Abstract
Two cases of infiltrative lipomas compressing the spinal cord and causing nonambulatory paraparesis in 2 large-breed dogs are reported. Magnetic resonance imaging (MRI) revealed severe extradural spinal cord compression by inhomogenous masses that infiltrated the adjacent tissues and the muscles of the spine in both dogs. The presumptive clinical diagnoses were infiltrative lipomas, which were confirmed by histopathology. In rare cases infiltrative lipomas are able to compress the spinal cord by the agressive growth of invasive adipocytes causing neurological deficits.
Résumé
Lipome infiltrant comprimant la colonne vertébrale chez 2 chiens de grande race. Deux cas de lipomes infiltrants comprimant la colonne vertébrale et causant une paraparésie non ambulatoire chez 2 chiens de grande race sont signalés. L’imagerie par résonance magnétique (IRM) a révélé une compression extradurale grave de la colonne vertébrale par des masses inhomogènes qui infiltraient les tissus adjacents et les muscles de la colonne vertébrale des 2 chiens. Les diagnostics cliniques présumés étaient des lipomes infiltrants, ce qui a été confirmé par histopathologie. Une croissance agressive des cellules adipeuses a causé les déficits neurologiques.
(Traduit par Isabelle Vallières)
Subcutaneous tissue adipocytes are considered to be the cells of origin of infiltrative lipomas and of well-demarcated lipomas. Both tumor types, therefore, have similar histological appearance (1). Typically, lipomas are benign neoplasms that have no tendency to metastasize. Nevertheless, they are able to invade adjacent tissue such as muscles and connective tissue, bones, and in rare cases peripheral nerves and the spinal cord (2,3). In general, lipomas are poorly defined and grow slowly but they can also exhibit periods of rapid growth (4). The definitive diagnosis of infiltrative lipoma can only be made by histological evaluation (5). It is not known why some lipomas develop a locally invasive behavior and form the infiltrative lipomas (2). The recurrence rate, even after aggressive surgical resection ranges from 36% to 50% in contrast to simple lipomas in which the local recurrence rate is < 2% (2,3,6). Therefore, adjuvant therapy such as radiation or chemotherapy is recommended (7,8).
This case report describes the neurological signs, magnetic resonance imaging (MRI), and histopathological findings in 2 large-breed dogs suffering from infiltrative lipomas compressing the spinal cord.
Case descriptions
Case 1
A 5-year-old, female, intact Bernese mountain dog was presented to the Department of Small Animal Medicine and Surgery of the University of Veterinary Medicine Hannover, Germany, with a history of falling from a first floor balcony missing a balustrade the night before. Since this accident the dog was unable to walk. The referring veterinarian treated the dog with glucocorticosteroids. In addition to the acute trauma, the dog had a history of progressive weakness of the hind limbs for the past 4 to 5 wk.
The physical examination was normal with the exception of an elevated body temperature (39.9°C). The neurological examination revealed a nonambulatory spastic paraparesis with accentuation to the left side and with severe deficits in the postural reactions of the hind limbs. The patellar reflex was decreased in both hind limbs. The postural reactions in the front limbs were unremarkable. The dog was not in pain when the spine was palpated. The cranial nerves were unremarkable. The neuroanatomical localization was L3–L4 because of the reduced patellar reflex and T3–L3 because of the spastic paraparesis. A chronic progressive disease such as neoplasia, degenerative myelopathy, or disc disease was suspected with a superimposed spinal cord trauma. The dog had a mild to moderate leukocytosis [20.8 × 109/L, reference interval (RI): 6.0 to 17.0 × 109/L] with neutrophilia 18.0 × 109/L (RI for segmented neutrophils: 3.0 to 11.5 × 109/L), a mild lymphopenia (0.97 × 109/L, RI: 1.0 to 4.8 × 109/L) and a monocytosis (1.7 × 109/L, RI: 0.18 to 1.35 × 109/L). The electrolyte profile showed a mild decrease in ionized calcium concentration (1.22 mmol/L, RI: 1.25 to 1.47 mmol/L). Total protein was mildly increased to 71.3 g/L (RI: 60 to 70 g/L). The base excess (BE) of the venous blood gas was −6.2 (RI: −4.0 to 4.0). The partial thromboplastin time (PTT) was 12.4 s (RI: 14.5 to 19.0 s).
Thoracic radiographs showed a mild radiolucency of the spinous process of Th9. A MR scan (Philips Achieva 3.0 T; Philips Healthcare, PC Best, The Netherlands) revealed an inhomogenous hyperintense area in T1-and T2-weighted sequences at the level of the 9th thoracic vertebra (Th9) extending from the right Musculus longissimus and M. multifidus to the spinal cord. The hyperintense area reached extradurally into the vertebral canal to the right side of the spinal cord and displaced the spinal cord to the left side. The adjacent muscles in the right paravertebral region, the M. longissimus thoracis, M. multifidus lumborum, M. spinalis et semispinalis thoracis et cervicis, and the M. intercostalis internus and externus showed hyperintense areas in T2- and T1-weighted sequences. The extent of the whole hyperintense lesion was 2.6 × 3.4 cm (length × width) with a height of 4.0 cm. The vertebral body of Th9 exhibited a central inhomogenous hyperintense area (Figures 1a, 1b). In a T2-weighted Spectral Adiabatic Inversion Recovery (SPAIR) sequence, which is a fat saturation sequence, all alterations appeared hypointense compared to the spinal cord and isointense compared to the subcutaneous fat tissue (Figure 2). After intravenous administration of contrast medium (Gadolinium 64Gd, “Dotarem®;” Guerbet, Roissy, France) the mass lesion showed a mild contrast enhancement. With these MR findings the infiltrating tissue was identified as fatty tissue and a tentative diagnosis of infiltrative lipoma was made.
