Targeted LVA guided by ultra-high frequency ultrasound and ICG lymphography: A pilot study

Abstract

Introduction

Breast cancer-related lymphedema (BCRL) is a chronic sequela that significantly impairs quality of life. Lymphovenous anastomosis (LVA) is a promising microsurgical treatment, but its success relies on accurate identification of functional lymphatic vessels and adjacent venules. While indocyanine green lymphography (ICG-L) provides functional lymphatic mapping, venous visualization is often limited. Ultra-high frequency ultrasound (UHFUS), offering high-resolution imaging, may complement ICG-L in preoperative planning. This feasibility study investigated the combined use of ICG-L and UHFUS for vessel identification prior to LVA.

Method

Ten female patients with BCRL were included. Preoperative mapping combined ICG-L for lymphatics and UHFUS (45–70 MHz) for venules before LVA surgery. The primary outcome was agreement between preoperatively identified and intraoperatively confirmed lymphatic and venous vessels, as well as between planned and performed anastomoses. Secondary outcomes included changes in arm volume, quality of life, and L-Dex scores over 12 months.

Results

Preoperative mapping identified a median of two lymphatic and two venous vessels per patient, leading to a median of two planned anastomoses. Intraoperative findings confirmed most vessels, with a median of two anastomoses performed per patient (p = 0.65). Arm volume did not decrease significantly over the 12 months. Quality-of-life and L-Dex scores showed a trend towards improvement.

Conclusion

Combining UHFUS with ICG-L for preoperative planning of LVA is feasible and enables accurate identification of suitable lymphatic and venous vessels. Although no significant reduction in arm volume was observed, improvements in quality of life and L-Dex scores suggest a possible clinical benefit.

Introduction

Breast cancer-related lymphedema (BCRL) is a chronic and often debilitating sequela of breast cancer treatment, particularly among patients who undergo axillary lymph node dissection and radiotherapy.¹ Damage to the lymphatic system disrupts normal drainage, leading to accumulation of protein-rich fluid in the interstitial tissues of the arm and hand.¹ This causes swelling, heaviness, pain, altered sensation, reduced mobility, and recurrent infections, affecting quality of life and body image.1, 2, 3

Lymphovenous anastomosis (LVA) was first introduced in the 1960s but did not initially achieve widespread recognition due to inconsistent clinical success.4, 5, 6 However, LVA has re-emerged as a promising surgical procedure for BCRL, driven by advancements in microsurgical technology, imaging techniques, and our evolving understanding of lymphedema.⁷ These developments offer the potential to make LVA more reliable treatment.8, 9, 10 However, a key challenge remains: accurate preoperative planning.

Successful LVA requires the identification of both functional lymphatic vessels and adjacent small-caliber veins suitable for anastomosis.¹¹ While indocyanine green lymphography (ICG-L) effectively maps superficial lymphatic channels, locating appropriate veins is often difficult and typically performed intraoperatively.¹² This can prolong surgery and compromise outcomes. Inadequate preoperative venous mapping is a significant barrier to optimizing LVA success.¹¹ To address this challenge, ultra-high frequency ultrasound (UHFUS) is a promising imaging modality.12, 13, 14, 15 Unlike conventional ultrasound, UHFUS operates at frequencies of up to 70 MHz, offering superior resolution of superficial tissues and microvasculature. It allows detailed, real-time visualization of small veins, including their diameter and depth, and may also visualize lymphatic structures and flow when combined with Color Doppler imaging.12, 13, 14, 15 Its non-invasive nature and portability further support its clinical utility in surgical planning.12, 13, 14, 15 By providing high-resolution anatomical detail, UHFUS has the potential to complement ICG-L and improve preoperative identification of suitable veins for LVA surgery.¹¹ This could lead to more targeted, efficient procedures, better anastomosis outcomes, and improved clinical results for patients undergoing LVA for BCRL.

The aim of this feasibility study was to determine whether the combined use of ICG-L and UHFUS allows for the correct identification of lymphatic vessels and adjacent venules in preoperative planning for LVA.

Method

Study design and setting

This was a prospective pilot study at the Department of Plastic Surgery at Odense University Hospital from 2022 to 2025. All participants provided written informed consent before enrollment. This study is written according to the STROBE guidelines.¹⁶

Study population

The study population consisted of ten female patients with BCRL, who were able to understand Danish and provide informed consent. Patients were evaluated as candidates for LVA surgery in cases where ICG-L revealed lymphatic vessels ending in areas of obstruction, also called dermal backflow, where pitting edema was evident. Patients who had untreated or uncontrolled cancer or axillary surgery in the contralateral axilla were excluded; see Table 1 for full inclusion criteria.

