Trends in First-Line Glaucoma Treatment from 2013 to 2024: A Multi-institutional and Multinational Cohort Study

Abstract

Introduction

Glaucoma remains a leading cause of adult blindness worldwide, highlighting the need for updated insights into contemporary first-line treatment patterns.

Methods

We conducted a retrospective cohort study using electronic health records from the TriNetX Global Collaborative Network (2013–2024). Patients aged ≥ 40 years with a diagnosis of open-angle glaucoma (OAG) or angle-closure glaucoma (ACG) were included. Monotonic trends in first-line glaucoma treatment were assessed over 12 years using Mann–Kendall tests, with Kendall’s τ and a false discovery rate-adjusted P value (the Q value) reported. A positive τ indicated an increasing trend and a negative τ a decreasing trend. First-line treatments were classified into three categories: medication (single or combination therapy), surgery, and laser therapy.

Results

We included 322,910 patients with OAG and 40,119 patients with ACG. In OAG, the use of medication as initial therapy declined over time (91.0% in 2013, 87.7% in 2018, and 85.8% in 2024; τ = − 0.515, Q = 0.036), whereas surgery increased (4.8%, 7.9%, and 9.0%, respectively; τ = 0.606, Q = 0.021), and laser treatment remained stable (4.2%, 4.5%, and 5.2%; τ = 0.303, Q = 0.193). In ACG, medication use increased (72.4%, 79.3%, and 88.6%; τ = 0.818, Q < 0.001), while both surgery (9.8%, 5.3%, and 2.6%; τ = − 0.879, Q < 0.001) and laser therapy (17.8%, 15.4%, and 8.8%; τ = − 0.788, Q < 0.001) declined.

Conclusions

Medication therapy accounted for most first-line treatment in both OAG and ACG between 2013 and 2024. Over time, treatment patterns shifted toward greater use of surgery in OAG and increased reliance on medication in ACG. These trends highlight the need for future studies to evaluate long-term outcomes and inform subtype-specific glaucoma care.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40123-026-01327-y.

Key Summary Points

Reduction of intraocular pressure through medications, laser therapy, and surgery remains the cornerstone of glaucoma management, although the choice of initial treatment often differs by glaucoma subtype.

There is limited evidence on how initial treatment patterns for glaucoma have evolved over time, particularly from multi-institutional and multinational longitudinal evaluations.

In this TriNetX Global Collaborative Network study, conducted from 2013 to 2024, the use of medications as initial therapy for open-angle glaucoma (OAG) declined over time, while surgical interventions increased and laser therapy remained stable. By contrast, medication use for angle-closure glaucoma (ACG) rose, whereas both surgery and laser therapy decreased.

Prostaglandin analogues were the most commonly prescribed first-line drug class, for both OAG and ACG.

The observed changes in initial treatment strategies for OAG and ACG likely reflect advances in pharmacotherapy and evolving clinical practice guidelines, and further studies are needed to evaluate long-term outcomes and guide subtype-specific glaucoma care.

Introduction

Glaucoma is a group of optic neuropathies characterized by progressive optic nerve damage, often associated with elevated intraocular pressure (IOP) [1]. Without timely intervention, it can lead to irreversible vision loss, making it a leading cause of blindness worldwide [2]. In 2020, an estimated 79.6 million people were affected by glaucoma, with open-angle glaucoma (OAG) being the most prevalent form in developed countries, while angle-closure glaucoma (ACG) is more common in Asia [3]. The aging population and disparities in healthcare access are expected to drive a significant rise in glaucoma incidence and prevalence [2, 4–6].

