Thoracic ultrasonography-guided biopsy: Radiological and radiometabolic factors affecting diagnostic success

Abstract

OBJECTIVE

Investigate radiologic and radiometabolic features affecting the diagnostic success of transthoracic biopsy (TTB) procedures guided by Thoracic Ultrasonography (TUS).

METHODS

The electronic medical records of all patients who underwent TUS-guided TTB procedures (TUS-TBB) were retrospectively reviewed. Patients with a final diagnosis of primary lung cancer (LC), established via TUS-TBB or other diagnostic modalities, were included in the study. The cohort was stratified into two groups: those in whom the TUS-TBB procedure was diagnostic and those in whom it was non-diagnostic. Clinical, demographic, radiological [Chest Computed Tomography (CT)], and radiometabolic (PET-CT) characteristics were compared between the two groups.

RESULTS

A total of 107 patients with a final diagnosis of lung cancer were included in the study. Of these, 91 (85%) were male, with a mean age of 67.8±11.4 years. The mean maximum diameter of the mass lesion on chest CT was 68.2±31 millimeters (mm), and the mean maximum standardized uptake value (SUV_max) on PET-CT was 14.1±6.9. A necrotic component/ametabolic area within the lesion on PET-CT was present in 43 (40.2%) patients and absent in 64 (59.8%). The most frequent histopathological subtype was adenocarcinoma, observed in 46 (42.9%) cases. While 41 (38.3%) patients underwent transthoracic fine-needle aspiration biopsy (TTFNAB) under TUS guidance, 66 (61.7%) patients underwent tru-cut biopsy (TCB). The TUS-TBB procedure was diagnostically successful in 87 (81.3%) of the cases. The presence of a necrotic component within the lesion on PET-CT was identified as the sole independent risk factor significantly impacting the diagnostic success of the procedure (p=0.010).

CONCLUSION

TUS-TBB procedures demonstrate high diagnostic success. In carefully selected patients, pre-procedural evaluation of lesion metabolic activity using PET-CT may enhance the diagnostic success of the biopsy.

Highlight key points

• High Diagnostic Yield: Thoracic ultrasonography (TUS)-guided transthoracic biopsy (TTB) procedures achieved a diagnostic success rate of 81.3%, with no major complications reported.

• Necrotic Components on PET-CT Reduce Diagnostic Accuracy: The presence of necrotic or non-metabolic areas within the lesion on PET-CT was identified as the only independent factor significantly associated with lower diagnostic success (p=0.010).

• Needle Type Did Not Affect Diagnostic Success: There was no statistically significant difference in diagnostic yield between fine-needle aspiration biopsy (FNAB) and Tru-Cut core biopsy methods.

• Lesion Size, SUV_max, and Clinical Factors Were Not Determinant: Lesion size, SUV_max value, and clinical or demographic variables did not significantly influence diagnostic outcomes.

• PET-CT Should Be Considered for Biopsy Planning: Pre-procedural PET-CT evaluation of metabolic activity may help guide biopsy targeting and optimize diagnostic success, particularly by avoiding necrotic zones.

Ultrasonography (US) platforms, which transmit sound waves with wavelengths beyond the human hearing range into tissues via a probe and generate images from the echoes received, were first discovered in the 1920s and have been used in medical practice since the 1940s. As solid organs typically reflect these sound waves, the application of US in lung imaging emerged later. Except for the pleura, healthy lungs containing air cannot be directly visualized sonographically. However, pleural effusions, consolidations, and soft tissue tumors of the lung that are in full contact with the chest wall without intervening air can be readily visualized using thoracic ultrasonography (TUS) [1, 2].

Today, TUS is widely utilized not only for diagnostic purposes such as imaging pleural effusions, evaluating pathological pleural lesions, detecting malignant or benign thoracic wall pathologies, and visualizing peripherally located lung tumors, but also as a guidance tool for various interventional procedures. These include thoracentesis, pleural biopsy, pleural catheter insertion, biopsy of thoracic wall or supraclavicular metastases, and biopsy of peripheral lung lesions [3].

Several studies have demonstrated that TUS-guided transthoracic biopsies (TUS-TTB) of peripherally located pulmonary tumors possess high diagnostic success. However, TTB may occasionally be insufficient for pathological diagnosis. It is particularly well-recognized that diagnostic performance is reduced in lung lesions with necrotic or cavitary architecture [4].

