Wait-and-see and other factors associated with recurrence risk in central serous chorioretinopathy patients: A systematic review and meta-analysis of cohort studies

Protocol and registration

We registered a protocol prior to authoring this review and meta-analysis, and it was registered in the International prospective register of systematic reviews (PROSPERO) on April 13, 2023 (CRD42023413852).

Study design

Our study utilizes a systematic review design method and conducts a meta-analysis of cohort studies. The clinical research in this study focuses on prognosis, specifically investigating various exposures to recurrence in CSC patients. Subsequently, a meta-analysis is performed on a homogenous type of study. We specifically opted for cohort studies as they provide the highest level of evidence for prognostic studies, alongside systematic reviews and meta-analyses of cohort studies.

Eligibility criteria

The inclusion criteria used in this study were: (a) Cohort research study type, (b) Suitability to the topic of recurrence risk in central serous chorioretinopathy patients, (c) The study had at least one control group and one exposure group, and (d) clear extraction and statistical analysis method.

The exclusion criteria used in this study were: (a) Full text was not available, and (b) Use of languages other than English.

Search strategy

This study used the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines.¹³ Four electronic databases were used including PubMed, ScienceDirect, Embase and Cochrane. We used the term “((Central serous chorioretinopathy) AND (Recurrence) AND (Risk))”.

Study selection and data extraction

Four researchers screened the title and abstract independently and excluded the irrelevant studies. After conducting title and abstract screening, four authors (AAV, AD, ARY, KAS) independently evaluated each entry based on the inclusion and exclusion criteria, responding with ‘Yes’, ‘No’, or ‘Maybe’. Any discrepancies were resolved through discussion and mutual agreement. The final retrieved studies were screened for duplicates and systematically evaluated according to the inclusion and exclusion criteria. The subject characteristics (central serous chorioretinopathy patients), outcome of the study (recurrence risk), and other relevant information were assessed by the four researchers, each researcher worked independently. The search was conducted using a systematic review method using the PubMed, Cochrane, and ScienceDirect databases. The keywords “((Central serous chorioretinopathy) AND (Recurrence) AND (Risk))” were used. After synthesizing the various factors that cause recurrent central serous chorioretinopathy, a meta-analysis was carried out using the ReviewManager, Revman 5.4 (RRID: SCR_003581), application related to the wait-and-see factor for the risk of recurrence in CSC patients. The use of the term ‘wait-and-see’ refers to CSC patients who do not receive therapeutic interventions in the form of photodynamics, anti-vascular endothelial growth factor (VEGF), and laser therapy, or a combination of them.¹⁴ In our study, the measured effect is the risk ratio of wait-and-see compared to therapy, along with a 95% confidence interval, regarding the recurrence events of CSC patients. This will be combined through a forest plot using data from various studies. Additionally, the effect of adjusting measurement factors on CSC recurrence will be discussed in the study.

Risk of bias and certainty assessment

Researchers assessed each risk of bias from the retrieved study using the ReviewManager (RevMan 5.4; Cochrane, London, UK) based on the Cochrane Risk of Bias for cohort studies. Each study was classified as having a low, unclear, or high risk of bias. Any dissent was discussed until a final decision was reached. The GRADE (grading of recommendations assessment, development and evaluation) approach was employed to evaluate the certainty of evidence in this study. The assessment was conducted independently by all authors, and any disagreements among assessors were resolved through discussions among them.

Study selection

A total of 625 studies were retrieved based on the keywords from PubMed (n = 50), ScienceDirect (n = 467), Embase (n = 107) and Cochrane Library (n = 1). After screening the abstract and title, 620 studies were excluded. The studies were then compared for duplicates and no studies were excluded. The remaining five studies were screened for the inclusion and exclusion criteria. The final screening resulted in five studies that met the inclusion and exclusion criteria. The flowchart of the study selection can be seen in Figure 1.

Figure 1. Article search and selection flowchart.

