Optical coherence tomography findings in bipolar disorder - A hospital-based case-control study

Abstract

Background:

Studies on bipolar disorder indicate the existence of structural brain abnormalities. There is a paucity of Indian studies in this area.

Aim:

To evaluate the retinal nerve fiber layer (RNFL), macular, and subfoveal choroidal thickness (SFCT) in patients with bipolar disorder type-1 (BD-1) and healthy controls.

Materials and Methods:

This was a cross-sectional hospital-based study in which 49 patients of BD-1 were in different phases of illness, and an equal number of normal healthy controls were taken. All the cases and controls were subjected to RNFL, ganglion cell layer (GCL), and estimation of macular thickness using spectral-domain optical coherence tomography (SD-OCT) and choroid layer thicknesses manually using the depth imaging mode of SD-OCT.

Results:

The mean age of patients and the control group was 39.55 ± 5.71 and 37.91 ± 6.17 years, respectively. The mean duration of illness in different phases was 14.55 ± 10.71 years. The total number of attacks in the mania group was 4.40 ± 3.30; in the depression group, it was 3.98. ±2.82; and in the remission group, it was 4.0. ±2.20. This study found that patients with BD-1 had significantly reduced RNFL thickness, SFCT, and GCL layers compared to the healthy control group (P < 0.05). The reduction in RNFL thickness was significantly associated with the longer duration of illness.

Conclusion:

The BD-1 patients evaluated in our study present structural alterations in the retina. The search for biomarkers of the disease through different ophthalmological examinations such as OCT enhances the usefulness of ophthalmological examination as a way to access the brain.

Bipolar disorder (BD) is a common, chronic, enduring mood disorder characterized by manic, depressive, and mixed episodes.[1] The lifetime prevalence of BD varies from 1% to 6.5% in the general population.[2] BD generally begins in late adolescence or early adulthood, leading to psychotic, affective, behavioral, and cognitive impairment, thus affecting the most productive years of life.[3] The precise mechanism that causes BD is not well understood. BD is thought to be associated with abnormalities in the structure and function of specific brain areas responsible for cognitive tasks and the processing of emotions.[4] Structural neuroimaging of BD has shown reductions in gray and white matter in various brain regions.[5] However, there are some limitations in the brain imaging methods, such as low image resolution, cost, and patient burden.[6] To overcome the limitations of neuroimaging, such as low image resolution and high investigation cost, a non-invasive retinal imaging technology, that is, optical coherence tomography (OCT), can prove a usual alternative. OCT provides tissue morphology imagery at much higher resolution (less than 10 μm axially and less than 20 μm laterally than other imaging modalities such as MRI or ultrasound.[7] OCT is an imaging technique that uses interferometry with short-coherence-length light to obtain micrometer-level depth resolution and uses transverse scanning of the light beam to form two- and three-dimensional images from light reflected from biological tissue or other scattering media.[8] OCT has been used in a wide spectrum of ophthalmology to assess different retina layers’ thickness, volume, and structure.[9]

As neurodevelopmentally, the brain and the retina develop from the same tissue (neuroectoderm), knowing the structural and functional brain changes in the retina can prove an easy and non-invasive option.[10] This has prompted the use of OCT for structural analyses of the retina and retinal nerve fiber layer (RNFL) to detect and monitor axonal loss and neurodegeneration in patients with neurodegenerative diseases, including Alzheimer’s disease, Parkinson’s disease, and multiple sclerosis.[11] BD is thought to be associated with abnormalities in the structure and function of specific brain areas. Visualizing the retina and demonstrating thinning of the RNFL with OCT can indirectly show gray matter loss in BD.[12] A few studies have explored the relationship between neurodegeneration in BD and OCT findings. Neurodegeneration observed throughout BD, such as decreases in the RNFL thickness and ganglion cell layer (GCL), may have implications for understanding the disease’s progression and developing therapeutic strategies. This study investigated the potential correlation between OCT findings and clinical features in patients with bipolar disorder type-1 (BD-1). We compared these findings with healthy controls to assess potential differences and implications.