In this study, we investigated the factors influencing the diagnostic success of TUS-TTB procedures.

MATERIALS AND METHODS

All cases who underwent TUS-TTB between 2022 and 2024 in the Departments of Pulmonology and Interventional Radiology were retrospectively reviewed in electronic records. This study was conducted in accordance with the Declaration of Helsinki and received approval from our institutional ethics committee (Number: 10-10 Date: 11/06/2025). Only cases whose final diagnosis was primary lung cancer (LC) were included. Patients with extrapulmonary malignancies or extrathoracic metastasis to the lung were excluded. Inclusion and exclusion criteria are presented in Table 1.

Table 1.

Inclusion and exclusion criteria of the study

Inclusion criteria
1-Patients aged ≥18 years
2-Patients who underwent transthoracic biopsy (TTB) for a mass lesion in thorax
3-Patients with a final diagnosis of primary lung cancer
4-Patients for whom a PET-CT report was available
Exclusion criteria
1-Patients with a diagnosis of extrapulmonary cancer
2-Patients with radiological suspicion of pulmonary metastasis
3-Patients with a mass lesion in one hemithorax who underwent biopsy of cervical, supraclavicular, or superficial chest wall lesions instead of the main lesion
4-Patients for whom a PET-CT report was not available
5-Patients who did not undergo TTB, had a lesion size <1 cm, had contraindications for the procedure, or did not consent to the procedure
6-Patients with a platelet count <20,000/mm3 or an INR >1.3 on coagulation testing

PET-CT: Positron emission tomography-computed tomography; INR: International normalized ratio; TTB: Transthoracic biopsy.

For all included patients, clinical, demographic, radiological (Thoracic computed tomography, CT), and radiometabolic (Positron emission tomography–CT, PET-CT) characteristics were recorded, along with the type of biopsy performed [Fine-needle aspiration biopsy (FNAB) or Tru-Cut core biopsy (TCB)], the distribution of final histopathological subtypes in primary LC cases, cancer stages, and the definitive diagnostic methods employed in cases in which TUS-TTB was non-diagnostic. Patients were divided into two groups: those in whom TUS-TTB yielded a diagnostic specimen (Group 1) and those in whom TUS-TTB was non-diagnostic (Group 2). The two groups were compared with respect to all recorded variables.

TTB Procedures (FNAB-TCB) under TUS Guidence

TUS-guided FNAB procedures are performed with TUS, a 3.5 MHz convex probe, and abdominal mode by following the steps below:

1. The entire thorax is systematically scanned, beginning in the seated position over the radiologically identified lesion site and, where necessary, in the supine, oblique, and lateral decubitus positions to localize the target and plan the biopsy trajectory.

2. After sonographic localization of the target lesion, its dimensions, the anticipated needle trajectory through the lesion, and any intralesional or perilesional vascular structures are identified using color Doppler imaging.

3. Informed consent is obtained from all patients prior to the procedure.

4. The biopsy site is disinfected with an iodine–alcohol solution, and the ultrasound probe used for guidance is similarly sterilized.

5. A 22-G spinal needle attached to a 20-mL syringe is advanced through the marked entry site.

6. Once real-time ultrasound monitoring confirms the needle tip within the target lesion, aspiration is performed by fanning the needle through different regions of the lesion to obtain multiple tissue samples. The aspiration is terminated when grossly visible material or hemorrhagic content is observed in the syringe.

7. Collected specimens are then appropriately prepared and submitted to the pathology laboratory.

TUS-guided TCB procedures are performed with TUS, a 3.5 MHz convex probe, and abdominal mode by following the steps below:

1. Similar to fine-needle aspiration biopsy procedures, the entire thorax is scanned, and once the lesion is identified sonographically, local anesthesia is administered with 2% lidocaine hydrochloride to the targeted TCB site prior to the procedure.

2. Under thoracic USG guidance, an 18- or 20-G Chiba needle is advanced, and once it is confirmed that the TCB needle has reached the target lesion and that tumor tissue is present more than 3 cm distal to the intended biopsy area, the biopsy is performed to obtain tissue samples measuring either 2 cm or 1 cm in length.

3. The collected specimens are then appropriately placed in formalin solution and sent to the pathology laboratory in accordance with standard protocols (Fig. 1).

Figure 1.