Study characteristics and outcome

In this systematic review, a total of five studies met the inclusion criteria. A description of the search results can be seen in Table 1.^14–18

A meta-analysis was also performed based on three studies involving 1025 eyes. There was a statistically significant difference between CRC patients who did not receive therapy or were told to wait and see and those who received specific therapeutic interventions, with an RR of 1.82 (95% CI 1.49–2.22 I2 95%, P < 0.00001). The forest plot can be seen in Figure 2.

Figure 2. Forest plot for three studies.

Risk of bias and certainty

Figure 3 shows details of the risk of bias of the included studies. Figure 3a is a plot of risk of bias according to the Cochrane risk of bias tool for cohort studies. Figure 3b is a summary table of the risk of bias of each study. The study by Yu et al.,¹⁵ has a high risk of bias in assessment of exposure due to its high-risk self-reporting bias method. The study by Mohabati et al.¹⁴ has a high risk of bias in adjusting the possible prognostic factors. Three studies by Mohabati et al.,¹⁴ Yu et al.,¹⁵ and Zhou et al.¹⁸ have a unclear method in outcome assessment due to unclear blinding methods. The study by Mohabati et al.¹⁴ also has an unclear co-intervention bias. The evaluation of evidence certainty for all outcomes in this meta-analysis yielded a moderate level of certainty.

Figure 3. Risk of bias for the included studies.

(a) Plot of risk of bias per category; (b) Summary table of risk of bias for each study.

Summary of main results

In this systematic review, five studies demonstrated the recurrence of CSC in patients in various populations. These five studies used a total of 1,325 patients with CSC including patients from the Netherlands, China, Hong Kong, and Japan.

The first study from Mohabati et al. (2020) was a prospective cohort study with a minimum follow-up of 12 months. This study showed that 29% of patients experienced a recurrence of symptoms of CSC. Of the population that experienced recurrences, 24% of them were patients who were not treated immediately or were advised to wait-and-see, whereas in patients who underwent initial therapy, only 4% experienced recurrences. Initial therapy for patients included photodynamic therapy, conventional lasers, and subthreshold micropulse diode lasers.¹⁴ Therefore, patients with CSC had a better outcome if they had initial therapy with these three options compared to no therapy or wait and see. The next study by Yu et al. (2019), which was a retrospective cohort study providing questions in the form of questionnaires to 138 patients relating to various risk factors for CSC. Of all the patients, 85.5% had recurrent or persistent conditions. Various factors that were significantly associated with recurrence or persistence compared to only acute conditions included male sex (OR = 5.63), older age (OR = 1.14), and a higher Insomnia Severity Index score with an odds ratio of 1.30.¹⁵ Therefore, this second study showed that male sex, age, and degree of insomnia are risk factors for the recurrence of CSC in patients.

The study by Lai et al. (2015) was a retrospective cohort study comparing two groups of patients, namely a group that received half-dose verteporfin photodynamic therapy (PDT) and the group that did not receive any therapy. It was found that the group of patients without therapy had a higher percentage of recurrences than those with the therapy (53.8% vs 20%).¹⁶ Therefore, this study showed an increased risk of recurrence of CSC patients without therapy. The next study by Zhang et al. (2022) was a prospective cohort study of 162 patients with CSC and 19.14% of patients had recurrences of CSC. Various factors were significantly associated (P < 0.001) with recurrence according to this study including pigment epithelium detachment (OR = 1.78), retinal pigment epithelium damage (OR = 1.13), baseline best-corrected visual acuity (OR = 0.96), and subfoveal choroidal thickness level (OR = 1.18).¹⁷

The last study by Zhou et al. (2022), a retrospective cohort study of 538 CSC patients, showed that recurrence had a significant relationship with corticosteroid use history (OR = 5.52), bilateral disease (OR = 3.94), chronic manifestations (OR = 7.12), non-intense fluorescein leakage (OR = 3.34) and initial central retinal thickness (OR = 0.997).¹⁸ This study showed four risk factors for recurrence and one factor that can reduce the risk of recurrence in CRC patients, namely initial central retinal thickness.

