Ultrasound Applicability in the Assessment of Patients With Fibrosing Interstitial Lung Disease

Introduction Fibrosing interstitial lung diseases (F-ILD) represent a heterogeneous disease category involving many subtypes with different clinical, radiological and histological characteristics . Idiopathic pulmonary fibrosis (IPF) is the most common subtype of F-ILD, and IPF is the prototype of other F-ILD phenotypes, known as progressive pulmonary fibrosis (PPF). Despite their diverse presentation, the common denominator defining them as F-ILD patients is due to similarities regarding irreversible symptom worsening, decline in lung physiological parameters and progression of pulmonary fibrosis on thoracic imaging . Accomplishment of early F-ILD diagnosis and disease progression is crucial to initiate treatment at early stage of disease. Chest high-resolution computed tomography (HRCT) is considered the golden standard to diagnose and quantify pulmonary fibrosis. However, repeated HRCT performance is time-consuming, and repeated HRCT scanning is hampered by the cumulated radiation exposure with the increasing risk of cancer development.

Therefore, it is necessary to propose new modalities for the early identification and monitoring of disease progression in F-ILD patients.

Thoracic ultrasound (TUS) is superior to diagnose various conditions of the lung parenchyma and pleura compared to chest X-ray (CXR) . B-line artefacts (BLA) and the interstitial syndrome (IS) are well described TUS indicators related to ILD, and several studies have described a linear correlation between an increasing number of BLAs and the severity degree of fibrotic HRCT findings based on a Warrick score . Additionally, a significant negative correlation between an increased distance between BLAs and decreasing lung physiological parameters has also been demonstrated. Not only parenchymal, but also pleural abnormalities (PA) occur in patients with F-ILD. Several studies have proposed that increasing thickness and irregularity of the pleura can be used as a tool to assess disease severity in patients with F-ILD.

Less investigated ultrasound (US) modalities within F-ILD are diaphragm excursion (DE) and lung ultrasound share wave elastography (LUSWE). Different methods have been used to evaluate DE . According to current evidence, IPF patients were found to have a lower DE compared to healthy individuals, and restrictive ventilation pattern with lower values of forced vital capacity (FVC) was also associated with lower DE . These findings may be explained by a combination of lung compliance reduction ("lung stiffness") and development of diaphragm muscle atrophy over time in patients presenting with F-ILD traits where further development of hypoxemia and immobilization in severe disease stages may develop to sarcopenia.

LUSWE is based on measuring the velocity of the US waves' propagation along the lung surface from a given surface point (intercostal space), where the velocity is dependent on the change in wave phase with distance. The velocity of the US waves increases with tissue stiffness as consequence of reduced elastic properties. In two observational studies distinct higher LUSWE velocities were found in ILD patients compared to healthy individuals, and furthermore a negative correlation between LUSWE velocities and FVC and interstitial lung abnormalities was observed . Therefore, LUSWE may be a valuable supplementary and clinical tool not only to identify F-ILD but also to separate F-ILD from non-fibrotic ILD.

As previously described patients with F-ILD will in addition to increased dyspnea over time, also develop weight loss due to sarcopenia. Sarcopenia is defined as loss of muscle strength, endurance (muscle quality), and muscle mass (muscle quantity), and part of the frailty syndrome defined by a combined loss of functional reserves across different domains (e.g. physical ability, cognition, and energy) resulting in negative implications on health status in multimorbid and especially older patients . The European Working Group on Sarcopenia in Older People 2 (EWGSOP-2) has established diagnostic consensus criteria in which the diagnosis of sarcopenia is made by assessing 1) physical function either by chair stand test or hand-grip strength, by hand dynamometer and 2) estimating the appendicular lean mass. The latter can be done by using B-mode US and anthropometric measurements the appendicular lean mass (ALM) can be estimated. The established thresholds for ALM for men and women have been defined (\<20 kg and \<7, 0 kg/m2 for men and \<15 kg and \< 5,5kg/m2 for women).

The main challenge in comparing study results on TUS/US in patients with F-ILD is the heterogeneity in the methodology with use of different TUS/US protocols in incomparable cohorts with a mixture of different ILD subtypes, where some subtypes present with a more fibrotic pattern than others. In this context, it is problematic to determine the diagnostic value of TUS/US for a specific F-ILD category. Some studies have investigated on TUS use in patients with IPF , but most of the studies on patients with F-ILD are of observational nature with cross-sectional designs and primarily performed in patients with underlying autoimmune rheumatic diseases . At present, no studies have accomplished real-life prospective TUS/US follow-up, including promptly investigated TUS/US modalities such as BLA, IS, and PA, and less investigated TUS/US modalities such as DE, LUSWE and ALM, and dealt with these TUS/US modalities as a monitoring tool for F-ILD disease activity or treatment response, and is thus an area encumbered with a significant knowledge gap. The latter was emphasized in the recent European Respiratory Society statement on TUS that warrants studies assessing the impact of these points as areas of future TUS/US research .