Ultrasonographic appearance of a posteriorly located right upper lobe lesion identified on PET-CT. Although the lesion is predominantly hypoechoic, it contains focal heterogeneous (necrotic) areas. Necrotic regions are indicated by red arrows.

Positron Emission Tomography/Computed Tomography

PET/CT imaging was performed using a PET/CT scanner (Philips Gemini TF 64, Philips Medical Systems, Cleveland, OH, USA). Patients fasted for at least 8 hours prior to imaging, and only those with normal blood glucose levels were included. Following intravenous injection of 370–550 MBq (5–15 mCi) F-18 FDG, imaging was performed after a standardized uptake period of 60 minutes. PET data were acquired from the vertex to the upper thigh with a 5-mm slice thickness, using a time-of-flight–enabled OSEM reconstruction algorithm (3 iterations and 33 subsets) and a 144 × 144 matrix. Low-dose CT scans were obtained for attenuation correction and anatomical localization using the following parameters: 120 kVp, 50–150 mAs, 0.8–1.0 pitch, and 0.5–0.8 s rotation time. An oral contrast agent was administered prior to imaging.

Statistical Analysis

Statistical analysis was performed using the SPSS 24.0 software program (IBM IncReleased-SPSS Statistics for Windows, Chicago, USA). In descriptive statistics, continuous variables were expressed as mean±standard deviation, and categorical variables as percentages. The Kolmogorov-Smirnov test was used to assess the normality of distribution. Data from the groups were evaluated using the Independent Samples T-Test and Chi-Square test, and when necessary, the Mann-Whitney U test. Multivariate regression analysis was employed to assess the significance of factors associated with the diagnostic success of TUS-TTB procedures, within a 95% confidence interval (CI). A p-value of <0.05 was considered statistically significant for all tests.

RESULTS

For the study, the records of 149 patients who underwent TUS due to a lung mass lesion were retrospectively reviewed. Seventeen cases were excluded because their final diagnosis was extrapulmonary cancer, 16 cases had biopsies performed from sites other than the lung mass (e.g., cervical/supraclavicular lymph nodes), and 9 cases were excluded due to the unavailability of PET-CT reports. The study proceeded with a total of 107 patients who met the inclusion criteria and had a final diagnosis of primary lung cancer. Of these, 91 patients (85%) were male and 16 (15%) were female, with a mean age of 67.8±11.4 years. The mean smoking history was 40.8±19.2 pack-years. While 56 patients (52.3%) had comorbidities other than cancer, 51 patients (47.7%) had no additional diseases. The most common comorbidity was hypertension (HT), observed in 32 patients (29.9%) (Table 2).

	Biopsy is diagnostic (n=87)	Biyopy is not diagnostic (n=20)	p
Demographic features
Age (avg±SD)	68.4±10.8	65±13.6	0.443
Sex (f/m %)	13.7/86.3	20/80	0.494
Comorbidities (yes) (%)	56.3	35	0.085
HT (yes)	32.2	20	0.283
DM (yes)	19.5	15	0.760
CHD (yes)	18.4	10	0.516
CRD (yes)	20.7	10	0.354
CND (yes)	10.3	10	1
CKD (yes)	10.3	5	0.406
Radiologic features (thoracic CT)
Lesion’s long axis (mm) (avg±SD)	67.9±28.6	69.7±40.6	0.898
Distance from skin to lesion (avg±SD)	30.3±11.3	28.3±8.5	0.424
Emphysematous appearance on the same side (yes) (%)	62.1	75	0.276
Sequela lesion on the same side (yes) (%)	57.5	65	0.537
Preferred biopsy method (yes) (%)			0.495
TUS- FNAB	36.8	45
TUS- TCB	63.2	55

Avg: Average; DM: Diabetes mellitus; HT: Hypertension; CHD: Chronic heart disease; CRD: Chronic respiratory disease; CND: Chronic neurologic disease; CKD: Chronic Kidney disease; mm: Millimeter; CT: Computed tomography; SD: Standard deviation; TUS-TTB: Thoracic ultrasound-guided transthoracic biopsies; FNAB: Fine needle aspiration biopsy; TCB: Tru-cut biopsy.

