FNAB technique
	FNAB is preferably carried out in an outpatient setting. The patients is informed of the limitations of the diagnostic procedure (possibility of inadequate samples or “indeterminate” diagnoses) and reassured that the procedure is almost painless and free of major complications. The patient is then invited to assume a recumbent position with a pillow under the neck to relax the neck muscles. Thyroid nodules are identified by palpation and the patient is advised to swallow once or twice before inserting the needle but not during the procedure. Most authors use a 23-gauge needle attached to a 20-ml plastic syringe hold with a special pistol grip (Cameco pistol). The needle is then inserted into the nodule and moved back and forth while maintaining a slight aspiration. The aspiration is released when the aspirated material appears in the needle hub and just before withdrawing the needle, in order to avoid contaminating the sample with material coming from the needle tract. FNAB has no major complications. Subcutaneous hematoma, limited intranodular bleeding or slight local pain occasionally occur. Rarely, aspiration of a cystic lesion may favor acute inflammation. Isolated reports exist of acute thyroid swelling, transient bradycardia and transient paralysis of the vocal cord. Variables that may affect adequacy and quality of cytologic smears A number of variables may affect the yield of cellular material and the quality of cytological smears: Use of real time echography. The use of ultrasounds (US) to assist or guide FNAB (US-FNAB) has been shown to be helpful in increasing cellular yield while avoiding excessive blood or cystic fluid aspiration and increasing diagnostic accuracy19-21. US-FNAB is used by more than 40 % European authors but only infrequently in the States. Although the use of US assistance is unnecessary for palpable nodules, it is precious for localizing difficult-to-palpate nodules and identifying solid parenchymatous areas in partially cystic or partially calcified nodules. The higher cost of US-FNAB is counterbalanced by the reduction of inadequate samples and the increase of diagnostic accuracy19. Clinical examination. The use of US should not lead to neglect palpation, which allows evaluation of nodule mobility and consistency. Small, mobile nodules should be carefully fixed manually whether or not US assistance is used. Furthermore, assessing nodule consistency is very useful to graduate the suction applied to the syringe and the number of needle movements. In most cases it is preferable to apply a slight suction or no suction at all. This technique, “non aspiration FNAB”, will reduce blood aspiration while still yielding a cellular sample and works well except for hard- consistency/fibrotic nodules22. Choice of needle and syringe. A 23 gauge (0.6 mm) needle is mostly used. The use of thinner needles (27 gauge, 0.4 mm) may limit the aspiration of peripheral blood that can dilute the cellular yield. The use of larger needles (21 gauge, 0.7 mm) may result in a more abundant cellular yield but also in excessive peripheral blood aspiration and should be reserved in selected cases (e.g. presence of very viscous colloid, fibrotic nodules). A 20ml syringe is used when employing Cameco pistol. This has the advantage to allow the operator to fix the nodule with one hand. When examining small nodules, especially with US assistance, it may be easier to use a 5ml syringe or a needle without a syringe. Number of punctures per nodule. The structure of a thyroid nodule is often heterogeneous. In order to examine different areas of the same nodule and to increase the cellular yield, each nodule should be subjected to more than one aspiration. Some authors have found that the number of aspirations is directly related to the adequacy of the specimen23,24. In one study, the inadequate sampling rate decreased from 16 % (only one aspiration at the nodule center) to 5.3 %, 4% or 2.6 % with increasing number of aspirations (2, 3 or 4 more aspirations in different nodule areas)24. Accordingly, some authors perform as many as 6–12 aspirations. We and others25-27, however, find that 2–4 punctures per nodule result in an acceptably low inadequacy rate and are well tolerated by the patient, whereas several aspirations carried out in one setting are often poorly tolerated and increase the risk of local edema and hemorrhage. Cystic or complex nodules. Approximately one third of all palpable thyroid nodules have a partially cystic structure (complex nodules) originating by ischemic necrosis of solid areas. In cystic nodules a progressive suction is applied in order to allow an evacuation of the cystic fluid as complete as possible. The cystic fluid is then centrifuged and the sediment smeared onto slides. However, the fluid sediment may only contain scanty follicular cells. Aspiration of solid areas in the context of complex nodules, preferably under US assistance, may result in a better cellular yield28. Smear preparation. Direct smear of the aspirated material onto glass slides is most commonly used. Care should be taken to avoid the sample to dry, as cells may become fragile. The aspirated material should be immediately smeared after expulsion and, preferably, in a single course. The smears are then fixed according to the staining technique. Recently, thin layer processing of FNAB aspirates by Thin-Prep® processor has gained increasing popularity29,30. According to this technique, the aspirated material is collected in a buffered fixative, spun down and resuspended in a second buffer. Aspecial device, Thin-Prep® processor, is then used to prepare a thin cell monolayer on a glass slide. Thin layer processing may be carried out with the material that remains in the needle hub after preparation of direct smears. The interpretation of thin-layer slides requires specific experience since the cytomorphology may substantially differ from direct smears29,30. In general, colloid appears disperse in droplets and is less evident in thin-layer preparations. Tissue fragments are less frequently observed that in direct smears, although other major architectural features (microfollicles, honeycomb arrangement and papillae) are conserved. Nuclear details may be better preserved but cytoplasm is more often disrupted as compared to direct smears. The rate of inadequate samples may increase when using thin-layers as the sole processing technique. However, Thin-Prep® processing may be a useful addition to direct smears. Staining technique. Papanicolau and May-Grunwald-Giemsa are the two most used staining techniques. Papanicolau method is preferable, in my opinion, because it best evidences details of nuclear chromatin including the ground glass nuclei and the nuclear grooves characteristics of papillary cancer. Colloid is also better evidenced with this method. On the other hand, lymphoid cells are better recognized with May-Grunwald-Giemsa staining. Immediate microscopic evaluation of sample adequacy. Some authors use Diff-Quick staining for rapid assessment of sample adequacy while the patient is waiting. The aspiration can be repeated if the smears are judged poor or inadequate, thus reducing to a minimum the proportion of inadequate sampling.