|Year : 2019 | Volume
| Issue : 1 | Page : 22-28
Correlations Between Macula Vessel Density and Macula Thickness in Early Stage of Primary Open-Angle Glaucoma
Usman Bosso Abubakar1, Marchenko Ludmilla Nikolaevna2, Dzhumova Marina Fedrovna2, Kachan Tatsiana Vladimerovna2, Dalidovich Anastasiya Aleksandrovna2, Mushtina Tatsiana Anatolivna3
1 Department of Eye Diseases, Belarusian State Medical University, Minsk, Belarus; Eye Centre, General Hospital, Minna, Nigeria
2 Department of Eye Diseases, Belarusian State Medical University, Minsk, Belarus
3 Urban Ophthalmological Consultative and Diagnostic Center, 3rd Hospital, Minsk, Belarus
|Date of Web Publication||4-Jul-2019|
Dr. Usman Bosso Abubakar
Centre, General Hospital, Minna, Nigeria; Prospect Dzershenskovo 83, korpus 8. 220045, Minsk, Belarus
Source of Support: None, Conflict of Interest: None
Background: Glaucomatous optic neuropathy is associated with both structural and vascular changes. The objective of this study was to assess the possible relationship between the vascular and structural changes in the macula region in early stage of primary open-angle glaucoma (POAG). Purpose: The aim of this study was to evaluate correlations between superficial macula vessel density and parafoveal inner macula thickness in patients with early-stage POAG using optical coherence tomography angiography (OCT-A). Materials and Methods: Sixty-three patients were included in a prospective, cross-sectional study who were divided into three groups: early glaucoma (EG) group (50 eyes of 29 patients with early POAG), glaucoma suspects group (36 eyes of 21 patients), and control group (25 eyes of 13 healthy individuals). The mean age of the respective groups was 65.58 ± 9.53, 56.54 ± 9.32, and 51.62 ± 4.13. All patients underwent OCT-A scanning using RTVue-100 “Оptovue” (Optovue, Inc. Fremont, California, USA). The retina map vessel density 3 × 3 mm and retina thickness map 5 × 5 mm scans were used to obtain parameters of the macula vessel density and the parafoveal inner macula thickness, respectively. Parameters analyzed were the total, superior, inferior, nasal, and temporal superficial macula vessel densities and parafoveal inner macula thicknesses. Although area under the receiver operating characteristic curves were calculated for all available parameters, Spearman rank-order correlations were used to evaluate possible correlations between the macula vessel density and the structural changes. Results: The early glaucoma group had significant reduction in both the superficial macula vessel densities and the parafoveal inner macula thicknesses compared to the control group (P < 0.05), but parafoveal inner macula thickness parameters had higher area under the receiver operating characteristic curves. The macula density and the macula thickness parameters were positively correlated but none was statistically significant (P > 0.05). Conclusions: In early POAG, there is no statistically significant correlation between the superficial macula vessel densities and the parafoveal inner macula thicknesses using OCT-A RTVue-100. The parafoveal inner macula thicknesses are diagnostically more significant than superficial macula vessel densities.
Keywords: Optical coherence tomography angiography, primary open-angle glaucoma, superficial macula vessel density
|How to cite this article:|
Abubakar UB, Nikolaevna ML, Fedrovna DM, Vladimerovna KT, Aleksandrovna DA, Anatolivna MT. Correlations Between Macula Vessel Density and Macula Thickness in Early Stage of Primary Open-Angle Glaucoma. Niger J Ophthalmol 2019;27:22-8
|How to cite this URL:|
Abubakar UB, Nikolaevna ML, Fedrovna DM, Vladimerovna KT, Aleksandrovna DA, Anatolivna MT. Correlations Between Macula Vessel Density and Macula Thickness in Early Stage of Primary Open-Angle Glaucoma. Niger J Ophthalmol [serial online] 2019 [cited 2020 Jun 1];27:22-8. Available from: http://www.nigerianjournalofophthalmology.com/text.asp?2019/27/1/22/262062
| Introduction|| |
Glaucoma is a multifactorial chronic optic neuropathy with associated nerve fiber loss due to retinal ganglion cells (RGCs) damage. The final common pathway is usually cupping of optic nerve head (ONH) and irreversible vision loss. The hypotheses of the RGCs loss in glaucoma among other theories are “biomechanical factor” as a result of rise in intraocular pressure (IOP) on the axons and “vascular factor” due to reduced ocular blood flow as a consequence of vasospasm and/or autoregulatory dysfunction., In recent times, optical coherence tomography angiography (OCT-A), a high-speed, high-resolution spectral-domain OCT that can evaluate both structural and microvascular changes of the retina, scanned images analyzing both structural and vascular changes that are presented on the same page for comparison.
