Appraisal of Aspiration Cytology in Management of Hodgkin ’s Disease

M. FRIEDMAN, MD, U. KIM, MD, K. SHIMAOKA, MD, A. PANAHON, MD, T. HAN, MD, AND L. STUTZMAN, MD

The value of aspiration cytology in the management of Hodgkin’s disease is shown in this study of 228 pa- ients and 403 aspirations; 385 from lymph nodes and 18 from extranodal masses. In all patients the initial diagnosis was established on surgical biopsy. Aspirates were helpful in staging, defining extension of unusual radiation fields, and in recognizing residual disease and relapses after therapy. Adequate mate- rial was obtained in 80% of aspirations. The diagnosis of Hodgkin’s disease could not be established in the adequate cytologic sample in 9.9% of cases. In 5.5%’ the diagnosis was that of benign reactive hy- perplasia and in 4.4%’ non-Hodgkin’s lymphoma. Unsatisfactory material was usually obtained from nodes less than 1 cm in diameter or from residual lesions following radiation or chemotherapy. Only 14 of 93 such lesions proved to have active disease during follow up. There were no significant complications. Characteristics of the varied aspects of aspirated tumor cells found in Hodgkin’s disease are described.

Cancer 451653-1663, 1980.

HE DIAGNOSTIC VALUE of needle aspiration of T lymph node metastases originating from a var- iety of malignant tumors has been gaining recogni- tion and its advantages and limitations have been described.6.7,9,17,19.21.24.2” Since GuthrieH reported his experience in 1921 with needle aspiration of lymph nodes in the cytodiagnosis of Hodgkin’s disease, others have described the technique and morphologic aspects of this disease found on aspirated smears and im-

While the classification of Hodgkin’s disease into four s~b types ’~ . ’~ relies on histologic sectioning for dem- onstrating changes in tissue architecture and patterns of cellular infiltrates and correlating them with progno- sis, the most essential criterion for the diagnosis of Hodgkin’s disease is based on cytologic details, espe- cially the presence of Reed-Sternberg cells or its vari- ants.I2 Therefore, the cytologic smears obtained from needle aspirations may provide useful information dur- ing the course of the disease to confirm clinical findings without recourse to repeated surgical biopsies. With the current emphasis on obtaining objective histologic evidence for verifying response following therapy, the utilization of aspiration procedures for cytodiagnosis may become of even greater value.

prints ~2-4,10.11,15,16,22,23

From the Departments of Radiation Medicine, Pathology and Medicine B, Roswell Park Memorial Institute, New York State De- partment of Health, Buffalo, New York.

Address for reprints: M. Friedman, MD, Roswell Park Memorial Institute, 666 Elm Street, Buffalo, New York 14263.

The authors thank Dante B. Terrana and his staff for photographic assistance and Sue Hurley for technical assistance.

Accepted for publication April 30, 1979.

In this report, we present our ten-year experience with needle aspiration in Hodgkin’s disease and discuss the clinical indications and limitations of the technique.

Materials and Methods

Between 1967 and 1977, needle aspirations were per- formed in 228 patients with Hodgkin’s disease, ranging in ages from 15 to 65 years. In each case, standard histo- logic sections were obtained during the initial workup, on which the original diagnosis was based.

Needle aspiration was used either at the time of initial workup to identify additional regions of involvement, to help in staging the extent of disease, or later to con- firm suspected relapses during the follow-up period. Aspirations of both nodal and extranodal lesions were included in this study. Surgical biopsies were always obtained in patients with strong clinical evidence of active disease but with repeated negative cytologic findings or when cytologic examinations were sugges- tive of other malignant lymphomas.

