Equipment

Syringes and syringe holder

Standard disposable plastic syringes mounted in a syringe holder/pistol grip are suitable for conventional aspiration biopsy. The Cameco Syringe Pistol (Cameco AB, Taby, Sweden) is made to fit 10-cc plastic syringes. The holder leaves one hand free to fix and to feel the target, which allows better precision in placing the needle (Fig. 2.1).

Fig. 2.1 FNB with aspiration (thyroid)

Syringe and needle mounted in a pistol grip is operated by one hand, leaving the other free to feel and fix the target. Note the thumb supporting the syringe.

Patient preparation

The procedure should be clearly explained to the patient to assure his/her consent and cooperation.⁵ A formal written consent may be required, at least for deep biopsies. The procedure is usually carried out with the patient supine on an examination couch with easy access from either side. A couch with stirrups is preferable for transrectal and transvaginal biopsy, and an examination chair with adjustable headrest for biopsy of lesions in the head and neck.

Simple skin disinfectant using prepacked swabs for injections is adequate for biopsy of superficial lesions. Preparations as for minor surgical procedures (surgical skin disinfectant, fenestrated sterile cloth, sterile gloves) are recommended for transpleural, transperitoneal and bone biopsies.

Insertion of the needle

With frequent practice and growing experience, the operator acquires the ability to feel the consistency of the tissue through the needle. This helps considerably to position the needle accurately without technical aids. The ‘fingertip sensitivity’ is much greater when the needle is held directly without the interposition of a syringe and holder, as in the non-aspiration technique. A near vertical pathway tends to be less painful and allows better appreciation of depth; a tangential path is preferable in superficial skin lesions and for lesions in the chest wall.

The use of radiological imaging techniques to guide deep biopsies is described in Chapter 3. Ultrasonographic (US) guidance may be of value even in palpable lesions. It defines the lesion exactly, gives the optimal depth for biopsy, guides the needle to a solid portion of a complex lesion, and shows the relationship to other anatomical structures such as major vessels, pleura, etc.

FNB with aspiration (Figs. 2.1 and 2.2)

The aspiration technique is illustrated diagrammatically in Figure 2.2. The negative pressure does not tear cells from the tissue but merely holds the tissue against the sharp cutting edge of the needle, which scrapes or cuts softer tissue components along the track as the needle advances through the tissue.⁶ Highly cellular tissue components are softer and more friable than the supporting stroma and are selectively sampled. Fibrous stromal components are poorly represented, whereas myxoid stroma is more easily sampled. To increase the yield, the needle should be moved back and forth within the lesion with the negative pressure maintained, more vigorously in fibrous tissues with low cell content. Several passes may be necessary to sample a sufficient number of cells. In highly cellular and vascular tissues such as spleen, lymph nodes, liver and thyroid, a few rapid passes usually suffice. Additional passes mainly increase the amount of blood aspirated, causing dilution of the cellular component. Admixture with blood tends to be less if the needle is moved along the same track rather than in multiple directions. One should never wait to see material enter the hub of the needle, except when evacuating a cyst or an abscess. The ideal aspirate has a creamy consistency due to high cell content in a small amount of fluid and remains inside the needle.

Fig. 2.2 FNB with aspiration

Diagrammatic stepwise illustration of biopsy procedure: (A) needle positioned within the target tissue; (B) plunger pulled to apply negative pressure; (C) needle moved back and forth inside target; (D) negative pressure released while needle remains in target tissue; (E) needle withdrawn; (F) needle detached and air drawn into syringe; (G) sample blown onto microscopy slide.

The negative pressure must be released before the needle is withdrawn. Even so, part of the aspirate is often drawn up into the hub of the needle (see below). A maintained negative pressure may draw the aspirate into the syringe, which must then be rinsed with fluid to recover the specimen. It can also cause contamination by material aspirated along the track during withdrawal of the needle. Aspiration of US gel in guided FNB of breast lesions (Chapter 7) is a good example.

