3.1. Demographics and visual measurements
There were significant differences in best-corrected visual acuity (P < 0.001) between two groups. There were no significant differences in sex, age, between two groups. More details are shown in Table 1.
Table 1
Demographics and visual measurements between two groups
Condition
|
PACG group
|
HC group
|
T-Value
|
P-Value
|
Gender (male/female)
|
23/21
|
19/25
|
N/A
|
0.370
|
Age (years)
|
54.21 ± 10.60
|
52.36 ± 11.32
|
0.574
|
0.354
|
BCVA-OD
|
0.34 ± 0.42
|
1.13 ± 0.16
|
−6.789
|
< 0.001*
|
BCVA-OS
|
0.33 ± 0.34
|
1.05 ± 0.35
|
−8.054
|
< 0.001*
|
Note: Chi-square test for gender; Independent t-test was used for other normally distributed continuous data. Data are presented as mean ± standard deviation; PACG: primary angle-closure glaucoma; HC, healthy control; BCVA, best-corrected visual acuity; OD, oculus dexter; OS, oculus sinister; N/A, not applicable; *p < 0.001.
3.2. Global Metrics
In the analysis of brain networks, various global metrics were scrutinized, encompassing small-world properties and network efficiency properties. small-world properties included clustering coefficient (Cp), (Fig. 2A) normalized clustering coefficient (Gamma) (Fig. 2B), normalized characteristic path length (Lambda) (Fig. 2C), characteristic path length (Lp) (Fig. 2D), and small-worldness (Sigma) (Fig. 2E). Network efficiency properties comprised global efficiency (Eg) (Fig. 2F) and local efficiency (Eloc) (Fig. 2G).
However, all p-values for global metrics, as revealed by two-sample t-tests, exceeded 0.05, as depicted in the figure. This indicates that, in this study, no significant differences in either network efficiency properties or small-world properties were observed between the PACG patient group and the healthy control group.
3.3. Nodal Metrics
Significant disparities were observed in the left frontal superior medial region (ORBsupmed-L), right frontal superior medial region (ORBsupmed-R), right posterior central brain region (MOG-R), and right temporal superior gyrus region (PAL-R) between the PACG group and the healthy control group. First. for the ORBsupmed-L region, a notable increase in both network node centrality (t-value = 4.984800, p-value = 0.000003) and network node efficiency (t-value = 4.955469, p-value = 0.000004) was observed. This implies a stronger network connectivity in the left frontal superior medial region for the PACG group compared to the healthy control group.Second,in the ORBsupmed-R region, similar significant increases were found in network node centrality (t-value = 4.039719, p-value = 0.000116) and network node efficiency (t-value = 3.872781, p-value = 0.000209). This indicates a significant enhancement in network connectivity in the right frontal superior medial region for the PACG group.Third,the analysis of the MOG-R region showed a significant increase in network node centrality (t-value = 3.738446, p-value = 0.000333), further emphasizing the strengthened network connectivity in the PACG group.Finally, in the PAL-R region, significant decreases in both network node centrality and network node efficiency were observed, corresponding to t-values of -3.659073 and − 3.869785, with p-values of 0.000436 and 0.000211, respectively. This suggests a significant reduction in network connectivity in the right temporal superior gyrus region for the PACG group. (Fig. 3 and Table 2)
Table 2
Differences in nodal metrics between patients with PACG and healthy controls.
|
Brain Regions
|
Degree Centrality
|
Nodal efficiency
|
t-Values
|
p-Values
|
t-Values
|
p-Values
|
PACG > HC
|
ORBsupmed-L
|
4.984800
|
0.000003
|
4.955469
|
0.000004
|
GPACG > HC
|
ORBsupmed-R
|
4.039719
|
0.000116
|
3.872781
|
0.000209
|
PACG > HC
|
MOG-R
|
3.738446
|
0.000333
|
-
|
-
|
PACG < HC
|
PAL-R
|
-3.659073
|
0.000436
|
-3.869785
|
0.000211
|
Abbreviations: ORBsupmed-L, left frontal superior medial region; ORBsupmed-R , right frontal superior medial region; MOG-R, right posterior central brain region; PAL-R, right temporal superior gyrus region; PACG, primary angle-closure glaucoma; HC, healthy control.