MATERIAL AND METHODS

Study setting and design

This case-control, hospital-based study was conducted at the Department of Psychiatry, Government Medical College, Srinagar, between April 1 and September 30, 2023. The patients in the study were diagnosed as BD-1 per the Diagnostic and Statistical Manual of Mental Disorders, 5^th edition (DSM-5), and were in different phases of the BD-1. The patients were in the age group of 18–60 years and included both males and females. These patients were selected through a purposive sampling technique and were free to accept or refuse participation in the study. Informed written consent was taken from the patient or immediate caregivers and controls. The normal healthy controls for this study were chosen from the hospital staff and through public advertisements. The controls were age-matched, normal, healthy people unrelated to the cases. Patients with concurrent psychiatric illnesses and medical disorders such as diabetes mellitus (DM), hypertension (HT), and intellectual disability (ID) were excluded from the study. Other exclusion criteria included patients with neurodegenerative diseases, such as Alzheimer’s dementia, Parkinson’s disease, or multiple sclerosis; history of surgery or retinal laser or acute infection such as uveitis; IOP of greater than 21 mm/Hg; and powers of ± 3.0 D for spherical lenses and a value of ± 2.0 D for cylindrical lenses on refractometer. Controls with a prior history of any mental illness or with a signal quality of 6 or less in the SD-OCT were excluded from the study.

Assessment

Detailed history, current mental status examination, and review of old hospital records were done and recorded in a predesigned structured proforma. For the assessment of symptom severity, the Young Mania Rating Scale (YMRS) and Hamilton Depression Scale (HDS) were used. Detailed history and ophthalmological examination of cases and controls were done and recorded by two independent expert ophthalmologists. Autorefractometer Accuref R-800 was used to objectively measure refractive errors and a prescription for glasses or contact lenses.

Fundus examinations, biomicroscopic examination, Goldmann applanation tonometry, and retinal layer thickness measurements were done in cases and controls. The retinal layer thicknesses of cases and controls were measured using a Zeis Cirrus 5000 model OCT device. It can take spectral-domain images, has a scan speed of 27000–68000 A-scan/second, an A-Scan depth of 2 mm, an axial resolution of 5 μm, and a transverse resolution of 15 μm.

Spectral-domain optical coherence tomography imaging

In cases and controls, the choroidal layer thickness was measured manually. However, peripapillary RNFL, GCL, and macular thicknesses used in the study were measured automatically by the operating device. The choroidal layer was imagined using SD-OCT in the enhanced depth imaging (EDI) mode. In many studies, OCT using EDI has been used to measure choroidal thickness in normal populations and eyes with ocular diseases.[13,14]

It has been postulated that variations in choroidal thickness may indicate the presence of retinal diseases and may be used to monitor its prognosis and treatment progress.[15] The manual quantification of choroidal thickness is performed at the fovea or specific points in the superior, inferior, temporal, and nasal regions around the fovea.[16] Choroid thickness measurements were taken separately by two ophthalmologists who were unaware of which group the patient belonged to. The aforementioned experts carried out the SD-OCT imaging procedures between 11:00 and 12:00 am each day to prevent diurnal variations in the choroid thicknesses. Moreover, the patients were not allowed to smoke in the 3 hours before OCT imaging, as smoking before OCT imaging may affect choroidal thicknesses.

Statistical analysis

Data were entered into a Microsoft Excel sheet, and analysis was done using Statistical Package for the Social Sciences version 23.0 (IBM SPSS Corp.; Armonk, NY, USA). The means of the case and control groups were compared using an independent samples t-test. A paired samples t-test was used to compare the SD-OCT measurements between the right and left eyes. Pearson correlation analysis was applied to explore the relationships between sociodemographic factors, disease variables, and SD-OCT measurements. Chi-square tests were utilized to analyze categorical data such as gender, marital status, and education across the groups. A significance level of 0.05 was set for all statistical tests.

RESULTS

A total of 63 patients with BD-1 were approached for the study, out of which three patients had IOP of greater than 21 mm/Hg and uncontrolled diabetes, two patients had senile macular degeneration, three patients had uncontrolled myopia with a history of retinal laser, and three patients had poor OCT scans and thus were excluded. An equal number of patients (n = 49) and controls (n = 49) were taken for the study, with a mean age of 39.55 ± 5.71 years for patients and 37.91 ± 6.17 years for controls. Out of 49 patients who were in different phases of illness, 12 (24.49%) patients were in mania, 16 (32.65%) patients were in depression, and 21 (42.85%) patients of BD-1 were in remission. There were no statistically significant differences in sociodemographic characteristics between the patient and control group except in educational attainment [Table 1].

