This was a retrospective observational study that analyzed the results of lung cryobiopsies performed in the Respiratory Endoscopy Unit of a tertiary hospital between January 2011 and January 2014. Patients had been previously assessed by an ILD multidisciplinary committee, where all cases referred from the different hospital departments, outpatients, and district primary care clinics were evaluated. Clinical decisions were taken by consensus among the members of the multidisciplinary team, which was composed of pulmonologists, thoracic surgeons, radiologists, and specialists in systemic and pathological autoimmune diseases.

Patients with an ILD pattern on HRCT defined as “possible UIP” or “inconsistent with UIP”, in whom histological confirmation would be required (in accordance with ATS/ERS guidelines1) were included. The specific lobe where the cryobiopsy was performed was determined following the radiologists’ recommendations, based on ATS/ERS1 criteria for surgical lung biopsy. In summary, lung biopsy should cover a wide spectrum of lung disease, including honeycomb foci, as this finding is a criterion for UIP. However, if there are signs of very severe fibrosis with gross honeycombing, the biopsy should be taken from less affected honeycomb areas, as non-specific changes are usually found in the most affected areas. If the lung does not show any signs of fibrosis or honeycombing, the sample should be taken from radiologically abnormal areas. Following this protocol, biopsies were taken in all cases from different sub-segments of the target lobe and, when possible, from various lobes. All patients had a complete blood count with coagulation study, echocardiography and lung function study that included forced spirometry, determination of lung volumes and carbon monoxide diffusing capacity. SLB was proposed in patients in whom the cryobiopsies were not diagnostic. All patients signed a specific informed consent for cryobiopsies.

The cryobiopsies were performed in an examination room equipped with hemodynamic monitoring equipment, a ventilator and mechanical radioscopy arm. The procedure was carried out under general anesthetic with propofol (3–6mg/kg/h) and remifentanil (0.05–0.1mg/kg/h) infusion. Patients were intubated with a flexible, cuffed tube (7.5mm Bronchoflex, Rüsch, Teleflex Medical, Durham, NC, US). This tube has a side channel that allows the introduction of a Fogarty™ occlusion balloon catheter (4F/5F/6F; Edwards Lifesciences Corporation™, Irvine, CA, US), which was placed at the entry of the lobe to be biopsied and inflated immediately after performing the cryobiopsy to stem any possible bleeding from the target airway. An EB-1970K video bronchoscope (outer diameter: 6.2mm, diameter of working channel: 2.8mm; Pentax Medical, US) was used. After examining the bronchial tree, bronchoalveolar lavage was carried out in the area selected by HRCT imaging in patients in whom it had not previously been performed. TBB was then performed on the selected segments using a 1.9mm cryoprobe (Erbokryo CA, Erbe, Germany) under fluoroscopic guidance. The cryoprobe was introduced through the working channel of the bronchoscope and, once the correct position had been confirmed by fluoroscopy, the cryoprobe pedal was activated for 3–4s. The probe was then withdrawn, together with the bronchoscope, with the specimen adhered to the tip of the cryoprobe. At least 3 samples were taken, although the total number of biopsies depended on their size, the patient's tolerance and possible complications. After the procedure had been completed and the patient ex-tubated, a post-procedure chest X-ray was performed. Any complications observed were recorded, particularly bleeding and pneumothorax. Pulmonary bleeding was classified according to its severity as: grade I: mild bleeding which does not require endoscopic intervention; grade II: moderate bleeding which stops within 3min after endoscopic intervention (bronchial occlusion and/or instillation of cold serum), and grade III: severe bleeding that cannot be controlled endoscopically, causing hemodynamic or respiratory instability, and making it necessary to halt the procedure.20

The specimens were fixed in 4% formaldehyde and sent to the histopathology laboratory, where they were measured before being embedded in paraffin. Each sample was embedded in a separate block and stained with hematoxylin–eosin and Masson's trichrome stain. The specimens were evaluated by a pathologist from the multidisciplinary team with experience in respiratory disease pathology. The sample was not considered diagnostic when the findings did not allow a specific diagnosis to be made. Samples with insufficient alveoli for analysis were considered invalid.

The clinical, radiological, and histopathological parameters were reviewed by specialists who regularly participated in our hospital's ILD multidisciplinary committee. A sample was considered to be diagnostic when the histopathological findings enabled the specific diagnosis of the particular disease. Once the histopathological report had been obtained, the members of the multidisciplinary committee established the definitive diagnosis of the patient's interstitial disease based on the report and the patient's clinical and radiological characteristics.

For the economic analysis, the cost of the different diagnostic processes was calculated in euros. We used data based on the Catalonian Government taxes and public prices Act approved by Legislative Decree 33/2008 of 25 June and Act 26/2010 of 3 August, and data reported by other authors in the same setting.21,22 The final cost per process was calculated by totaling the cost of the different procedures required to reach the definitive diagnosis.

The results of the quantitative variables were expressed as mean±standard deviation. Percentages and absolute frequencies were used for the qualitative variables. Statistical data analysis was performed using the SPSS statistics package version 17.0 for Windows® (SPSS, Chicago, IL, US).

