Evaluating the association between autoimmune disease and normal tension glaucoma: a retrospective case-control study

Abstract

Background

Limited population-based data as well as proposed mechanisms of retinal ganglion cell (RGC) loss suggest autoimmune disease may be a risk factor for glaucoma, the leading cause of irreversible blindness worldwide. Though intraocular pressure (IOP) is the leading risk factor for glaucoma onset and progression, a subset of glaucoma referred to as normal tension glaucoma (NTG) may be more likely to be associated with IOP-independent mechanisms of RGC injury including those of an inflammatory or immune nature.

Methods

This retrospective case-control study enrolled 277 patients with NTG and the same number of age- and sex-matched controls to determine whether autoimmune disease diagnosis, treatment thereof, or relevant laboratory markers are associated with NTG.

Results

There was no significant difference between the two groups in frequency of autoimmune disease overall, autoimmune disease catagorized by mechanism or organ involvement, or individual autoimmune disease including psoriasis (6% vs. 5%), rheumatoid arthritis (5% vs. 4%), inflammatory bowel disease (2% vs. 3%), Sjögren’s syndrome (1% vs. 1%), sarcoidosis (1% vs. 1%), autoimmune thyroiditis (1% vs. 0%), type 1 diabetes (1% vs. 0%), or systemic lupus erythematosus (1% vs. 0%). There was also no significant difference in laboratory values or treatment of identified autoimmune conditions.

Conclusions

Our study found no significant association between autoimmune disease and NTG, suggesting that other factors may play a more significant role in the pathogenesis of NTG.

Background

Glaucoma is the most common cause of irreversible vision loss and the second leading cause of overall blindness worldwide [1]. It is a neurodegenerative disorder characterized by retinal ganglion cell (RGC) loss and axon degeneration which may lead to progressive visual field loss and eventually blindness [2]. Intraocular pressure (IOP) is the leading and only reliable treatable risk factor for the glaucomas including primary open angle glaucoma (POAG). Normal tension glaucoma (NTG) is a subtype of open angle glaucoma without a history of IOP over a measured value of 21 mmHg [3]. Though IOP is still the major risk factor for onset and progression in NTG [4], mechanisms of RGC loss that are independent or indirectly dependent of IOP are being explored [5]. For instance, the vascular theory suggests that glaucomatous optic neuropathy results from inadequate blood supply, caused by elevated IOP or other factors that impair ocular blood flow [6]. Other risk factors currently under investigation include genetic susceptibility, structural and mechanical stress, excessive glutamate stimulation, mitochondrial dysfunction, oxidative stress, and aberrant immune and inflammatory responses [7, 8].

Several studies have shown that neurons in glaucoma are subject to autoimmune and inflammatory stress [9–11]. Activation of immune cells, including microglia and astrocytes, are triggered by various factors and results in the release of pro-inflammatory cytokines, chemokines, and other inflammatory mediators, initiating a series of neuroinflammatory events [12, 13]. Prolonged exposure to these inflammatory molecules leads to dysfunction, apoptosis, and degeneration of RGCs, contributing to the neural damage observed in glaucoma [14, 15]. Neuroinflammation may further disrupt the integrity of the blood retinal barrier allowing immune cells and inflammatory molecules to infiltrate the retina, further exacerbating neuroinflammation and compromising the microenvironment necessary for RGC survival [16]. The association between neuroinflammation and glaucoma is complex and multifactorial. While neuroinflammatory processes have been identified in glaucoma, the precise mechanisms and clinical implications are still being studied [12].

Prior reports have studied immune system-related risk factors and glaucomatous optic neuropathy in NTG, although the literature on this topic is not in agreement. NTG has been associated with systemic autoimmune disease [17, 18] and heightened humoral immune responses in some studies [19, 20], while others refute this association [21, 22]. A recent study described a higher prevalence of T-cell mediated autoimmune disease in POAG compared to controls [23], an area that has not been previously examined in NTG. Comprehensive studies examining various aspects of autoimmunity in NTG are lacking. Therefore, the goal of this study was to determine whether systemic autoimmune disease diagnosis, treatment thereof, or relevant laboratory markers are associated with NTG using a case control study design.

