Severe tracheal and bronchial collapse in adults with type II mucopolysaccharidosis

PFTs were obtained in all patients as part of routine follow-up. Spirometry was performed in accordance with American Thoracic Society standards [10]. Values of percent-predicted for spirometry were calculated using reference values based on age, height, sex and race [11].

The basic parameters used to properly interpret lung function were the vital capacity (VC), forced expiratory volume in 1s (FEV1), FEV1/VC ratio and total lung capacity (TLC) [11]. The carbon monoxide diffusing capacity (DLCO) and adjusted DLCO for the measured lung volume (DLCO/VA) assisted to diagnose the underlying disease, when interpreted in conjunction with the lung volumes assessment. Airway obstruction was defined by reduction of the FEV1/VC ratio (also known as Tiffeneau- index) below 70 %. A Tiffenau-index below 70 % depicts a disproportionate reduction of maximal airflow from the lung in relation to the maximal expiratory volume (VC), implying airway narrowing during exhalation. The severity of lung function impairment was based on the forced expiratory volume in 1s as a percentage of the normal predicted value, corrected for sex, age, height and race (FEV1%pred). A restrictive ventilatory defect is characterized by a reduction in total lung capacity (TLC) below 80 % of the predicted value. TLC was measured by Helium [He] dilution (TLC-He) in all patients and additionally by whole-body plethysmography (TLC-Pleth) in a single patient. In body plethysmography, breathing manoeuvres are performed in an airtight box of known volume with fixed barometric pressure. The two techniques for measuring TLC yield similar results for those with normal lung mechanical function but can yield differential results in patients with airflow obstruction. Whole-body plethysmography is more accurate because of the detection of poorly or non-ventilated lung areas, also known as trapped air. A higher TLC measured by bodyplethysmography compared to that measured by Helium dilution is therefore indicative of air trapping [12].

We performed in- and expiratory chest CT scans in all patients. Chest CT was done on different multi-detectors CT scanners, with detectors ranging from 64 (SOMATON Definition AS+, SIEMENS Healthcare, Forchheim, Germany) to 192 (SOMATOM Force, SIEMENS Healthcare, Forchheim, Germany) detectors. CT protocol included an end-inspiratory and -expiratory scan starting from 2 cm above the lung apices to the bases. Instructions for voluntary breath holds were given before scanning by the CT technician. Scanning parameters were: collimation 0.625 mm, 50 % overlap, slice thickness 1 mm, pitch 1, 110 kV tube voltage and tube current adapted to recommended volumetric computer tomography dose index (CTDI_vol), which is 1.6 mGy per adult subject. For expiratory CTs, image quality was considered sufficient using a tube current fixed at 25 mA with an effective tube current-time product of 10 mAs; producing a lower radiation dose than the inspiratory protocol. No contrast agent was administered. Images were reconstructed with a 2 mm slice thickness as multiplanar reformats (MPR) in the coronal and sagittal planes. Sharp reconstruction kernel filter without under- or overshoot at edges were used for central airway measurements.

Images were transferred to the Myrian dicom viewer (Intrasense, Montpellier, France). Cross-sectional area (CSA) was measured using a manual tracing tool perpendicular to the tracheal midline. To obtain images perpendicular to the midline, double-oblique reformats were obtained by the axial MPR with sharp kernel (i.e. B50f, I70f, B75f) [13] (Ciet et al. 2015). Measurements of CSA were obtained at five levels throughout the central airways: 1 cm above the aortic arch, at the level of the aortic arch, 2 cm above the carina, 1 cm below carina in the main left and right bronchi (Fig. 1a). CSA measurements were performed with constant lung window (WL -600; WW 1200) both in the end-inspiratory and at end-expiratory images by a single radiologist with extensive experience in thoracic radiology. The percentage airway collapse was calculated according the formula:

Where Δ % is the percentage collapse, and CSAins and CSAexp are the CSA in the end-inspiratory and expiratory images respectively.

Five adult MPSII patients were enrolled in the study (mean age 40 years, age range 29–50). All patients were male and had normal cognitive function. Four out of five patients were diagnosed during childhood (age range 3 to 9 years), one patient was diagnosed at adult age (age 45 years). In the latter patient the disease was the least severe and only at age 45 the cluster of ENT problems, recurrent pulmonary infections, cardiac valve pathology and skeletal problems were recognized as being caused by MPS II. In all five patients, enzyme replacement therapy (ERT) was started and all but one received ERT at the time of the pulmonary evaluation. Duration of ERT at time of evaluation ranged from 3 months to 7 years. The single untreated patient discontinued ERT 4 years prior to the evaluation because of recurrent anaphylactic reactions during enzyme infusion. At the time, he had been treated for 2 years. Patient characteristics are summarized in Table 1. Three patients received nightly ventilatory support, in one this was bilevel positive airway pressure (BiPAP) and in the other two continuous positive airway pressure (CPAP) (Table 2).

Table 2 shows the results of the PFT measurements, performed at a time point closest to the airway imaging date (median time gap 14 days). In patients 1, 2, 4 and 5 the lung function analysis showed airway obstruction (FEV1/VC <0.7). In three patients this was accompanied by restriction as the total lung capacity (TLC) measured was below 80 % of predicted. Thus, mixed obstructive and restrictive airway disease was present in these patients. Patient 1 showed only obstructive airway disease. Patient 3 appeared to have only restrictive airway disease but the marked difference between TLC measured by body plethysmography (60 %) and TLC-He (49 %) was suggestive of trapped air, which is usually seen in the context of airway obstruction. The concave shape of the expiratory part of the flow volume loop was indeed indicative of airflow obstruction (Additional file 1: Figure S1). Thus, patient three also had obstructive airway disease.

In patient 4 the airway impairment was very severe (FEV1%pred <35). In patients 1, 3 and 5 the airway impairment was severe (FEV1%pred <50). Patient 2 had moderate airway impairment (FEV1%pred 61). This was the patient diagnosed at adulthood, with an overall mild MPSII phenotype.

Diffusing capacity was normal in all patients. The low DLCO but normal DLCO/VA excludes parenchymal disease and indicates extraparenchymal abnormalities. In MPSII these extraparenchymal abnormalities are the anomalous thorax shape and mechanics caused by thoracic spine and rib deformities [4].

All patients successfully completed the CT protocol. In the inspiratory imaging, all patients had diffusely narrowed and thickened trachea and main bronchi. The trachea showed an abnormal triangular shape with flattening of the cartilaginous portion and enlargement of the pars membranacea (Fig. 1b). As a result, the CSA was already significantly reduced in the end-inspiratory images.

Severe expiratory collapse of the trachea was seen in all patients. Table 3 described the central airways measurements at five different locations at end-inspiration and -expiration and the percentage collapse for each patient. Median percentage collapse at different levels of the airway were: 68 % at 1 cm above the aortic arch), 64 % at the level of the aortic arch, 58 % 2 cm above the carina. Main bronchi collapse measured at 1 cm below carina was 58 % for the left main bronchus and 44 % for the right main bronchus (Fig. 1b-e). Because of the already diffusely narrowed central airway lumen in inspiration, the remaining airway lumen at end expiration is much smaller than if the collapse would have occurred in a normally shaped trachea. In the most severely affected patient, there was an almost complete closure of the trachea during expiration. The expiratory chest scan also showed the presence of trapped air in all but one patient (Fig. 1f).