Predictors of aspiration pneumonia in oral cavity cancer patients undergoing radiation therapy: A focus on skeletal muscle index

Abstract

Purpose

Aspiration pneumonia is a life-threatening adverse event of radiotherapy in patients with oral cavity cancer. This study aimed to evaluate the association between muscle mass changes and aspiration pneumonia after radiotherapy.

Materials and methods

This study comprised 1,122 patients with oral cavity cancer who underwent surgery and post-operative radiotherapy between 2010 and 2021 at two tertiary centers. Changes in skeletal muscle index (SMI) were measured using pre- and mid-radiotherapy computed tomography scans at the C3 vertebral level. The primary and secondary outcomes were aspiration pneumonia and overall survival, respectively.

Results

With a median follow-up of 5.6 (interquartile range: 2.9–9.2) years, 125 (11.1%) patients developed aspiration pneumonia. Patients with aspiration pneumonia tended to have older age, poorer performance status, smoking, and higher mean dose to swallowing organs. Patients with aspiration pneumonia lost more SMI during radiotherapy than those without it (-3.4% vs. -1.0%; p < 0.001). After adjusting for clinical and dosimetric factors, an increase in SMI change during radiotherapy was independently associated with a lower risk of aspiration pneumonia in the overall population (hazard ratio: 0.78 per 1% increase, p < 0.001) or subgroups stratified by pre-radiotherapy SMI tertile. The occurrence of aspiration pneumonia was independently associated with poorer overall survival (hazard ratio: 2.81, p < 0.001). A 1% increase in SMI change during radiotherapy was independently associated with better overall survival (hazard ratio: 0.78, p < 0.001).

Conclusions

Increased muscle mass during radiotherapy was associated with a lower risk of aspiration pneumonia, which in turn was associated with overall survival.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00520-026-10394-7.

Introduction

Oral cavity cancer (OCC) is the most common head and neck malignancy in terms of incidence and mortality, with 389,485 new cases and 188,230 deaths worldwide in 2022 [1]. Surgery is the primary treatment, while post-operative radiotherapy improves locoregional control and survival in the presence of high-risk pathological features [2–5]. However, radiotherapy is associated with aspiration pneumonia, a life-threatening complication and a leading cause of noncancer-related mortality in patients with OCC [6–12]. Identifying modifiable factors to reduce this risk is essential.

Skeletal muscle mass is a recognized prognostic marker in OCC [13–20]. Loss of muscle mass during radiotherapy may result from toxicity, nutritional decline, or inactivity and is associated with impaired swallowing (sarcopenic dysphagia) [21]. This may increase the risk of aspiration pneumonia. Although lower pre-radiotherapy muscle mass has been associated with higher pneumonia risk in head and neck cancer, whether changes in muscle mass during radiotherapy affect this risk remains unclear [22–24].

Skeletal muscle mass can be assessed using computed tomography (CT) scans at the third cervical vertebra (C3), a validated proxy for whole-body muscle mass [13–15]. In adaptive radiotherapy, interim CT scans can be used to monitor changes in muscle mass [25–27]. Therefore, this study aimed to evaluate changes in muscle mass during post-operative radiotherapy and determine their association with aspiration pneumonia in patients with OCC (Fig. 1).

Fig. 1.

Investigating the association between muscle mass change during radiotherapy and aspiration pneumonia in patients with oral cavity cancer. Skeletal muscle was assessed on a transversal computed tomography slice at the level of C3. The patient had left tongue cancer and underwent surgery and left modified radical neck dissection. Red: right sternocleidomastoid muscle and paravertebral muscle

Methods

Study population

This study was approved by the Institutional Review Boards of the participating institutions, and the requirement for informed consent was waived due to its retrospective design. This study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement. We reviewed data from 1,356 consecutive patients newly diagnosed with OCC, who underwent surgery and post-operative radiotherapy at MacKay Memorial Hospital or Changhua Christian Hospital between January 2010 and September 2021. Patients were excluded if they had (i) prior malignancy, (ii) insufficient clinical data, (iii) no CT scans for adaptive radiotherapy during treatment, (iv) pneumonia prior to radiotherapy, or (v) insufficient follow-up. Based on these criteria, a final cohort of 1,122 patients was included in the analysis (Supplementary Fig. 1).