International guidelines consistently emphasize IOP reduction as the cornerstone of glaucoma management [7–9]. Treatment options include medications, laser therapy, and surgery. Prostaglandin analogues have become the preferred first-line treatment for OAG owing to their efficacy and safety, while beta-blockers remain a common alternative [10, 11], particularly in Asia, as a result of cost considerations [12]. When medications fail, laser and surgical interventions, including selective laser trabeculoplasty (SLT), minimally invasive glaucoma surgery, and traditional filtration surgery, are crucial, especially in ACG management [7–9]. SLT is gaining recognition as a first-line therapy for OAG, as demonstrated by the LiGHT trial [13, 14], which found SLT to be non-inferior to eye drops in IOP control, with superior cost-effectiveness.

While some studies have examined the treatment patterns for glaucoma [15, 16], comprehensive longitudinal evaluations based on multi-institutional global networks examining the trends in initial treatment changes remain limited. Specifically, there is a notable gap in large-scale, real-world evidence tracking the evolution of first-line therapeutic choices for both OAG and ACG over an extended period. This study aimed to describe global trends in first-line glaucoma management with medication, laser therapy, and surgery between 2013 and 2024 using a large healthcare database, and to provide crucial insights into contemporary clinical practice.

Methods

Database Information

We analyzed the TriNetX database, a global, federated health research network that provides real-time access to de-identified electronic medical records from diverse healthcare organizations [17]. Data analysis was performed in February 2025, at which time the database included records from 140 healthcare organizations spanning 21 countries, including Australia, Belgium, Brazil, Bulgaria, Estonia, France, Georgia, Germany, Ghana, Israel, Italy, Japan, Lithuania, Malaysia, Poland, Singapore, Spain, Taiwan, UAE, UK, and USA. This platform aggregates various types of clinical data, including diagnoses, medication profiles, procedures, and laboratory results to support clinical research and improve patient outcomes. Diagnoses, medications, laser and surgical treatments in the dataset are coded, respectively, using the International Classification of Diseases, Tenth Revision (ICD-10), Anatomical Therapeutic Chemical (ATC) classification system, Current Procedural Terminology (CPT), and ICD-10 Procedure Coding Systems. This data has been used in previous studies to generate high-quality evidence, including research in ophthalmology [18–21]. This study was performed in accordance with the Helsinki Declaration of 1964 and its later amendments, with the exception that this trial was retrospectively registered. It was approved by the Institutional Review Board of Chang Gung Medical Foundation (IRB number 202401821B1). The results are reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines [22].

Cohort Definition

The retrospective cohort study included patients aged 40 or older, diagnosed between 2013 and 2024 with OAG (ICD-10-CM: H40.10 to H40.12) or ACG (ICD-10-CM: H40.20 to H40.24).

Outcome Measures

We categorized glaucoma treatments into three groups: (1) Medication, including prostaglandin analogues, beta-blockers, carbonic anhydrase inhibitors, alpha-agonists, Rho kinase inhibitors, parasympathomimetics, and combination therapy. We defined combination therapy as the use of more than two antiglaucoma drug classes on the same prescription day, reflecting the clinician’s intended initial therapeutic regimen based on the patients’ condition upon treatment. (2) Surgery, including trabeculotomy, trabeculectomy, iridectomy, and other fistulization procedures. (3) Laser therapy, including selective laser trabeculoplasty, argon laser trabeculoplasty, and laser peripheral iridotomy (LPI). A complete list of glaucoma treatment codes is provided in Supplemental Tables 1–3. We calculated the initial glaucoma treatment among patients aged 40 years and older. For each patient, an index date was defined as the first recorded diagnosis of glaucoma (ICD-10 H40.x) within the study period. First-line treatment was defined as the initial glaucoma treatment (medication, laser, or surgery) occurring on or after the index date, following a 12-month lookback period during which no glaucoma medication prescriptions were recorded. We extracted annual treatment data to analyze the distribution of initial treatment modalities. For each study year, the denominator for incidence proportion included patients aged 40 or older with glaucoma within the year and no prior record of glaucoma medication prescription during the lookback period (up to 1 year before the start of that year). The numerator comprised patients from this denominator who were newly prescribed any glaucoma medication within that year.

