Case report
Primary spinal extranodal Hodgkin’s disease at two levels
Ferda C
¸ a˘gavi
∗, Murat Kalaycı, ˙Ishak ¨
Ozel Tekin, Gamze Numano˘glu,
Zeynep C
¸ a˘gavi, S¸anser G¨ul, Bektas¸ Ac¸ıkg¨oz
Zonguldak Karaelmas ¨Universitesi, Tıp Fak¨ultesi Hastanesi N¨oro¸sir¨urji AD, Zonguldak 67600, Turkey
Received 30 January 2004; received in revised form 18 October 2004; accepted 26 November 2004
Abstract
About 90% of Hodgkin’s disease cases originate from lymph nodes whereas 10% from extranodal regions. Patients rarely present with spinal cord compression due to epidural Hodgkin’s disease. Primary spinal extradural Hodgkin’s disease which does not have any other organ involvement in the body is even rarer.
A 39-year-old male patient who complained of lumbar pain had normal findings upon neurological examination. Radiological examination revealed a mass on the epidural space at level L3 and the involvement of the vertebral corpus accompanied by the involvement of C6 vertebral body. Primary focus could not be identified despite further investigation. The patient underwent L3 laminectomy and posterior decompression and biopsy was obtained from the lesion extending to epidural space. The pathological result was reported as lymphocyte dominant type Hodgkin’s disease. Flow cytometry was performed to the lesion, also. The patient was evaluated as Stage 4A according to Ann Arbor classification. Postoperative radiotherapy was applied to lumbar and cervical region.
In the literature we have not come across any case of primary spinal extranodal Hodgkin’s disease with involvement at two levels. In conclusion, although it might be extremely rare, primary spinal extranodal Hodgkin’s disease with involvement at two levels might be observed.
© 2004 Elsevier B.V. All rights reserved.
Keywords: Hodgkin’s disease; Primary extranodal lymphoma; Spinal cord
1. Introduction
Hodgkin’s disease constitutes 0.5–1% of all adult tumors
[1]. Spinal involvement is observed in 5.8% of all Hodgkin’s cases [2]. The 90% of Hodgkin’s disease cases originate from lymph nodes whereas 10% from extranodal regions
[3]. Epidural space is one of the extranodal regions. Patients rarely present with spinal cord compression due to epidural Hodgkin’s disease. Limited cases with such presentation have been reported in the literature[2,4–13]. Primary spinal ex-tradural Hodgkin’s disease with no further organ involvement is extremely rare. Four publications and ten case reports are available in the literature reporting such cases[5,8,10,12].
∗Corresponding author. Tel.: +90 372 261 0169/1588;
fax: +90 372 261 0155.
E-mail address: cagavif@yahoo.com (F. C¸ a˘gavi).
We present a 39-year-old male patient, with lumbar pain and diagnosed as Hodgkin’s disease at lumbar and cervical regions without any other organ involvement. We have not come across any case in the literature with two level involve-ments due to primary spinal extranodal Hodgkin’s disease as observed in our patient.
2. Case report
A 39-year-old male patient with a complain of lumbar pain for the last 2 months is presented in this case report. He had normal neurological examination and did not reveal any pathology in two-dimensional lumbosacral films. Rou-tine laboratory tests were normal: WBC: 5400l−1, ESR: 5 mm/h, Brucella agglutination negative. So, he received con-servative treatment. As his complaints did not resolve, lum-bar magnetic resonance images were obtained and
involve-0303-8467/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.clineuro.2004.11.023
Fig. 1. Sagittal T1-weighted (A), sagittal T2-weighted (B), sagittal gadolinium-enhanced T1-weighted (C), axial gadolinium-enhanced T1-weighted (E) lumbar magnetic resonance imaging revealed the involvement of the L3 vertebral body and a mass in the epidural distance. After surgery and radiotherapy sagittal gadolinium-enhanced T1-weighted (D) and axial gadolinium-enhanced T1-weighted (F) lumbar magnetic resonance imaging showed that the epidural mass at L3 level had disappeared and a little contrast enhancement was present on L3 vertebral body.
