Changes in retinal nerve fiber layer, ganglion cell complex, and ganglion cell layer thickness in thyroid eye disease: A systematic review

Abstract

Thyroid eye disease (TED) is characterized by progressive orbital inflammation. In severe cases, it can lead to dysthyroid optic neuropathy (DON), involving death of retinal ganglion cells and permanent vision loss. Imaging with optical coherence tomography (OCT) has enabled noninvasive examination of the structural impact of the disease, including of the retinal nerve fiber layer (RNFL), the ganglion cell layer (GCL), and the inner plexiform layer (IPL). In this study, we systematically review 20 studies (897 patients, 1430 eyes) to evaluate changes in retinal thickness in patients with TED, stratified by severity. Most studies found a decrease in thickness in peripapillary RNFL (pRNFL), non-pRNFL, combined RNFL/GCL/IPL, or combined GCL/IPL when comparing patients with DON to healthy controls or to patients without DON. However, results were equivocal regarding differences between non-DON TED patients and controls. In addition, several studies reported a postoperative decrease in pRNFL thickness after orbital decompression surgery. Future studies examining OCT changes of retinal thickness along with disease progression are needed to assess the potential of OCT in diagnosis and management of patients with TED.

Introduction

Thyroid eye disease (TED) is an autoimmune condition characterized by swelling of orbital contents, such as extraocular muscles, retro-orbital fat, and connective tissue.[1] While most patients have a mild course, about 5% develop dysthyroid optic neuropathy (DON), a serious complication that can lead to death of retinal ganglion cells and irreversible vision loss.[2] In these patients, orbital tissue swelling leads to orbital apical crowding resulting in optic nerve compression and injury.

Optical coherence tomography (OCT) is a noninvasive imaging modality that can obtain cross-sectional images of the layers of the retina and optic nerve head. It is commonly used to monitor glaucomatous and compressive optic neuropathies.[3,4,5] Layers of interest include the retinal nerve fiber layer (RNFL), the ganglion cell layer (GCL), and the inner plexiform layer (IPL), which correspond to the axons, cell bodies, and dendrites of retinal ganglion cells, respectively.[6,7] The ganglion cell complex (GCC) is composed of the three innermost retinal layers: the RNFL, the GCL, and the IPL. A decrease in thickness in these layers indicates loss of retinal ganglion cells which may correspond to vision loss. While the use of OCT in glaucoma is well established in clinical practice, there is a growing interest in examining OCT measurements of people with TED given the pathophysiology of compressive optic neuropathy involved in DON.

The aims of this study were to perform a systematic review of literature on RNFL, GCL, IPL, and GCC measurements in TED. Specifically, we aimed to evaluate the differences in thickness in four groups: mild TED, moderate TED, DON, and healthy controls. By elucidating the differences between these groups, we hoped to discern whether OCT would be a viable clinical tool in evaluating patients with TED and their disease activity and progression.

Methods

Identification of studies

A systematic search of PubMed, Embase, and Cochrane databases was performed in January 2022. Keywords and MESH terms for “Graves’ disease,” “dysthyroid,” or “thyroid-associated” were combined with “orbitopathy” or “eye disease” and with “OCT” or similar terms. Then, duplicate results were removed. Two reviewers (L. C. and C. C. G.) independently screened the titles and abstracts of these articles. Afterward, full-text articles of relevant articles were obtained and underwent review by the same two reviewers for final selection of included articles. The reviewers discussed all discrepancies until reaching a mutual decision about inclusion.

Inclusion and exclusion criteria

This study included both prospective and retrospective analytical observational studies (case series, case–control, and cross-sectional). Studies were included if they had measurements of RNFL, GCC, or GCL thickness using OCT in patients with TED. We excluded studies that included patients with comorbidities that could impact thickness measurements (e.g., high myopia, glaucoma, or optic disc atrophy). We excluded conference abstracts, review articles, case reports, and non-English language papers. Studies were also excluded if they overlapped patient populations with another included study.