Table 2.Demographic, radiological and methodological characteristics of the groups in which TUS-TTB procedures were diagnostic and non-diagnostic

When the radiological and radiometabolic characteristics of the entire patient group were examined, the mean long-axis diameter of the mass lesion on thoracic CT was 68.2±31 millimeters (mm), and the mean distance from the skin to the lesion was 30±10 mm. Emphysematous appearance in the area targeted for biopsy on CT was identified in 69 patients (64.5%), and ipsilateral parenchymal sequela lesions were observed in 63 patients (58.9%) (Table 2). On PET-CT, the mean SUV_max value of the lesions was 14.1±6.9. A necrotic component or non-metabolic area within the lesion was detected on PET-CT in 43 patients (40.2%), whereas 64 patients (59.8%) showed no such features. Mediastinal involvement was present in 75 patients (70.1%) and absent in 32 patients (29.9%) on PET-CT. Distant metastases were observed in 56 patients (52.3%) and not observed in 51 patients (47.7%) (Table 3).

Table 3.

Radiometabolic and histopathological characteristics of diagnostic and non diagnostic TUS-TTB procedures

	Biopsy is diagnostic (n=87)	Biyopy is not diagnostic (n=20)	p
Radio-metabolic properties (PET-CT)
Lesion SUVmax value	14±6.8	14.2±7.4	0.767
Presence of necrotic component within the lesion (yes) (%)	47.1	10	0.002
Mediastinal lymph node involvement (yes) (%)	70.1	70	0.992
Presence of distant organ metastasis (yes) (%)	54	45	0.446
Histopathological distribution of cancer (yes) (%)
Adenocarcinoma	40.2	55	0.229
Squamous cell carcinoma	34.5	25	0.415
Non-small cell carcinoma, NOS	16.1	15	1
Small cell carcinoma	5.7	5	0.896
Others*	3.4	–	1
Stages of cancer (yes) (%)
Stage 1-2	18.7	2	0.512
Stage 3-4	81.3	100	0.746
Extensive stage small cell lung cancer	4.5	10	0.312

^*: 1 case is Adenoid Cystic Carcinoma, 1 case is Large Cell Carcinoma, 1 case is Liposarcoma. SUV_max: Standard uptake value-maximum; SD: Standard deviation; PET-CT: Positron emission tomography-computed tomography; NOS: Not otherwise specified.

Upon examination of the patients’ final histopathological diagnoses and disease stages, the most frequently observed histopathological subtype was adenocarcinoma, identified in 46 cases (42.9%). The most commonly detected disease stage was advanced stage (stage 3–4), found in 80 cases (81.6%) (Table 3, Fig. 2, 3).

Figure 2.

Graph showing the final histopathological distribution of the cases (the x-axis of the graph shows histopathologic results, while the y-axis shows cumulative percentages).

Figure 3.

Graph showing the distribution of lung cancer stages of cases (the x-axis of the graph represents tumor stages, while the y-axis displays the cumulative percentages for each stage.

Upon analysis of the biopsy method subgroups, 41 patients (38.3%) underwent FNAB under TUS guidance, while 66 patients (61.7%) underwent TCB under TUS guidance (Table 3). No major or minor complications (such as pneumothorax requiring chest tube placement, minimal pneumothorax, or hemoptysis) were recorded during or after the procedures.

Among the patients who underwent TUS-TTB, the procedure was diagnostic in 87 cases (81.3%) and non-diagnostic in 20 cases (18.7%). The clinical, demographic, radiological, and radiometabolic characteristics of the diagnostic and non-diagnostic groups are presented in detail in Tables 2, 3.

Among the 20 non-diagnostic cases, 12 (60%) were diagnosed via endobronchial ultrasonography (EBUS), 7 (35%) via CT-guided transthoracic biopsy (CT-TTB), and 1 (5%) via surgical biopsy. To identify independent risk factors affecting the diagnostic success of TUS-TTB procedures, a regression model was constructed that included lesion long-axis diameter, lesion SUV_max value, presence of a necrotic component within the lesion, and the biopsy method used. Within this model, the presence of a necrotic component in the lesion on PET-CT was found to be the only independent risk factor significantly associated with reduced diagnostic success (p=0.010) (Table 4).

Table 4.