All patients underwent radical excision and neck dissection, followed by post-operative intensity-modulated radiotherapy (60–66 Gy in 30–33 fractions to the tumor bed and dissected nodal region; 54–56 Gy in 30–33 fractions to the elective region) [17–19]. CT scans were acquired after surgical wound healing for radiotherapy planning (pre-RT CT) and again at the fifth week of radiotherapy for adaptive planning (mid-RT CT). The decision to add concurrent platinum-based chemotherapy was based on major pathological risk factors.

Skeletal muscle measurement

Pre- and mid-RT CT scans were retrieved for analysis. A single axial CT slice at C3, displaying the transverse processes and entire vertebral arch, was selected for skeletal muscle measurement [14]. One radiologist (Y.T.J., 18 years of experience), blinded to patient information, assessed skeletal muscle area using 3D Slicer software (version 4.11) in a semi-automatic manner. Muscle tissue was defined as − 29 to 150 Hounsfield units and included paravertebral and sternocleidomastoid (SCM) muscles. Patients with OCC often undergo ipsilateral neck dissection, involving mobilization or resection of the SCM, whereas the contralateral SCM typically remains intact. To account for potential distortion or absence of the ipsilateral SCM, the cross-sectional area (CSA, cm²) was calculated as the sum of the paravertebral muscles and twice the area of the contralateral SCM, following a validated algorithm (Eq. 1) [14]. The total CSA at C3 was then converted to the estimated CSA at the third lumbar vertebra (L3) using Eq. 2. The conversion was unadjusted for age, sex, or other covariates to minimize potential bias in the longitudinal analysis [19]. The skeletal muscle index (SMI, cm²/m²) and the percentage change in SMI were calculated using Eqs. 3 and 4. Pre-RT SMI was stratified into sex-specific tertiles for subsequent analysis.

$$CSAatC3\left({\mathit{cm}}^2\right)=paravertebralmusclearea+oneSCMmusclearea\times 2$$ 1

$$CSAatL3\left({\mathit{cm}}^2\right)=24.078+2.789\times CSAatC3\left({\mathit{cm}}^2\right)$$ 2

$$SMI\left({\mathit{cm}}^2,/,m^2\right)=\frac{CSAatL3\left({\mathit{cm}}^2\right)}{{\mathit{Height}}^2\left(m^2\right)}$$ 3

$$SMIchange\left(\%\right)=\frac{{\mathit{SMI}}_{mid-RT}-{\mathit{SMI}}_{pre-RT}}{{\mathit{SMI}}_{pre-RT}}\times 100$$ 4

Outcome measures

The primary outcome was the occurrence of aspiration pneumonia, defined as the time from the start of radiotherapy to the date of diagnosis of aspiration pneumonia or last follow-up. Patients were considered to have aspiration pneumonia if they met the following criteria: (i) clinical symptoms accompanied by characteristic radiographic findings, and (ii) aspiration identified on videofluoroscopy or fiberoptic endoscopic evaluation of swallowing, or clinically suspected based on symptoms such as choking or delayed swallowing. The secondary outcome was overall survival (OS) and 30-day mortality after the development of aspiration pneumonia. Overall survival was defined as the time from cancer diagnosis to death from any cause or last follow-up. Thirty-day mortality was defined as death from any cause within 30 days after aspiration pneumonia, whether in-hospital or after discharge (determined via hospital and follow-up records). Data on the occurrence and date of aspiration pneumonia, as well as the date of last follow-up, were extracted from the institutional database.

Covariates

Clinical data collected included age, sex, Eastern Cooperative Oncology Group (ECOG) performance status, smoking status, primary tumor site, American Joint Committee on Cancer pathological stage, chemotherapy, and type of neck dissection. The mean radiation doses to the superior, middle, and inferior pharyngeal constrictor muscles, cricopharyngeus muscle, supraglottic larynx, glottic larynx, esophagus inlet muscle, and cervical esophagus were derived from the original radiotherapy plans based on previous literature [7–10, 28–30]. Swallowing structures were delineated according to published guidelines [31].

Statistical analyses

Continuous data are presented as means with standard deviations (SDs) or as medians with interquartile ranges (IQRs), while categorical data are presented as frequencies and percentages. Comparisons between groups were assessed using the independent t-test or Mann–Whitney U test for continuous variables, and the chi-squared test for categorical variables, as statistically appropriate. Paired t-test was used to evaluate changes in SMI.