Quality assessment and data extraction

Quality assessment was performed using the Joanna Briggs Institute critical appraisal checklist for cross-sectional studies, case series, or case–control studies. Each checklist includes eight or ten questions that evaluate the potential risk of bias on a study. For each checklist item, each potential study was graded as “yes,” “no,” “unclear,” or “not applicable.” Studies that had a “yes” answer to at least six of eight or seven of ten checklist items were deemed to be of acceptable quality and were included in the systematic review. Data abstraction included author, publication year, study location, study design, number of participants, demographics (e.g., age and gender), visual measurements (e.g., refractive error, intraocular pressure, and proptosis), and thicknesses of RNFL, GCC, and GCL.

Two authors (L. C. and C. C. G.) independently reviewed the papers, performed data abstraction, and completed the quality assessment. All discrepancies were discussed, and a mutual decision was made. Meta-analysis was deemed not feasible due to high heterogeneity of the results and methodological differences between studies. Thus, findings were summarized using tables and narrative synthesis. Mild or moderate–severe TED was defined using the European Group of Graves’ Orbitopathy (EUGOGO) classification system.[8] Per EUGOGO, patients with mild TED usually have one or more of the following: minor lid retraction (<2 mm), mild soft-tissue involvement, exophthalmos <3 mm above normal, and no or intermittent diplopia or corneal exposure. Moderate-to-severe TED was defined as nonsight-threatening TED but with at least one of the following features: with lid retraction >2 mm, moderate or severe soft-tissue involvement, exophthalmos ≥3 mm above normal, or frequent diplopia.

Results

Study selection

We initially identified 186 papers from the three literature databases and removed 55 duplicates. After initial screening of titles/abstracts, 22 potential studies were identified for full-text review. We excluded two additional studies: one study did not report any outcomes of interest and one study overlapped data with another included study.[9,10] In the two studies with overlapping data, the decision was made to include the study that presented subgroup analyses on TED severity (mild vs. moderate–severe) rather than smoking status, as this was more in line with our study objectives.[9,11] Of the remaining 20 studies that underwent quality assessment, 12 were cross-sectional studies, 4 were case series, and 4 were case–control studies. Eleven of the 12 cross-sectional studies scored 8 out of 8 points, while one scored 6. Of the four case series, one scored 9 out of 10, one scored 8, and two scored 7. Of the four case–control studies, one scored 9 out of 10 and the other three scored 8. Therefore, all studies were deemed of acceptable quality and the review included 19 studies. A diagram of the article selection process is shown in Figure 1.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram of the selection process of studies screened and evaluated for eligibility

Description of included studies

Six studies were performed retrospectively, while the remaining 14 studies were prospective. Seven studies were conducted in China; three each in Iran and Turkey; two each in India, Italy, and Korea; and one in Israel. Studies included anywhere from 20% to 80% male patients. The mean age of subjects in most studies was between 40 and 50 years. Studies included a wide range of TED presentations, ranging from mild disease to DON. A full summary of demographic data is found in Table 1.

Table 1.

Patient demographics of included studies

Study	OCT type	Number of eyes (subjects)		Mean age		Sex (male/female)		IOP
		TED	Controls	TED	Controls	TED	Controls	TED	Controls
Abdolalizadeh et al.[23]	SD	79 (36)	39 (20)	40.79±10.34	37.60±8.35	17/19	7/13	17.04±3.5	14.26±2.17
Casini et al.[12]	SD	40 (40)	40 (40)	39.57±5.60	36.28±5.79	5/35	11/29	-	-
Cheng et al.[21]	HD	52 (37)	-	48	-	9/28	-	19.71±4.95	-
Dave et al.[22]	OCTA	126 (126)	52 (52)	41.60±13.25	47.74±7.33	59/67	31/21	-	-
Forte et al.[18]	OCT/SLO	60 (30)	80 (40)	56.0±6.3	58.3±7.5	12/18	20/20	24.0±0.9	19.1±0.4
Guo et al.[13]	HD	145 (75)	70 (35)	46.4±13.1	53.3±11.7	25/50	13/22	18.5±3.6	15.1±2.1
Jamshidian et al.[9]	OCTA	98 (55)	65 (35)	43.9±13.3	38.9±14.1	-	-	16.9±3.8	15.3±2.7
Jian et al.[15]	OCTA	82 (43)	52 (26)	39.88±12.17	43.39±11.09	17/26	10/16	19.74±3.83	15.44±2.62
Kurt et al.[29]	SS	58 (29)	60 (30)	40.66±11.83	42.60±12.38	15/14	12/18	16.88±2.28	15.02±2.56
Luo et al.[14]	SD	116 (58)	116 (58)	43.8±12.4	44.7±11.4	35/23	35/23	-	-
Meirovitch et al.[17]	Stratus OCT 3	42 (21)	73 (41)	44.19±14.21	42.90±14.65	4/17	9/32	-	-
Mugdha et al.[19]	HD	40 (40)	40 (40)	-	-	18/22	18/22	-	-
Ogmen et al.[30]	SD	44 (23)	46 (18)	38.8±10.8	35.9±8.7	8/15	7/11	15.7	14.5
Park et al.[16]	HD	52 (30)	94 (47)	54±11	48±14	15/15	26/21	-	-
Rajabi el al.[25]	SD	26 (15)	-	50.48±8.77	-	14/12	-	-	-
Sayın et al.[20]	Stratus	132 (66)	72 (36)	40.1±12.6	38.3±9.7	23/43	11/25	15.5±2.5	12.9±1.9
Seo et al.[24]	SD	40 (30)	-	36.83±2.15	-	12/18	-	15.1±0.53	-
Wu et al.[28]	SD	74 (44)	38 (38)	44.6±9.4	45.2±10.8	23/21	19/19	17.6±5.4	14.4±2.3
Yu et al.[21]	HD	53 (53)	29 (29)	40.9±11.4	38.7±12.5	14/39	5/24	18.8±3.7	15.4±2.6
Zhang et al.[27]	OCTA	71 (46)	23 (19)	49.5±9.0	46.7±13.0	21/25	10/9	18.2±4.7	14.6±2.7