Table 1.

Demographic characteristics of patients and controls

Characteristics	Cases group (n)	Control group (n)	P
Age (SD)	39.55±5.71	37.91±6.17	0.35
Gender
Male	27 (55.10)	22 (44.90)	0.41
Female	22 (44.90)	27 (55.10)
Marital status
Unmarried	15 (30.61)	15 (30.61)	0.63
Married	32 (65.31)	34 (69.39)
Divorced/widowed	02 (4.08)	0
Education
Illiterate	2 (4.08)	2 (4.08)	0.001
Primary	10 (20.41)	7 (14.29)
Higher secondary	28 (57.14)	17 (34.69)
College	4 (8.16)	21 (42.86)
University	5 (10.20)	2 (4.08)
Smoking status
Yes	21 (42.86)	13 (26.53)	0.13
No	28 (57.14)	36 (73.47)

The mean duration of illness was 14.55 ± 10.71 years. Total number of attacks in the mania group was 4.40 ± 3.30; in the depression group, it was 3.98. ±2.82; and in the remission group, it was 4.0. ±2.20; there were no statistically significant differences between the groups. The clinical characteristics of the patients are given in Table 2.

Table 2.

Clinical characteristics of the studied patients

Characteristics	Mania group (n=12)		Depression group (n=16)		Remission group (n=21)
	Mean	SD	Mean	SD	Mean	SD
Duration of Illness	15.80	12.10	11.30	8.80	12.20	9.20
No. of attacks	4.40	3.30	3.98	2.82	4.0	2.20
YMRS Score	30.20	10.80	………	………	1.54	1.42
HDS Score	…….	…….	22.28	8.80	3.30	1.20
Mood stabilizer	n	%	n	%	n	%
Valproic acid n=(20)	13	26.53	4	8.16	3	6.12
Lithium n=(29)	14	28.57	10	20.41	5	10.20

HDS, Hamilton Depression Scale; n, number; SD, standard deviation; YMRS, Young Mania Rating Scale

The study showed that RNFL thickness, subfoveal choroidal thickness (SFCT), and GCL measured in both eyes were statistically significantly more in cases than in healthy controls. A negative association was found between the duration of the disease and retinal and choroidal layers. There was statistically no difference in the different measures of OCT [Tables 3 and 4].

Table 3.

Comparison of the chorioretinal layer thickness of patients and controls in the right eye

Characteristics	Group	n	Mean±SD	t	P
Right Tonus	Case	49	14.98±2.19	−0.59	0.55
	Control	49	14.73±1.90
Right NCT	Case	49	277.24±10.41	−0.68	0.49
	Control	49	278.77±11.66
Right SFCT	Case	49	301.43±14.11	3.81	0.002
	Control	49	311.43±11.73
Right TCT	Case	49	260.61±21.25	1.35	0.09
	Control	49	267.14±26.45
Right RNFL	Case	49	79.53±3.53	16.33	0.001
	Control	49	90.08±2.82
Right CMT	Case	49	256.12±10.71	−0.60	0.72
	Control	49	254.83±10.52
Right GCL	Case	49	76.32±2.32	3.30	0.007
	Control	49	77.85±2.26

NCT, nasal choroid thickness; SFCT, subfoveal choroidal thickness; TCT, temporal choroidal thickness; RNFL, retinal nerve fiber layer, CMT, centrale macular thickness; GCL, ganglion cell layer; n, number; SD, standard deviation; t, t value

Table 4.