Fig. 2. Sagittal T1-weighted (A), T2-weighted (B) and gadolinium-enhanced T1-weighted (C) cervical magnetic resonance imaging revealed the involvement of the C6 vertebral corpus. After radiotherapy sagittal gadolinium-enhanced T1-weighted (D) cervical magnetic resonance imaging showed that no contrast enhancement on C6 vertebral body.
Fig. 3. Photomicrograph demonstrated the membrane staining of Reed–Sternberg’s cells with CD15 (Neomarkers Clone 15CD2).
ment of the L3 vertebral body and a mass in the epidural dis-tance were identified (Fig. 1A–C, and E). Lumbar CT did not demonstrate any lytic appearance on vertebral body. Physical examination of the patient did not reveal any lymphadenopa-thy or other pathological findings. He did not complain about any fever, weight loss or night sweats. He did not have any special finding in his personal and family history. The result of human immunodeficiency virus antibody test was nega-tive, and tumor markers (CA-125, CA-15-3, CA-19-9, carci-noembryonic antigen, alpha-feto protein) were normal. There were no abnormal findings in abdomino-pelvic ultrasonog-raphy (USG). Abdominal, pelvic and thoracic CT’s of the patient did not reveal any pathology. Considering further in-volvement other than lumbar region, cerebral, cervical and thoracic MRIs were obtained. In the cervical MRI, there was involvement of the C6 vertebral corpus (Fig. 2A–C). Cranial and thoracic MRI did not reveal any pathology. Whole body bone scintigraphy (20 mCi Tc 99 m-MDP) did not demon-strate any involvement. Peripheral blood smears and bone marrow aspirates did not reveal any pathological appear-ance. Bone marrow biopsy was normocellular. Biopsy was obtained by performing L3 laminectomy and posterior de-compression. Pathology result was reported as lymphocyte dominant type Hodgkin’s disease (Fig. 3). Flow cytome-try was also performed (Fig. 4). CD3/8, CD3/19, CD2/20, CD3/16 + 56, CD3/HLA-DR antibody pairs were used for differential typing of CD45 positive cells. Immunophenotyp-ically, most of the cells have B-cell phenotype with surface expression of CD19 and CD20. The percentage of the CD19 positive cells was 61% on the lymphocyte region. On the other hand, the proportion of T cell with surface expression of CD3 was 13.6%. HLA-DR expressed T-cell percentage was 87. In accordance with these results, we decided on a lym-phoid cell accumulation and lymphoma. The patient did not develop any neurological deficit in the postoperative period. Gallium scintigraphy was performed in an attempt to
iden-tify the primary focus. In the imaging of whole body planar, thorax and abdomen SPECT (8 mCi Gallium-67) no involve-ment was observed including the vertebrae. The patient was evaluated as Stage 4A according to Ann Arbor classification. In the postoperative period, spinal region was irradiated with 2000 cGy, lumbar and cervical regions were irradiated with additional 2000 cGy radiotherapy. On the fourth month of ra-diotherapy, the new MRI revealed that the epidural mass at L3 level has disappeared and a little contrast enhancement was present on L3 vertebral body (Fig. 1D and F). There was no contrast enhancement on C6 vertebral body (Fig. 2D).
Fig. 4. Immunophenotypically, the percentage of the CD19 positive cells (B lymphocytes—y-axis) was 61% and surface expression of CD3 (T lymphocytes—x-axis) was 13.6%.
3. Discussion
Ultmann and DeVita reported that 10% of Hodgkin’s disease cases originated from extranodal region[3]. Spinal involvement is observed in nearly 10% of all extranodal lymphomas[14]. Presentation with spinal cord compression due to spinal extradural Hodgkin’s disease is observed in 0.2–0.4% of the cases[2,7]. In a research study conducted at Yale University School of Medicine, only one out of 600 Hodgkin’s disease cases was found to have findings of spinal cord compression[7]. Corrale reported spinal cord compres-sion in 25 of 426 Hodgkin’s disease cases; two of these cases (0.4%) had spinal findings on first presentation[2]. Spinal epidural Hodgkin’s disease cases presenting with spinal find-ings are presented inTable 1.