The Kaplan–Meier curves were generated, and the log-rank test was conducted for comparisons between groups. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Multivariable models were selected using backward elimination with a 0.10 significance level of inclusion. The data were analyzed using IBM SPSS software (version 21.0; IBM Corp., Armonk, NY, USA). A p < 0.05 was considered statistically significant.

Results

A total of 1,122 patients were included (146 [13.0%] women and 976 (87.0%) men; median age: 55 years). Most had ECOG 0 (68.3%), smoking at diagnosis (65.7%), and pathological stage III–IVB disease (83.0%) (Table 1). The mean pre-RT and mid-RT SMI were 51.6 (8.4) and 50.9 (8.5) cm²/m², respectively, reflecting a 1.3% decrease of SMI during radiotherapy (95% CI, –1.4 to –1.1; p < 0.001). Pre-RT SMI tertile thresholds were 47.8 and 54.7 cm²/m² for men and 43.6 and 49.6 cm²/m² for women.

Table 1.

Patient characteristics

Total number of patients	n = 1122
Age (years), median (IQR)	55 (48–62)
Sex
Female	146 (13.0)
Male	976 (87.0)
ECOG
0	766 (68.3)
1	356 (31.7)
Pre-RT SMI (cm2/m2)	51.6 (8.4)
Mid-RT SMI (cm2/m2)	50.9 (8.5)
SMI change (%)	−1.3 (2.4)
Smoking status
Never/former	385 (34.3)
Yes	737 (65.7)
Tumor location
Tongue	400 (35.7)
Buccal	432 (38.5)
Other subsites	290 (25.8)
Pathological stage
I-II	191 (17.0)
III-IVB	931 (83.0)
Neck dissection type
SOND	485 (43.2)
MRND	637 (56.8)
Adjuvant treatment
RT alone	423 (37.7)
Chemoradiotherapy	699 (62.3)
Mean dose (Gy) to swallowing organs
Superior PCM	51.7 (5.1)
Middle PCM	48.4 (4.4)
Inferior PCM	43.4 (4.6)
Cricopharyngeus muscle	39.9 (5.9)
Esophagus inlet muscles	36.8 (5.9)
Cervical esophagus	35.2 (6.4)
Supraglottic larynx	45.4 (5.2)
Glottic larynx	41.8 (5.6)

Abbreviations: ECOG, Eastern Cooperative Oncology Group; IQR, interquartile range; MRND, modified radical neck dissection; PCM, pharyngeal constrictor muscle; RT, radiotherapy; SMI, skeletal muscle index; SOND, supraomohyoid neck dissection

Data are number (%) or mean (standard deviation)

All the patients completed adjuvant radiotherapy. With a median follow-up of 5.6 (IQR: 2.9–9.2) years, 125 (11.1%) patients developed aspiration pneumonia, in a median onset time of 10.7 (IQR: 5.9–16.8) months post-radiotherapy. Cumulative aspiration pneumonia incidence was 5.4% and 12.0% at 1 and 5 years, respectively (Fig. 2A). Compared to those without aspiration pneumonia, patients with aspiration pneumonia displayed lower pre-RT SMI (47.3 vs. 52.1 cm²/m²; p < 0.001) and mid-RT SMI (45.7 vs. 51.6 cm²/m²; p < 0.001) and greater SMI loss (–3.4% vs. –1.0%; p < 0.001). In the lowest, middle, and highest pre-RT SMI groups, 5-year incidences were 20.3%, 9.4%, and 6.8%, respectively (p < 0.001; Fig. 2B).

Fig. 2.

