Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) has become widely accepted as one of the various methods currently in existence for obtaining mediastinal samples in a range of pathological states.1–7 It is regarded as a minimally invasive technique and is not associated with serious complication.3,5,7 EBUS-TBNA is particularly useful in the mediastinal staging of non-microcytic lung-cancer (NMLC), for which its sensitivity has been estimated at between 89% and 98%, and its specificity 100%, with a negative predictive value (NPV) in excess of 92%.2,8–13 The EBUS-TBNA has proven useful in diagnosing malignancy in lymph nodes, with a sensitivity of 91% and a NPV of 93%.14 Benign tumours, especially sarcoidosis, can also be easily diagnosed by means of EBUS-TBNA.15–17

The success of EBUS-TBNA depends on the adequate training of the endoscopy team, the availability of an experienced cytopathologist to carry out a rapid evaluation (ROSE), and the use of an appropriate technique during the procedure.18–20 The conventional sampling technique for EBUS-TBNA is based on the application of negative pressure or aspiration with a syringe.20 It has been suggested that continuous aspiration optimises the amount of tissue obtained for diagnostic purposes, and keeps the needle in the target lesion.21,22 All the studies presented so far are based on EBUS-TBNA as the preferable procedure for obtaining samples.

Biffod was the first researcher to describe an alternative sampling technique, with the use of a needle and without the application of external pressure, for breast lesions.23 This simplified sampling technique is based on the physical properties of the liquids, which allows a smooth extraction of the material with no need for aspiration. The technique, which is known as fine needle capillary sampling or FNC has also been tried on other organs24–26 and on lymph nodes. FNC can provide cytology samples of a higher quality, and it may be associated with a lesser degree of contamination of the sample due to haemorrhage than is the case with TBNA. In different studies, it has been described as the technique of choice for sampling in endoscopic ultrasound-guided (EBUS). EBUS-FNC is a simplified technique which obviates the need to withdraw the stylet or to aspirate with a syringe. The aims of the present study are to describe this technique and determine its performance capacity in a random group of patients on whom EBUS mediastinal examination is carried out.

The study included a total of 44 consecutive patients, on whom a EBUS examination of the mediastinum was performed in the Clinic of the University of Navarre in Pamplona, Spain, between 1 January and 31 August 2011. All the patients concerned had been sent for EBUS following the results obtained in image diagnoses in which mediastinal adenopathies were identified or because they were eligible for mediastinal staging, due to the presence of lung cancer. A medical evaluation was carried out by an endoscopist on all the participants in the scheme, with the aim of recording the relevant demographic data, confirming the advisability of such an intervention, obtaining the informed consent of the patients, and excluding any other possible contraindication. In the same way, the researchers reviewed the existing diagnostic images (CT and or PET-TC) before the intervention, with the aim of identifying the presence of mediastinal lesions that could be aspirated, and planning the procedures to be applied.

The study included 44 patients (75% male) with an average age of 62.2 (±9.8) years. The characteristics of the patient are shown in Table 1.

All patients presented mediasinal lymph node damage, visible in the CT or PET-CT scan images. In total, the clinical suspicion of malignant mediasinal damage was 82% (36/44) prior to the EBUS, in a population in which the incidence of malignant disease was 31.8% (14/44). The most prevalent malignancies, in decreasing order of frequency, were in renal neoplasms, cancer of the urinary bladder and of the lung. 18% of the patients presented mediastinal disease whose aetiology was uncertain. The endobronchial ultrasound-guided examination of the mediastinum revealed lymph nodes or mediastinal mass with a diameter of more than 5mm in 38 patients (86.4%). Four patients presented an apparently normal mediastinum, despite indications to the contrary in the CT scan, 1 patient had a lymph node that was too small for aspiration and 1 patient had a lesion which could not be clearly identified for technical reasons, and from which samples could therefore not be obtained. Samples were obtained from more than one lymph node in 23 patients (52.3%). The lymph nodes from which most samples were obtained are shown in Table 2, with a total of 192 FNC passes and an average of 5 passes per patient.

In those cases in which aspiration was not carried out on any of the lymph glands, the sample was obtained by means of conventional bronchoscopy (bronchial biopsy, TBNA, broncial brushing and/or bronchoalveolar lavage) in the same intervention.

In only 2 patients (4.5%) were any complications recorded, including a light to moderate haemorrhage at the location of the aspiration. The bleeding could be brought under control locally with an ice-cold saline solution and aspiration, and neither further invasive measure nor intubation was required. In one of the patients with bleeding, uncontrolled hypertension occurred, which was attributable to failure to comply with the habitual anti-hypertensive treatment prior to the intervention. No more sever complications, and no deaths, occurred during the study period.

The analysis of the samples by the Cytopathology Department confirmed that all the aspirations carried out on lymph glands using the capillary technique provided sufficient representative material for an interpretation (Table 3), with a diagnostic result of 86.8%. The diagnostic sensitivity of the EBUS-FNC was 88% for interventions which produced sufficient samples, and 84.1% for the total number of interventions.

The final diagnosis was lung cancer in 61.4% (27/44) of the patients, with a greater incidence in men (81.5%), and in patients with tobacco-related antecedents (89.9%). Other diagnoses were metastasis in the lymph nodes, renal, skin and breast neoplasm, nontuberculosis mediastinal granulotamous lymphadenitis (2 cases), plasmocytoid cell tumour (1 case) and reactive lymphadenopathy with an obvious focus of infection (2 cases).

In 4 of the 7 cases in which EBUS-FNC did not permit a definitive diagnosis, the procedure applied to establish diagnosis was correlation with clinical antecedents, diagnostic image examination and reports from the Cytopathology Department in relation to the samples.