Methods

The patient cohort included in this study were based on our prior report studying the association between vascular risk factors and NTG [7]. Briefly, patients with NTG seen at Mayo Clinic, Rochester, MN, between January 1, 2005, and December 31, 2015, were retrospectively identified using ICD-9 code 365.12. Inclusion criteria for the study group included a visit billing code diagnosis of NTG and subsequent verification of the billing diagnosis with clinical diagnosis via chart review and age of 40 years or older at time of visit. The clinical diagnosis of NTG was made based on the presence of open angles, glaucomatous optic neuropathy evidenced by optic nerve head notching and/or cupping, and an IOP history that did not exceed 21 mmHg. Patients were excluded if not meeting inclusion criteria, lacking a comprehensive medical visit on record, having a history of nonglaucomatous type optic neuropathy, or any form of glaucoma other than NTG. No other exclusion criteria were employed, including patients with history of cataract or refractive surgery.

An age-matched and sex-matched control group was also selected. Inclusion criteria for the control group was an age of 40 years or older at time of visit with a billing-code diagnosis that was subsequently verified by chart review for refractive error and refraction. Exclusion criteria were the same as the NTG group plus no prior diagnosis of any form of glaucoma. This study was conducted in compliance with the Health Insurance Portability and Accountability Act, and the Mayo Clinic Institutional Review Board deemed the study exempt.

There were initially 370 patients identified with NTG. Upon confirmatory chart review, after excluding 93 patients (62 excluded for lack of information on eye exam, 17 excluded due to exclusion of matched control, 7 excluded for other causes of optic neuropathy, and 7 excluded for an alternate type of glaucoma), 277 remained in the NTG group. For the control group, 19,523 patients met the inclusion criteria. Using a 1:1 matching process based on sex, age (± 5 years), and visit date (within 1 year), 277 control patients were randomly selected using a SAS function that employed the GREEDY algorithm. The selected controls were reviewed to determine if there was a reason for exclusion as previously outlined. Any excluded control patient was substituted with a new, appropriately matched control to maintain each group at a total of 277 patients.

All eligible medical records were retrospectively reviewed for patient demographics, systemic and organ-specific autoimmune disorders, serum autoantibodies, systemic immunosuppressive medications, and baseline eye exam. For ocular findings, both eyes of a patient were included and these values were recorded in each eye separately and averaged into a single value to be used in the study. Data collected from the baseline eye examination included IOP (in mmHg) measured by applanation or rebound tonometry (Icare), refractive error (in diopters), cup-to-disc ratio and presence of a disc hemorrhage as determined by slit lamp examination with a 90 diopter lens, and central corneal thickness (CCT) measurements, when available.

Autoimmune disorders were grouped by predominant immune mechanism as previously described [23]. T-cell mediated disorders included psoriasis, alopecia areata, vitiligo, uveitis, spondyloarthritides, multiple sclerosis, sarcoidosis, autoimmune gastritis, type 1 diabetes, polymyositis. B-cell mediated disorders included systemic lupus erythematosus and bullous pemphigoid. Combined T-cell and B-cell mediated disorders included rheumatic fever, inflammatory bowel disease, rheumatoid arthritis, Sjogren’s syndrome, juvenile idiopathic arthritis, autoimmune thyroiditis, systemic sclerosis, autoimmune hemolytic anemia, autoimmune hepatitis, chronic inflammatory demyelinating polyradiculoneuropathy (CIDP).

The presence of the following serum autoantibodies was recorded: antinuclear antibodies (ANA), antineutrophil cytoplasmic antibodies (ANCA), rheumatoid factor (RF), cyclic citrullinated peptide antibodies (CCP), anticentromere antibodies (ACA), antihistone antibodies (AHA), extractable nuclear antigen (ENA) antibodies (including Sjögren’s syndrome (SS) including SS-A/Ro and SS-B/La, Sm, RNP, Scl 70, Jo 1), immunoglobulins (Igs) (including IgA, IgM, IgG), antiphospholipid (aPL) antibodies (anticardiolipin, anti-beta2-glycoprotein I, lupus anticoagulant). Due to the retrospective nature of the review, there was not a standardized timing for the serum assays of each patient.

Patient data was summarized as raw counts and proportions for categorical variables and mean and standard deviation (SD) for continuous variables. Comparisons among the NTG and control group were performed using 2-sample t tests for continuous variables and Fisher’s exact test for categorical variables. The data set was normally distributed, and a P-value < 0.05 was considered significant. Analysis was completed using SAS, version 9.4 (SAS Institute Inc., Cary, NC).