Fine needle sampling without aspiration (Figs 2.3 and 2.4)

As mentioned, the negative pressure plays a relatively minor role compared to the scraping or cutting effect of the advancing oblique needle tip. Fine needle biopsy without aspiration was introduced by Zajdela in 1987.⁷ This technique is based on the observation that the capillary pressure in a fine needle is sufficient to keep the detached cells inside its lumen. A 27–23-gauge standard needle is held directly with the fingers, inserted into the target tissue, moved back and forth in several directions for a few seconds depending on the cellularity and the vascularity of the tissue, and is then withdrawn. Using this technique, the operator gets an excellent feel of the consistency of the tissues. This is a valuable piece of diagnostic information and improves precision when sampling small lesions. Admixture with blood is generally less than with aspiration. The technique is particularly well suited for biopsy of the thyroid and other vascular tissues. The cell yield may be smaller than with aspiration but not significantly so.^8,9 We use sampling without aspiration routinely in superficial biopsies except in cystic lesions and in fibrotic paucicellular tumors in the breast and soft tissues. Aspiration with 22-gauge needles is used in most deep biopsies in order to obtain a maximum volume of cells with a minimum number of passes, in view of the frequent demand for ancillary tests. However, non-aspiration biopsy sometimes produces better samples in highly vascular lesions, for example in renal tumors.

Fig. 2.3 FNB without aspiration (thyroid)

Needle held with finger tips, other hand feels and fixes the target.

Fig. 2.4 FNB without aspiration

Diagrammatic stepwise illustration of biopsy procedure: (A) needle inserted into target tissue; (B) needle moved back and forth, varying the angle; (C) needle withdrawn; (D) needle attached to syringe and sample blown onto microscopy slide.

After biopsy of superficial lesions, pressure should be applied over the biopsy site to minimize bruising or post-biopsy hematoma. Patients should be kept under observation for a couple of hours after biopsy of deep sited lesions.

Failure to obtain a representative sample

Possible causes of failure to obtain a representative sample are illustrated diagrammatically in Figure 2.5. If the needle narrowly misses the target (Fig. 2.5B), only the adjacent tissues are sampled. An inflammatory reaction around the tumor may lead to an erroneous diagnosis. If the sample is derived from a central focus of necrosis, hemorrhage or cystic change (Fig. 2.5C), no diagnostic elements may be included. A dominant benign lesion like a cyst or lipoma can hide a small adjacent malignancy (Fig. 2.5D), for example in the breast or thyroid. Repeat biopsy of any remaining palpable abnormality after evacuation of a cyst is important. Finally, adequate samples may be difficult to obtain from desmoplastic tumor tissue in which cells are firmly held in a dense collagenous framework (Fig. 2.5E).

Fig. 2.5 Causes of unsatisfactory yield

(A) Needle well positioned within target tissue; (B) needle has missed the target tangentially; (C) needle in central cystic/necrotic/hemorrhagic area devoid of diagnostic cells; (D) needle sampling a dominant benign mass but missing a small adjacent malignant lesion; (E) fibrotic/desmoplastic target tissue giving a scant cell yield.

Direct smearing (Figs 2.7–2.12)

Smear quality is highly dependent on the smear being thin and evenly spread, ideally as a monolayer of cells. Perfect smearing is not easily learned. This is one of the reasons why FNB generally has a higher success rate when the biopsy procedure is attended by laboratory staff. Nuclear detail is poorly shown and confusing artifacts are common if smears are thick, uneven and dry slowly (Figs 2.9–2.11). Sometimes, even well prepared smears, particularly of lymphoid cells, may show a peculiar raisin-like distortion of the nuclei probably caused by moisture on the slide (Fig. 2.12). Exposure of air-dried smears to formalin vapor, which can occur during transport of material to the laboratory, can affect nuclear staining and cause loss of morphologic detail.¹⁰

Fig. 2.7 Direct smearing

Diagrammatic illustration of smearing of ‘dry’ and ‘wet’ samples. Top line: a ‘dry’ sample is smeared with the flat of a slide exerting a well-balanced pressure; middle line: two-step smearing of a ‘wet’ sample on one slide; bottom line: two-step smearing of a ‘wet’ sample with plenty of fluid moving the concentrated cells to a second slide. For details see text.