Table 1.

Inclusion- and exclusion criteria. ICG-L: Indocyanine green lymphography, UHFUS: Ultra-high frequency ultrasound.

Inclusion criteria	Exclusion criteria
•Age >18 •Female •Unilateral arm lymphedema secondary to breast cancer treatment •Active pitting lymphedema •Presence of dermal backflow in ICG-L •Identifiable lymphatic vessel(s) in the affected arm using ICG-L •Able to provide informed consent •Able to read, understand, and complete Danish questionnaires	•Smoker •Untreated or uncontrolled primary cancer •Previous surgery on lymph nodes in the contralateral axilla •No applicable lymphatic vessels identified, using ICG-L. •No applicable venous vessels identified using UHFUS.

Intervention

Experienced microsurgeons at the Department of Plastic Surgery at Odense University Hospital, Denmark, using conventional or robot-assisted techniques, performed all LVA procedures. Preoperatively, lymphatic mapping with ICG-L was conducted using indocyanine green injected intradermally at the first and third web spaces of the hand and the wrist.¹⁷ Lymphatic drainage patterns and vessel placement were visualized with a near-infrared imaging system (FLUOPTICS© (FLUOPTICS SAS, 44 rue des Berges, 38000 Grenoble, France)), to identify obstructed lymphatic vessels and their location on the arm. Following this, an additional assessment was performed using UHFUS (Vevo MD, FUJIFILM VisualSonics Inc., Amsterdam, The Netherlands) with a 48 or 70 MHz linear probe to locate appropriate venous vessels nearby the lymphatics mapped for surgery. This ultrasound allowed for precise visualization of the size and depth of these target vessels. Color Doppler imaging was used to evaluate the presence and characteristics of blood flow within the identified vessels. See Fig. 1 for illustration of the procedure. Preoperative planning was performed by the author and the surgeon, JBT.

Fig. 1.

Illustration of preoperative planning method for LVA surgery using ultra-high frequency ultrasound and ICG lymphography.

LVA surgery was performed under local or general anesthesia, depending on the number of anastomoses planned, and patient preference. Anastomoses were created using 11-0 nylon sutures in end-to-end, end-to-side, or sleeved configurations, depending on vessel alignment and caliber. Patency was assessed intraoperatively by observing immediate lymphatic flow into the venous system. After surgery, patients began compression therapy, rested until suture removal at day 14, then gradually resumed activities. Patients were discharged on the day of surgery and returned for follow-up at three-, six-, nine-, and twelve months for measurements and questionnaires.

Outcomes

Primary outcome

Our primary outcome was the number of lymphatic and venous vessels planned preoperatively and identified during LVA surgery, as well as the number of planned and performed anastomoses.

Secondary outcome

The secondary outcome of this study was changes in:

• 1)Arm volume measured with water displacement.
• 2)Arm volume measured with tape circumference.
• 3)Health-related quality of life, measured with LYMPH-Q.
• 4)L-Dex score measured with bioimpedance.

Arm volume

The changes in arm volume were defined as the difference between the affected and unaffected arm, measured in absolute volume (mL) and as relative volume (%):

$$Absolutearmvolumedifference\left(mL\right)=\left(Lymphedemaarmvolume\left(mL\right)-Healthyarmvolume\left(mL\right)\right)$$

$$Relativearmvolumedifference(\%)=\left(\frac{Lymphedemaarmvolume\left(mL\right)-Healthyarmvolume\left(mL\right)}{Healthyarmvolume\left(mL\right)}\right)*100$$

Arm volumes were measured using a water-displacement test: patients’ arms were submerged in a basin of water, and the amount of water displaced was measured. The amount of displaced water was measured in grams and converted to milliliters. To ensure comparability with existing literature, arm volumes were also determined through tape circumference measurements, a widely accepted method in lymphatic research.¹⁸ Both the affected arm and the unaffected arm were measured.¹⁹ The difference in volume between the affected and unaffected arm was calculated, and this volume was compared from baseline to three-, six-, nine-, and twelve months follow-up.