Results

A total of 277 patients were included in the NTG group. Mean age at presentation was 70 years and 57% were female. An age and sex-matched control group of 277 patients was selected. There were no differences in the demographic characteristics between these two groups (Table 1). Patients in the NTG group had a lower mean IOP (14 vs. 15 mmHg, P < 0.001), greater myopic refractive error (− 1.6 vs. −0.1 D, P < 0.001), increased cup-to-disc ratio (0.7 vs. 0.3, P < 0.001), and increased presence of disc hemorrhage (21 vs. 0, P < 0.001) compared to patients in the control group (Table 1). There was no difference in the central corneal thickness between groups (Table 1). One patient in the NTG group had a history of refractive surgery, specifically laser-assisted in situ keratomileusis (LASIK), while no patients in the control group had this history.

Table 1.

Demographics and baseline eye examination in NTG and control groups

Characteristics	NTG, n = 277 (%)	Control, n = 277 (%)	p-value
Demographics
Age in years, mean (SD)	70 (11)	69 (10)	0.478
Sex: female	159 (57)	159 (57)	1
Race: white	250 (90)	259 (94)	0.213
Baseline eye examination, mean (SD)
Intraocular pressure (mmHg)	14 (3)	15 (3)	< 0.001
Refractive error (D)	−1.6 (2.8)	−0.1 (2.6)	< 0.001
Cup-to-disc ratio	0.7 (0.1)	0.3 (0.1)	< 0.001
Disc hemorrhage	21 (8)	0 (0)	< 0.001
Central corneal thickness (µM)	545 (40)	585 (13)	0.08

The prevalence of autoimmune disorders in NTG and control groups are listed (Table 2). No significant difference was noted between these two groups in the prevalence of all autoimmune disorders grouped together (23 vs. 18%, P = 0.21) or of autoimmune disorders when classified by specific organ involvement including endocrine (2 vs. 0%, P = 0.12), gastrointestinal (3 vs. 3%, P = 1.0), neurologic (0 vs. 1%, P = 1.0), hematologic 0 vs. 1%, P = 0.50), dermatologic (8 vs. 6%, P = 0.62), or ocular (1 vs. 0%, P = 0.50). Furthermore, no significant difference was seen when looking at specific autoimmune disorders including rheumatoid arthritis (5 vs. 4%, P = 0.84), Sjögren’s syndrome (1 vs. 1%, P = 1.0), systemic lupus erythematosus (1 vs. 0%, P = 0.62), sarcoidosis (1 vs. 1%, P = 0.69), autoimmune thyroiditis (1 vs. 0%, P = 0.12), type 1 diabetes (1 vs. 0%, P = 1.0), inflammatory bowel disease (2 vs. 3%, P = 0.77) psoriasis (6 vs. 5%, P = 0.46), or vitiligo (1 vs. 1%, P = 1.0). There was also no difference between the two groups with further subgroup analysis into predominant immune mechanism including T-cell mediated (11 vs. 9%, P = 0.48), B-cell mediated (2% vs. 0%, P = 0.22), and combined T-cell and B-cell mediated (10 vs. 9%, P = 0.77) autoimmune conditions.

Table 2.