Statistical Analysis

We assessed monotonic trends in treatment patterns for OAG and ACG over time using Mann–Kendall tests, reporting τ and P values [23]. The null hypothesis assumed no trend, and a significant result would indicate sufficient evidence to reject it. A positive Kendall’s τ would indicate an increasing trend, while a negative τ would indicate a decreasing trend [24]. To control the false discovery rate and minimize type I errors due to multiple comparisons within each cohort (i.e., multiple medication categories per year), we applied the Benjamini–Hochberg procedure [25]. To account for multiple comparisons, statistical significance was defined as a false discovery rate-adjusted P (the Q value) < 0.05, corresponding to a false discovery rate of 5% under the Benjamini–Hochberg procedure [26].

Results

Patient Characteristics

The study included a total of 363,029 patients with glaucoma. Table 1 presents the demographics of the OAG (n = 322,910) and ACG (n = 40,119) cohorts. Patients with OAG were older (mean age 68.5 ± 10.2 years), less often female (55.2%), and had more comorbidities compared with those with ACG (mean age 66.4 ± 10.3 years; 63.9% female).

Table 1.

Demographics of the study population

	OAG (n = 322,910)	ACG (n = 40,119)	p value
Age, mean (SD)a	68.5 (10.2)	66.4 (10.3)	< 0.001
Sex, %b
Male	44.8%	36.1%	< 0.001
Female	55.2%	63.9%
Race, %b
White	44.7%	44.5%	< 0.001
Black or African American	26.0%	19.7%
Asian	4.2%	9.0%
Other	19.8%	22.3%
Comorbidity, %b
Diabetes mellitus	19.5%	18.0%	< 0.001
Hypertension	34.8%	30.7%	0.317
Dyslipidemia	29.1%	26.2%	0.840
Obesity	8.2%	7.4%	0.252
Sleep disorder	8.2%	7.5%	0.093
Ischemic heart disease	8.8%	8.0%	0.527
Stroke	4.7%	4.5%	0.403
Depression	5.2%	5.6%	0.439
Schizophrenia	0.3%	0.4%	0.001
Epilepsy	1.0%	0.9%	0.752
Migraine	1.6%	2.0%	0.949
COPD	3.7%	3.8%	0.887
Asthma	4.3%	4.2%	0.906
Hypothyroidism	6.9%	6.7%	0.654
Cancer	15.4%	13.4%	0.907
Cataract	10.8%	11.7%	0.835

OAG open-angle glaucoma, ACG angle-closure glaucoma, SD standard deviation, COPD chronic obstructive pulmonary disease

^aIndependent t test

^bChi-squared test

First-Line Treatment Strategy for Glaucoma

Medications consistently remained the predominant first-line treatment for both OAG and ACG from 2013 to 2024. In OAG (Fig. 1a), 85.8–91.0% of patients initiated therapy with medication, showing a decreasing trend over time (91.0% in 2013, 87.7% in 2018, and 85.8% in 2024; τ = − 0.515, Q = 0.036). Surgical treatment accounted for 4.8–9.0% of initial therapies, with an increasing trend (4.8%, 7.9%, and 9.0% across the same years; τ = 0.606, Q = 0.021). Laser therapy represented 3.6–5.2% of initial treatments, with a non-significant change (4.2%, 4.5%, and 5.2%; τ = 0.303, Q = 0.193).

Fig. 1.