The aspiration procedure is simple, and rapidly and inexpensively performed on both hospitalized and out- patients. Sedation or local anesthesia are unnecessary. Local lidocaine anesthesia was used only when deep lesions were aspirated. A 20-cc glass syringe with an 18- gauge, 1.5-inch long needle was used. For deep-seated soft-tissue, transthoracic; or transabdominal masses, 35-inch spinal needles were used. The depth of the lesion and its proximity to large vessels or vital organs were evaluated in advance. Local tenderness or fluctu- ation of the mass sometimes forewarned the need for

0008-543W80/040111653 $1.05 0 American Cancer Society

1653

1654 CANCER April I 1980 Vol. 45

.PREAURICULAR 9 ,SUPRACLAVICULAR 31

RETROAURICULAR 8

INTRABUCCAL I

SUBMANDIBULAR 24

NECK I77 ...... . . . . .-.

9

65

3

FIG. 1. Anatomic distri- bution of 385 needle- aspirated lymph nodes.

bacteriological tests in addition to cytologic examina- tion. The skin was cleaned with alcohol, the area to be aspirated precisely defined; and mobile masses held be- tween fingers of one hand. With the other hand, the needle attached to the syringe was inserted into the node. Suction was applied to the node by retracting the plunger. Mild suction usually produced a drop or two of fluid containing fairly pure tissue cells which, however, remained in the lumen of the needle and were rarely seen in the syringe. Excessive suction sometimes caused dilution of the tumor material by blood which flowed into the syringe. After several suction move- ments, the plunger was released to its initial position and the needle withdrawn. The material was expelled onto a glass slide by repeated pumping of the plunger and smeared gently on as many slides as possible by means of the usual technique for blood smears. Caution was exercised to minimize cell damage and preserve cell distribution. The air-dried smears were stained by

TABLE 1. Anatomic Distribution of Needle-Aspirated Extranodal Lesions in Hodgkin's Disease Patients

9

32

17

the Wright-Giemsa technique. Some variations of stain- ing time by trial and error were sometimes necessary to obtain optimal staining. Aspirations of the same lesion were repeated if necessary in order to obtain satisfac- tory materials.

Results

Evaluation of Technique

The results of needle aspirations obtained from 403 lesions in 228 patients were analyzed. Single lesions were aspirated in 177 patients, two lesions in 36 pa- tients; and three or more lesions in 15 patients. During the follow-up period, 87 patients from the initial group presented with new findings and their lesions were aspirated a second time and in 22 cases, a third time. A total of 403 aspirations were performed; 385 from lymph nodes and 18 from extranodal masses.

The anatomic distribution of the 385 nodal lesions is shown in Fig. 1. Most were in the neck, axilla, inguinal, submandibular, and supraclavicular areas. The evalua- tion of size and depth of axillary lymph nodes, particularly the ones located deep in the apex of the axilla, was sometimes difficult. Similarly, small deep- Location Number of lesions

Chest wall Oropharynx Arm Gluteal muscle Subcutaneous Abdominal mass Intrathoracic mass

TABLE 2. Clinical Conditions Associated with Inadequate Needle Aspirations in Hodgkin's Disease Patients

seated groin and femoral nodes in obese patients were hard to aspirate. The distribution of extranodal lesions is shown in Table 1. Intrapulmonary lesions were aspirated under fluoroscopy.

Two major criteria were analyzed: 1) the rate of suc- cess in obtaining sufficient material for interpretation; and 2) the comparison of diagnosis from aspirates with that of surgical biopsy specimens and/or clinical course of the disease.

Lymph nodes > 1 cm Rate of Successful Aspiration Lymph nodes Nodular High Post-radiation Material adequate for interpretation was obtained < 1 cm sclerosing axillary residue Others Total from 182 patients (80%) and 310 lesions (77%).