As the quest for the precise relationship between the structural and the vascular changes especially in the macula region in patients with early primary open-angle glaucoma (POAG) is still ongoing, the possible correlation between the parafoveal inner macula thickness and the superficial macula vessel density in the early stage of POAG using the OCT-A was evaluated in the current study.
| Material and methods|| |
This prospective, cross-sectional, hospital-based observational study was performed between December 2015 and June 2016 at the 3rd City Hospital, Minsk, Republic of Belarus. The research protocol was approved by the Belarusian State Medical University Review Board and was performed in accordance to the moral, ethical, and scientific principles of the clinical trials on human beings, as reflected in the Declaration of Helsinki of 1975 (as revised in 2000). Written informed consent was obtained from each participant after giving detailed explanations about the test.
The participants, totaling 63 (111 eyes), mainly Caucasians (90.5%), were categorized into three groups. The first group [early glaucoma (EG)] group comprised 29 patients (50 eyes), all with early glaucomatous optic neuropathy (GON). The second group comprised glaucoma suspects (GS) totaling 21 patients (36 eyes), and the third group was the normal (N) control group of 13 healthy individuals (25 eyes). The EG and the GS groups were patients referred from various clinics and polyclinics in the city whereas the N group had volunteered health personnel of the 3rd City Hospital, Minsk, and their relatives.
All individuals included in the study were subjected to a comprehensive ophthalmic evaluation at the 3rd City Hospital, Minsk. These included initial medical, family, and ocular histories, best-corrected distance visual acuity at 6 m using the Golovin–Sivtsev table (decimal fraction), dilated fundoscopy using the Heine Beta 200 LED ophthalmoscope (optotechnik Herrschiog, Deutsch land) for the direct ophthalmoscopic assessment of the ONH, slit-lamp biomicroscopy including the use of 78D lens (Volk Optical) for the indirect binocular ONH assessment, gonioscopy using the three mirror universal diagnostic lens of 18 mm (Ocular Instr., USA), IOP measurement using the Goldman tonometry, pachymetry using the Humphrey instruments Inc., model 850 (San Leandro, California, USA), and central visual field (CVF) testing on the standard automated perimetry using Humphrey Field Analyzer model 745, employing the Swedish Interactive Threshold Algorithm 30-2 (Carl Zeiss Meditec Inc., Dubling, California, USA). IOP was recorded after 14 days wash-off period of glaucoma drugs wherever indicated, and analyzed after adjusting for central corneal thickness wherever required.
CVF, a second reliable testing, results were used for analysis.
Case definition for the EG group were eyes that fulfilled any one of the following criteria: (1) early signs of glaucomatous ONH changes on dilated fundoscopic examination (vertical cup/disc ratio of 0.6–0.7 or asymmetry in the fellow eye >0.2; a localized notch in the rim; presence of minimal neuroretinal rim narrowing between 11-1 and 5-7 clock hours); (2) glaucomatous CVF defects [repeatable mean deviation (MD) of −1.5 to −6.0 dB; a glaucoma hemifield test (GHT) outside normal limits at 95% confidence limits confirmed on at least two visual field examinations]; and a persistent repeated difference of >2 mmHg in the IOP of any figure between the pair of eyes on follow-up presence of disc hemorrhage.
Case definition for GS group was as follows: (1) eyes of individuals with a normal CVF findings (MD within 95% limits of the normal reference and GHT within normal limits), but with a history of occasional rise in IOP >21 mmHg; and (2i) an asymmetry in the color saturation of the ONH, neuroretinal rim findings, violation of the inferior superior nasal temporal rule.
These are mainly individuals with ocular hypertension (IOP > 21 mmHg) who neither met any other criteria for stage 1 POAG, nor were considered healthy, that is, glaucoma stage 0 based on Hodapp-Anderson-Parrish scale. The inclusion of the EG group was aimed at getting any morphological and vascular changes of the central retinal region and their possible correlations at the earliest stage of the GON.