3.3. Modular Analysis
In the intra-module analysis revealed that only Module 5 displayed significant internal connectivity (p=0.003796, t=2.975600) among brain regions between PACG and HC groups, which suggests that Module 5 in PACG patients exhibits a unique connectivity pattern, potentially playing a specific role in the development of PACG disease. However, significant correlations in connectivity were not observed within other modules.
The inter-module analysis between the PACG disease group and the healthy control group, significant differences were observed in connection strength between Module 1 and Module 7 (p=0.000549, t=-3.590259) (Figure 4A and Table 3), as well as between Module 1 and Module 8 (p=0.000947, t=-3.424143) (Figure 4B and Table 3), which suggest that significant differences in functional connections between modules in the PACG disease group and the healthy control group, potentially reflecting changes in brain network organization under PACG disease conditions.
Table 3: Intra-module and Inter-module connection difference
|
Module Regions
|
p-Value
|
t-Value
|
Intra-module
|
5
|
0.003796
|
2.975600
|
Inter-module
|
1
|
7
|
0.000549
|
-3.590259
|
|
1
|
8
|
0.000947
|
-3.424143
|
Note: In the intra-module analysis, only module 5 showed significant internal connectivity between the PACG and HC groups; While in the inter-module analysis, significant differences in connection strength were found between Module 1 and Module 7, as well as between Module 1 and Module 8.
3.4. Network-Based Statistics Analysis
Functional brain network connectivity strength was compared between the normal control group (HC group) and the glaucomatous angle-closure glaucoma group (PACG group) by NBS analysis method. The results showed that the PACG group displayed a significant alteration in functional brain network connectivity strength in the case of comparison with the HC group (p-value less than 0.05). This involved 36 network nodes and 46 functional connections including prefrontal, frontal, occipital, temporal, parietal, and subcortical brain regions. In terms of positive connectivity, enhanced connectivity was mainly located between prefrontal and occipital lobe, occipital and occipital lobe, temporal and temporal lobe, and occipital and temporal lobe. In contrast, in terms of negative connectivity, attenuated connections were mainly found between prefrontal and frontal lobe, subcortical brain regions and frontal lobe, prefrontal and temporal lobe, frontal and temporal lobe, subcortical brain regions and temporal lobe, and subcortical brain regions and prefrontal lobe. (Figure 5 and Table 4)
Table 4
Significant alterations in functional connectivities observed in PACG patients compared to HCs
|
Region 1
|
Category
|
Region 2
|
Category
|
p-Values
|
t-Values
|
Positive
|
REC-R
|
Prefontal
|
SOG-L
|
Occipital
|
0.00019000
|
3.90037214
|
|
REC-L
|
Prefontal
|
SOG-R
|
Occipital
|
0.00018032
|
3.91526811
|
|
REC-R
|
Prefontal
|
SOG-R
|
Occipital
|
0.00027781
|
3.79109548
|
|
ORBsup-L
|
Prefontal
|
MOG-L
|
Occipital
|
0.00098555
|
3.41179491
|
|
REC-L
|
Prefontal
|
MOG-L
|
Occipital
|
0.00026751
|
3.80204347
|
|
REC-R
|
Prefontal
|
MOG-L
|
Occipital
|
0.00079732
|
3.47707279
|
|
CAL-R
|
Occipital
|
MOG-L
|
Occipital
|
0.00027458
|
3.79448871
|
|
PHG-L
|
Temporal
|
TPOsup-L
|
Temporal
|
0.00001536
|
4.58394565
|
|
MOG-L
|
Occipital
|
TPOsup-L
|
Temporal
|
0.00024943
|
3.82227780
|
Negative
|
ORBinf-L
|
Prefontal
|
SMA-L
|
Frontal
|
0.00023375
|
-3.84099941
|
|
INS-L
|
Subcortical
|
DCG-L
|
Frontal
|
0.00006252
|
-4.21020869
|
|
INS-R
|
Subcortical
|