The diagnostic performance of the EBUS-TBNA depends on a number of factors, including the following: the availability of recent diagnostic images for an adequate planning of the intervention, adequate experience on the part of the staff involved, the availability of ROSE, obtaining an adequate representative sample from the lymph gland.18–20,35–38 The conventional technique for obtaining samples using EBUS or TBNA requires the application of negative pressure.20 The sample is obtained primarily by means of aspiration. It has been said that this method reduces the presence of inadequate samples by 5% or more.22,38 Concerns about the influence of FNA on the quality of the sample were first voiced by Zajdela et al.24 This author described contamination of the sample resulting from bleeding attribute to the wounding of the neighbouring tissue, which complicated the cytopathological evaluation of mammary samples obtained by means of FNA. FNC was recommended as an alternative means of obtaining samples. This method was introduced by Bradford in 1982 in a study of mammary carcinoma samples. This key study highlighted the fact that there is a good correlation between cytological samples obtained by means of FNC and histology.23

FNC is based on capillarity, a physical property of liquid and semi-liquid media which permits the surface of a liquid to flow spontaneously towards the interior of a capillary tube. This effect is the result of the cohesion forces, known as surface tension, which are present between the molecules of the liquid. The analysis of the liquid is directly proportional to the surface tension, and inversely proportional to the calibre of the tube.39 FNC can be useful as a means of obtaining lymph node samples, as the cells are not bonded in a compact ultrastructure, which means that material can rise through the interior of the needle by capillarity as the stylet is withdrawn.25 FNC can provide cellular material and its diagnostic usefulness is comparable to that of conventional FNA, according to data obtained in patients with breast cancer in the Curie Institute in France.24 In comparative studies carried out on thyroid glands and the liver, a distortion of the cellular architecture of the samples obtained by conventional FNA has been described,25,26 although no significant differences were observed in the diagnostic usefulness of FNC as opposed to FNA (sensitivity of 90% compared with 80%–90%).25,40 FNC can be considered superior to FNA as contamination resulting from blood is minimal, and the cellular morphology is better preserved.41,42 This serves to explain the greater diagnostic usefulness compared with FNA.24,26,41–43

The question of whether FNC is more advantageous than conventional FNA in EUS remains to be answered. Wallace and Storch did not observe any significant differences in terms of the quality or diagnostic usefulness of the samples obtained by means of EUS-FNA compared with FNC.28,29 In Wallace's study, FNA was more likely to produce sanguineous samples, but it also produced more cellular samples than FNC.28 The authors arrived at the conclusion that FNC results in a greater overall sample quality as less blood is present, although there were no differences in terms of the diagnostic usefulness. Puri et al., by contrast, described a greater sensitivity and negative predictive value in samples obtained by means of EUS-FNA than in those obtained by EUS-FNC (85.7% as opposed to 66.7%).30

One of the advantages of EUS-FNC compared with EUS-FNA could be the need for fewer needle passes. It has been pointed out that the optimal number of needle passes with conventional FNA is 5 for lymph glands and 7 for solid tumours, with an overall sensitivity of 77% and 83%, respectively, and a specificity of 100%.44,45 Wallace, by contrast, suggests that 3 is the optimal necessary number for a diagnosis based on samples obtained by means of EUS-FNC.28 Given that each needle pass takes between 3 and 5min from beginning to end, fewer passes would substantially reduce the length of time required for the intervention and therefore the length of the sedation required, which could result in a smaller number of possible EUS complications.

Our study supports the use of FNC as a viable alternative to TBNA in the samples obtained by means of EBUS. Comparisons with historical data published so far suggest that the techniques are, to say the least, comparable, as regards their effectiveness in obtaining adequate samples and as regards diagnostic accuracy, given that 77%–100% of the samples obtained with EBUS-TBNA form, are adequate and/or diagnostic.13,27,46 It is possible to argue that FNC is easier to as it avoids the need for needle aspiration, which simplifies the process of obtaining lymph gland samples in the mediastinum. There is a consensus among practitioners with experience in EBUS and conventional TBNA that EBUS-TBNA is a laborious technique which often requires at least two specialists in order to obtain a single sample, although there are some bronchoscopists who are used to applying the aspiration while they hold the needle and the bronchoscope with one hand. Simplifying the technique could be useful in EBUS examination of the mediastinum in an overburdened working environment in which the endoscopist has to deal with sedation, bronchoscopy, monitoring of the patient and the obtaining of tissue all at the same time.

To the best of our knowledge, this is the first study which systematically focuses on the obtaining of mediastinal samples with EBUS using FNC as opposed to conventional TBNA. One of the limitations of our study is the lack of comparisons between EBUS-FNC with conventional EBUS-TBNA. In this respect it should be seen as a hypothesis-generating study and proof of concept. Another serious limitation is the prevailing uncertainty regarding what constitutes an adequate sample. It should be pointed out that there are no established criteria in this regard. Consequently, the evaluation of a given sample as adequate or inadequate can be considered subjective; in that it depends on the criteria of the specialist, and may not be reproducible. By and large, little coincidence among observers appears to be the order of the day.31

Our study indicates that EBUS-FNC is a viable alternative to conventional EBUS-TBNA for obtaining mediastinal samples. The modified technique is safe and comparable, in terms of its effectiveness and its success in obtaining adequate samples, to everything that has been published in relation to the performance of EBUS-TBNA. It is arguable that EBUS-FNC is simpler than EBUS-TBNA and can be better in the sense that it can reduce the amount of time required for obtaining samples, and in terms of the complexity of the procedure. Whether EBUS-FNC provides better quality samples remains to be demonstrated by means of a prospective and randomised study.

The authors have no conflict of interest to declare.