-=Information not reported, HD=High definition, IOP=Intraocular pressure, OCT=Optical coherence tomography, OCTA=Optical coherence tomography angiography, SD=Spectral domain, SS=Swept-source, TED=Thyroid eye disease, SLO=Scanning laser ophthalmascope

Association between thyroid eye disease and peripapillary retinal nerve fiber layer

Six studies reported peripapillary RNFL (pRNFL) outcomes. This measurement was taken in 611 eyes in 357 patients. The average pRNFL of patients with TED ranged from 28.4 to 116.1 μm, with a median of 102.7 μm. The average pRNFL for healthy controls ranged from 28.4 to 118.0 μm, with a median of 103.2 μm. Of these six studies, one study did not stratify TED patients by disease severity. This study found no difference in pRNFL thickness between TED patients and healthy controls. they found no difference in pRNFL thickness between patients with TED and controls.[12] When looking at subgroups, the average thicknesses ranged from 97.8 to 118.7 μm (median: 106.1) for mild TED, 101.1–122.8 μm (median: 110.6) for moderate TED, and 63.5–111.4 μm (median: 99.7) for DON. Three studies found no difference between patients with mild TED and controls, and two studies found no difference between patients with moderate TED and controls.[13,14] For patients with DON, three studies found a significant difference in thickness between patients with DON and controls up to 7.4%, while one study found no difference.[10,12,13,14] In addition, Jian et al. also found a significant decrease in thickness in patients with DON and both mild TED and moderate TED by 46.5% and 48.3%, respectively.[15]

Three of these studies reported pRNFL outcomes in the four quadrants. Kurt et al.[29] found a significant decrease in the superior pRNFL thickness between mild TED and controls by 4.8%. Luo et al.[14] found a significant decrease between moderate TED and controls in the temporal and nasal regions by 7.0% and 10.5%, respectively.[11,12] Finally, Park et al. found a significant decrease between both chronic and acute DON and controls in the temporal region by 13.1%.[16] Results are summarized in Table 2.

Table 2.

Peripapillary retinal nerve fiber layer thickness in patients with thyroid eye disease and controls