The cumulative incidence of aspiration pneumonia is illustrated in (A) whole population (n = 1,122), and (B) patients stratified by the tertile of pre-radiotherapy skeletal muscle index

The results of the Cox proportional hazard regression analysis for aspiration pneumonia are presented in Table 2. Univariable analysis demonstrated that older age, ECOG 1, smoking, pathological stage III-IVB, modified radical neck dissection, the lowest pre-RT SMI tertile, SMI change, and higher mean doses to swallowing organs, were associated with the risk of aspiration pneumonia. Multivariable analysis demonstrated that the lowest pre-RT SMI tertile was independently associated with increased risk of aspiration pneumonia (HR, 2.85; 95% CI, 1.76–4.60; p < 0.001), and increased SMI during radiotherapy was associated with a lower risk of aspiration pneumonia (HR, 0.78; 95% CI, 0.73–0.84; p < 0.001). At subgroup analysis according to the pre-RT SMI tertile, an increase in SMI during radiotherapy was also independently associated with a lower risk of aspiration pneumonia (the highest tertile: HR, 0.81; 95% CI, 0.69–0.95; p = 0.01; middle tertile: HR, 0.85; 95% CI, 0.76–0.96; p = 0.01; the lowest tertile: HR, 0.70; 95% CI, 0.62–0.78; p < 0.001) (Supplementary Table 1).

Table 2.

Cox proportional hazards model for aspiration pneumonia

	Univariable		Multivariablea
	HR (95% CI)	p	HR (95% CI)	p
Age (continuous)	1.04 (1.02–1.06)	< 0.001	1.04 (1.02–1.06)	< 0.001
Male (ref: Female)	1.02 (0.60–1.73)	0.93	-	-
ECOG 1 (ref: ECOG 0)	2.51 (1.77–3.56)	< 0.001	-	-
Smoker (ref: never/former)	2.23 (1.44–3.45)	< 0.001	2.02 (1.29–3.17)	0.002
Tumor location (ref: tongue)
Buccal	0.62 (0.40–0.96)	0.03	-	-
Other subsites	1.10 (0.73–1.67)	0.65	-	-
Pathological stage III-IVB (ref: stage I-II)	1.78 (1.02–3.10)	0.04	-	-
MRND (ref. SOND)	1.62 (1.12–2.35)	0.01	-	-
CRT (ref: RT alone)	1.28 (0.88–1.85)	0.20	-	-
Pre-RT SMI groups (ref: highest tertile)
Middle tertile	1.39 (0.82–2.35)	0.23	1.61 (0.94–2.75)	0.08
Lowest tertile	3.37 (2.12–5.37)	< 0.001	2.85 (1.76–4.60)	< 0.001
SMI change (per 1% increase)	0.65 (0.61–0.69)	< 0.001	0.78 (0.73–0.84)	< 0.001
Mean dose to swallowing organs
Superior PCM	1.28 (1.23–1.32)	< 0.001	1.14 (1.09–1.19)	< 0.001
Middle PCM	1.28 (1.24–1.32)	< 0.001	-	-
Inferior PCM	1.19 (1.15–1.24)	< 0.001	-	-
Cricopharyngeus muscle	1.14 (1.10–1.18)	< 0.001	-	-
Esophagus inlet muscles	1.10 (1.06–1.14)	< 0.001	1.04 (1.01–1.08)	0.04
Cervical esophagus	1.00 (0.98–1.03)	0.75	-	-
Supraglottic larynx	1.24 (1.21–1.27)	< 0.001	1.10 (1.06–1.14)	< 0.001
Glottic larynx	1.21 (1.17–1.26)	< 0.001	-	-

Abbreviations: CRT, chemoradiotherapy; CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; HR, hazard ratio; MRND, modified radical neck dissection; PCM, pharyngeal constrictor muscle; RT, radiotherapy; SMI, skeletal muscle index; SOND, supraomohyoid neck dissection

^a Multivariable models were selected using stepwise backward selection with a 0.10 significance level of inclusion

Subsequent evaluations focused on aspiration pneumonia-associated morbidity and mortality. Among the 125 patients who experienced aspiration pneumonia, 108 (86.4%) required hospitalizations, and the 30-day mortality rate following aspiration pneumonia development was 25.6%. Aspiration pneumonia was the cause of death in 61 (48.8%) patients, and those that experienced aspiration pneumonia had significantly reduced OS compared with those who did not (Fig. 3). The 5-year OS rates after cancer diagnosis among those with and without aspiration pneumonia were 24.4% versus 73.0% (p < 0.001).

Fig. 3.