Comparison of the chorioretinal layer thickness of cases and controls in the left eye

Characteristics	Group	n	Mean±SD	t	P
Left Tonus	Case	49	14.55±2.05	−1.29	0.20
	Control	49	14.06±1.68
Left NCT	Case	49	276.33±10.59	1.35	0.9
	Control	49	278.18±10.42
Left SFCT	Case	49	287.55±10.51	7.26	0.001
	Control	49	303.87±11.69
Left TCT	Case	49	260.61±21.25	1.35	0.09
	Control	49	267.14±26.45
Left RNFL	Case	49	79.43±2.93	17.33	0.0001
	Control	49	90.08±2.83
Left CMT	Case	49	254.20±10.71	−0.70	0.78
	Control	49	253.73±10.52
Left GCL	Case	49	75.25±2.54	2.80	0.008
	Control	49	76.50±2.58

On applying paired sample statistics to compare right and left eye measurements in all the cases and controls, it was found that there was no statistically significant difference between measured parameters, as can be seen in Table 5. It was also found that there was no statistically significant difference between means of retinal and choroidal layers in depression, mania, or remission. No relationship was found between HDS and YMRS and retinal and choroidal measurements.

Table 5.

Both-eye OCT measurements

Parameter	n	Mean	SD	t	df	P
Right Tonus	49	14.98	2.19	0.61	116	0.57
Left Tonus	49	14.55	2.05
Right NCT	49	277.24	10.41	0.71	116	0.63
Left NCT	49	276.33	10.59
Right SFCT	49	301.43	14.11	0.41	116	0.45
Left SFCT	49	287.55	10.51
Right TCT	49	260.61	21.25	0.86	116	0.96
Left TCT	49	260.61	21.25
Right RNFL	49	79.53	3.53	0.52	116	0.84
Left RNFL	49	79.43	2.83
Right CMT	49	256.12	10.71	0.91	116	0.94
Left CMT	49	254.20	10.71
Right GCL	49	76.32	2.32	0.72	116	0.86
Left GCL	49	75.25	2.54

df, differentiation factor

DISCUSSION

OCT is a non-invasive imaging technique that provides high-resolution cross-sectional images of biological tissues.[17] Its application in psychiatric disorders is an exciting area of research for determining neuronal degeneration.[18] The researchers proposed that in the early stages of BD, neurodegeneration may be observed through a reduction in the thickness of the retinal GCL.[18] As the disorder progresses, this degeneration may extend to other retinal layers. This progression highlights a potential biomarker for tracking the advancement of neurodegenerative changes in BD through retinal imaging.[18]

This study found that patients with BD-1 had significantly reduced RNFL thickness, SFCT, and GCL compared to the healthy control group. Our results are in unison with previous studies, which have shown that RNFL and GCL are thinner in patients with BD-1 compared to normal healthy controls.[18,19,20] Our results also showed significant and independent efficacy in differentiating decreased volumes of GCL and SFCT in patients compared to the control group.[21] The reduced thickness of RNFL in patients with BD-1 can be explained by neurodegeneration, which plays an important role in the pathophysiology of BD. A study conducted in Iran also found that the superior, inferior, and total RNFL measurements were thinner in individuals with BD type-1 compared to healthy controls, which aligns with the findings of our study.[22]

In this study, we found that a reduction in RNFL thickness was significantly associated with the longer duration of BD-1, with a greater reduction in patients with a longer duration of illness. Similar trends with a significant correlation between gray matter loss and the number of mood episodes were reported by authors across the globe.[23] Our results also resonate with some previous studies that have shown that the thickness of RNFL decreases as the duration of the disease increases.[20,23,24] In contrast to this study, structural brain changes occur even in the first episode of BD patients; thus, monitoring RNFL thickness might provide insights into the disease’s onset and progression.[22]

Another parameter that has been evaluated in this study was the GCL thickness. Our study results showed that GCL thicknesses of patients with BD-1 were significantly thinner than controls. Our results were supported by numerous previous studies, which have shown that GCL thicknesses in the BD patient group were significantly less in comparison to the healthy control group.[19,20,25]

Consistent with other authors,[26] our study results showed a significant reduction in GCL thickness in the superior quadrant and the average thickness for patients with BD-1, suggesting that the neurodegenerative process in BD-1 starts in the GCL and then gradually progresses to axonal degeneration. These focal GCL changes may reflect focal affectation of the visual pathway in BD patients, as was observed previously in other neurological processes.[27]

Another researcher reported similar findings of a global reduction in RNFL, with statistically significant results.[18] In their study, the reduction was more prominent in the GCL volume than in the RNFL.