First-line treatment strategy for OAG and ACG from 2013 to 2024. a The tests for monotonic trends in the time series for medication, laser, and surgical treatment for OAG yielded τ = − 0.515 (Q = 0.036*), τ = 0.303 (Q = 0.193), and τ = 0.606 (Q = 0.021*), respectively. b The tests for monotonic trends in the time series for medication, laser, and surgical treatment for ACG yielded τ = 0.818 (Q < 0.001*), τ = − 0.788 (Q < 0.001*), and τ = − 0.879 (Q < 0.001*), respectively. The Benjamini–Hochberg procedure was applied to control the type 1 error. The significant values are marked with an asterisk (*)

In the ACG group (Fig. 1b), 72.4–89.2.% of the patients initiated therapy with medication, showing an increasing trend over time (72.4% in 2013, 79.3% in 2018, and 88.6% in 2024; τ = 0.818, Q < 0.001). Surgical treatment accounted for 2.6–9.8% of initial therapies, with a decreasing trend (9.8%, 5.3% and 2.6% across the same years; τ = − 0.879, Q < 0.001). Laser therapy accounted for 8.2–17.8% of first-line treatments, also decreasing over time (17.8%, 15.4%, and 8.8%; τ = − 0.788, Q < 0.001).

Proportion of Individual Medications Prescribed as Initial Treatment

Among patients with OAG initially treated with medication (Fig. 2a), 63.0–66.5% received a single drug, showing a non-significant, decreasing trend over time (63.8% in 2013, 66.5% in 2018, and 63.0% in 2024; τ = − 0.485, Q = 0.079). By contrast, 33.5–37.0% received combination therapy, with a non-significant, increasing trend (36.2%, 33.5%, and 37.0% across the same years; τ = 0.485, Q = 0.079). As regards individual drug classes, 42.4–42.6% received prostaglandin analogues, with a non-significant trend (τ = − 0.424, Q = 0.075); 8.0–11.0% of patients received beta-blockers, with a declining trend (11.0% in 2013, 9.5% in 2018, and 9.0% in 2024; τ = − 0.576, Q = 0.039), and 3.4–5% received carbonic anhydrase inhibitors with a non-significant trend (τ = 0.455, Q = 0.082). Furthermore, 5.6–6.6% of patients received alpha-agonists with a non-significant trend (τ = 0.455, Q = 0.082), while 0.7–1.0% received parasympathomimetics with a non-significant trend (τ = − 0.424, Q = 0.075). Use of Rho kinase inhibitors increased from 0.1% in 2013 to 1.0% in 2024 (τ = 0.697, Q = 0.014), which was significant.

Fig. 2.

Proportion of individual antiglaucoma drugs used for OAG and ACG from 2013 to 2024. a The tests for monotonic trends in the time series for prostaglandin analogues, beta-blockers, carbonic anhydrase inhibitors, alpha-agonists, parasympathomimetics, Rho kinase inhibitors, and combination therapy uses for OAG yielded τ = − 0.424 (Q = 0.075), τ = − 0.576 (Q = 0.039*), τ = 0.455 (Q = 0.082), τ = 0.455 (Q = 0.082), τ = − 0.424 (Q = 0.075), τ = 0.697 (Q = 0.014*), and τ = 0.485 (Q = 0.079), respectively. b The tests for monotonic trends in the time series for prostaglandin analogues, beta-blockers, carbonic anhydrase inhibitors, alpha-agonists, parasympathomimetics, Rho kinase inhibitors, and combination therapy uses for ACG yielded τ = − 0.455 (Q = 0.066), τ = 0.152 (Q = 0.537), τ = − 0.394 (Q = 0.100), τ = − 0.485 (Q = 0.060), τ = − 0.788 (Q < 0.001*), τ = 0.667 (Q = 0.014*), and τ = 0.636 (Q = 0.012*), respectively. The Benjamini–Hochberg procedure was applied to control the type 1 error. The significant values are marked with an asterisk (*)