Study	Peripapillary RNFL thickness					Significant difference in thickness? (P<0.05)
	Mild TED	Moderate TED	DON	All TED	Control
Casini et al.[12]	-	-	-	28.42±2.16	28.37±2.79	No significant difference between all TED and control
Cheng et al.[21]	-	-	-	101.54±17.73	-	Significant reduction in thickness after decompression surgery, globally and in all four quadrants
Guo et al.[13]	97.8±9.2	110.6±34.2	100.3±6.3	102.7±19.4	103.2±6.6	No significant difference between mild TED and control, moderate TED, and control Significant difference between DON and control
Jian et al.[15]	118.68±11.08	122.79±15.33	63.47±15.81	116.60±13.67	118.01±13.07	Significant difference between DON and mild TED, moderate TED, and control
Kurt et al.[29]	107.14±11.02	-	-	107.14±11.02	107.18±8.73	Significant difference between mild TED and control in superior region No significant difference between mild TED and control globally and in inferior, temporal, and nasal regions
Luo et al.[14]	105.17±13.21	101.05±10.28	111.38±7.74	104.23±10.68	103.15±9.29	No significant difference between mild TED and control Significant difference between moderate TED and control in temporal and nasal regions Significant difference between DON and control in temporal inferior, nasal superior, and global areas
Park et al.[16]	-	-	99±11	99±11	96±8	Significant difference between chronic DON and both acute DON and control in temporal region No significant difference between acute DON, chronic DON, and controls globally
Rajabi et al.[25]	-	-	-	108	-	Significant reduction in thickness after decompression surgery, globally and in all four quadrants
Seo et al.[24]	-	-	-	98.55±12.76	-	No significant change in thickness after decompression surgery

-=Information not reported, DON=Dysthyroid optic neuropathy, TED=Thyroid eye disease, RNFL=Retinal nerve fiber layer

Association between thyroid eye disease and retinal nerve fiber layer (nonpapillary)

Overall, nine studies included measurements of RNFL thickness. This measurement was taken in 647 eyes in 441 patients. The average thickness in patients with TED across studies ranged from 12 to 112.2 μm, with a median of 99.0 μm. The average thickness in healthy controls across studies ranged from 13 μm to 131.4 μm, with a median of 101.8 μm. Results comparing patient groups were mixed. One study found a significant decrease in RNFL thickness between TED patients without DON and controls by 9.1%, two studies found a decrease that did not reach significance, and one study found a significant increase by 14.3%.[17,18,19,20] No studies found any difference between mild TED patients and controls.[21] However, there was a significant relative decrease in RNFL thickness when comparing moderate–severe to mild patients or controls.[21,22] Finally, patients with DON had a significantly thinner RNFL than both mild TED patients and controls by 3.4% and 3.0% respectively.[23]

Six of these studies reported RNFL outcomes in the four quadrants. When comparing TED patients with controls, two studies found a significant decrease in inferior RNFL thickness by 5.0%–7.9%, one found a decrease in the remaining three quadrants as well by 8.0%–10.0%, and one study found a significant increase in the superior, inferior, and nasal RNFL by 15.4%–27.6%.[17,19,20,21,22,23] When comparing patients with moderate–severe TED and mild TED, two studies found a significant difference in inferior RNFL thickness by 12.0%–25.0%, one found a difference in superior RNFL thickness by 23.5%, and one found a difference in temporal RNFL thickness by 14.3%.[21,22] When considering patients with DON, Abdolalizadeh et al. found a significant decrease by 3.7% compared to mild TED in their superior RNFL thickness and, when compared to both mild and controls, a decrease in inferior RNFL by 4.0% and 5.9%, respectively.[23] Results are summarized in Table 3.

Table 3.

Retinal nerve fiber layer thickness in patients with thyroid eye disease and controls

Study	Global RNFL thickness					Significant difference in thickness? (P<0.05)
	Mild TED	Moderate TED	DON	All TED	Control
Abdolizadeh et al.[23]	102.32±4.14	95.97±10.00	98.86±26.13	98.99±14.96	101.97±8.93	Globally, significant difference between DON and both mild TED and control Superiorly, significant difference between mild TED and DON Inferiorly, significant difference between control and both moderate–severe TED and DON. Also difference between mild TED and DON
Dave et al.[22]	35.55±7	45.11±18.3	-	37.37±10.02	36.28±7.89	Significant difference between active and inactive TED globally, superiorly, and inferiorly Significant difference between active TED and controls globally and inferiorly No significant difference between inactive TED and controls
Forte et al.[18]	-	-	-	112.2±9.1	131.4±14.2	Decreased RNFL thickness in TED compared to control, but not significant (P=0.2)
Meirovitch et al.[17]	-	-	-	110.06±33.3	96.25±9.42	Significant increase in RNFL thickness in TED compared to controls globally, superiorly, inferiorly, and nasally No significant difference temporally
Mugdha et al.[19]	-	-	-	92.06±12.44	101.28±6.64	Significant decrease in RNFL thickness in TED compared to controls globally and in all four quadrants
Ogmen et al.[30]				12 (10–16)*	13 (9–17)*	No significant difference in RNFL thickness in TED compared to controls
Sayın et al.[20]	-	-	-	98.3±29.4	101.7±30.4	Inferiorly, significant decrease in RNFL thickness in TED compared to controls No significant difference globally or in the other three quadrants
Yu et al.[21]	101±8	92±7	-	-	102±8	Significant decrease in active TED compared to inactive TED and controls, globally, temporally, and inferiorly No significant difference between inactive TED and controls
Zhang et al.[27]	101.36±8.64		98.30±10.77	100.06±9.46	104.91±7.99	Overall significant differences between the three groups No significant difference when comparing non-DON TED patients with controls and comparing DON with non-DON patients