Primary extranodal presentation of Hodgkin’s disease is extremely rare. Wood and Coltman reported that primary ex-tranodal presentation was present in less than 0.25% of all Hodgkin’s disease cases[15]. There are 4 studies reporting 10 patients with spinal findings due to primary spinal epidural Hodgkin’s disease[5,8,10,12]. Citow et al. reported one case with extension to epidural space on fourth and fifth vertebra and having cord compression. This patient underwent two biopsies under computed tomography guidance and defini-tive diagnosis could not be reached. Following decompres-sion, acute and chronic inflammation were identified in the biopsy, so the patient was accepted as having tuberculosis and antituberculosis treatment was initiated. As spinal cord com-pression reappeared 5 months later, computed tomography-guided biopsy was performed revealing lymphoid infiltra-tion [5]. Moridaira et al. reported a case of Hodgkin
hav-ing T9 involvement and demonstrathav-ing cord compression at T8-10 epidural space[10]. Rao et al. published 21 primary spinal epidural lymphoma cases. Only one of the patients had Hodgkin’s disease. Specific information pertaining to this pa-tient has not been mentioned clearly in the article [12]. In 1954, Love et al. reported seven cases in similar manner[8]. Seven of the 39 epidural lymphoblastomatous cases were re-ported as Hodgkin’s disease, yet information concerning this group has not been provided[8]. Citow criticized the series of Love presented in 1954 claiming that disease at other sites might not have been diagnosed due to insufficient diagnostic techniques of those days[5]. In our case, the diagnosis of the lesion at L3 epidural space was accomplished with biopsy. As MRI appearance of the lesion at C6 corpus was identi-cal to that of L3, this lesion was also accepted as Hodgkin’s disease without another biopsy. In the literature we could not come across any other cases having no further involvement than spinal involvement at two levels.
The two level involvements in our case might be due to metastasis from a primary focus. However, further screening could not identify any systemic disease. Thus, the diagno-sis of the presented case appears to be a primary spinal ex-tranodal Hodgkin’s disease. Several claims have been made concerning the pathogenesis of primary spinal extranodal Hodgkin’s disease. Love underlined the possibility of hav-ing hematogenous spread from an unidentified lymph node resulting in Hodgkin’s disease at an extranodal site[8]. Our case has been screened in depth as much as the current tech-nology allowed. Although it is a small possibility, there still can be an unidentified focus despite detailed search. For the identification of Hodgkin’s disease in the body, gallium-67
Table 1
Spinal epidural Hodgkin’s disease cases with spinal findings
Ref. no. Study Year N Age
(year)/sex Location MRI findings Treatment Survival (month) Clinical im-provement Complete remission Stagea [8] Love et al.b 1954 Pr 7 79/M? ? – S, R? d 4, 8, 12, 36 yr ? ? – a 6, 12, 18 yr [6] De Freitas et al. 1980 1 21/M T9 – S d 4, 8, 12, 36 + ? ? [12] Rao et al. 1982 Pr 1 ? ? – S, R a 6, 12, 18 yr ? ? ? [4] Aabo and Walbom-Jorgensen 1986 6 ? ? ? Sc, C, R m10 91% – ?
[9] Mikhael and Paige 1987 1 33/M T6d E, V S, C ? + ? ?
[2] Correale et al. 1990 2 ? ? ? ? ? ? ? ?
[11] Perry et al. 1993 2 30/F T5 E S, R, C a 32 Same + 4A
42/F C8 E S, R, C a 79 Same + 2B
[10] Moridaira et al. 1994 Pr 1 56/M T8-10 E, V S, C ? ? + 1E
[7] Higgins and Peschel 1995 1 46/M C5-T8 E, P S, C, R a 21 + + ?
[13] Toprak et al. 1997 1 20/M L4-S1 E, V, P B, C a 12 Same + 4
[5] Citow et al. 2001 Pr 1 54/F T4-5 E, V, P S, R a 24 + ? 1E
C¸ a˘gavi 2004 Pr 1 39/M L3, C6 E, V, P S, R a 13 Same ? 4A
Ref. no.: reference number; Pr: primary spinal epidural Hodgkin’s disease. N: number of patients; E: epidural; V: vertebral; P: paraspinal; S: surgery; B: biopsy (bone marrow and lymph node); R: radiotherapy C: chemotherapy; d: dead; a: alive, yr: year, m: median survival.
aAnn Arbor stage at diagnosis.
b Seven of the 39 epidural lymphoblastomatous cases were reported as Hodgkin’s disease, yet information concerning this group has not been provided.