DCG-L
|
Frontal
|
0.00044215
|
-3.65475433
|
|
INS-L
|
Subcortical
|
DCG-R
|
Frontal
|
0.00017423
|
-3.92504221
|
|
ORBsup-R
|
Prefontal
|
PHG-L
|
Temporal
|
0.00068066
|
-3.52529441
|
|
SFGmed-R
|
Prefontal
|
PHG-L
|
Temporal
|
0.00079742
|
-3.47703415
|
|
REC-R
|
Prefontal
|
PHG-L
|
Temporal
|
0.00060239
|
-3.56224914
|
|
ACG-L
|
Prefontal
|
PHG-L
|
Temporal
|
0.00008362
|
-4.13045544
|
|
DCG-L
|
Frontal
|
PHG-L
|
Temporal
|
0.00004193
|
-4.31836569
|
|
DCG-R
|
Frontal
|
PHG-L
|
Temporal
|
0.00000168
|
-5.14340453
|
|
SFGmed-R
|
Prefontal
|
PHG-R
|
Temporal
|
0.00021262
|
-3.86821361
|
|
ACG-L
|
Prefontal
|
PHG-R
|
Temporal
|
0.00013433
|
-3.99853342
|
|
HIP-R
|
Temporal
|
CAU-L
|
Subcortical
|
0.00005023
|
-4.26965626
|
|
PHG-R
|
Temporal
|
CAU-L
|
Subcortical
|
0.00004855
|
-4.27886605
|
|
AMYG-R
|
Temporal
|
CAU-L
|
Subcortical
|
0.00010186
|
-4.07584198
|
|
HIP-R
|
Temporal
|
CAU-R
|
Subcortical
|
0.00003945
|
-4.33475971
|
|
AMYG-R
|
Temporal
|
CAU-R
|
Subcortical
|
0.00072912
|
-3.50438173
|
|
IFGoperc-L
|
Prefontal
|
PAL-R
|
Subcortical
|
0.00080491
|
-3.47417286
|
|
IFGoperc-R
|
Prefontal
|
PAL-R
|
Subcortical
|
0.00024082
|
-3.83240910
|
|
IFGtriang-L
|
Prefontal
|
PAL-R
|
Subcortical
|
0.00001324
|
-4.62248467
|
|
ACG-R
|
Prefontal
|
PAL-R
|
Subcortical
|
0.00033039
|
-3.74059766
|
|
CAU-L
|
Subcortical
|
PAL-R
|
Subcortical
|
0.00016507
|
-3.94035808
|
|
DCG-L
|
Frontal
|
HES-R
|
Temporal
|
0.00079435
|
-3.47821537
|
|
PUT-L
|
Subcortical
|
HES-R
|
Temporal
|
0.00005118
|
-4.26461328
|
|
DCG-L
|
Frontal
|
STG-L
|
Temporal
|
0.00041216
|
-3.67555557
|
|
DCG-R
|
Frontal
|
STG-L
|
Temporal
|
0.00074879
|
-3.49626732
|
|
IFGoperc-L
|
Prefontal
|
TPOsup-L
|
Temporal
|
0.00063546
|
-3.54611350
|
|
IFGtriang-L
|
Prefontal
|
TPOsup-L
|
Temporal
|
0.00037356
|
-3.70456213
|
|
SMA-L
|
Frontal
|
TPOsup-L
|
Temporal
|
0.00000073
|
-5.34554657
|
|
SMA-R
|
Frontal
|
TPOsup-L
|
Temporal
|
0.00002106
|
-4.50136340
|
|
DCG-L
|
Frontal
|
TPOsup-R
|
Temporal
|
0.00040338
|
-3.68191916
|
|
SFGmed-L
|
Prefontal
|
TPOmid-L
|
Temporal
|
0.00000221
|
-5.07540202
|
|
SFGmed-R
|
Prefontal
|
TPOmid-L
|
Temporal
|
0.00005564
|
-4.24192212
|
|
SFGdor-L
|
Prefontal
|
TPOmid-R
|
Temporal
|
0.00009326
|
-4.10031897
|
|
IFGtriang-R
|
Prefontal
|
TPOmid-R
|
Temporal
|
0.00071447
|
-3.51056492
|
|
SFGmed-L
|
Prefontal
|
TPOmid-R
|
Temporal
|
0.00004710
|
-4.28701995
|
|
SFGmed-R
|
Prefontal
|
TPOmid-R
|
Temporal
|
0.00003434
|
-4.37187929
|
Note: NBS analysis results show that in the PACG group compared to the HC group, a significantly altered network consisting of 36 nodes and 46 connections was identified. Specifically, the analysis revealed a positively altered network with 8 nodes and 9 connections, as well as a negatively altered network with 30 nodes and 38 connections. |
Abbreviations: REC-R, Right Rectus Gyrus; SOG-L, Left Superior Occipital Gyrus; REC-L, Left Rectus Gyrus; SOG-R, Right Superior Occipital Gyrus; ORBsup-L, Left Superior Orbital Gyrus; MOG-L, Left Middle Occipital Gyrus; CAL-R, Right Calcarine Fissure and Surrounding Cortex; PHG-L, Left Parahippocampal Gyrus; TPOsup-L, Left Superior Temporal Pole; ORBinf-L, Left Inferior Orbital Gyrus; SMA-L, Left Supplementary Motor Area; INS-L, Left Insula; DCG-L, Left Dorsal Cingulate Gyrus; INS-R, Right Insula; DCG-R, Right Dorsal Cingulate Gyrus; PHG-R, Right Parahippocampal Gyrus; SFGmed-R, Right Medial Superior Frontal Gyrus; ACG-L, Left Anterior Cingulate Gyrus; HIP-R, Right Hippocampus; CAU-L, Left Caudate Nucleus; AMYG-R, Right Amygdala; CAU-R, Right Caudate Nucleus; IFGoperc-L, Left Inferior Frontal