Kaplan–Meier curves illustrate the overall survival from the date of diagnosis of oral cavity cancer among patients with (n = 125) or without aspiration pneumonia (n = 997). OS, overall survival

Multivariable analysis controlling for potential confounders revealed that the occurrence aspiration pneumonia was independently associated with a poorer OS (HR, 2.81; 95% CI, 2.20–3.58; p < 0.001) compared with the absence of aspiration pneumonia (Table 3). In addition, increased SMI during radiotherapy was independently associated with a better OS (HR, 0.78; 95% CI, 0.77–0.80; p < 0.001).

Table 3.

Cox proportional hazards model for overall survival

	Univariable		Multivariablea
	HR (95% CI)	p	HR (95% CI)	p
Age (continuous)	1.02 (1.01–1.03)	< 0.001	1.02 (1.01–1.03)	0.002
Male (ref: Female)	1.17 (0.86–1.60)	0.31	-	-
ECOG 1 (ref: ECOG 0)	1.90 (1.56–2.31)	< 0.001	-	-
Smoker (ref: never/former)	1.71 (1.37–2.15)	< 0.001	1.53 (1.21–1.92)	< 0.001
Tumor location (ref: tongue)
Buccal	0.82 (0.65–1.03)	0.09	-	-
Other subsites	0.93 (0.73–1.19)	0.57	-	-
Stage III-IVB (ref: stage I-II)	1.89 (1.39–2.57)	< 0.001	-	-
MRND (ref. SOND)	1.53 (1.25–1.88)	< 0.001	1.32 (1.06–1.64)	0.01
CRT (ref: RT alone)	1.72 (1.38–2.13)	< 0.001	1.57 (1.24–1.99)	< 0.001
Pre-RT SMI groups (ref: highest tertile)		< 0.001		< 0.001
Middle tertile	1.97 (1.50–2.60)		2.42 (1.85–3.17)
Lowest tertile	2.71 (2.09–3.52)		2.92 (2.23–3.81)
SMI change (per 1% increase)	0.67 (0.65–0.70)	< 0.001	0.78 (0.77–0.80)	< 0.001
Aspiration pneumonia (ref: none)	4.38 (3.48–5.52)	< 0.001	2.81 (2.20–3.58)	< 0.001

Abbreviations: CRT, chemoradiotherapy; CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; HR, hazard ratio; MRND, modified radical neck dissection; RT, radiotherapy; SMI, skeletal muscle index; SOND, supraomohyoid neck dissection

^a Multivariate analysis was adjusted for age, ECOG, smoking status, stage, type of neck dissection, use of chemotherapy, pre-RT SMI, and SMI change. Multivariable models were selected using stepwise backward selection with a 0.10 significance level of inclusion

Discussion

To the best of our knowledge, this is the first study to explore the association between muscle mass change during post-operative radiotherapy and aspiration pneumonia in patients with OCC. We found that increased SMI during radiotherapy was associated with a lower risk of aspiration pneumonia, independent of clinical factors and mean dose to the swallowing organs, in the overall population or subgroups stratified by pre-RT SMI tertile. Moreover, the occurrence of aspiration pneumonia was independently associated with poorer OS.

Previously, a study involving individuals with head and neck cancers revealed a 5-year cumulative post-radiotherapy aspiration pneumonia incidence of 19.9% in patients with OCC [6]. Compared with the result of that study, we observed a lower rate of aspiration pneumonia, possibly because all our patients underwent intensity-modulated radiotherapy. Moreover, as we also observed, aspiration pneumonia was the major cause of mortality, and its occurrence was independently associated with poorer OS. Therefore, it is critical to identify potentially modifiable risk factors of aspiration pneumonia.

Age, male gender, smoking, tumor location, pre-RT SMI, and radiation dose to the swallowing organs, are risk factors of aspiration pneumonia following radiotherapy for head and neck cancer [6–12]. However, none of the previous studies have incorporated these factors together into a single analysis, to evaluate their association with aspiration pneumonia. Our study builds on these studies by further demonstrating that age, smoking, pre-RT SMI, and mean dose to the superior pharyngeal constrictor muscle, esophagus inlet muscles, and supraglottic larynx were independent risk factors of aspiration pneumonia. Moreover, we found that SMI change during radiotherapy was independently associated with aspiration pneumonia either in the overall population or subgroup analyses. A possible explanation is that increasing muscle mass might be associated with favorable overall health and less impaired swallowing function, which may lower the risk of aspiration pneumonia [21]. However, the swallowing function was not routinely evaluated in this population; further studies are required to evaluate the association between muscle mass and swallowing function. In addition, this retrospective study was unable to establish a definitive causal relationship between SMI and aspiration pneumonia. A prospective study is needed to validate the predictive value of SMI for aspiration pneumonia by stratifying patients by the pre-RT SMI or SMI change. Such a study may potentially confirm whether patients with lower pre-RT SMI or greater SMI decline during radiotherapy are actually at a higher risk aspiration pneumonia. Ultimately, a threshold for SMI that could be used for clinically actionable risk prediction may also be established in the future study.