Health-related quality of life

To assess health-related quality of life, patients completed the LYMPH-Q Upper Extremity Module, a translated and validated questionnaire comprising six scales: symptoms, arm function, appearance, psychological well-being, lymphedema knowledge, and compression garment satisfaction. Each scale generates a raw score that is Rasch-transformed to a standardized score ranging from 0 to 100, with a higher score indicating better quality of life.²⁰

L-Dex score

To assess patients' L-Dex Score, the SOZO ® device (Impedimed, Brisbane, Australia) was used. The device measures each arm’s extracellular fluid to generate a lymphedema index (L-Dex), with scores above 10 considered indicative of lymphedema.^21,22 Measurements were taken at baseline and at three-, six-, nine-, and twelve months postoperatively.

Data collection

Patients were initially assessed in the outpatient clinic for evaluation of their lymphedema, eligibility for LVA surgery and inclusion in the study. Baseline assessments were conducted within 14 days prior to the scheduled LVA surgery. Follow-up assessments were performed at three-, six-, nine- and twelve months postoperatively to evaluate secondary outcomes. Data were collected and stored in a secure Research Electronic Data Capture (REDCap) system via OPEN.²³

Statistical analysis

Given the nature of this feasibility study, power calculations were not conducted. A sample size of 10 patients was chosen. This sample size was considered sufficient to assess participant compliance with the study protocol, to evaluate the feasibility of performing combined ICG-L and ultrasound assessments in routine clinical practice, and to ensure completion of inclusion within the planned study timeline.^24,25

Descriptive statistics

Baseline demographic and clinical characteristics were summarized using descriptive statistics. Categorical variables are presented as frequencies and percentages, while continuous variables are reported as means (SD) or medians (IQR) based on data distribution. Normality was assessed using Q-Q plots.

Evaluation of targeted planning

For each patient, the number of lymphatic and venous vessels identified during preoperative planning was recorded, along with the number of these vessels identified and utilized intraoperatively. Mapping accuracy was defined as the proportion of preoperatively planned vessels that were confirmed during surgery. The same was done for the number of planned and performed anastomoses. Vessel counts were summarized with descriptive statistics. For patients whose confirmed perioperative vessels were not used, we provide details on the excluded vessels and the reasons anastomosis was not performed. Normality was assessed using Q-Q plots. Normally distributed data are shown as mean ± SD; non-normal data are shown as median.

To evaluate the agreement between preoperative planning and intraoperative execution, the number of planned anastomoses was compared to the number actually performed using a paired t-test for normally distributed data or the Wilcoxon signed-rank test for non-normally distributed data. The same statistical tests were applied to assess the agreement between the number of planned and rediscovered lymphatic and venous vessels, respectively. Differences in baseline BMI and lymphedema duration between patients who had all planned anastomoses performed versus those who did not were evaluated using the Mann-Whitney U.

Mixed-effects linear regression

Secondary outcomes were analyzed using mixed-effects regression to account for repeated measurements. Time was modeled as a categorical variable (baseline, 3, 6, 9, and 12 months), with baseline as the reference. Model fit was assessed using likelihood ratio tests, and model assumptions were checked to ensure validity.

Software

All analyses were performed in STATA. Results are presented with 95% confidence intervals, standard errors, and p-values < 0.05 considered statistically significant.

Results

Participant characteristics

The pilot study included 10 female patients with breast cancer-related lymphedema. Participants had a mean age of 58.5 years (SD± 12.12), a median lymphedema duration of 16.8 months (IQR 10.48–35.59), and a mean BMI of 29.7 kg/m² (SD± 5.09). Only one patient reported a history of erysipelas in the affected limb. All patients had undergone sentinel node biopsy followed by axillary lymph node dissection and radiotherapy. Prior lymphedema treatments commonly involved compression sleeves (n = 10) and compression gloves (n = 9). For the full demography, please see Table 2.

Table 2.

Patient characteristics. SN=Sentinel Node, ALND: Axillary Lymph Node Dissection.