Case definition for N group was as follows: (1) normal CVF findings (MD within 95% limits of the normal reference and GHT within normal limits); (2) absent ONH and neuroretinal rim abnormalities; (3) normal anterior segment findings, IOP of <21 mmHg in both the eyes on the day of investigation and with no history of raised IOP >20 mmHg; and (4) no chronic ocular or systemic corticosteroid use in the past.
Inclusion criteria for all participants were as follows:
- age 40 years and above;
- best-corrected distance visual acuity of 0.6 and better;
- spherical and cylindrical correction not higher than ± 2.5D;
- a fully opened anterior chamber angle in its entire circumferences on gonioscopy devoid of pigmentations and/or exfoliative materials; and
- absence of vascular pathology, diabetes, or vasoactive medication.
All patients who met the inclusion criteria underwent a posterior retinal segment scans using OCT-A, RTVue-100, Optovue (Optovue, Inc. Fremont, California, USA), employing the split-spectrum amplitude-decorrelation angiography algorithm with tracking. This was performed by a well-trained examiner. The “retina thickness map 5 × 5 mm” and the “retina map-vessel density 3 × 3 mm” scans were used to obtain the parameters of the parafoveal inner macula thickness and the superficial macula vessel density, respectively. The total average, superior, inferior, nasal, and temporal parafoveal macula thicknesses (in micrometer) and their respective quadrants vessel density (in %) were the parameters analyzed. The macula vessel density is considered the percentage area occupied by the vessels in these quadrants as calculated by the software of the OCT-A.
Only scans with signal strength index of ≥50 were included for the analyses.
| Statistical analysis|| |
The results were collected in an excel sheet (Microsoft Office 2013) and analyzed using STATISTICA (Version 12) software (Stat Soft Inc., Texas, USA). The distribution was charted on the histogram. A Spearman rank-order correlation was used to compare between the vessel density and the structural changes taking a P-value < 0.05 as statistically significant. Area under the receiver operating characteristic curves (AUCs), sensitivity, and specificity tests were calculated for all available parameters using receiver operating characteristic curves analysis (Atte Stat package, Atte Stat inc. Moscow, Russia).
| Results|| |
[Table 1] shows the sociodemographic characteristic of the study groups. The mean age was 64.45 ± 9.53 years in the EG group, 56.54 ± 9.32 years in the GS group, and 51.62 ± 4.13 years in the N group.
|Table 1 Main clinical and demographic characteristics of the study groups|
Click here to view
Dense microvascular network of the macula region was visible in the OCT-A images of most eyes [[Figure 1]]. Although some focal defects of avascular areas were noticed in some eyes of the EG group [[Figure 2]], we could not appreciate any differences in the calibers of the blood vessels between the groups.
|Figure 1 Patient S, 54 years. Example of an OCT-A image of retinal structures and vessel networks (3 × 3-mm scan area), with reports of full retinal thickness and superficial vessel density. In the vessel network image, the inner circle (1 mm in diameter) represent the fovea areas, whereas the outer circle (3 mm in diameter) covers the parafoveal area of the macula that is further divided into six subsections: superior, inferior, nasal, temporal, superior-hemifield, and inferior-hemifield. At the center is the foveal avascular region. Note: The values of the structural measurements in the figure were not used for analyses in our study as these represent measurements for full retina thickness (ILM-RPE). We used values for the inner macula thickness (ILM-IPL) of the 5 × 5 mm retina thickness map (not shown here). ILM = inner limiting membrane, IPL = inner plexiform layer, OCT-A = optical coherance tomography angiography. RPE = retinal pigment epithelial.|
Click here to view
|Figure 2 Patient D, 61 years. Angio-OCT images of retinal structures and vessel networks (3 × 3-mm scan area), with reports of full retinal thickness and superficial vessel density. In the vessel network image, a defect in vascular network is noted (red circle). OCT = optical coherence tomography.|
Click here to view
The mean values of both the superficial macula vessel density and the parafoveal inner macula thicknesses are presented in [Table 2] and [Table 3], respectively. The EG group was noticed to have reduced superficial macula vessel density and thinner parafoveal inner macula thickness as compared to the GS and N (P < 0.05) groups in most of the analyzed parameters.