Unsatisfactory aspirations were recorded in 46 patients (20%) and 93 lesions (23%). These included specimens

93 44 16 11 8 14

No. 7 ASPIRATION CYTOLOGY . Friedman et al. 1655

TABLE 3 . Accuracy of Needle Aspiration-Hodgkin’s Disease Cytodiagnosis vs. Histologic Type -. -

Cytology

H istopathology

Hodgkin‘s disease lymphocytic predominance (n = 11) Nodular sclerosis (n = 79) Mixed cellularity (n = 51) Lymphocytic depletion (n = 19) Undetermined (n = 19)

Lymphoma histiocytic type (n = 3 ) 182

1 11.5% False-negative = S.S% False histologic type = 6%

Lymphoma Lymphoma Hodgkin’s Lymphoma well poorly

disease Reactive histiocytic differentiated differentiated ~-

8 2 (18%) 1 71 6 (7.5%) 1 I 48 2 (4%) 1 17 2 17 1 1

3 161 10 6 4 1

-_ * Transformation during relapse in previously treated patients who

initially presented with typical Hodgkin’s disease histopathology.

containing only peripheral blood or scattered tissue elements.

Needless to say, the rate of diagnostic accuracy de- pended largely on the quality of smears. In most opti- mal aspirates. only a drop or two of tissue material, sometimes pus-like and occasionally semisolid or vis- cous, could be expelled from the needle. Smears made from such material show a high cell density, and when

homogeneously spread on slides they frequently repro- duced the cell distribution pattern seen in the tissue section.

The conditions in which 93 negative nodal aspirations were recorded are shown in Table 2. Small nodes less than 1 c m in diameter were the most frequent cause of unsuccessful aspiration. It was assumed that many of these nodes were missed by the inserted needle but

FIG. 2. Hodgkin’s disease general view. Photomicrograph showing high den5ity of lymphoid cells and scattered Hodgkin’s and Reed- Sternberg cells. ( In th i s and all following figures, Wright-Giemsa stains of lymph node aspirates are shown ~ 2 2 5 ) .

1656 CANCER April I 1980 VOl. 45

FIG. 3 . Binucleated Reed-Sternberg cell. Photomicrograph showing large cell with pale blue cytoplasm, reticulated chromatin and multiple blue nucleoli ( X 875).

most nodes of this size proved benign in subsequent follow-up examinations. In 16 lesions of nodular scle- rosing type, the aspirate was either dry or contained only blood. In some patients with deep axillary nodes and others with residual fibrotic nodes after irradiation, the aspirations were unsatisfactory. Follow-up exami- nation of these 93 lymph nodes showed that active disease eventually developed in only 14 of these previously unsuccessful aspiration sites.

Diagnostic Accuracy

Comparison of cytologic interpretation with the his- tologic diagnosis for 182 patients from whom satisfac- tory specimens were obtained is summarized in Table 3. In 5.5%, the diagnosis was benign reactive hyper- plasia. These were shown to be positive by later surgi- cal biopsy obtained after the persistence or progression of lesions suggested the discrepancy between clinical signs and the cytologic diagnosis.

In recurrent Hodgkin’s disease, particularly in those cases in which the disease progressed rapidly despite multimodality treatments, the cytologic picture was

rarely reminiscent of the original disease but often pre- sented the cellular features of histiocytic lymphomas. Eleven aspirations (6%) were thus classified as malig- nant non-Hodgkin’s lymphoma and surgical biopsy in 3 confirmed the transformation of cell type.

Morphologic Findings

The correct diagnosis of Hodgkin’s disease was made in 161 of 182 cases. Criteria by which it was achieved in- cluded: 1) high cell density; 2 ) mononuclear Hodgkin’s cells and/or Reed-Sternberg cells; and 3) a lymphoid background with reactive elements and occasional eo- sinophilia. The cell density was often comparable to that seen on tissue sections (Fig. 2). Bi- and multi- nucleated cells (Figs. 3,4) were found in most Hodgkin’s disease patients. In some instances, the presence of large mononuclear cells, approximately 25-50 microns in diameter, with scattered cytoplasmic vacuoles and re- ticulated nuclei with one or two large nucleoli, characteristic of the so-called Hodgkin’s cell, were considered of diagnostic value in the absence of classic Reed-Sternberg cells (Fig. 5) . The lacunar cells seen in