Demography
Population, n		10
Age (years), mean (±SD)		58.5 (±12.12)
Lymphedema duration (months), median (IQR)		16.8 (10.48–35.59)
BMI (kg/m2), mean (±SD)		29.7 (±5.09)
Comorbidities, n (%)
	Hypertension Depression Diabetes	4 (40.00%) 3 (30.00%) 1 (10.00%)
Dominant arm, n (%)
	Right Left	10 (100.00%) 0 (0.00%)
Lymphedema arm, n (%)
	Right Left	4 (40.00%) 6 (60.00%)
Afflicted with erysipelas, n (%)
	Yes No	1 (10.00%) 9 (90.00%)
Cancer surgery, n (%)
	Mastectomy Lumpectomy	7 (70.00%) 3 (30.00%)
Breast reconstruction, n (%)
	Autologous tissue reconstruction Implant based None	5 (50.00%) 2 (20.00%) 3 (30.00%)
Axillary surgery, n (%)
	SN ALND	0 (0.0%) 10 (100.00%)
Cancer treatment, n (%)
	Radiotherapy Chemotherapy Neo-adjuvant chemotherapy Anti-hormone Biological treatment	10 (100.00%) 5 (50.00%) 3 (30.00%) 8 (80.00%) 2 (20.00%)
Conservative lymphedema treatment, n (%)
	Compression sleeve Compression glove Compression sleeve (night) Previously bandages by physiotherapist Lymphedema exercise Self-drainage Intermittent pneumatic compression	10 (100.00%) 9 (90.00%) 5 (50.00%) 7 (70.00%) 7 (70.00%) 6 (60.00%) 6 (60.00%)

Primary outcome

Surgical data

Preoperative planning identified a median of two lymph vessels (IQR: 1–3) per patient. Similarly, UHFUS was used to locate a median of two venous vessels (IQR: 1–3) for anastomosis. Consequently, a median of two anastomoses were planned and successfully performed for each patient, see Table 3, Table 4. Only patient 10 did not receive an LVA, as the lymphatic vessel was within the cutis, preventing dissection. In patient 3, one anastomosis was not performed due to a very small lymphatic vessel, and in patient 8, one was not performed because of vessel size mismatch and delayed ICG-L visualization, likely from retrograde backflow. All other patients received at least one anastomosis. No multi-LVA was performed. There were no significant differences in lymphedema duration or BMI amongst the patients who received all planned anastomoses compared with those who did not receive all planned anastomoses, see the appendix for full results.

Table 3.

Overview of the surgical outcomes and pre-operative planning using indocyanine green lymphography and ultra-high frequency ultrasound. Medians have been tested using Wilcoxon signed-rank test for non-normally distributed data. IQR: Interquartile range.

Surgical Outcomes
		Pre-operative	Peri-operative	p-value
No. lymph vessels, median (IQR)		2 (1–3)	2 (1–3)	0.16
No. veins, median (IQR)		2 (1–3)	2 (1–3)	0.65
No. anastomoses, median (IQR)		2 (1–3)	2 (1–3)	0.65
Type of anastomoses, n
	End-to-end Sleeved End-to-side	12 3 2
Surgical duration (mean, ±SD), minutes		137.44 (± 53.18)
Duration per anastomoses (mean, ±SD), minutes		72.5 (± 16.18)

Table 4.

Overview of all patients' pre-operative planning findings from indocyanine green lymphography and ultra-high frequency ultrasound.

Overview of patients
	Pre-operative			Peri-operative			Comments
Patient	Lymph vessel	Venule	Anastomsis planned	Lymph vessel	Venule	Anastomosis performed
1	3	3	3	3	3	3
2	4	4	3	3	3	3
3	3	2	2	2	2	2
4	3	3	3	3	3	2	Due to the small caliber of lymph vessel, the third LVA was not performed.
5	2	2	2	2	2	2
6	1	1	1	1	1	1
7	2	2	2	2	2	2
8	2	2	2	2	1	1	Due to the great caliber of vein, the second LVA was not performed.
9	1	1	1	1	1	1
10	1	1	1	1	.	0	The lymph vessel planned for surgery was located within the dermis, and therefore not able to freely dissect for anastomosis. The nearby vein was therefore not located and anastomosis not performed.

With just 10 patients and 1–3 anastomoses each, median values are relatively imprecise. For clarity, both median (IQR) and mean (±SD) are reported: the median proportion of performed anastomoses was 100% (IQR 66.6–100%), and the mean was 81.7% (±33.7).

Secondary outcomes

Arm volume

Mixed-effects analysis showed no significant changes in relative or absolute arm volume over 12 months when measured by water displacement. Tape circumference similarly showed no significant changes overall, although a small reduction in absolute arm volume was observed at 9 months (−123.8 mL, SE 56.7, p = 0.03). The likelihood-ratio test supported the mixed-effects regression over a simple linear model for both outcomes. Full results are presented in Fig. 2 and Table 2a in the appendix.

