6A4) polymorphism is associated with
thalmol. 1999;128:63e68. reduced risk of irritable bowel syndrome in American and
13. Mansuetta CC, Mason JO, Swanner J, et al. An association Asian population: a meta-analysis. PLoS One. 2013;8:
between central serous chorioretinopathy and gastroesopha- e75567.
geal reflux disease. Am J Ophthalmol. 2004;137:1096e1100. 29. Schürks M, Rist PM, Kurth T. 5-HTTLPR polymorphism in
14. Buysse DJ, Reynolds CF, Monk TH, et al. The Pittsburgh the serotonin transporter gene and migraine: a systematic re-
Sleep Quality Index: a new instrument for psychiatric practice view and meta-analysis. Cephalalgia. 2010;30:1296e1305.
and research. Psychiatry Res. 1989;28:193e213. 30. Quirk SE, Berk M, Pasco JA, et al. The prevalence, age
15. Bortner RW. A short rating scale as a potential measure of distribution and comorbidity of personality disorders in
pattern a behavior. J Chronic Dis. 1969;22:87e91. Australian women. Aust N Z J Psychiatry. 2017;51:141e150.
16. Arora S, Tierney T, Sevdalis N, et al. The Imperial Stress 31. Blom RM, Samuels JF, Riddle MA, et al. Association between
Assessment Tool (ISAT): a feasible, reliable and valid a serotonin transporter promoter polymorphism (5HTTLPR)
approach to measuring stress in the operating room. World J and personality disorder traits in a community sample.
Surg. 2010;34:1756e1763. J Psychiatr Res. 2011;45:1153e1159.
17. Rosen RC, Cappelleri JC, Smith MD, et al. Development and 32. Gordon OM, Salkovskis PM, Oldfield VB, Carter N. The
evaluation of an abridged, 5-item version of the International association between obsessive compulsive disorder and
Index of Erectile Function (IIEF-5) as a diagnostic tool for obsessive compulsive personality disorder: prevalence and
erectile dysfunction. Int J Impot Res. 1999;11:319e326. clinical presentation. Br J Clin Psychol. 2013;52:300e315.
18. Carvalho-Recchia CA, Yannuzzi LA, Negrão S, et al. Corti- 33. Taylor S. Disorder-specific genetic factors in obsessive-
costeroids and central serous chorioretinopathy. Ophthal- compulsive disorder: a comprehensive meta-analysis. Am J
mology. 2002;109:1834e1837. Med Genet Part B Neuropsychiatr Genet. 2016;171:
19. Conrad R, Weber NF, Lehnert M, et al. Alexithymia and 325e332.
emotional distress in patients with central serous chorioretin- 34. Lovallo WR, Whitsett TL, Al’Absi M, et al. Caffeine stim-
opathy. Psychosomatics. 2007;48:489e495. ulation of cortisol secretion across the waking hours in
20. Conrad R, Geiser F, Kleiman A, et al. Temperament and relation to caffeine intake levels. Psychosom Med. 2005;67:
character personality profile and illness-related stress in central 734e739.
serous chorioretinopathy. Sci World J. 2014;2014:631e687. 35. Kudielka BM, Kirschbaum C. Sex differences in HPA axis
21. Dagli N, Turgut B, Tanyildizi R, et al. QT interval dispersion responses to stress: a review. Biol Psychol. 2005;69:113e132.
in the patients with central serous chorioretinopathy. Int J 36. Vural AD, Kara N, Sayin N, et al. Choroidal thickness changes
Ophthalmol. 2015;8:61e65. after a single administration of coffee in healthy subjects.
22. Misiuk-Hoj1o M, Micha1owska M, Turno-Krecicka A. Heli- Retina. 2014;34:1223e1228.
cobacter pylori- a risk factor for the development of the central 37. Vilarim MM, Rocha Araujo DM, Nardi AE. Caffeine chal-
serous chorioretinopathy. Klin Oczna. 2009;111:30e32. lenge test and panic disorder: a systematic literature review.
23. Roshani M, Davoodi NA, Seyyedmajidi MR, et al. Associa- Expert Rev Neurother. 2011;11:1185e1195.
tion of Helicobacter pylori with central serous chorioretinop- 38. McMahon CG. Premature ejaculation. Indian J Urol. 2007;23:
athy in Iranian patients. Gastroenterol Hepatol Bed Bench. 97e108.
2014;7:63e67. 39. Karadimas P, Bouzas EA. Glucocorticoid use represents a risk
24. Tsai D-C, Huang C-C, Chen S-J, et al. Increased risk of factor for central serous chorioretinopathy: a prospective,
erectile dysfunction among males with central serous cho- caseecontrol study. Graefe’s Arch Clin Exp Ophthalmol.
rioretinopathy- a retrospective cohort study. Acta Ophthalmol. 2004;242:800e802.
2013;91:666e671. 40. Larkin GL, Hooker RS. Patient willingness to be seen by
25. McMurray RG, Hardy CJ, Roberts S, et al. Neuroendocrine physician assistants, nurse practitioners, and residents in the
responses of type A individuals to exercise. Behav Med. emergency department: does the presumption of assent have
1989;15:84e92. an empirical basis? Am J Bioeth. 2010;10:1e10.
26. Lahousen T, Painold A, Luxenberger W, et al. Psychological 41. Shin MC, Lim JW. Concentration of cytokines in the aqueous
factors associated with acute and chronic central serous cho- humor of patients with central serous chorioretinopathy.
rioretinopathy. Nord J Psychiatry. 2016;70:24e30. Retina. 2011;31:1937e1943.
27. Bousquet E, Dhundass M, Lehmann M, et al. Shift work: a risk 42. Franz CE, York TP, Eaves LJ, et al. Genetic and environ-
factor for central serous chorioretinopathy. Am J Ophthalmol. mental influences on cortisol regulation across days and con-
2016;165:23e28. texts in middle-aged men. Behav Genet. 2010;40:467e479.