|Table 2 Means and standard deviations of the superficial vessel density of the parafoveal macula region by angio-OCT among the study groups|
Click here to view
|Table 3 Means and standard deviations of the parafoveal inner macula thickness by angio-OCT among the study groups|
Click here to view
AUCs, sensitivity/specificity test of the parameters of the superficial macula vessel density, and the parafoveal inner macula thicknesses are presented in [Table 4] and [Table 5]. Higher values of the AUCs were recorded among parameters of the parafoveal inner macula thickness as compared to the superficial macula vessel density with average temporal parafoveal inner macula thickness having the highest AUCs (0.927).
|Table 4 AUCs/sensitivity/specificity test for the superficial macula vessel density|
Click here to view
|Table 5 AUCs/sensitivity/specificity test for the parafoveal inner macula thickness|
Click here to view
[Table 6] and [Table 7] present the Spearman correlation analysis between the parameters of the parafoveal inner macula thickness and the superficial macula vessel density in the EG and the GS groups, respectively. Most of the parameters were positive but poorly correlated to each other in both the groups with none having a statistically significant value (P > 0.05).
|Table 6 Spearman correlations between parameters of parafoveal inner macula thickness and superficial parafoveal macula vessel density in the early glaucoma group|
Click here to view
|Table 7 Spearman correlations between parameters of parafoveal inner macula thickness and superficial parafoveal macula vessel density in the glaucoma suspect group|
Click here to view
| Discussions|| |
RGCs damage with a subsequent retinal nerve fiber loss is considered an important step in the pathogenesis of GON. In addition to the mechanical factor due to elevated IOP, vascular factors may play a critical role in the development of glaucoma.,,, Reduction in retinal vessel density, especially on the disc and peripapillary areas, in patients with POAG has been reported in some previous studies.,, These findings may support a role for focal ischemia as a causative factor for glaucoma. As the macula is shown to be involved in early GON, precise relationship between the nerve fiber/neuronal loss and reduction in the vessel density in the macula region in early GON is still inconclusive.
The OCT has remained the main noninvasive imaging device in the evaluation of structural damages of the retina in glaucoma. Although the damage of the RGC with a subsequent death/or atrophy of its axons is revealed on OCT imaging as thinning of the retinal nerve fiber layer, retinal vessel density measurements can be used as parameter for the evaluation of the retinal vascular dysfunction. The vessel densities are calculated by the software of the device as the ratio of the area occupied by the vessels.
Understanding the relationship between retinal vessel density and structural changes in early glaucoma may increase our knowledge of the role of retinal blood flow in the glaucoma cascade and in the pathophysiology of the diseases. It may also improve in the early diagnosis and treatment of glaucoma.
OCT-A is one of the recent generation spectral-domain OCT with a good intraretinal layer separation capability that can evaluate both structural and vascular parameters of the retina noninvasively., The imaging capability of the retinal vessels by the device is due to the incorporated angiographic software, and it is based on the flow characteristics of the vessel and the intrinsic motion contrast present in the vascular network hence not requiring the introduction of the dye externally. High-resolution images of both the deep and superficial retinal capillary networks are produced by the OCT-A.,
The retina thickness map program of the Optovue OCT-A allows good sampling points of the macula and provides values of three different layers of the macula region: full, inner, and outer macula thicknesses of the fovea, the parafovea, and the perifovea [with the inner plexiform layer (IPL) as the reference point for the division between the inner and outer segments]. Parameters of each of these sections can then be analyzed independently.
Similarly, the “retina map-vessel density” scans of the device allow evaluation of the retinal deep and the superficial vessel plexus independently (also obeying the IPL reference point for the division)., In the macula region, the superficial vessel plexus are the vascular networks embedded between the layers of the inner limiting membrane and the IPL. By the OCT-A nomenclature of the macula area, that portion is referred to as the inner macula. Consequently, the superficial capillary plexus as imaged using the OCT-A includes vasculature within the retinal nerve fiber layer and RGC layer. The deep capillary plexus therefore constitutes vasculature of the intermediate and outer retinal layers., Subsequently, any vascular factor from the respective sources of the superficial and the deep vessel or in the respective retinal plexus may have influence on the corresponding layers of the retina.
