Trial design and participants
A four-year spontaneous/non-profit single-center prospective observational design has been conducted from December, 2015, up to January, 2020. The study was approved by the Ethics Committee (EC) of the IRCCS Fondazione Ca' Granda (EC approval number: 55_2018bis).
The primary goal of the study was to evaluate the effectiveness of LCI, measured by nitrogen MBW, as early marker of lung damage in patients with PADs, even in the absence of bronchiectasis. Secondarily the study aimed to evaluate the overtime trend of LCI in patients with PADs during follow-up of respiratory complications, verifying the usefulness of MBW in monitoring the respiratory function and in the early highlighting of the onset of pulmonary complications.
Inclusion criteria in the study were: patients suffering from PADs diagnosed according to the European Society for Immunodeficiencies (ESID) criteria [26]; patients in follow-up of the respiratory complications according to the AIEOP/IPINet recommendations [13]; availability of chest HRCT performed in well-being at least once in the last four years; age > five years; compliance with the study protocol and ability in performing Pulmonary Function Tests (PFTs) according to the criteria of reliability. Exclusion criteria were considered age < five years; patients not cooperating and/or not able to carry out the test correctly.
Seventeen patients (10 males and 7 females, age range between 5.7 and 14.8 years) with PADs were involved in the study: 11 (6 males and 5 females) with Common Variable Immunodeficiency (CVID), 2 males with X-linked Agammaglobulinemia (XLA) and 1 male with Autosomal Recessive Agammaglobulinemia (ARA), 2 female patients with Selective IgA Deficiency (SIgAD) with respiratory complications and 1 male with Hyper IgM Syndrome (HIGM). 12 out of the 17 patients (8 CVID, 2 XLA, 1 ARA and 1 HIGM) were undergoing Ig substitutive therapy at the time of the study, all with adequate pre-infusion IgG levels (median 915 mg/dl; IQR 872-964). All patients were also following a personalized respiratory physiotherapy program. The median age of symptoms onset in all patients involved was 1.5 years (IQR 1.0-3.3). The median age at diagnosis was 6.6 years (IQR 4.3-9.8). The median diagnosis delay was 3.2 years (IQR 1.3-6.7). The patients median age at baseline was 10.7 years (IQR 8.5-12.6). The median duration of follow-up for this study was 4.0 years (IQR 3.2-4.0) (Table 1).
Intervention
According to the study design, MBW with LCI determination has been included in the standard follow-up of patients with PADs in order to monitor the progression of respiratory complications. At the time of enrollment each patient has been subjected to a global baseline assessment of lung involvement by performing spirometry, nitrogen MBW and chest HRCT. For 13 out of the 17 patients with stable conditions the Pulmonary Function Tests (spirometry and MBW), performed annually, have been compared with chest HRCT, performed every four years. For the other 4 patients, enrolled later, the follow-up was too limited to check repeated chest HRCT and comparison has been possible only at baseline time.
HRCT scan studies were scored adopting the Bhalla score [27] in a random order by a pediatric radiologist and a pediatric pneumologist, blinded to patients’ identities and clinical information, respectively. Readers evaluated each examination independently and reporting in consensus [28,29].
The Bhalla score [27] is a morphological scoring system usually applied to monitor pulmonary disease progression and stage of chronic broncho-pneumopathies. Bhalla Score can vary between 0 and 25 and includes presence, severity and extent of different pulmonary alterations, including bronchiectasis, peri-bronchial thickening, mucous plugging, sacculation/abscesses, bullae, emphysema and collapse/consolidation [12,30].
Each MBW test consists of a wash-in and a wash-out phase. This respiratory test is feasible in two ways using an extrinsic inert tracer gas, inhaled until equilibrium is reached and then eliminated by breathing ambient air, or by inhaling 100% oxygen for the wash-out of nitrogen, an intrinsic inert gas normally present in the airways, respectively. In our survey 100% oxygen was used for the nitrogen wash-out in order to increase sensitivity in identifying abnormalities being nitrogen present in all areas of the lung [20,21].
Parameters derived from MBW are obtained on the analysis of the breath-by-breath concentrations and volumes of gas. The LCI may be considered the most reliable MBW parameter since it reflects the inhomogeneity of pulmonary ventilation and is defined as the number of lung turnovers required to washout the tracer gas up to 2.5% of its initial concentration. In a healthy person, this takes about 5-7 turnovers. According to a 2011 review by Fuchs and Gappa [15], which reports the most relevant studies conducted to define LCI normal values and Upper Limit of Normal (ULN) in pediatric age, LCI is constant during childhood and independent of age, weight, height and gender during adolescence. The value of 7.00 is reported as ULN for LCI in childhood. In case of ventilatory inhomogeneity, LCI increases with the number of turn-over needed for the tracer gas to be eliminated from the lungs [15,20,22,31].
Statistical methods
The median age of the 17 patients at the time of baseline HRCT was 10.7 years (IQR 8.5-12.6), while the median age at the HRCT control after four years was 14.8 years (IQR 13.7-16.5) (Table 1).
Parameters of the respiratory function tests considered for the study were FEV1 as a percentage of the predicted value (FEV1% mis/ T) and LCI at 2.5% of the gas initial concentration (LCI2.5%) calculated with nitrogen MBW, respectively.
Regarding the cross-sectional evaluation, LCI and FEV1 parameters were compared, taking the HRCT as gold standard, evaluated by Bhalla score. Given the non-normal and asymmetrical distribution of the Bhalla score, in order to better analyze the data, HRCT was transformed into a dichotomous variable and statistical analyzes were accordingly performed by splitting the patients in two groups. Accordingly, HRCT has been considered normal (N-CT) for patients with Bhalla score = 0, while it has been considered altered (A-CT) for patients with Bhalla score > 0. Fisher's exact test has been employed to test the dependency on gender, while age, FEV1 and LCI were compared using the Wilcoxon-Mann-Whitney test. Univariate and multivariate (adjusting for gender and age) logistic regression analyses were used to study the association between FEV1, LCI and N-CT/A-CT.