Fig. 2.8 Macro appearance of air-dried smears

(A) Optimal smear of ‘dry’ sample (carcinoma of prostate); cell clusters seen as blue dots, evenly spread; (B) Smear of ‘wet’ sample by two-step smearing; mainly blood at top end, cell clusters concentrated and evenly spread in the thin mid portion; (C, D) Examples of poorly prepared smears of bloody material, partly dried or clotted before smearing.

Fig. 2.9 Optimal and suboptimal air-dried smears

(A) Optimal spread and fixation of ‘dry’ sample (carcinoma prostate); (B) Artifacts caused by slow drying of thick bloody smear (prostatic hyperplasia) render diagnosis impossible in spite of good cell yield (MGG, HP).

Fig. 2.10 Optimal and suboptimal air-dried smears

(A) Thinly spread, well-fixed smear of papillary carcinoma of thyroid. Typical nuclear morphology well demonstrated; (B) Cells in a thick, bloody part of the same smear. Nuclei show shrinking artifacts: small and dark, structural detail not visible (MGG, IP).

Fig. 2.11 Artifacts caused by smearing

(A) Smudging artifacts at tail of smear caused by heavy pressure; fibroadenoma of breast (MGG, HP); (B) Clumping of cells and shrinkage artifacts due to slow drying/fixation of thick and bloody smear; lymphoid tissue (MGG, IP); Interpretation difficult or impossible in both in spite of good cell yield.

Fig. 2.12 Fixation artifacts

Peculiar fixation artifacts in a smear of lymphoma. Raisinoid nuclei with a ‘prickly’ contour, nuclear chromatin appears irregular (MGG, HP).

The optimal sample, obtainable from cell-rich tissues, has a creamy consistency due to high cellularity with little or no blood or fluid (’dry’ sample). A ‘dry’ sample is best smeared with the flat of a second slide exerting a light pressure as it is moved along the specimen slide (Fig. 2.7, top). The pressure must be carefully adjusted to achieve a thin, even spread without causing disruption of tissue fragments with loss of micro-architecture, or smudging artifacts as in Figure 2.11A. Optimal smearing is a fine balance between too thick and too thin. Smears of ‘dry’ aspirates dry quickly, resulting in a milky, finely granular film on the slide.

A ‘wet’ aspirate consists mainly of blood or fluid containing smaller numbers of cells. The cells can be concentrated and separated from the fluid using a two-step smearing technique as illustrated in Figure 2.7. The smearing slide is held against the specimen slide at a blunt angle near one end of the slide, allowing the fluid to accumulate in the angle. The smearing slide is then rapidly moved along the specimen slide, half way or all the way depending on the amount of fluid. Most of the fluid is left behind while the cells tend to follow the smearing slide. The concentrated cells are then smeared with the flat of the slide as for a ‘dry’ aspirate, either on the same specimen slide, or swiped to another slide.

If a larger volume of blood or fluid is obtained, it can be spread on a slide or watch glass using the needle, or by tilting the slide. With a suitable background, tiny tissue fragments become visible and can be picked up with a needle or a slide, moved to another slide and smeared. Or fragments can be placed in a drop of blood, thrombin added, to form a clot for processing as a cell block. It is critical that bloody samples are processed quickly before coagulation occurs as clotting blood makes it nearly impossible to produce optimal smears.

Examples of the macroscopical appearances of smears are shown in Figure 2.8. Figure 2.8A is an optimal smear of carcinoma showing numerous cell aggregates seen as granules spread evenly over the slide. Figure 2.8B is a two-step smear of carcinoma showing a film of blood at the top and concentrated cells at the middle, seen as a granular material. Figure 2.8C and 2.8D are examples of unsatisfactory smears from external sources.