In the current study, the structural and the corresponding vascular changes of the parafoveal region of the macula in patients with early POAG using Optovue OCT-A and RTVue-100 was evaluated.
There was thinning of the parafoveal inner macula thicknesses in the EG group compared to the GS and the N groups. The thinning in the inner macula thickness in patients with the early stage of POAG as observed in our study confirms damage to the macula region in the initial stage of the glaucoma, as was revealed in the previous studies. There was also reduction in superficial vessel density of the parafoveal region of the macula in the EG group with compared to the GS and the N groups. The greatest reduction of superficial macula vessel density was found in the inferior inner macular sector in the EG group (52.16 ± 5.26%) versus the N group (56.12 ± 3.15%). These findings indicate a possible disturbance in the blood circulation of the macula region in patients with early POAG and are consistent with some previous studies that also identified reduction in macula vessel density in glaucoma patients.,,,
In our study, it was observed that parameters of the inner macula thickness [[Table 5]] have higher AUCs than those of the macula vessel density [[Table 4]]. This indicates higher diagnostic significance of the parafoveal inner macula thickness parameters as compared to those of the parafoveal superficial macula vessel density in the early stage of POAG using the OCT-A. Rao et al., however, argue that this relationship may change with increase in severity of the glaucoma stage. In our study, we compared sensitivity at 95% specificity value to find out if the variables significantly deviate from a normal distribution. It was observed that the parameters of the parafoveal inner macula thickness have higher chance of ruling out glaucoma than those of the vessel density.Spearman correlation analysis of the structural and the vascular changes of the parafoveal macula zone were positive but very poorly correlated in both the EG [[Table 6]] and the GS [[Table 7]] groups. The only moderate correlation (R = 0.277) in the EG group was between average inferior inner macula thickness and total average macula vessel densities. Absence of significance in correlation is in contrast to the study of Xu and Kong in which there was statistically significant positive correlation between macular perfusion (flow index/vessel area density) and the macular retina thickness (full/inner thickness) (R = 0.35–0.71, all P < 0.01). The outcome of the correlation result in their study probably may not be unconnected to the use of full macula thickness parameter for the correlation analyses. Similarly, the work of Yang et al. also showed a statistically significant correlation between the superficial macula vessel density and the Ganglion cell complex but in young healthy myopic patients.
Uniqueness of our study is the correlation of the inner retinal segment and its corresponding superficial microvasculature in the parafoveal area of the macula in patients with early POAG.
From this study, it may be postulated that there is alteration in the superficial vascular network of the macula region in the early stage of POAG. And that those vascular changes are poorly correlated to the structural changes in the macula region in early GON.
The findings may provide additional information to the clinicians in the diagnosis and management of glaucoma.
Further research in the area of our study is a multiethnic approach, a larger size of participants, and a longitudinal follow-up including the influence of the use of vasoprotectors on the vessel density.
Limitations of our study
In our analysis, there is absence of the blood flow index of the macula region. Relatively small number and ethnically homogeneous (90.5% Caucasians) study participants may limit its universal applicability. A larger sample size and a multiethnicity approach is suggested for further studies.
| Conclusion|| |
There is no statistically significant correlation between the superficial macula densities and parafoveal inner macula thickness, although there is thinning in the parafoveal inner macula thicknesses and a reduction in the superficial macula vessel density in early stage of POAG using the OCT-A RTVue-100. The parameters of the superficial parafoveal vascular density, compared with the parafoveal inner macula thickness, have limited diagnostic significances in the early stage of POAG.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Agarwal R, Gupta SK, Agarwal P, Saxena R, Agrawal SS. Current concepts in the pathophysiology of glaucoma. Indian J Ophthalmol 2009;57:257-66.
] [Full text]
Leske MC, Connell AM, Wu SY, Hyman LG, Schachat AP. Risk factors for open-angle glaucoma. The Barbados Eye Study. Arch Ophthalmol 1995;113:918-24.
Flammer J. The vascular concept of glaucoma. Surv Ophthalmol 1994;38(suppl):S3-6.
de Carlo TE, Romano A, Waheed NK, Duker JS. A review of optical coherence tomography angiography (OCTA). Int J Retina Vitreous 2015;1:5.
Mills RP, Budenz DL, Lee PP, Noecker RJ, Walt JG, Siegartel LR et al.