Among patients with ACG initially treated with medication (Fig. 2b), 43.4–53.4% received a single drug, showing a decreasing trend (52.6% in 2013, 49.1% in 2018, and 44.7% in 2024; τ = − 0.636, Q = 0.012). By contrast, 46.6–56.6% received combination therapy, with an increasing trend (47.4%, 50.9% and 55.3% across the same years; τ = 0.636, Q = 0.012). As regards the individual drug classes, 17.7–28.3% received prostaglandin analogues with a non-significant trend (τ = − 0.455, Q = 0.066), 6.0–11.4% received beta-blockers with a non-significant trend (τ = 0.152, Q = 0.537), 6.9–10.4% received carbonic anhydrase inhibitors with a non-significant trend (τ = − 0.394, Q = 0.100), 3.7–8.0% received alpha-agonists with a non-significant trend (τ = − 0.485, Q = 0.060), and 1.0–3.7% of patients received parasympathomimetics, with a marked decline (2.9% in 2013, 3.6% in 2018, and 1.0% in 2024; τ = − 0.788, Q < 0.001). Use of Rho kinase inhibitors increased from 0% in 2013 to 0.3% in 2024 (τ = 0.677, Q = 0.014), which was significant.

Discussion

This study examined trends in first-line treatment for OAG and ACG using data from the TriNetX multi-institutional global research network. Our results, primarily observed in a North American cohort, found that medication remained the predominant initial therapy for both subtypes. Surgical interventions for OAG increased significantly, whereas for ACG both surgery and laser use declined. Prostaglandin analogues were the most commonly prescribed first-line drug class, with a significant rise in the use of Rho kinase inhibitors observed for both OAG and ACG.

Our study found that prostaglandin analogues remained the dominant first-line drug class for both OAG and ACG from 2013 to 2024, with no significant change over time. This finding aligns with previous studies and current clinical guidelines [7–9, 15, 27], which emphasize the effectiveness and safety of prostaglandin analogues as first-line therapy, owing to their superior IOP reduction, minimal systemic adverse effects, and convenient once-daily dosing. Our results also showed that beta-blockers remain widely used in clinical practice, likely reflecting their cost-effectiveness and complementary mechanism of action. Regional variation was evident: while prostaglandin analogues represent the global first-line choice, beta-blockers continue to be commonly prescribed in Asia, largely as a result of lower cost and greater accessibility [12]. The continued reliance on these drug classes underscores the need for policies that ensure their accessibility, particularly through availability of generic formulations and high reimbursement rates. At the same time, the growing use of newer agents such as Rho kinase inhibitors as first-line therapy for both OAG and ACG reflects advances in pharmacotherapy and a shift away from older treatments, including parasympathomimetics for ACG and beta-blockers for OAG [28, 29]. However, it is important to note that despite the increasing trend in Rho kinase inhibitor use as first-line therapy, the overall proportion thereof remains relatively low (1.0% for OAG and 0.3% for ACG by 2024). This low overall utilization suggests that while they are gaining traction, they have not yet become a widespread first-line choice, potentially because of factors such as newer market entry, cost, or limited formulary availability in many healthcare systems. It is worth noting that Rho kinase inhibitors were first approved in Japan in 2014, in the USA in 2017, and from 2020 onwards in Europe. Future research should examine long-term adherence and evaluate how newer agents can be integrated into evolving treatment paradigms for glaucoma. Notably, combination therapy was more frequently required for ACG, whereas monotherapy predominated for OAG, underscoring the greater need for aggressive IOP reduction in patients with ACG [30]. Our findings also showed an increasing trend in combination therapy for ACG, consistent with previous research [29]. These results support incorporating glaucoma subtype and treatment response into clinical guidelines to optimize management.