Regression analysis showing independent risk factors for TUS-TTB

Variables	B	Bias	SE	Sig. (2-tailed)	95% CI
					Lower	Upper
Lesion’s long axis	-0.005	0.001	0.012	0.630	-0.023	0.023
Lesion SUVmax value	-0.005	0.000	0.039	0.878	-0.078	0.077
Presence of necrotic component within the lesion	-2.182	-2.432	6.271	0.010	-20.698	-0.864
Preferred biopsy method	-0.343	-0.088	0.687	0.564	-1.869	0.894
Constant	3.644	2.444	6.422	0.003	1.481	22.989

TUS-TTB: Thoracic ultrasound-guided transthoracic biopsies; SE: Standard error; B: Beta; Sig (2-tailed): Two-tailed p-value; CI: Confidence interval.

DISCUSSION

In this study, the diagnostic success of TUS-TTB procedures in primary LC cases and the factors influencing diagnostic success were investigated. The overall diagnostic success rate of the procedures was calculated as 81.3%. No significant complications were observed as a result of the procedures. The only independent factor found to significantly affect diagnostic success was the presence of a necrotic or non-metabolic component within the tumoral lesion on PET-CT. Apart from this, the clinical, demographic, and histopathological characteristics were similar between the diagnostic and non-diagnostic groups.

Despite certain limitations imposed by the anatomical structure of the lungs, TUS is increasingly being utilized in the diagnosis of pulmonary diseases. Compared to other imaging modalities, its major advantages include real-time imaging capability, absence of radiation exposure, portability, and bedside applicability. Lung tumors that are in contact with the chest wall without intervening aerated lung can be visualized and sampled using TUS [3, 5–7]. Lung cancers remain one of the most common cancer types today. Histopathological examination of stained tissue biopsies continues to be the gold standard for the diagnosis [8]. In a large and recent meta-analysis including 12 original studies (encompassing 3,830 TUS-guided transthoracic biopsies), Li S et al. [9] described TUS-TTB as a highly accurate, technically well-performing, and safe diagnostic tool. The diagnostic success of TUS-TTB procedures varies across a wide spectrum, ranging from 66.7% to 97.1% [10]. Numerous studies have investigated the factors influencing the diagnostic success of TUS-TTB. These studies particularly highlight the high diagnostic accuracy and low complication rates in peripherally located pulmonary lesions. Among the clinical factors affecting diagnostic success, patient cooperation, respiratory control, and comorbid conditions (such as COPD and emphysema) are of particular importance. From a radiological perspective, lesion size—particularly in lesions ≥2–3 cm—has been associated with higher diagnostic success. The presence of atelectatic areas within the lesion and the distance from the skin to the lesion are also among the radiological factors influencing procedural outcomes. Demographically, factors such as age, sex, and smoking history have a limited direct impact on diagnostic success. However, some studies have reported decreased tolerability of the procedure in elderly patients. Operator experience, the type of needle used (FNAB vs. TCB), and the use of real-time imaging during the procedure are among the procedural factors that enhance diagnostic success [1, 11–15]. In our study, consistent with the literature, the diagnostic success rate of the procedures was calculated as 81.3%. One of our notable findings was that the type of needle used did not have a statistically significant impact on diagnostic success. Most studies on this topic, however, suggest that TCB procedures are associated with higher diagnostic success [1, 16]. In this context, the discrepancy between our findings and those in the literature may be attributed to the relatively small sample size of our study or may simply be due to chance. Alternatively, it may be explained by the inherently high diagnostic success of TUS-TTB procedures, which are performed in real time, thereby minimizing the number of factors that could adversely affect diagnostic accuracy.

In our analysis, we found no significant differences between the diagnostic and non-diagnostic groups in terms of clinical, demographic, radiological characteristics, needle type used, cancer stage, or histopathological subtype. Although our results may not be generalizable, they suggest that TUS-TTB is a procedure with an acceptably high diagnostic success, and that the clinical, demographic, and radiological parameters recorded in our study do not appear to have a direct impact on diagnostic success.