Footnotes and Financial Disclosures

4
Originally received: February 5, 2017. Department of Ophthalmology, Universidade Estadual de Londrina,
Accepted: February 22, 2017. Londrina, Brazil.
Available online: ---. Manuscript no. ORET_2017_35. 5
Department of Ophthalmology, University of Arkansas for Medical Sci-
1
Department of Ophthalmology, American University of Beirut, Beirut, ences, Little Rock, Arkansas.
Lebanon. 6
Department of Ophthalmology, Mansoura University, Mansoura, Egypt.
2
Department of Ophthalmology, Rafic Hariri University Hospital, Beirut, 7
Biostatistics Unit, Clinical Research Institute, American University of
Lebanon.
Beirut, Beirut, Lebanon.
3
Department of Ophthalmology and Visual Sciences, Federal University of 8
Psychology Department, Lebanese American University, Byblos,
Sao Paulo, Sao Paulo, Brazil. Lebanon.

10
 Mansour et al 
CSCR Risk Factors
9
Department of Ophthalmology, Memorial Medical Center, Springfield, Manuscript preparation: Mansour, Koaik, Lima, Casella, Uwaydat, Shahin,
Illinois. Tamim, Sanchez-Ruiz, Mansour, Dodwell
Financial Disclosure(s): Abbreviations and Acronyms:
The author(s) have no proprietary or commercial interest in any materials CI ¼ confidence interval; COMT ¼ catechol-O-methyltransferase;
discussed in this article. CSCR ¼ central serous chorioretinopathy; HPA ¼ hypothalamic-pituitary-
Author Contributions: adrenal; OR ¼ odds ratio; RPE ¼ retinal pigment epithelium.
Research design: Mansour Correspondence:
Data acquisition and/or research execution: Mansour, Koaik, Lima, Casella, Ahmad M. Mansour, MD, Department of Ophthalmology, AUB- POB 113-
Uwaydat, Shahin, Tamim, Sanchez-Ruiz, Mansour, Dodwell 6044, Beirut, Lebanon. E-mail: ammansourmd@gmail.com.
Data analysis and/or interpretation: Mansour, Koaik, Lima, Casella, Shahin,
Tamim, Sanchez-Ruiz, Mansour, Dodwell

11