The results of the present study indicated that the global volume of the GCL was significantly and independently effective in distinguishing individuals with BD-1 from the control group. Although the specificity of these findings is low, the results of the current study support the hypothesis that neurodegeneration in BD-1 initially affects the GCL, which contains neuronal cell bodies and subsequently progresses to axonal degeneration. Notably, the study highlights that neuronal degeneration in BD can be characterized by thinning of the GCL. Additionally, OCT findings may serve as high-sensitivity markers for monitoring BD-1 patients over time.

Examination of the data has shown that structural brain abnormalities such as a progressive decline in grey matter volume in the cerebellum, fusiform gyrus, and hippocampus, together with temporal lobe ventricular enlargement in the first episode, are considered neurobiological markers for BD-1.[23,28]

While studying neurodegeneration in BD, it was observed that the reduction in GCL volume was more pronounced than the thinning of the RNFL, indicating that the neurodegenerative process in BD begins in the GCL and gradually progresses to axonal degeneration.[18]

In our study, no significant difference was found between means of retinal and choroidal layers in depression, mania, or remission. However, another author found that there was no difference between patients with depression and healthy controls about OCT measures.[29] However, ganglion cell layer and RNFL were correlated with the duration of the latest depressive episode. An author from Turkey found a negative correlation between the disease duration, YMRS score, clinical global impression scale, number of hospitalizations, and GCL volume in the euthymic phase of the disease at the time of assessment.[18] This correlation suggested a progressive degenerative process or inflammation.[18] No relationship was found between HDS and YMRS and retinal and choroidal measurements; however, a study done in the USA linked these findings to the longitudinal course of illness and severity of mood episodes.[30]

The results of the present study did not find any difference between right and left eye measurements. Other authors published similar reports, and no difference was found between the right and the left eye except for a single study.[31]

Limitations

The study’s limitations include a small sample size and a cross-sectional design, which hindered the ability to assess causality. Another limitation is the potential impact of smoking and medications, including antipsychotics and mood stabilizers, which were not considered. For instance, valproate has been shown to cause gray matter loss in patients with epilepsy, while lithium has demonstrated a neuroprotective effect in individuals with BD. Another limitation of the study was that specific natural age-related changes in OCT may resemble changes in the eyes of patients with BD.

CONCLUSION

We conclude that BD-1 is a neurodegenerative disease that causes structural alterations in the brain. The patients with BD-1 evaluated in our study present structural modifications in the retina. These patients showed thinning of RNFL and GCL detected by OCT, and these changes were related to the duration of the illness but not to its severity. The search for disease biomarkers through different ophthalmological examinations such as OCT enhances the usefulness of ophthalmological examination to access the brain and determines the need to find a safe, fast, and reliable means, such as OCT, capable of observing and quantifying these alterations. Early detection and understanding of biological changes in clinical progression might help to identify therapeutic targets to promote better outcomes in BD.

List of abbreviations

BD-1, bipolar disorder type-1, SD-OCT, spectral domain optical coherence tomography HDS, Hamilton Depression Scale; n, number; SD, standard deviation; YMRS, Young mania rating scale; NCT, nasal choroid thickness; SFCT, subfoveal choroidal thickness; TCT, temporal choroidal thickness; RNFL, retinal nerve fiber layer, CMT, centrale macular thickness; GCL, ganglion cell layer; n, number; SD, standard deviation; t, t value, df, differentiation factor.

Authors’ contributions

Concept, design, literature search: SAD, RAW, R. Data acquisition, SAD, RAW, R, IAS. Data analysis: AH. Manuscript preparation, editing, and review: SAD, RAW, R., IAS, AH.

Data availability

All the data is reported in the manuscript.

Ethical statement

The study was approved by the institutional ethics committee of Government Medical College, Srinagar, vide number 123/IEC dated 23/03/2023

Conflicts of interest

There are no conflicts of interest.

Declaration of patient consent

The authors declare that they have obtained consent from patients. Patients have given their consent for their images and other clinical information to be reported in the journal. Patients understand that their names will not be published and due efforts will be made to conceal their identity but anonymity cannot be guaranteed.

Funding Statement

Nil.