No. 7 ASPIRATION CYTOLOGY . Friedman et al. 1657

1658 CANCER April 1 1980 Vol. 45

FIG. 6 . Lacunar cell. Photomicrograph of large cell with clear pale cytoplasm and convoluted nucleus ( ~ 8 7 5 ) .

the formalin-fixed sections of nodular sclerosis type' were recognized as large cells, 20-40 microns in diameter, with clear or pale blue cytoplasms and distinct cellular borders (Fig. 6). The nuclei were either lobated, indented, or displayed overlapping multiple nuclear segments, in which the distinction between mono- and multinucleation was quite difficult. Interme- diately sized mononuclear cells with plasmacytoid features including eccentric nuclei with reticulated chromatin, blue nucleoli, granulated basophilic cyto- plasms with a few vacuoles, and clear space near one pole of the nucleus were found in many specimens (Fig. 7A). Cells which were similar but lacking large blue nucleoli were often seen on cytologic smears of benign reactive nodes which are often described as immuno- blasts.

At higher magnifications, the nuclei of Hodgkin's cells often revealed bud-like extensions of the chromatin mass which in more atypical cells appear lobated or seg- mented. In some Reed-Sternberg cells, filaments con- nected multiple nuclei (Fig. 7B). Mitotic figures of the large mononuclear cells (Fig. 7C) were often found and on several occasions multinucleated cells were de- tected in mitosis (Fig. 7D). Prophases and polyploid

metaphases were usually seen and atypical forms with bridges and aberrent pairs lagging behind the synchron- ized sets of chromosomes were somtimes present.

The search for diagnostic cells in the cytologic prepa- ration failed in 2 of 11 patients previously classified by lymph node biopsy as having lymphocytic predomi- nance type, 6 of 79 patients with nodular sclerosing type, and 2 of 51 with mixed cellular type.

The background usually consisted of small lympho- cytes admixed with varying numbers of reactive cells such as immunoblasts, immature lymphocytes, and plasma cells. Occasionally, the number of eosinophils and/or neutrophils approached that of lymphoid cells but in most cases, only a few granulocytic elements were found and quite often no eosinophils.

When small lymphocytes were predominant, Reed- Sternberg cells and large mononuclear cells with blue nucleoli were rarely seen in the aspirated material and this correlated with the lymphocytic predominant type (Fig. 8). On specimens with a variety of reactive cells, (immunoblasts, plasma cells, eosinophils, medium and large lymphoid cells) bi- and multinucleated cells were seen more often and such a finding correlated with mixed cellularity type (Fig. 9). Cellular compositions

No. 7 ASPIRATION CYTOLOGY . Friedman et ul. 1659

FIG. 7. Variants of Hodgkin's cell and Reed-Sternberg cells in interphase and mitosis. ( f o p left) Hyperbasophilic Hodgkin's cell, a large cell with a high NIC ratio, dark blue cytoplasm with vacuoles, eccentric nucleus, and giant blue nucleolus (~875). (fop r i g h f ) Trinucleated Reed- Sternberg cell, at prophase or end of telophase. Nuclei are connected by chromatin filaments ( ~ 8 7 5 ) . (borrorn left) Hodgkin's cell in mitosis; hyperdiploid metaphase ( ~ 5 6 0 ) . (hofforn right) Reed-Sternberg Cell in mitosis; asynchronous mitosis with two nuclei in prophase and one nucleus in interphase ( ~ 5 6 0 ) .

similar to the mixed cell type were seen on specimens obtained from lesions classified as nodular sclerosing type; scattered mono- and/or multinucleated cells with clear cytoplasms were often encountered on such spec- imens (Fig. 10). A preponderance of large pleomorphic cells with indented nuclei and sparse lymphocytic back- ground was obtained from patients classified as having the lymphocytic depletion type (Fig. 11). In this type, cells having denuded nuclei with large blue nucleoli were also frequently seen.