By comparing the pre- and mid-RT CT scans, we showed that patients demonstrated SMI loss during radiotherapy. Schaeffers et al. also revealed that patients demonstrated continuously decreasing muscle mass from the start to the fifth week of radiotherapy [32]. It may be necessary to implement prehabilitation intervention at the start of radiotherapy, to preserve muscle mass [33]. Supervised resistance, aerobic, and impact exercises may enhance muscle mass during radiotherapy [34]. Individualized nutritional interventions could improve the nutritional status, body weight, and quality of life during radiotherapy. However, the effect of this approach on muscle mass requires further evaluation [35]. On the other hand, pretreatment swallowing education and exercise programs may help maintain or improve swallowing function, thereby supporting nutritional status, preserving muscle mass, and reducing the risk of aspiration [36]. Therefore, future studies are also required to assess whether multidisciplinary prehabilitation interventions, including exercise, nutrition, and swallowing therapy would benefit these patients in preserving muscle mass during radiotherapy.

This study had some limitations. Its retrospective design may limit the inclusion of several covariates potentially predictive of aspiration pneumonia, including pretreatment swallowing function, dysphagia, and smoking intensity. This information was available only for a few patients and could not be analyzed. Aspiration pneumonia might have been under-recognized by the clinicians and under-reported by patients in this study. Therefore, this study was unable to establish a causal relationship between SMI and aspiration pneumonia, but only identified an association. Moreover, this study only included patients with OCC because the treatment, including surgery and radiotherapy, for OCC may differ from those for pharyngeal and laryngeal cancers [7]. A study focusing on patients with OCC is needed to identify the risk factors of aspiration pneumonia among them [8]. While aspiration pneumonia is also a life-threatening complication for patients with pharyngeal and laryngeal cancers receiving radiotherapy [6–12], future studies are also required to investigate the association of muscle mass with aspiration pneumonia in this population. Despite these limitations, this study had a large sample size and included clinical and dosimetric factors for evaluating the association between SMI and aspiration pneumonia in patients with OCC. The study had an adequate follow-up period with outcomes comparable to those of previous studies [2–4].

Conclusion

Increased SMI during post-operative radiotherapy was independently associated with a lower risk of aspiration pneumonia in patients with OCC, regardless of the baseline SMI. Aspiration pneumonia was associated with poorer OS, while increased SMI during radiotherapy was associated with better OS. Future prospective studies are required to confirm the causal relationship between SMI and aspiration pneumonia, and to assess whether prehabilitation strategies can enhance muscle mass during radiotherapy.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contribution

Study concept and design: Jie Lee Collection and assembly of data: Jhen-Bin Lin, Yu-Jen Chen, and Yi-Shing Leu Data analysis and interpretation: Jhen-Bin Lin, Yu-Jen Chen, Yi-Shing Leu, Ya-Ting Jan, and Jie Lee Manuscript writing: Jhen-Bin Lin and Jie Lee Critical revision of the manuscript for important intellectual content: All authors. Final approval of manuscript: All authors.

Funding

This work was supported by the National Science and Technology Council of Taiwan (grant number Contract No. MOST 110–2314-B-195–033 and NSTC 113–2314-B-195–011-MY3), and MacKay Memorial Hospital (grant number: MMH-114–22).

Data availability

The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval

This study was conducted in accordance with the principles of the Declaration of Helsinki. This study was approved by the Institutional Review Boards of MacKay Memorial Hospital (23MMHIS437e) and Changhua Christian Hospital (IRB No.180503).

Conflict of interest

The authors declare no competing interests.

Declaration of generative ai and ai-assisted technologies in the writing process

The authors declare no Generative AI and AI-assisted technologies in the writing process.