The meta-analysis study was carried out based on three studies comparing CSC patients with the wait and see treatment, which did not provide specific interventions to patients and controls in the form of patients receiving certain therapies such as PDT, conventional laser, micropulse laser, and anti-VEGF. The first study by Lai et al. (2015) showed a significant risk of wait and see with an RR of 2.69 (95% CI 1.66–4.36).¹⁶ The second study by Mohabati et al. (2020) also showed a significant risk with an RR of 5.33 (95% CI 2.93–9.68).¹⁴ The last study by Zhou et al. (2022) showed non-significant results with an RR of 1.04 (95% CI 0.82–1.31). However, after calculations from the forest plot, a statistically significant risk was found with an RR of 1.82 (95% CI 1.49–2.22 I2 95%, P < 0.00001).¹⁸ Therefore, patients with CRC should be given an alternative therapy to avoid recurrence.

Discussion of other studies

The probability of CSC recurrence is strongly correlated with increased choroidal thickness, according to a recent study by Matet et al.¹⁹ Although there weren’t enough choroidal thickness measurements in our data set to evaluate this apparent association, we noticed a thicker choroid was commonly found in patients with recurrent CSC. When compared to the wait-and-see group, subretinal fluid (SRF) resolved more quickly in the treatment group, and the prevalence of SRF recurrence was lower in the treatment group.¹⁴ Moreover Ozkaya et al. observed a two-fold greater probability of recurrence in 41 wait-and-see treatments in acute CSC (aCSC) patients (51%) than in 36 aCSC patients who received low-fluence PDT treatment.²⁰ Therefore, to prevent the disease re-occurring it is better for patients to be given initial therapy compared to no therapy especially observation only.

Overall discussion based on main result

Our study showed that there are four broad stroke risk factors for the recurrence of CSC in patients, including the patient’s clinical condition, therapy administration, age and gender factors, and patient history factors. The risk factors for the patient’s clinical condition include pigment epithelium detachment, retinal pigment epithelium damage, baseline best-corrected visual acuity, subfoveal choroidal thickness level, chronic manifestations, bilateral disease, initial central retinal thickness, and non-intense fluorescein leakage.^17,18 The recurrence factor based on patient therapy, namely patients who receive initial therapy in the form of photodynamic therapy, conventional lasers, and subthreshold micropulse diode lasers, can reduce the risk of recurrence in CRC patients.^14,16 Furthermore, the risk based on the patient’s age and gender, namely male sex and older age, can increase the incidence of recurrence in CRC patients.¹⁵ Finally, patient history factors of CRC recurrence include a higher degree of insomnia and history of corticosteroid use.^15,18 Sleep disruptions have been linked to heightened functioning of the hypothalamic-pituitary-adrenal axis and the sympathetic-adrenal medullary system, leading to changes in the release of stress hormones such as cortisol and catecholamines.²¹ These hormonal imbalances have been identified as contributing factors to the underlying mechanisms of CSC.²²

Overall completeness and quality of evidence

Our study was a two-stage review consisting of a systematic review that synthesized five cohort studies regarding factors that influence the recurrence of CSC patients. The five studies were divided into three studies with a retrospective cohort design and two studies with a prospective cohort design. We subsequently performed a meta-analysis of three studies comparing the recurrence risk of patients on wait and see treatment and patients receiving initial therapy. The three studies consisted of two retrospective cohort studies and one prospective cohort study. In terms of level of evidence, cohort studies are a good reference for studies with prognostic question types. Our study also conducted a risk of bias analysis, which showed that there were studies that had low, high, and an unclear risk of bias in the five studies reviewed.

Applicability of evidence

The study found results showing that the recurrence of CSC increases if patients only receive treatment in the form of wait and see compared to patients receiving various initial therapies including laser, anti-VEGF, and photodynamics. Therefore, as the application of this evidence, doctors who have patients with CSC should recommend patients for initial therapy rather than suggesting wait and see treatment to patients, which can risk CSC recurrence.

Limitation and potential bias

Our study certainly has limitations. Several studies had a high risk of bias and some aspects were unclear. Our study also performed a meta-analysis but only comparing wait and see management with initial treatment of CSC patients. It is hoped that other studies will perform a meta-analysis of various other recurrence factors.