Gyrus, Opercular Part; PAL-R, Right Pallidum; IFGoperc-R, Right Inferior Frontal Gyrus, Opercular Part; IFGtriang-L, Left Inferior Frontal Gyrus, Triangular Part; ACG-R, Right Anterior Cingulate Gyrus; PUT-L, Left Putamen; HES-R, Right Heschl's Gyrus; STG-L, Left Superior Temporal Gyrus; TPOsup-R, Right Superior Temporal Pole; SFGmed-L, Left Medial Superior Frontal Gyrus; TPOmid-L, Left Middle Temporal Pole; SFGdor-L, Left Dorsal Superior Frontal Gyrus; TPOmid-R, Right Middle Temporal Pole; IFGtriang-R, Right Inferior Frontal Gyrus, Triangular Part; TPOmid-R, Right Middle Temporal Pole; SFGmed-L, Left Medial Superior Frontal Gyrus; PACG, primary angle-closure glaucoma; HC, healthy control.
3.5. Validation Analysis via AAL116 Template
3.5.1. Global Metrics
In the examination of global indicators, a detailed investigation was conducted into the small-world property and network efficiency property. However, the statistical results indicate that these two indicators did not demonstrate significant differences, with p-values exceeding 0.05. This suggests that the similarities between the PACG group and the healthy control group cannot be disregarded with regard to these global indicators. (Supplementary Figure 1).small-world properties and network efficiency properties. small-world properties included clustering coefficient (Cp), (Supplementary Figure 1A) normalized clustering coefficient (Gamma) (Supplementary Figure 1B), normalized characteristic path length (Lambda) (Supplementary Figure 1C), characteristic path length (Lp) (Supplementary Figure 1D), and small-worldness (Sigma) (Supplementary Figure 1E). Network efficiency properties comprised global efficiency (Eg) (Supplementary Figure 1F) and local efficiency (Eloc) (Supplementary Figure 1G).
3.5.2. Nodal Metrics
The study of local metrics revealed some interesting findings. In the region of the right middle temporal gyrus (MTG.R), there was a significant difference between the PACG group and the healthy control group. Specifically, there was a significant increase in network node centrality in the MTG.R region (t-value = 4.195465, p-value = 0.000066), as well as a significant increase in network node efficiency (t-value = 3.939366, p-value = 0.000166). These data imply that the PACG group exhibited stronger network connectivity in the right middle temporal gyrus region compared to the healthy control group. (Supplementary Figure 2, Supplementary Table 4).
3.5.3. Network-Based Statistics Analysis
NBS analysis revealed a significant alteration in the strength of functional brain network connectivity in the PACG group compared to healthy controls, with a p-value of less than 0.05. This alteration involved 38 network nodes and 64 functional connections in the prefrontal, frontal, occipital, temporal, parietal, subcortical brain regions and cerebellum.
In terms of positive connectivity, the enhanced connections were mainly located in Prefrontal, Temporal, Occipital, and Cerebellum regions. In particular, the connections between Prefrontal zone and Occipital zone, Prefrontal zone and Temporal zone, Temporal zone and Cerebellum zone, and Occipital zone and Cerebellum zone were significantly enhanced. (Supplementary Figure 3A, Supplementary Table 5).
In contrast, in terms of negative connections, weakened connections were mainly found between Prefrontal area and Temporal area, Prefrontal area and Subcortical area, Frontal area and Temporal area, Frontal area and Cerebellum area, Temporal area and Subcortical area, Temporal zone with Occipital zone. (Supplementary Figure 3B, Supplementary Table 5).