Autoimmune disorders in NTG and control groups

Characteristics	NTG, n = 277 (%)	Control, n = 277 (%)	p-value
Systemic autoimmune disorders	26 (9)	21 (8)	0.542
Rheumatoid arthritis	13 (5)	11 (4)	0.835
Sjogren’s syndrome	4 (1)	4 (1)	1
Systemic lupus erythematosus	3 (1)	1 (0)	0.624
Sarcoidosis	2 (1)	4 (1)	0.686
Spondyloarthritides	2 (1)	1 (0)	1
Systemic sclerosis	1 (0)	0 (0)	1
Autoimmune myopathies	1 (0)	0 (0)	1
Antiphospholipid antibody syndrome	0 (0)	0 (0)	-
Vasculitides	0 (0)	0 (0)	-
Endocrine autoimmune disorders	6 (2)	1 (0)	0.123
Autoimmune thyroiditis	4 (1)	0 (0)	0.124
Type 1 diabetes	2 (1)	1 (0)	1
Gastrointestinal autoimmune disorders	7 (3)	7 (3)	1
Inflammatory bowel disease	5 (2)	7 (3)	0.772
Pernicious anemia and autoimmune gastritis	1 (0)	0 (0)	1
Autoimmune hepatitis	1 (0)	0 (0)	1
Celiac disease	0 (0)	0 (0)	-
Neurologic autoimmune disorders	1 (0)	2 (1)	1
Multiple sclerosis	1 (0)	1 (0)	1
CIDP	0 (0)	1 (0)	1
Myasthenia gravis	0 (0)	0 (0)	-
Hematologic autoimmune disorders	0 (0)	2 (1)	0.499
Autoimmune hemolytic anemia	0 (0)	2 (1)	0.499
Dermatologic autoimmune disorders	22 (8)	18 (6)	0.623
Psoriasis	18 (6)	13 (5)	0.460
Vitiligo	2 (1)	2 (1)	1
Alopecia areata	0 (0)	3 (1)	0.249
Bullous pemphigoid	2 (1)	0 (0)	0.499
Ocular autoimmune disorders (Uveitis)	2 (1)	0 (0)	0.499
All autoimmune disorders	64 (23)	51 (18)	0.209
Predominant immune mechanism
T-cell mediated	31 (11)	25 (9)	0.48
B-cell mediated	5 (2)	1 (0)	0.22
Combined T-cell and B-cell mediated	28 (10)	25 (9)	0.77

CIDP chronic inflammatory demyelinating polyradiculoneuropathy

The presence of serum autoantibodies in NTG and control groups was also analyzed (Table 3). Compared to the control group, patients in the NTG group had similar levels compared to patients in the control group of serum antibodies including ANA (18 vs. 21%, P = 0.83), ENA (3 vs. 7%, P = 0.39), CCP (11 vs. 10%, P = 1.0), Igs (12 vs. 17%, P = 0.72), and rheumatoid factor (31 vs. 11%, P = 0.21). Mean immunoglobulin levels were also similar between groups (Table 4). Though there was a trend towards higher IgA levels in the NTG group (2.38 vs. 1.69 g/L, P = 0.055), this was not statistically significant. No difference was seen in IgM levels (1.36 vs. 4.20, P = 0.70) or IgG levels (12.76 vs. 13.76, P = 0.95). There was also no difference in the use of systemic immunosuppressive medications (Table 5) to treat these conditions (15 vs. 10%, P = 0.16), including corticosteroids (8 vs. 5%, 0.24), disease-modifying antirheumatic drugs (4 vs. 3%, P = 0.64), and B-cell inhibitors (1 vs. 1%, P = 1.0).

Table 3.

Serum autoantibodies in NTG and control groups

Characteristics	NTG, n = 277 (%)	Control, n = 277 (%)	p-value
Serum antibodies (elevated/total)
ANA	9/50 (18)	20/94 (21)	0.828
ENA	2/63 (3)	3/42 (7)	0.387
CCP	5/44 (11)	5/48 (10)	1
Igs	5/30 (12)	4/34 (17)	0.723
RF	5/16 (31)	2/18 (11)	0.214
aPL	1/6 (17)	0/2 (0)	1
ACA	1/3 (33)	0/3 (0)	1
ANCA	0/3 (0)	0/2 (0)	1
AHA	0/0 (0)	0/1 (0)	1

ANA antinuclear antibodies, ENA extractable nuclear antigen antibodies, CCP cyclic citrullinated peptide antibodies, Igs immunoglobulins, RF rheumatoid factor, aPL antiphospholipid antibodies, ACA anticentromere antibodies, ANCA antineutrophil cytoplasmic antibodies, AHA antihistone antibodies

Table 4.

Mean immunoglobin levels in NTG and control groups

Characteristics	NTG, n = 277 (%)	Control, n = 277 (%)	p-value
Immunoglobulin levels in g/L
IgA level, mean (SD)	2.38 (1.43)	1.69 (1.07)	0.055
IgM level, mean (SD)	1.36 (1.77)	4.20 (8.54)	0.704
IgG level, mean (SD)	12.76 (4.06)	13.76 (7.27)	0.952

IgA normal range: 0.61–3.56 g/L; IgM normal range: 0.37–2.86 g/L, IgG normal range: 7.67–15.90 g/L

Table 5.