Clinica I Indications

The clinical conditions under which needle aspiration was found to be indicated were as follows:

1) Stuging M9orkup: In 12 cases, some doubt remained after routine clinical and pathologic staging as to the sig- nificance of small nodes to which little importance had been attached. Such nodes, less than 1 cm in diameter, were found in the groin (4 patients), neck (5 patients) and axilla (3 patients) with clear evidence of Hodgkin's disease with Reed-Sternberg cells on the aspirated

smears. These nodes were either on the opposite side of the diaphragm from the known disease or were in pa- tients otherwise considered to have Stage I disease. The findings on aspiration, therefore, changed the stage and treatment.

2 ) Extension c$radiationfields: In 9 cases, although the stage of the disease was not changed, the positive findings in small nodes indicated the need for extension of the radiation field. Thus, discovering extended disease in pre- or retroauricular nodes in 4 patients, in lower axillary nodes in 3, and in the lateral iliac areas in 2, necessitated either enlarging the radiation port or applying separate radiation fields.

3) Recurrences in previously irradiated areas: Surgi- cal biopsy specimens of suspected recurrences in irradi- ated areas with advanced post-radiation changes, such as cutaneous and subcutaneous fibrosis, chronic lymph- edema, induration, and telangiectasia are not routinely performed, since the healing process in such areas is impaired. Needle aspiration was successful in revealing active lymphomatous disease in 7 such patients.

4) Residual nodes: Residual disease due to an incom-

1660 CANCER April 1 1980 Vol. 45

FIG. 8. Aspirate from histologically proven lymphocytic-predominance type. Dense monotonous lymphocytic background with scattered Hodgkin’s cells (~350).

plete resolution of nodular sclerosing-type lymph nodes after therapy, or persistent hyalin or collagenous stroma, may produce enlarged or palpable nodes. Therefore, it is important to ascertain whether repeated negative aspirates are truly indicative of no active dis- ease. In 4 of our patients, active disease was found by aspiration of supposedly stable residual nodes six to 12 months following radiation and in 11 patients following chemotherapy.

5 ) Soft tissue musses: Unlike enlarged lymph nodes, diffuse ill defined soft-tissue lesions were aspirated at random sites, but a successful needle aspiration may help us avoid an extensive surgical procedure to obtain tissue specimens. Aspiration smears containing Reed- Sternberg cells were obtained in 15 patients with soft- tissue masses.

6) Nodal relapses detected during follow-up: Recur- rent disease most commonly involves lymph nodes of non-irradiated areas. Small palpable nodes (0.51 cm) of doubtful clinical significance may be found frequently in the groin and axilla. The surgical biopsy of such nodes as a routine screening procedure is impractical, whereas the needle aspiration is a simple procedure and may be performed repeatedly during routine clinical

follow-up examinations. Recurrent disease was dis- covered through aspiration of small lymph nodes during the routine follow-up examination in 103 patients.

Complicutions

Needle aspiration was well tolerated by all patients. The discomfort caused by needle aspiration of superfi- cial lesions was usually no more than that of vena punc- ture. Occasional prolonged local bleeding was easily controlled by local pressure and rarely resulted in a small hematoma. N o local infection occurred, and tumor seeding along the needle tract was not seen.

Discussion

The analysis of needle aspirations in Hodgkin’s dis- ease has three major objectives: 1) evaluation of techni- cal procedure; 2) establishment of diagnostic criteria; and 3) clinical indications.

Technical Procedure

Although tissue aspiration is a simple procedure, careful attention to each step, from obtaining cells to

No. 7 ASPIRATION CYTOLOGY 1 Friedman et al. 1661

FIG. 9. Aspirate from histologically proven mixed-cellular type. Lymphocytes, neutrophils, eosinophils, histiocytes, and two Reed-Sternberg cells (~560).

spreading and staining them, is necessary in order to obtain high-quality smears. In most cases, the needle aspirate of an enlarged lymph node yielded smears rep- resentative of the nodal cell population, equivalent to bone marrow aspiration smears. While an adequate number of cells is usually obtained, some aspirates, particularly from nodular sclerosing type of lesions, may contain only a few lymphoid cells and fibrocytes or no cellular elements.