Categorizing the stage of glaucoma from pre-diagnosis to end-stage disease. Am J Ophthalmol 2006;141:24-30.
Quigley HA, Addicks EM, Green WR, Maumenee AE. Optic nerve damage in human glaucoma. II. The site of injury and susceptibility to damage. Arch Ophthalmol 1981;99:635-49.
Bonomi L, Marchini G, Marraffa M, Bernardi P, Morbio R, Varotto A. Vascular risk factors for primary open angle glaucoma: the Egna-Neumarkt Study. Ophthalmology 2000;107:1287-93.
Yarmohammadi A, Zangwill LM, Diniz-Filho A, Suh MH, Manalastas PI, Fatehee N et al.
Optical coherence tomography angiography vessel density in healthy, glaucoma suspect, and glaucoma eyes. Invest Ophthalmol Vis Sci 2016;57:451-9.
Yarmohammadi A, Zangwill LM, Diniz-Filho A, Saunders LJ, Suh MH, Wu Z et al.
Peripapillary and macular vessel density in patients with glaucoma and single-Hemifield visual field defect. Ophthalmology 2017;124:709-19.
Wang X, Jiang C, Ko T, Kong X, Yu X, Min W et al.
Correlation between optic disc perfusion and glaucomatous severity in patients with open-angle glaucoma: an optical coherence tomography angiography study. Graefes Arch Clin Exp Ophthalmol 2015;253:1557-64.
Hood DC, Raza AS, de Moraes CG, Liebmann JM, Ritch R. Glaucomatous damage of the macula. Prog Retin Eye Res 2013;32:1-21.
Pagliara MM, Lepore D, Balestrazzi E. The role of OCT in glaucoma management. Prog Brain Res 2008;173:139-48.
Jia Y, Tan O, Tokayer J. Split-spectrum amplitude-decorrelation angiography with optical coherence tomography. Opt Express 2012;20:4710-25.
Spaide RF, Klancnik JM, Cooney MJ. Retinal vascular layers imaged by fluorescein angiography and optical coherence tomography angiography. JAMA Ophthalmol 2014;133:E1-6.
Hassan M, Sadiq MA, Halim MS, Afridi R, Soliman MK, Sarwar S et al.
Evaluation of macular and peripapillary vessel flow density in eyes with no known pathology using optical coherence tomography angiography. Int J Retina Vitreous 2017;3:27.
Jia Y, Wei E, Wang X, Zhang X, Morrison JC, Parikh M et al.
Optical coherence tomography angiography of optic disc perfusion in glaucoma. Ophthalmology 2014;121:1322-32.
Gadde SGK, Anegondi N, Bhanushali D, Chidambara L, Yadav NK, Khurana A et al.
Quantification of vessel density in retinal optical coherence tomography angiography images using local fractal dimension. Investig Ophthalmol Vis Sci 2016;57:246-52.
Anguelov B, Petrova K. Diagnostic accuracy of the parameters from ganglion cell complex map, evaluated with SD-OCT in primary open angle glaucoma. Ophthalmol Russia 2014;11:28-32.
Lommatzsch C, Rothaus K, Koch JM, Heinz C, Grisanti S. OCTA vessel density changes in the macular zone in glaucomatous eyes. Graefes Arch Clin Exp Ophthalmol 2018;256:1499-508.
Triolo G, Rabiolo A, Shemonski ND, Fard A, Di Matteo F, Sacconi R et al.
Optical coherence tomography angiography macular and peripapillary vessel perfusion density in healthy subjects, glaucoma suspects, and glaucoma patients. Invest Ophthalmol Vis Sci 2017;58:5713-22.
Rao HL, Pradhan ZS, Weinreb RN, Reddy HB, Riyazuddin M, Dasari S et al.
Regional comparisons of optical coherence tomography angiography vessel density in primary open-angle glaucoma. Am J Ophthalmol 2016;171:75–83.
Xu H, Kong XM. Study of retinal microvascular perfusion alteration and structural damage at macular region in primary open-angle glaucoma patients. Chin J Ophthalmol 2017;53:98-103.
Yang S, Zhou M, Lu B, Zhang P, Zhao J, Kang M et al.
Quantification of macular vascular density using optical coherence tomography angiography and its relationship with retinal thickness in myopic eyes of young adults. J Ophthalmol 2017;2017:1397179.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7]