Hypotheses The investigators hypothesize that TUS/US modalities such as BLA, IS, PA, DE, LUSWE, and ALM are clinically valuable, both individually and in combination for identifying F-ILD and monitoring disease activity. These modalities are expected to correlate with FVC, diffusion capacity of the lung for carbon monoxide (DLCO), HRCT findings, and clinical symptoms. In such, F-ILD patients with severe disease are hypothesized to present with TUS/US findings as increased number of BLAs, increased pleura thickness with irregularity, decreased DE, higher LUSWE velocities, and lower ALM values. Furthermore, these parameters are proposed to serve as dynamic indicators of F-ILD disease activity over time.

Statistical analyses To assess the reproducibility of our sample the investigators conducted descriptive statistics. Categorical outcomes were presented as numbers and percentages and non-categorical outcomes were presented as either median and interquartile range or mean and standard deviation, depending on the normality of the variable.

The primary objective where to evaluate the predictive value of the longitudinal measured TUS/US modalities on the outcomes HRCT, Warrick score and its subscales, PFT, 6MWT and PROMs. Hence the investigators utilize a mixed effect model including a random intercept. The evaluation of the model predictive performance will be assessed by way of reliability (including intra and inter rater reliability) based on the intra class correlation coefficient and agreement evaluated by Bland-Altman plots with 95% limits of agreement. The discrepancy will be defined as the difference between the predictive and true value of the outcome based on the mixed effect model.

For sensitivity purposes the investigators will also assess the predictive of specific characteristics of HRCT by way of descriptive statistics.

Missing data will be assumed to be missing either completely at random or at random, and thus will handled implicitly by maximum likelihood estimation.

All analyses will be performed using either Stata SE 16.1 (StataCorp, College Station, TX, USA) or the statistical software R, 95% confidence intervals will be reported and a p-value less than 0.05 will be considered statistically significant.

Ethics The study will be conducted in accordance with the Declaration of Helsinki with written patient consent of included patients after prior oral and written project information. The patient can withdraw her/his statement of commitment at any time. The law on the processing of personally identifiable information will be complied with and data from the project will be stored in accordance with Odense University Hospital's guidelines. The project is approved by the Danish Data Protection Agency (Ref. no. 22/45135) and the Science Ethics Committee for the Region of Southern Denmark (Ref. no. S-20220036).

Patient recruitment Respiratory physicians from SCILS, Department of Respiratory Medicine at Odense University Hospital will recruit patients who prior to follow-up in SCILS have received recruitment material. Patients, who are willing to participate in LORD study will be contacted by the PhD student, where an agreement for information and obtainment of detailed oral and written consent will be planned. This meeting will take place in the outpatient clinic of SCILS under quiet circumstances, and the patient will be given 24 hours to consider participation. The patients who are willing to participate will be included after obtain their oral and written consent. Other patients, who are not willing to participate will follow up in the rutine SCILS clinic.

HRCT A standard HRCT scan with inspiratory and expiratory image sections on average results in the patient having a radiation dose of approx. 6-7 mSv (Sievert definition: SI unit for the dosage of ionizing radiation equal to 100 rems. 1 Sv = 1 Sievert = 100 rem = 1 J/kg ((this is based on calculations from The National Institute of Radiation Hygiene (Danish: "Statens Institut for Strålehygiejne")). It is theoretically calculated that such a radiation dose entails a risk of developing cancer of 0.01-0.1%, meaning that the total lifetime risk of developing a cancer disease increases from 25% (which applies to the entire population) to 25.1%. In this population of patients this risk is considered negligible.

TUS/US In TUS/US examinations, sound waves are converted into images on a TV screen. No side effects have been reported with the use of US. Apart from a slight pressure experienced when the ultrasound head touches the chest, the US examination is painless with minimal discomfort for the patient (37). As there are no immediate disadvantages or risks associated with TUS/US, the study only seems to contribute with advantages in form of clinically relevant information.

Sample size The endpoint used for sample size calculation is the ability of TUS performed at 12 months follow-up to detect disease progression using HRCT findings at 12 months as the golden standard. With an expected overall prevalence of the F-ILD phenotype of 14-32% among patients with ILD(38, 39), significance level of 0.05, power of 80%, an expected TUS sensitivity of 70-80%, and assuming the clinical use of US would be as a screening test prior to HRCT confirmation, the calculated sample size is estimated to be between 31 (sensitivity 70%) and 107 patients (sensitivity 80%) (40). Patients with F-ILD are associated with increased mortality, why some patients will not be alive at 12 months follow-up. A conservative estimate is that 10% of the patients will not complete or have died within the 1-year follow-up following inclusion. Adjusting for a 10% drop-out at 12 months follow-up will increase the total sample size to be between 34 and 117 patients with F-ILD. Since the study is performed as part of a PhD programme and 12 months follow up time is needed, the inclusion period necessities to be restricted to 12 months to ensure feasibility within the three year time frame of the PhD programme. Based on the sample-size calculations and the inclusion period restrictions the study will aim to include up to 117 patients or as many as possible with at least 34 patients wi

• Measure the association and variability and observer variabilty. [Срок оценки: 12 months. The investigator group will examine patients in baseline (Day 0), after 6 months (day 180) and after 12 months (day 365).]