Our study revealed shifts in first-line treatment strategies and antiglaucoma medication use for both OAG and ACG from 2013 to 2024. In the OAG cohort, surgical interventions increased by 4.2%, while the use of medications remained stable. Although the absolute changes in first-line surgical use for OAG, from 4.8 to 9.0%, appear modest, Kendall’s τ reflects the consistency of the year-to-year trend rather than the magnitude of the overall difference. A moderate τ therefore indicates a stable upward trajectory across the study period. These subtle yet statistically significant shifts may suggest a gradual evolution in treatment paradigms. This observation aligns with recent trends favoring earlier surgical intervention to enhance long-term IOP control [31]. According to the European Glaucoma Society, surgery is recommended when combination therapy with two drug classes proves insufficient [7]. By contrast, clinical practice in Asia typically reserves surgery for cases requiring a fourth-line agent [9]. The emergence of minimally invasive glaucoma surgery techniques may have further driven this shift, owing to their favorable short-term safety compared to traditional procedures [32], making them an appropriate option for patients with mild to moderate glaucoma [33]. Minimally invasive glaucoma surgery also improves quality of life by reducing the reliance on medications [34], and some studies have reported a substantial rise in minimally invasive glaucoma surgery alongside a slight decline in conventional surgeries [35]. Future research should investigate the long-term outcomes and cost-effectiveness of emerging treatment modalities to further inform and refine clinical guidelines for glaucoma management.

In contrast to the OAG cohort, patients with ACG showed increased use of medication and a decline in both laser and surgical procedures, indicating a shift toward pharmacological management before procedural intervention. This observed decline in the ACG group is a significant finding that may be influenced by several factors not explicitly captured in our data. Although LPI remains the first-line treatment for primary angle-closure suspect (PACS) and PAC according to international guidelines [7–9], this reduction may partly reflect undercapture of procedures performed in separate laser suites. Also, the role of phacoemulsification in angle-closure should not be overlooked. While not classified as a glaucoma procedure in our dataset, following the EAGLE trial in 2016 [36], which showed clear-lens extraction provides superior long-term IOP control and quality-of-life outcomes compared with LPI, some clinicians favor early lens extraction, possibly explaining the decline in traditional glaucoma surgeries and laser therapy.

As regards laser treatment, the proportion remained largely unchanged during the study period. The LiGHT trial, a landmark study, showed that SLT could delay or reduce the need for medication in newly diagnosed primary OAG, with comparable IOP control at 3 years post-treatment [13, 14]. However, only a modest and non-significant increase in first-line laser use was observed in our data after 2018. This suggests that, despite the growing evidence base, SLT adoption as initial therapy did not change clinical practice patterns substantially within the study period, partly because drugs allow flexible titration and can be discontinued more easily. National guidelines also vary in their recommendations. For example, the American Academy of Ophthalmology advises SLT as a first-line or adjunctive option in patients with primary OAG, particularly when medication is poorly tolerated or adherence is suboptimal [8]. By contrast, the Japan Glaucoma Society recommends medication as the initial treatment, with laser therapy reserved as an adjunct or alternative option [9]. The European Glaucoma Society aligns closely with the American Academy of Ophthalmology in recommending SLT as a routine first-line option for patients with primary OAG or ocular hypertension [7]. Interestingly, previous reviews have noted that the advent of SLT, similar to the introduction of prostaglandin analogues, marked a shift in treatment patterns for OAG, with SLT contributing to a reduction in prescriptions for ocular hypotensive drugs [37]. Discrepancies between trial efficacy and real-world effectiveness highlight the need to integrate both sources of evidence to inform clinical decision-making.

Our results—derived primarily from a North American cohort—were compared with real-world data from Asian populations. In Japan and Korea, across all types of glaucoma, monotherapy with prostaglandin analogues remains the predominant initial treatment, accompanied by a gradual increase in fixed-combination therapy over time [27, 38]. Similarly, in our study prostaglandin analogues were also the most frequently used first-line medication. Fixed-combination use increased non-significantly for OAG and significantly for ACG. For ACG, Korean population-based data show that between 2007 and 2012, LPI rates decreased while cataract surgery increased [39]. Recent Japanese data also demonstrate a shift in treatment for primary ACG toward greater procedural intervention—particularly cataract surgery—with decreasing reliance on medications alone [29]. By comparison, our cohort showed a decline in LPI use, consistent with the Japanese findings. However, because cataract surgery was not included as a glaucoma procedure in our procedural category, we observed an overall decline in the proportion of surgical treatments for ACG. Correspondingly, our dataset showed an increase in medication-based therapy for ACG, in contrast to the decreasing medication reliance reported in the Japanese cohort [29]. These differences highlight the influence of database structure, coding practices, and regional variations in ACG management.