Diagnostic Criteria

Nuclear abnormalities and the size and number of nucleoli are the major characteristics of the diagnostic atypical giant cells seen in Hodgkin’s disease. Al- though it is generally assumed that the giant cell of Hodgkin‘s disease, whether mono-, bi- o r multi- nucleated, is a morphologic marker, the presence of a microenvironment of a lymphoid background with scattered Hodgkin’s cells and/or Reed-Sternberg cells represents the picture characteristic of the disease.

The diagnosis of Hodgkin’s disease could not be es- tablished in the adequate cytologic samples in 9.9% of cases. In 5.5%, the diagnosis was that of benign reac-

tive hyperplasia, and in 4.4%, of non-Hodgkin’s lym- phoma. Most of these patients had recurrent disease after intensive treatment and the smears probably indicated transformation to tumors mimicking other types of lymphoma.

Since cells similar to Reed-Sternberg cells have been described in infectious mononucleosis,13 other malig- nant tumors,20 and in benign lesion^,^ the question of specificity attributed to Reed-Sternberg cells as a diag- nostic marker for Hodgkin’s disease arises. Somewhat similar bi- and multinucleated cells are seen frequently on cytologic smears in a variety of malignant tumors, but as in histologic sections, thcse cells are interpreted in the overall context of the entire cytologic picture. In benign lesions, the magnitude of nuclear aberrations is more limited. The various reactive cell types seen in as- sociation with Hodgkin’s disease are easily recognized in aspirates stained with Wright-Giemsa. The nuclear and nucleolar changes of Hodgkin‘s and Reed- Sternberg cells are clearly visualized in well-prepared smears. The monotonous dense lymphoid background or the high proportion of atypical histiocytes may suggest the lymphocytic-predominance or lymphocytic-deple- tion type of this disease. The cytologic picture of

FIG. 10. Aspirate from histologically proven nodular-sclerosis type. Lymphocytes, histiocytes, and binucleated lacunar cell ( x 560).

FIG. 11. Aspirate from histologically proven lymphocytic-depletion type. Among the scanty lymphocytes is a preponderance of large atypical histiocytes with cleaved nuclei and large nucleoli ( ~ 5 6 0 ) .

No. 7 ASPIKATION CYTOLOGY . Friedinan et al. 1663

mixed cellular type is often indistinguishable from that of nodular sclerosing type, and only the histologic sec- tions can differentiate between them. Thus, cytologic specimens will reveal only the general diagnostic fea- tures of Hodgkin’s disease.

Even when the workup is completed and the histo- logic diagnosis of Hodgkin’s disease and its subtype es- tablished, various clinical situations developing either before treatment or during the follow-up period may necessitate further confirmation of suspicious findings. The process of screening such lesions and sampling them by needle aspiration has proven to be a most valuable procedure complementary to surgical biopsies in the staging and clinical follow-up of patients.

REFERENCES

1. Anagnostou. D., Parker, W. J. , Taylor, R. C. , Chir, B., Tindle, H. B.. and Lukes, J . R.: Lacunar cells of nodular sclerosing Hodgkin’s disease. Cirnc~r. 39: 1032- 1043, 1977.

2 . Bessis, M.: Living Blood Cells And Their Ultrastructure. New York, Springer-Verlag, 1973: p. 628.

3. Bloch, M.: Comparative study of lymph node cytology by punc- ture and histopathology. Actti Cytol. 2: 139- 144, 1967.

4. Desmet. V.. Beert. J . , and Keybrouck, G.: Value of the cytolog- ical diagnosis of Hodgkin’s disease in comparison with histological findings. Strngrc, 9:73-76, 1964.