Eleven of 39 patients got radiotherapy and the oldest patients were 79 years old.
cLaminectomy was performed only to the patients with acute paraplegia. d Multiple bone lesions from cervical region to sacral region.
scintigraphy has a sensitivity of 93% and a specificity of 100%[1]. In addition to Gallium scintigraphy, identifying no other focus in other parts of the body and diagnosing no lymph node involvement during the clinical follow-up (13 months) of the disease, removes us from the possibility of an unidentified focus. Another hypothesis is the spread of the tu-mor developing in the paraspinal region to the epidural space through vertebral foramen [7,16]. Some authors state that the tumor which develops from extradurally located cellular remnants may directly spread to paraspinal region and bones thereby to the epidural space [17,18]. Other authors claim that small lymphatic elements in the epidural space might evolve into lymphoma[16,19]. The lymphoma which devel-ops from these lymphatic elements might directly spread to epidural space and vertebral corpus. However, in the litera-ture we cannot find any publication which shows the pres-ence of these elements histopathologically. In our case, the possibility of simultaneously developing lumbar and cervi-cal tumors from lymphatic elements in the epidural space does not seem to be easily acceptable. Another possibility is the evolution of Hodgkin’s disease from vertebral bodies. In the literature we can find cases of Hodgkin originating from bone structures[10,20–22]. Similarly, in the presented case, the probability of the concomitant development of Hodgkin’s disease from cervical and lumbar vertebral bodies does not seem to be very high. Another mechanism in the pathogene-sis may be the development of the disease from lymphoid cell remnants or vertebral bodies at lumbar region and metastasiz-ing to cervical region. However, all these cannot go beyond speculation.
Vertebral destruction of lymphoma is in lytic fashion[23]. In the presented case CT did not demonstrate any lytic struc-tures on L3 vertebral body. The absence of lytic strucstruc-tures may be related to early diagnosis.
As the number of cases and publications are limited in number, the most appropriate treatment for the patients with Hodgkin’s disease who present spinal cord compression is not clear. Aabo and Walbom-Jorgensen underlined the fact that whether or not having laminectomy did not create any differ-ence in the clinical improvement rates of lymphoma patients with spinal cord compression[4]. Similarly, Correale stated that patients having laminectomy combined with chemother-apy and radiotherchemother-apy (favorable results in five of nine pa-tients), and patients having only chemotherapy and radio-therapy (favorable results in 11 of 21 patients) did not differ in their prognosis[2]. In Hodgkin patients with spinal find-ings, the general attitude is to perform decompression with laminectomy[5,7,9–11]. Remission in five Hodgkin patients with spinal findings has been reported, three of which had combined treatment (surgery, radiotherapy and chemother-apy) [9,11]. Other two cases received chemotherapy after surgery or biopsy[10,13].
The prognosis of the disease is in relation with its grade
[23]. In Hodgkin’s disease originating from bone, the progno-sis is unfavorable due to delays in the diagnoprogno-sis and treatment
[22]. Aabo and Walbom-Jorgensen reported the average
sur-vival of lymphoma case with spinal cord compression as 10 months[4]. Aabo and Walbom-Jorgensen in their series stated that patients presenting with spinal cord compression had a longer survival than patients developing cord compression during the course of the disease[4].
In conclusion, although being extremely rare, primary spinal extranodal Hodgkin’s disease with involvement at two levels may be observed.
References
[1] Van der Wall H, McLaughlin AF, Southee AE. Gallium scintigra-phy in tumor diagnosis and management. In: Murray IPC, Ell PJ, editors. Nuclear medicine in clinical diagnosis and treatment, vol. 2. Edinburgh: Churchill Livingstone; 1998. p. 813–29.
[2] Correale J, Monteverde DA, Bueri JA, Reich EG. Peripheral ner-vous system and spinal cord involvement in lymphoma. Acta Neurol Scand 1991;83:45–51.