Immunosuppressive medications in NTG and control groups

Characteristics	NTG, n = 277 (%)	Control, n = 277 (%)	p-value
Immunosuppressive medications	41 (15)	29 (10)	0.159
Corticosteroids	23 (8)	15 (5)	0.239
Traditional DMRADs	11 (4)	8 (3)	0.642
TNF inhibitors	1 (0)	1 (0)	1
Interleukin inhibitors	1 (0)	0 (0)	1
B-cell inhibitors	4 (1)	3 (1)	1
T-cell inhibitors	1 (0)	0 (0)	1
Plasmapheresis	0 (0)	2 (1)	0.499

ANA antinuclear antibodies, ENA extractable nuclear antigen antibodies, CCP cyclic citrullinated peptide antibodies, Igs immunoglobulins, RF rheumatoid factor, aPL antiphospholipid antibodies, ACA anticentromere antibodies, ANCA antineutrophil cytoplasmic antibodies, AHA antihistone antibodies, DMRADs disease-modifying antirheumatic drugs, TNF tumor necrosis factor

Discussion

NTG is a type of glaucoma with disease onset and/or progression of optic nerve damage despite IOP in the normal range [3]. Multiple mechanisms of RGC loss in NTG have been proposed in addition to IOP [4, 24]. While prior studies suggest the immune system plays a role in its pathogenesis, relatively little data is available, show different findings, and are limited in scope [17, 19, 20, 23]. In this retrospective case-control study, we did not find support to the hypothesis that the immune system plays a role in NTG pathogenesis as we found no difference between systemic autoimmune disease diagnosis, treatment thereof, or relevant laboratory markers in patients with NTG and an age and sex-matched control group.

Inflammatory and autoimmune processes have recently gained increased consideration in the pathogenesis of NTG [25–27]. It is postulated that immunoregulation plays a pivotal role in determining the fate of RGCs, with various stressors including aging, oxidative stress, altered vascular perfusion, structural changes in neuronal tissue, autoimmunity, and others exerting influence on this process [28, 29]. Moreover, heat shock proteins (HSPs) exhibit upregulation in glaucomatous eyes and have been implicated in autoimmune responses due to their antigenic properties [30]. Animal studies further support the potential involvement of autoimmunity in glaucoma by demonstrating that immunization with HSPs can induce RGC degeneration and axon loss [31]. Additionally, there is evidence of aberrant immunity in glaucoma patients, such as the presence of autoantibodies targeting various ocular components, immunoglobulin deposition in the RGC layer, and innate immune responses featuring complement activation [32–38]. While these findings collectively suggest a role for inflammation and autoimmunity in NTG, comprehensive population-based studies are currently lacking.

In our study, NTG was not associated with a statistically significant increase in diagnosis of immune related disease. These results contrast a case control study by Cartwright et al., which found that 20 (30%) out of 67 patients with NTG had one or more immune related diseases compared to 5 (8%) out of 67 patients with ocular hypertension [17]. While we found a similar proportion of patients with autoimmune disorders in the NTG group (29%), there was no difference when compared to patients in the control group (23%). This discrepancy may be explained by different patient populations used for the control group which make direct comparison challenging. To our knowledge, there were no other studies evaluating the prevalence of autoimmune disorders in NTG.

Furthermore, in subgroup analysis, NTG was also not associated with the humoral immune response. Prior reports on antibody levels are not in complete agreement. In 1994, Wax et al. found that NTG patients had higher levels of antibodies ENA, particularly SS-A, compared to POAG controls (30% vs. 2%) [19]. These results appear to contrast our study; however, a subsequent report from the same group determined that the ELISA used against anti-SS-A antibodies was cross-reactive against human protein HSP60 antibodies resulting in false positive results [32]. Another study by Hamman et al. partially confirmed our results [20]. They studied the serum of 95 patients and found no statistical significant difference in the levels of IgM, IgG, ENA, or ANA, which is in agreement with our study but in contrast, noted high levels of IgA in the NTG and POAG groups compared to healthy controls [20]. Our study showed a trend towards higher IgA levels in the NTG group but this did not reach statistical significance. Similar to our study, Skonieczna et al. found no association between serum antibodies and NTG [21]. They tested the serum of 105 patients, 35 with NTG, 34 with POAG, and 36 controls, and found similar levels of ANA, IgG, IgA, and IgM [21]. They also determined the median levels of RF and anti-citrullinated protein antibodies to be highest in the NTG group, but these were within normal laboratory values [21].