Fig. 2.

Overview of relative arm volume difference and its changes over time.

LYMPH-Q

There was observed an improvement in quality of life during the follow-up period in regards to all scales, however, only at limited follow-up time points. Symptoms improved after six- (coefficient=3.70, p = 0.03) and 12 months follow-up (coefficient=3.73, p = 0.03), arm function also improved after six and nine months (coefficient=3.01, p = 0.01 and coefficient=3.34, p ≤ 0.001) with twelve months showing a trend towards improvement (p = 0.06). The same tendency was seen for psychological function, information and satisfaction with arm sleeve, and also arm appearance. A likelihood-ratio test confirmed that the mixed-effects model provided a significantly better fit than the simple linear model (p < 0.05). See Fig. 3 and the appendix for full LYMPH-Q results.

Fig. 3.

Linear plots visualizing the changes of the mean scores for each LYMPH-Q scale over the study time. * = statistically significant.

L-Dex score

Little changes in the L-Dex score was observed at 3 and 6 months of follow-up. After 9 months, the L-Dex score improved significantly, with a reduction of 24.16 points compared to baseline (95% CI: [−45.95; −2.36], p = 0.03). At 12 months, the score increased slightly (−13.47, 95% CI: [−35.21; 8.27], p = 0.23) but remained lower than baseline, indicating reduced fluid in the lymphedematous arm relative to the healthy arm. See Fig. 4 and Table 5a in the appendix for all results.

Fig. 4.

Linear plots visualizing the changes of the mean L-Dex Scores over the study period. The dark area represents the area of normal L-Dex Score values. * = statistically significant.

Discussion

This pilot study evaluated the combined use of UHFUS and ICG-L for LVA planning and assessed its impact on arm volume, quality of life, and L-Dex in BCRL patients. While arm volume did not decrease significantly, improvements in quality of life and L-Dex were observed, consistent with prior studies showing patient-reported benefits even in the absence of measurable volume changes.¹⁸

Targeted planning

The use of targeted planning of combined UHFUS and ICG-L has been applied only to a limited degree over the last few years in preoperative planning of LVA. And to our knowledge, only a few studies have evaluated the use of ultrasound or UHFUS to identify venules or veins for surgery.12, 13, 14, 15^,26,27 The combination allows for both functional assessment of the lymphatics and precise detection of nearby venules or veins, without depending on exploratory surgery for venule detection. This provides surgeons with precise preoperative knowledge of the relevant anatomical structures, thereby facilitating more efficient surgery and potentially improving outcomes.

Previous studies have shown that UHFUS combined with color Doppler has significantly higher sensitivity for detecting lymphatic vessels compared with magnetic resonance lymphography (MRL) and ICG-L.¹² However, these ultrasound-based methods do not capture lymphatic function, which remains a key benefit of ICG-L. Therefore, the combined use of both modalities may provide complementary information: ICG-L provides functional insights, while UHFUS enables high-resolution structural evaluation, including vessel caliber and depth. And this study highlights the feasibility of targeted planning for LVA surgery using the combination of ICG-L and UHFUS, to combine the functional evaluation of the lymphatic system in the lymphedematous arm, and the UHFUS imaging modality with color doppler to evaluate anatomical location of venule nearby. Recent studies suggests that recipient veins with favorable flow characteristics, valves and utilization of the venturi effect, may improve lymphatic drainage and long-term anastomotic patency.^28,29 Although UHFUS has the potential to aid in the identification of suitable recipient veins, this aspect was not evaluated in the present study and warrants further investigation.