This study had several limitations related to the structure of the TriNetX database. Additionally, although it aggregated data from multiple international healthcare systems, most records originated from the USA, limiting geographic generalizability. The identification of OAG and ACG cases relied on ICD-10 codes without gonioscopic confirmation, which is not captured in the TriNetX platform, introducing potential misclassification, although prior validation studies suggest acceptable diagnostic reliability [40–42], and these limitations are unlikely to systematically bias the temporal trends observed in this study. Because the platform provides only aggregated, non-patient-level data, clinical variables such as IOP, disease severity, adherence, disease progression, or treatment course over time cannot be assessed. The database also does not distinguish MIGS from traditional incisional surgery, limiting interpretation of surgical trends. Defining treatment-naïve status using a 12-month lookback period may have misclassified previously treated patients, although this definition was consistent with prior work in glaucoma treatment trends. As such, our findings reflect prescribing initiation patterns within the observable data window, rather than strictly incident disease in the epidemiologic sense. Our analysis did not include cataract surgery data, and therefore may have underestimated clear-lens extraction for ACG. Also, the observed reduction in LPI may reflect undercapture or misclassification if laser treatments were performed in a separate laser clinic. Future studies incorporating multiple data sources and incorporating cataract data would help clarify whether these patterns represent a true shift. Finally, this study aimed to describe the trends in first-line treatment for glaucoma, and differences in patient characteristics may have informed treatment selection; therefore, our unadjusted trends should be interpreted with this consideration in mind.

Conclusion

Medications remained the mainstay of initial treatment for both OAG and ACG from 2013 to 2024. However, treatment patterns were shifting, and while prostaglandin analogues remained the mainstay monotherapy for both OAG and ACG, we observed increasing use of combination therapy for ACG and greater reliance on surgical intervention for OAG. These changes likely reflect advances in pharmacotherapy and clinical practice guidelines. Further studies are needed to evaluate long-term outcomes and guide subtype-specific glaucoma care.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Lan-Hsin Chuang, Yuan-Hsi Chan and Shih-Chieh Shao designed the study, Yuan-Hsi Chan and Shih-Chieh Shao performed the study, Daniel Hsiang-Te Tsai analysed the data, Yuan-Hsi Chan and Shih-Chieh Shao, Edward Chia-Cheng Lai, and Chi-Chun Lai drafted the manuscript, and all authors reviewed the manuscript. All authors read and approved the final version of the manuscript.

Funding

This study was supported in part by grants from the Keelung Chang Gung Memorial Hospital (CORPG2N0051) to Chi-Chun Lai, National Science and Technology Council of Taiwan (NSTC 114-2628-B-006-005-) and the National Health Research Institutes of Taiwan (NHRI-14A1-CG-CO-04-2225-1) to Edward Chia-Cheng Lai. Additionally, this research was supported in part by Higher Education Sprout Project, Ministry of Education to the Headquarters of University Advancement at National Cheng Kung University (NCKU). The journal’s Rapid Service fee was funded by the authors.

Data Availability

The data supporting the findings of this study are available from the TriNetX Analytics Network (https://trinetx.com).

Declarations

Conflict of Interest

Lan-Hsin Chuang, Daniel Hsiang-Te Tsai, Yuan-Hsi Chan, Shih-Chieh Shao, Chi-Chun Lai, and Edward Chia-Cheng Lai confirm that they have no conflicts of interest to declare.

Ethical Approval

This study was performed in accordance with the Helsinki Declaration of 1964 and its later amendments, with the exception that this trial was retrospectively registered. It was approved by the Institutional Review Board of Chang Gung Medical Foundation (IRB number 202401821B1).