NTG was also not associated with T cell mediated autoimmune disease in our study. Our results contrast a recent report by Lorenzo et al. [23]. They conducted a retrospective cross-sectional study of 172 patients with POAG and 179 controls who underwent ophthalmic surgery and found the prevalence of T-cell mediated disease to be higher in POAG patients than in controls (12% vs. 5.0%) [23]. The discrepancy between our studies may be explained by the study design and patient population. Our study was a case control study with age and sex-matched controls while theirs was a cross-sectional study of patients undergoing ophthalmic surgery. Their study showed differences in the baseline characteristics between the two groups, including that the POAG group was both older and had different racial demographics compared to the control group. Furthermore, they included giant cell arteritis (GCA) as a T-cell mediated autoimmune disease, while we excluded patients who had GCA and other causes of optic neuropathy. Given that their patient population consisted of subjects who were undergoing glaucoma surgery, cataract surgery, or both, it is likely that their results may not be directly comparable to patients in our study.

Overall, none of the immune-related metrics in our study showed an association with NTG. There are likely other risk factors and mechanisms contributing to the pathogenesis of NTG. For example, the vascular hypothesis suggests diminished perfusion of the optic nerve causes retinal ganglion cell stress and ultimately cell death and atrophy [7, 39]. Unlike POAG which is linked to elevated IOP, NTG occurs with IOP in the normal ranges [3]. The vascular hypothesis suggests that fluctuating or reduced blood flow to the optic nerve results in ischemic damage, contributing to the optic neuropathy observed in NTG patients [6, 39]. This impaired perfusion may result from systemic vascular dysregulation, leading to inadequate oxygen and nutrient delivery to the optic nerve [6, 39]. A study by our group utilized the same patient cohort as the current study and found multiple vascular conditions were associated with NTG including systemic hypertension, diabetes mellitus, peripheral vascular disease, migraine headache, Raynaud syndrome, anemia, systemic hypotension, and calcium channel blocker use [7]. Taken together, these studies suggest that unlike immune-related mechanisms, vascular-based mechanisms plays a role in NTG in our patient population.

This study has several limitations. Its retrospective design is limited by variable documentation in medical records and non-standardized assessments. Though confirmatory chart review validated NTG diagnosis, initial screening of NTG patients was via billing code; therefore, some NTG patients seen at our institution may not have been captured. Systemic autoimmune conditions may be correlated with one another [40], which can confound associations in our study. Furthermore, in order to limit the exclusion criteria, we did not exclude patients based on phakic status or a history of refractive surgery. One patient in the study had prior LASIK surgery, this history may underestimate IOP readings, resulting in an inappropriate NTG diagnosis or unreliable IOP measurements [41, 42]. Applanation tonometry is considered the gold standard for measuring IOP due to its greater accuracy and reliability compared to other methods including rebound tonometry [43]. In our clinical practice, it is standard practice for IOP to be assessed by applanation tonometry. For control patients, the IOP was assessed by either applanation or rebound tonometry which may have introduced bias [44].

Though our sample sizes were larger than other reports studying similar associations, the sample sizes, particularly in subgroup or disease-specific associations may be too small to show a statistically significant difference between groups. On the other hand, advanced statistical methods were not used to account for inter-eye correlations which may have resulted in artificially smaller P-values for the baseline eye examination [45, 46]. Finally, because the data were obtained from a single academic referral center, results may not be applicable to other institutions or patient populations.

Conclusions

In summary, this retrospective case-control study shows no statistically significant association between NTG and select autoimmune conditions, serum antibodies, or the use of immunosuppressive medications. Though mechanistically it would stand to reason that autoimmune risk factors may be associated with NTG, the results of our study do not support this association. Future larger, prospective studies may offer further clarification of this potential association.

Authors’ contributions

G.W.R contributed to conceptualization and supervision of the study; C.M.C.B and R.O.F developed the methodolody, C.M.C.B, R.O.F, and D.K. performed the data curation, D.O.H. ran the formal statistical analysis, C.M.C.B wrote the original draft. All authors reviewed the manuscript.

Funding

This research was funded Department of Ophthalmology, Mayo Clinic.

Data availability

The data generated and analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Patient consent was waived, the research involves no more than minimal risk to subjects. The study was conducted in compliance with the Health Insurance Portability and Accountability Act, and the Mayo Clinic Institutional Review Board deemed the study exempt.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.