*Median (minimum–maximum). -=Information not reported, DON=Dysthyroid optic neuropathy, TED=Thyroid eye disease, RNFL=Retinal nerve fiber layer

Effect of orbital decompression surgery on peripapillary retinal nerve fiber layer

Three studies were surgical case series that studied the effect of orbital decompression surgery on pRNFL thickness.[24,25,26] Two of these studies noted a significant reduction in pRNFL thickness after surgery by 8.6%–10.2%.[25,26] This effect held globally as well as on all four quadrants individually. They also noted a significant correlation between presurgical pRNFL thickness and postsurgical visual recovery. Eyes with normal preoperative pRNFL had greater postoperative improvements of visual acuity, color vision, and visual field parameters compared to eyes with thin preoperative pRNFL. On the contrary, Seo et al. noted no change in postoperative pRNFL thickness [Table 2].[24]

Association between thyroid eye disease and ganglion cell complex/ganglion cell layer

Five studies included measurements of GCC thickness.[11,14,23,27,28] This measurement was taken in 405 eyes in 236 patients. The average thickness in patients with TED ranged from 97.9–121.4 μm, with a median of 94.5 μm. The average thickness in healthy controls ranged from 100.3 to 123.3 μm, with a median of 100.6 μm. When considering patients with DON, three studies found a significant decrease in thickness between patients with DON and healthy controls by 9.8%–13.4% and two studies found that that decrease was also present when comparing patients with DON to TED patients without DON by (8.1%–11.6%).[23,27,28] Although one study found a significant decrease in thickness by 15.3% when comparing non-DON TED patients and healthy controls, two other studies found no difference in thickness when comparing these same groups.[12,26,27]

Similarly, six studies included measurements of GCL or combined GCL/IPL thickness. This measurement was taken in 487 eyes in 338 patients. One study found a statistically significant difference between TED patients and controls, while three studies found no difference.[12,22,29,30] Guo et al. found a significant decrease in combined GCL and IPL thickness between patients with DON or moderate TED and controls by 11.4% and 5.4%, respectively, but no difference between mild TED and controls.[13] Similarly, Wu et al. found a significant decrease in combined GCL and IPL thickness between patients with DON or non-DON TED and controls by 17.8% but no difference between DON and non-DON TED patients.[28]

Discussion

This systematic review identified 20 studies that commented on the correlation between inner retinal thickness and severity of TED. While DON patients showed markedly thinner inner retinal layer thickness, results were largely equivocal for TED patients with milder disease without DON when compared to healthy controls.

While the link between inner retinal thickness and severity of TED has been proposed, there was no clear correlation observed except in DON patients. Dave et al. speculated that this inconsistency may be because RNFL thickness varies with progression along the disease.[22] The mechanism behind TED and RNFL thickness has been hypothesized to be due to deposition of hydrophilic glycosaminoglycans by orbital fibroblasts, contributing to extraocular muscle and orbital fat enlargement and impairing vascular congestion. In the initial phase, increased RNFL thickness may occur secondary to optic disc swelling and edema. Over time, thinning of the RNFL occurs with compression as well as decreased axonal flow resulting in optic nerve hypoxia and ischemia. Patients with chronic DON had thinner mean pRNFL and temporal RNFL thickness compared to acute DON and control eyes.[16] It is also possible that an interplay between compressive mechanical and inflammatory mechanisms may mask RNFL changes.[17]