Indirect smearing

Thin watery samples are processed by centrifugation in a cytocentrifuge. Millipore or Nucleopore filtration is an alternative but has been less satisfactory in our hands. Some laboratories prefer to rinse needles and syringes routinely with saline or with a fixative, which is then centrifuged or filtered onto slides.^11,12 More recently, the ThinPrep technique developed for gynecological cytology has been increasingly applied also to FNB specimens.^13–16 These techniques offer alternative solutions to the frequent problem of suboptimal samples received from distant sources, when the laboratory has no control over the biopsy procedure. However, the ThinPrep technique has its specific problems, and established diagnostic criteria may have to be redefined for FNB samples.^17,18 In our opinion, direct smears expertly prepared by an experienced cytopathologist remain the optimal basis for FNAC diagnosis available today, and our first priority is to perfect this technique. ThinPrep preparations are a valuable supplement, particularly for immunocytochemical staining (see below).

Monolayered smears with optimal cell preservation are particularly important in the diagnosis of malignant lymphoma. For lymph node aspirates, we recommend that a cell suspension be prepared in addition to direct smears. Hank’s balanced salt solution with the addition of 10–20% fetal calf serum is ideal for this purpose. The suspension is spun on the cytocentrifuge at low r.p.m. Dilution may be necessary to achieve optimal dispersion of cells on the slide and to avoid clumping (Fig. 2.13). A number of slides can usually be made from one aspirate to allow immunocytochemical studies. Further details on techniques suitable for lymph node samples are given in Chapter 5.

Fig. 2.13 Cytocentrifuge smears of lymphoid cells

Cell suspension in Hank’s balanced salt solution of FNB sample of cells from lymphoma; cytocentrifugation smears. (A) Undiluted specimen; (B) Optimal dilution (MGG, HP).

Tissue fragments and cell blocks

Sometimes, a thin core or fragments of tissue may be obtained with a standard 22-gauge needle (Fig. 2.14). Tissue fragments are fixed in 5–10% buffered isotonic formalin and processed as for routine histology.

Fig. 2.14 Tissue core by 22-gauge lumbar puncture needle

Tissue section of a 22-gauge lumbar puncture needle core from a well-differentiated hepatocellular carcinoma (H&E, LP).

Some laboratories recommend the routine preparation of cell blocks for paraffin embedding of FNB samples. Cell blocks may give a better idea of tissue architecture and allow multiple sections for panels of immune markers with controls.^19–21 However, they are relatively time consuming and costly compared to routine smears.²² We use cell blocks selectively, mainly if a need for immunocytochemistry is anticipated. Cell blocks are helpful if samples are heavily admixed with blood. Surprisingly good tissue fragments are often found in a cell block even when smears show only blood.

More recently, we have developed a simplified technique for cell blocks that we call ‘cell buttons’, shown diagrammatically in Figure 2.15. It is applicable to cell-rich tissues such as lymph nodes and cellular neoplasms. A drop of thick, creamy material obtainable from such tissues using a 27–25-gauge needle without aspiration is gently expelled onto a glass slide as usual, but is not spread or smeared. After a few seconds to allow the drop to adhere to the slide, the slide is carefully immersed in 90% ethanol. The sample remains stuck to the slide as a drop (‘button’). Alcohol-fixation, unlike formalin, holds the sample together. After fixation, the ‘button’ is gently detached with a scalpel blade and processed like a small biopsy. The amount of tissue obtained in this way can be substantial, cell preservation and fixation is excellent, and the material is well suited to immunocytochemical studies (Fig. 2.16). An advantage over a conventional cell block is that the cell material is concentrated, whereas multiple sections may be necessary to find scanty tissue fragments in a cell block.

Fig. 2.15 Preparation of a ‘cell button’

(A) The FNB sample is blown onto a clean and dry microscopy slide; (B) The sample/drop is left untouched on the slide a few seconds to adhere without drying; (C) The slide with the sample is immersed in 95% ethanol and left to fix; (D) The solidified fixed ‘cell button’ is carefully removed from the slide with a scalpel and processed routinely like any small biopsy.

Fig. 2.16 FNB with ‘cell button’

FNB of peripheral lung tumor: (A) Air-dried direct smear (Diff-Quik, HP); (B, C) Tissue section of ‘cell button’; solid islands/cords of tumor tissue with a prominently vascular stroma (H&E, HP); positive synaptophysin (HP). Diagnosis of peripheral carcinoid tumor.