During the study period, 33 patients underwent lung cryobiopsy for confirmation of ILD. Their general characteristics are summarized in Table 1. The location of the biopsies, number of samples obtained, and complications observed are summarized in Table 2. The bronchoalveolar lavages were performed in the middle lobe or lingula to avoid technical difficulties in performing the cryobiopsy. More than one lung lobe was biopsied in 2 patients. In the remaining 31 patients, the biopsies were carried out in various segments of the lower lobes. A mean of 2.7 biopsies were obtained per patient, with a mean size of 0.4cm (SD±0.17). Biopsy specimens considered to be diagnostic were obtained in 26 patients (79%) (Fig. 1); the samples were considered non-diagnostic in 5 patients (15%) and invalid in only 2 patients (6%). There were no significant differences in the size of the diagnostic and non-diagnostic specimens (0.5±0.19cm and 0.34±0.15cm, respectively). In general, no artifacts due to compression of the alveoli were found in the samples analyzed. Fig. 2 summarizes the diagnostic orientation based on the HRCT images. The definitive diagnoses were obtained from the histopathology of the cryobiopsy samples and the opinion of the multidisciplinary team. Of the 5 patients with non-diagnostic specimens, a diagnostic SLB (peribronchiolar metaplasia) was performed in only 1 case (3%). Of the remaining 4 patients, 2 cases were diagnosed with idiopathic NSIP, mainly based on radiological criteria. Given that these patients had severe comorbidities that contraindicated the intervention, SLB was dismissed by the multidisciplinary team. The other 2 cases had radiological images, symptoms, and histopathological findings suggestive of ILD, but a more accurate diagnosis could not be reached, so they were diagnosed with ILD of unknown origin. Of the 2 patients with invalid samples, one had radiological images suggestive of organized pneumonia. Since SLB was also contraindicated due to severe comorbidities, the patient was treated with oral corticosteroids, with a good clinical and radiological response. In the other case, the radiological image and symptoms suggestive of ILD were mainly taken into account for the final diagnosis, suggesting ILD of unknown origin. In 1 patient, despite images characteristic of UIP on the HRCT, a cryobiopsy was performed due to clinical suspicion of hypersensitivity pneumonitis. Cryobiopsies were performed in patients with ILD associated with a systemic autoimmune disease (SAD) due to the lack of criteria for a firm diagnosis of SAD. These patients had lung disease at the outset and later presented extra-pulmonary symptoms, thus fulfilling sufficient criteria for a diagnosis of systemic sclerosis. As regards the safety of the technique, 4 cases of pneumothorax (12%) were identified; 3 resolved spontaneously while 1 required a chest drain for 2 days, after which the patient was discharged. In total, 9% of patients had grade I bleeding, and 21% had grade II bleeding. No cases of severe bleeding requiring suspension of the procedure were detected, and no additional medical or surgical measures were needed to control bleeding. Although 2 patients had mild pulmonary hypertension as comorbidity, neither of them had complications.

Fig. 1.

Optical microscopy image of a lung biopsy histological section, showing: (A) good conservation of the alveolar structures of the specimen, as well as the presence of predominantly lymphocytic inflammatory infiltrate located in the interstitium (hematoxylin-eosin, 40×). (B) The immunohistochemical study shows the presence of T lymphocytes (CD3 labeling; 100×) and (C) B lymphocytes (CD20 labeling, 100×). The patient was diagnosed with lymphocytic interstitial pneumonia, with no evidence of lymphoma.

(1.48MB).

Fig. 2.

Final diagnoses made by the multidisciplinary committee based on the radiological images, classified according to ATS/ERS guidelines for idiopathic pulmonary fibrosis into usual interstitial pneumonia (UIP), possible UIP and inconsistent with UIP, and histopathological results of the cryobiopsy specimens.

Hist, histopathology; MCD, multidisciplinary committee diagnosis; SAD, systemic autoimmune disease; ILD, interstitial lung disease; IPF, idiopathic pulmonary fibrosis; LAM, lymphangioleiomyomatosis; HP, hypersensitivity pneumonitis; LIP, lymphocytic interstitial pneumonia; NSIP, nonspecific interstitial pneumonia; UIP, usual interstitial pneumonia; OP, organized pneumonia; HRCT, high-resolution computed tomography. *Based on radiological images, the suspected diagnosis was OP. **The final diagnosis of peribronchiolar metaplasia was made by SLB.

(0.27MB).

Following the method used by Leiro Fernandez et al.22 and establishing the cost of FBS at €250.45, cryoprobe at €54 per case, and histopathology analysis at €253.69, the total cost of the cryobiopsies performed was [(€250.45+€54+€253.69)×33]=€18,426.54, giving a total cost of €304.69 per patient. In 26 of the 33 patients evaluated, the cryobiopsy provided the definitive diagnosis of the disease and SLB was not needed. The costs of this surgical technique in both the hospital and outpatient setting are summarized in Table 3. In the case of the 3 patients who presented pneumothorax with spontaneous resolution, the added cost was €1458.30 (€486.10 per patient). In the case of the patient who required a drainage tube to resolve the pneumothorax, the added cost was €2229.98. Therefore, the systematic use of lung cryobiopsy in the assessment of patients with suspected ILD reduced costs by €41,506.74−(€10,054.77+€3688.28)=€27,763.69 in the case of SLB without admission; €57,548.04−(€10,054.77+€3688.28)=€43,804.99 in the case of 24-h hospital stay, and €73,589.34−(€10,054.77+€3688.28)=€59,846.29 if hospital stay was extended to 48h.