Secondary outcomes

Our analysis of the secondary outcome, change in arm volume, did not demonstrate a statistically significant reduction in lymphedema volume at 3, 6, 9, or 12 months post-LVA. This finding contrasts with some existing studies that have reported volume reduction following LVA.¹⁸ In addition, the absolute arm volume difference at baseline differed between water displacement and tape circumference. This might be due to the tape circumference measurements not including the hand and is also associated with greater inter- and intrarater variability compared to water displacement, which is accepted as a highly valid measuring method.³⁰ However, using tape measurements of arm circumference allows the evaluator to focus on specific areas of volume change. Additionally, arm circumference measurements are commonly used in lymphedema surgery research, improving comparability across studies.¹⁸ The pathophysiology of lymphedema, characterized by chronic lymphatic dysfunction and tissue fibrosis, may limit the reversibility of volume changes.³¹ Arm volume, though a standard endpoint, may be insensitive or delayed in reflecting clinical improvement. The 12-month follow-up may be too short to capture full LVA effects, and the small sample size limits detection of subtle changes. The measurement methods also have inherent variability.¹⁹ Patient factors, such as day-to-day fluctuations in arm swelling, add to this variability.³²

The study found that LVA surgery generally improved quality of life for patients with breast cancer-related lymphedema across all LYMPH-Q scales. However, these improvements were not immediate or consistently sustained across all follow-up time points. Given the small sample size and the preliminary nature of these findings, some observed effects did not reach statistical significance; it is plausible that, with a larger cohort and longer follow-up, these differences might become significant. These results are consistent with previous studies reporting improvements in quality of life after LVA, suggesting that the procedure can offer meaningful benefits.¹⁸

Advantages and limitations

The main strengths of this combined planning approach are its safety and non-radioactive nature. Compared with MRL, it allows real-time, high-resolution visualization of anatomical structures and their precise locations. ICG-L has become the gold standard for evaluating lymphedema and visualizing lymphatic vessels. However, it has limitations: it can visualize vessels only to a depth of 1.5 cm, and vessels covered by dermal backflow may not be detected. Additionally, patients with hypersensitivity reactions to ICG cannot undergo ICG-L, making modalities such as UHFUS a possible alternative. Nonetheless, one could question whether patients with absent lymphatic uptake are ideal candidates for LVA, as surgical success depends on the presence of at least partially functional lymphatics capable of draining lymph into the anastomosed vein.³³

The limitations of ICG-L were also evident in this study, as the lymphatic vessels were not assessed with UHFUS. Consequently, the caliber of the mapped lymphatic vessels in patients 4 and 8, and the depth of the lymphatic vessel in patient 10 were not determined. These issues might have been avoided if UHFUS had been used following the ICG-L to evaluate both lymph vessel caliber and depth.³⁴ However, since the LVA procedure and its planning had only recently been established in the department, there was minimal basis for comparison with other approaches. Continued experience and refinement of the procedure after the study period have since improved our understanding, highlighting that lymph vessel and venous calibre, depth, and anatomical location should also be evaluated using UHFUS.

A limitation of UHFUS is its operator-dependency and difficulty in distinguishing lymphatic vessels from subcutaneous veins or nerves. Combining ICG-L for lymphatics with UHFUS for venules using color Doppler appears to be the most effective approach.

Lastly, the absence of a comparison group should be acknowledged as a limitation. This study reflects the department's initial experience with the surgical procedure and its preoperative planning, and consequently, no control group was available for comparison.

Future research

Future studies should explore the use of UHFUS to assess lymphatic vessel function, caliber, depth, and sclerosis. Long-term follow-up is needed to evaluate the durability of improvements, effects on arm volume, and the ongoing need for conservative therapy. Examining surgical factors, such as the number and patency of anastomoses, and assessing the cost-effectiveness of LVA versus long-term conservative management are also important areas for future research.³⁵

Conclusion

This study demonstrates that combining UHFUS with ICG-L in the preoperative planning of LVA is feasible and provides structural and functional information that may optimize vessel selection and improve surgical precision.

Institutional review board statement

This study has been approved by the Regional Committees on Health Research Ethics for Southern Denmark (S-20210169G) and by the Region of Southern Denmark register of research projects (21/59848).

Funding

This work was funded by the Danish Cancer Society and the Danish Research Center for Cancer Surgery (ACROBATIC), Danish Cancer Society (grant no. R317-A18255), and Danish Comprehensive Cancer Center; Danish Cancer Society (grant no. R374-A22446); Overlægerådets Forskningsfond; The Vissing Foundation; Ristrupfonden; Fabrikant Einar Willumsens Mindelegat; Syddansk Universitets Forskningsfond; Ristrupfondet; and the PhD grant from the Region of Southern Denmark (Region Syddanmarks Ph.d.-pulje). The funding covered equipment, publication costs, and the PhD student’s salary, and the funders had no role in the study design, data analysis, or publication decisions.

Declaration of competing interest

None.