5. Doyle, A. P.. and Hellstrom, H . R.: Mesantoin lymphade- nopathy morphologically simulating Hodgkin’s disease. Ann. Infcrt?. .Mod. 59:363-368, 1963.

6. Forkner. P. E.: Material from lymph node of man: Studies in living and fixed cells withdrawn from lymph nodes of man. A w h . In tern. M r d . 40:.532-537: 647-660, 1Y27.

7. Frable, W. J . : Thin needle aspiration biopsy. Am. . I . Clin. Ptrfhol. 65: 168- 182, 1976.

8. Guthrie, C. G.: Gland puncture as a diagnostic measure. Bull. Johris 1-lophin.r Hosp . 32:266-269, 192 1,

9. Hajdu, S. I . , and Melamed, M. R.: The diagnostic value ofaspi- ration smears. A m . J . Clin. Pnrliol. 59:350-356, 1973.

10. Loseke, L . . and Craver, L. F.: The diagnosis of Hodgkin’s disease by aspiration biopsy. Blood 1:76-82, 1946.

I I . Lukas, P. F.: Lymph node smears in the diagnosis of lymph- adenopathie: A review. Blood 10: 1030- 10.54, 1955.

12. Lukes, J . R.: Criteria for involvement of lymph nodes, bone marrow, qpleen and liver in Hodgkin’s disease. Carrc,er Res. 31: 1755-1767, 1971.

13. Lukes. J . R.. Tindle, H. B., and Parker, M. .: Reed-Sternberg- like cells in infectious moncinucleosis. Lancer 2: 1003- 1004, 1969.

14. Lukes, J . R . , Butler, J . , and Hicks, E.: Natural history of Hodgkin’s disease as related to its pathologic picture. Cuncer 19: 317-344, 1966.

15. Moore, R. D., Reagan, J . W.: Cellular study oflymph nodeim- prints. Crrnrur 6:606-618, 1953.

16. Morrison, M.. Samwick, A. A , , Rubinstein. J . . Stich, M . . and Loewe, L.: Lymph nodeaspiration. A m . J . Clin. Puthol. 22:255-262. 1952.

17. Pavlovski, A,: L a Puncion Ganglionar: Su contribucion al diagnostic0 clinico quiriirgico d e las afecciones ganglionares: Tesis, A. Lopez. Buenos Aires, 1934.

18. Mathe, G., and Rappaport, H.: Histological and cytological typing of neoplastic diseases of haematopoietic and lymphoid tissues. International Classification of Tumors, No. 14. Geneva, World Health Organization, 1976.

19. Soderstrom. N.: Fine Needle Aspiration Biopsy. Grune & Stratton. New York, 1966.

20. Strum, B. S., Park, K. J., and Rappaport. H.: Observations of cell resembling Sternberg-Reed cells in conditions other than Hodgkin‘s disease. C o n c ~ r 26: 176- 190, 1970.

21. Strunge, E.: La Ponction Des Ganglions Lymphatiques. Co- penhagen, E. Eynar Munksgaard, 1944.

22. Ultmann. J . E . , Koprowska. I., and Engle, R. L.: A cytologi- cal study of lymph node imprints. C r r / ? c ~ r I 1:507-524, 1958.

23. Weil. P. E . , Isch-Wall, P. and Perles, S.: L e diagnostic de la maladie d o Hodgkin par la ponction des ganglions. Pres.s ,Wed. 44: 1540-1543, 1936.

24. Zajdela, A , , Ennuyer, A,. Bataini, P . , and Poncet, P.: Valeur du diagnostic cytologique des adenopathies par ponction aspiration. H//II. Cancer 3:327-340, 1976.

25. Zajicek, J . : Lymph nodes of the neck. I n Aspiration Biopsy- Cytology, part I . New York, S. Karger; 1974: pp. YO-124.