Some of the inconsistency in inner retinal thickness may be due to progression of the disease over time; structural disease may precede functional disease. In other words, intraretinal layers are damaged while orbital symptoms remain subclinical. This trend has been observed in other disease processes such as glaucoma.[31,32] Nearly half of the glaucoma patients or suspects with preexisting RNFL thinness eventually developed visual field defects. Most interestingly, in glaucoma suspects, abnormal RNFL thickness could be used to predict future visual field conversion. Similar to glaucoma patients, for TED patients, changes in inner retinal thickness may indicate subclinical optic nerve injury. In addition, assessment of GCL/IPL parameter changes has also been suggested to enable early detection of compressive optic neuropathy secondary to parachiasmal tumors.[33]

Given the potential to use RNFL/GCL/IPL changes to predict future visual field changes in other disease processes, it is possible that OCT could also be used as a prognostic tool in TED. As most of the studies in this systematic review were cross-sectional, data from patients with TED were captured at varying stages of their disease course, which could lead to variability in thickness. As such, there would be a benefit for longitudinal studies examining the likelihood of developing visual field defects or other clinical symptoms in TED patients with inner retinal thickness changes. If this was the case, OCT measurements could become a useful functional prognostic tool for TED. In addition, OCT has been proposed as a supplemental tool for early diagnosis of DON, with mean pRNFL showing moderate diagnostic power.[14]

Previous studies have shown that inferior visual field defects are the most commonly observed defect in TED patients.[34,35] This likely indicates damage to the superior optic nerve. There are several hypotheses for this. Some postulate that the anatomy of the orbital apex leading to apical crowding, while others note preferential enlargement of the superior rectus-levator palpebrae muscle complex.[36,37,38,39,40] Both these hypotheses demonstrate a mechanism for compression of the superior optic nerve. In our study, we did not demonstrate preferential reduced RNFL thickness at the superior sectors. In fact, our results show that there are a similar number of studies showing thinning of the superior (six studies), inferior (seven studies), nasal (five studies), and temporal (six studies) RNFL in TED patients. It is possible that a superior sector damage is only apparent as certain stages of the disease course. Given the cross-sectional nature of most of these studies, this could not be captured. Future longitudinal studies would be useful in characterizing the changes to different sectors of the inner retina over time.

Aggressive management is required in cases of DON to prevent permanent vision loss. While there is no gold standard for DON treatment, orbital decompression serves as a viable option, particularly in steroid-resistant patients or those that are not candidates for treatment with teprotumumab.[41] The role of decompression surgery in both immediate and delayed visual recoveries has been well established.[42,43,44] Two studies in this systematic review which examined predictors of favorable postoperative recovery noted that greater preoperative pRNFL thickness was a significant prognosticator.[25,26] Since pRNFL thickness corresponds to the degree of axonal degeneration of retinal ganglion cells, OCT has been proposed as a means to determine prognosis in these patients.[16] The use of OCT as a prognostic factor for visual recovery following orbital decompression has been observed in other orbital pathologies. In patients with chiasmal compression, patients with normal preoperative RNFL thickness achieved better visual recovery.[45]

Our systematic review has several limitations. There was significant statistical and methodological heterogeneity, which precluded a meta-analysis. Furthermore, many of the included studies were cross-sectional without follow-up data which prevented assessment of retinal layer thickness with disease progression. Finally, many of the studies in this systematic review had a limited sample size, and more large-scale studies are needed.

In this review, we did not include studies looking at retinal vascularity using OCTA. While a number of these studies exist, measured outcomes vary greatly between studies, making a systematic review difficult at this time. However, given that changes in vascularity may be unrelated to changes in thickness, this remains a promising avenue for future study. Another interesting avenue for future study is to explore whether OCT can be used to follow up DON patients starting treatment, such as with corticosteroids or teprotumumab. If so, this could be a potentially promising clinical tool for monitoring progression of TED in the future.

Conclusion

Most TED patients have significant variations in inner retinal layer thickness despite increasing disease severity, possibly indicating the presence of subclinical optic neuropathy. DON patients had significantly thinner RNFL thickness compared to TED without DON. Future research monitoring OCT changes along with disease progression is needed to assess the potential of OCT in diagnosis and management of patients with TED.

Data availability statement

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Financial support and sponsorship

Nil.

Conflicts of interest

The authors declare that there are no conflicts of interest in this paper.