One of the key findings of our study is the impact of necrotic and ametabolic components observed on PET-CT on diagnostic success. In our cohort, biopsy samples obtained from lesions containing ametabolic–necrotic areas on PET-CT demonstrated a significantly lower diagnostic success rate (p=0.002). Regression analysis further revealed that the presence of a necrotic component within the lesion is an independent risk factor that adversely affects the diagnostic success of the procedure (p=0.010). Although TUS-TTB is a procedure with high diagnostic success, its diagnostic accuracy depends not only on the biopsy technique but also on the morphological and metabolic characteristics of the target lesion. This may be attributed to the inadequate cellular structure provided by necrotic tissue for histopathological evaluation. Necrotic components consist of hypoxic, acellular regions that replace viable tumor tissue, and biopsy needle targeting of these areas can complicate diagnostic acquisition [17, 18]. In the literature, it has been reported that sampling from areas demonstrating low metabolic activity on PET-CT is associated with suboptimal diagnostic success, due to the potential absence of viable tumor cells. Liu W et al. [19] in a study including 458 cases of PET/CT-guided biopsy of pulmonary lesions, found that the overall diagnostic success rate was significantly higher in the PET/CT group (93.0% versus 83.1%, p=0.001). The authors emphasized in their study that diagnostic failures in pulmonary lesions may occasionally occur due to factors such as fibrosis, necrosis, atelectasis, or obstructive pneumonia. They highlighted that PET/CT, as a metabolic imaging modality, has the ability to distinguish hypermetabolic areas that reflect the metabolic characteristics of lung cancers and the true biological behavior of the entire lesion. Therefore, they underlined that PET/CT may help reduce the risk of diagnostic failure. In another study, Dogan C et al. [20] investigated the impact of PET-CT on diagnostic success in 193 bronchoscopic biopsy procedures. They demonstrated that diagnostic success was lower in cases where the primary lesion contained ametabolic-necrotic areas or exhibited heterogeneous uptake on PET-CT. Numerous studies on this subject [21–23] have emphasized that PET-CT imaging should play a guiding role not only for diagnostic purposes but also in biopsy planning. We believe that the metabolic and anatomical data provided by PET/CT have the potential to enhance targeting accuracy, thereby maximizing diagnostic success and contributing to a reduction in non-diagnostic results through optimized needle guidance. PET-CT is highly sensitive in distinguishing viable tumor cells from necrotic or non-viable tumor tissue [24]. In clinical practice, targeting areas with the highest metabolic activity during biopsy planning, avoiding necrotic components of the lesion, and utilizing PET-CT to guide needle positioning within the lesion may enhance diagnostic success. Particularly in tumors with heterogeneous metabolic profiles, the effective use of PET-CT for biopsy mapping can improve overall diagnostic efficiency.

In our study, the diagnostic success of TUS-TTB procedures was found to be 81.3%. Our results are consistent with the literature. Studies on the subject underline that TUS-TTB procedures are quite reliable and have a high diagnostic rate [25, 26].

Study Limitations

This study has several limitations that should not be overlooked when interpreting the results or planning future research. Firstly, it is a retrospective study conducted with a small number of cases, which significantly limits its statistical power. Therefore, it is not as powerful as a multicenter, randomized prospective study. Additionally, in our country, according to the regulations of the Social Security Institution, PET-CT can only be requested after a histopathological diagnosis has been established. We believe that if PET-CT evaluation had been performed prior to diagnosis, it could have positively influenced the diagnostic success. Prospective studies involving TTB procedures in two distinct patient groups—those with and without pre-procedural PET-CT—would contribute more robustly to the literature on this topic.

Conclusion

TUS-TTB procedures, long utilized in the diagnosis of lung cancer, are highly successful diagnostic interventions. TUS is inexpensive, reliable, readily available at the bedside, does not involve radiation exposure, and most importantly, allows for real-time intervention. In selected cases where diagnostic success may be reduced, evaluating the metabolic activity of lesions prior to the procedure and mapping the biopsy to target areas with higher metabolic activity may further enhance diagnostic accuracy.

Ethics Committee Approval

This study was approved by The İstanbul Medeniyet University Faculty of Medicine review board (Number:10-10 Date:11/06/2025).

Informed Consent

Written informed consents were obtained from patient and his family.

Conflict of Interest

No conflict of interest was declared by the authors.

Financial Disclosure

The authors declared that this study received no financial support.

Use of AI for Writing Assistance

Not declared.

Authorship Contributions

Concept – CD, MA, RG; Design – CD, MA; Supervision – SG, CD; Fundings – SG, MA; Materials – GM, NK; Data collection and/or processing – GM, NK; Analysis and/or interpretation – GSY, ZNK; Literature review – GSY, ZNK; Writing – CR, DG; Critical review – CD, MA, SG.

Peer-review

Externally peer-reviewed.