[3] Ultmann JE, DeVita Jr VT. Hodgkin’s disease and other lymphomas. In: Petersdorf RG, Adams RD, Braunwald E, Isselbacher KJ, Martin JB, Wilson JD, editors. Harrison’s principles of internal medicine. New York: McGraw-Hill; 1983. p. 811–25.
[4] Aabo K, Walbom-Jorgensen S. Central nervous system complications by malignant lymphomas: radiation schedule and treatment results. Int J Radiat Oncol Biol Phys 1986;12:197–202.
[5] Citow JS, Rini B, Wollmann R, Macdonald RL. Isolated, primary extranodal Hodgkin’s disease of the spine: case report. Neurosurgery 2001;49:453–7.
[6] De Freitas MRG, Nascimento OJM, Cincinatus D, Praxedes H, Hahn MD. Spinal cord compression as initial manifestation of Hodgkin’s disease: report of a case. Arq Neuropsiquiatr 1980;38: 182–6.
[7] Higgins SA, Peschel RE. Hodgkin’s disease with spinal cord com-pression A case report and a review of the literature. Cancer 1995;1(75):94–8.
[8] Love JG, Miller RH, Kernohan JW. Lymphomas of spinal epidural space. Arch Surg 1954;69:66–76.
[9] Mikhael MA, Paige ML. Hodgkin’s disease of spine: computed tomography and magnetic resonance imaging. J Comput Tomogr 1987;11:174–7.
[10] Moridaira K, Handa H, Murakami H, Uchiyama T, Takeuchi T, Sato S, et al. Primary Hodgkin’s disease of the bone presenting with an extradural tumor. Acta Haematol 1994;92:148–9.
[11] Perry JR, Deodhare SS, Bilbao JM, Murray D, Muller P. The sig-nificance of spinal cord compression as the initial manifestation of lymphoma. Neurosurgery 1993;32:157–62.
[12] Rao TV, Narayanaswamy KS, Shankar SK, Deshpande DH. Primary spinal epidural lymphomas A clinico-pathological study. Acta Neu-rochir (Wien) 1982;62:307–17.
[13] Toprak A, Kodalli N, Alpdogan TB, Giral A, Celikel CA, Gurmen N, et al. Stage IV Hodgkin’s disease presenting with spinal epidural involvement and cauda equina compression as the initial manifesta-tion: case report. Spinal Cord 1997;35:704–7.
[14] Masaryk TJ. Neoplastic disease of the spine. Radiol Clin North Am 1991;29:829–43.
[15] Wood NL, Coltman CA. Localized primary extranodal Hodgkin’s disease. Annu Int Med 1973;78:113–8.
[16] Drake RL. Lymphosarcoma involving the epidural space. J Kans Med Soc 1941;42:212–22.
[17] Haddad P, Thaell JF, Kiely JM, Harrison EG, Miller RH. Lymphoma of the spinal extradural space. Cancer 1976;38:1862–6.
[18] Verda DJ. Malignant lymphomas of the spinal epidural space. Surg Clin North Am 1944;24:1228–44.
[19] Blakslee GA. Compression of the spinal cord in Hodgkin’s disease: a case thirteen years duration with recession of symptoms following Roentgen-ray therapy. Arch Neurol Psychiatr 1928;20:130–7. [20] Cowie F, Benghiat A, Holgate C. Primary Hodgkin’s disease of bone.
Clin Oncol (R Colloid Radiol) 1991;3:233–5.
[21] Fried G, Ben Arieh Y, Haim N, Dale J, Stein M. Primary Hodgkin’s disease of the bone. Med Pediatr Oncol 1995;24:204–7.
[22] Gross SB, Robertson Jr WW, Lange BJ, Bunin NJ, Drummond DS. Primary Hodgkin’s disease of bone A report of two cases in adoles-cents and review of the literature. Clin Orthop 1992;283:276–80. [23] Wolcott WP, Malik JM, Shaffrey CI, Shaffrey ME, Jane JA.
Differ-ential diagnosis of surgical disorders of the spine. In: Benzel EC, editor. Spine surgery, vol. 1. Philadelphia: Churchill Livingstone; 1999. p. 25–52.