Figure 1a.
T2-weighted sagittal magnetic resonance image of the thoracic portion of the vertebral column of the dog in case 1. A hyperintense mass is visible in the dorsal paraspinal region extending from Th8 to Th9 with consecutive compression of the spinal cord in this region. The spinal cord and the adjacent muscles are hyperintense compared to surrounding musculature (arrowhead) and spinal cord (star), respectively. The vertebral body of Th9 is characterized by a central hyperintense area (arrow).
Figure 1b.
T2-weighted transverse image of the mass in the region of Th9 of case 1. The hyperintense mass is compressing the spinal cord from the right to the left side (arrowhead). The adjacent tissue is also hyperintense (stars).
Figure 2.
T2-weighted transverse image with fat saturation (Spectral Adiabatic Inversion Recovery, SPAIR) of the mass in the region of Th9 of case 1. The mass (arrowhead) compressing the spinal cord (arrow) from the right side exhibits signal characteristics comparable to that of subcutaneous fat tissue (star). This finding suggests that the mass has a fatty origin.
Due to the guarded prognosis the owners elected euthanasia and an infiltrative lipoma was diagnosed at necropsy. In the region of Th9 to Th10 a focal, whitish-yellowish, soft tissue mass with dimensions of 4.0 × 1.0 × 0.4 cm (length × width × height) was detected that reached and broke into the vertebral canal and also infiltrated the neighboring muscles. The compression of the spinal cord resulted in degeneration in the region of Th9 to Th10. Histological examination of this region verified the macroscopic findings. The paravertebrally located lipoma focally infiltrated into the vertebral canal (Figure 3). Furthermore, acute hemorrhages were detected in the right axillar lymph node, the pancreas, and the lung in the macroscopic pathological examination. In the histological examination the right axillar lymph node and the lung showed mild, acute, multifocal hemorrhages. The pancreas had mild, acute, multifocal, interstitial hemorrhages. The hemorrhages were attributed to the acute trauma due to the falling from the balcony.
Figure 3.
Cross-section of Th10 after paramedian opening of the vertebral canal and removal of the spinal cord. The infiltrative lipoma is visible outside the vertebral canal lateral to the arcus vertebrae (arrowhead) and infiltrates into the vertebral canal (asterisk) through the vertebral arc (arrow). Hematoxylin and eosin (H&E), decalcified, bar 10 000 μm.
Case 2
A 10-year-old, male, entire crossbreed dog was presented to the Department of Small Animal Medicine and Surgery, University of Veterinary Medicine Hannover, Germany with a 10-day history of progressive gait abnormalities in the hind limbs finally leading to a nonambulatory paraparesis. The dog had received glucocorticosteroids 9 d prior to presentation as well as 200 mg carprofen (Rimadyl; Pfizer, Berlin, Germany) once a day for 7 d. Both popliteal lymph nodes were prominent and the urinary bladder was filled. The neurological examination revealed a nonambulatory paraparesis with reduced proprioception in both hind limbs. A mild hyperesthesia was noted when the dog was palpated at the thoracolumbar junction of the spine. The cranial nerves and the spinal reflexes were unremarkable. The neuroanatomical localization of the lesion was T3-L3 and a chronic progressive disease such as neoplasia or disc disease was suspected. The complete blood (cell) count (CBC) showed a mild leukopenia (5.3 × 109/L). Radiographs of the spinal column showed decreased widths of the intervertebral disc spaces Th10/11, Th11/12, and Th12/13. In T2-weighted MR images a hypointense mass was visible in the vertebral canal at Th10/11 with a length of 2 vertebral bodies, causing severe spinal cord compression. The adjacent muscles (M. spinalis and semispinalis, and M. multifidus) appeared inhomogeneous and hyperintense in the T2-weighted sequence. In the SPAIR sequence the mass itself was hypointense with hyperintense areas of the surrounding muscles and tissue (Figure 4). The compressing mass, the adjacent muscles and tissue showed a mild contrast enhancement (Figure 5).
Figure 4.
Transverse T2-weighted SPAIR-sequence in the region of Th10 of case 2. The spinal cord (arrow) is compressed by a hypointense mass to the right side (dotted line). The SPAIR-sequence decreases the signal intensity of the mass (dotted line) as the signal intensity of the subcutaneous fat is decreased (arrowheads). The adjacent musculature (star) is characterized by hyperintense signal intensity (compared with the surrounding musculature) which could be linked to the histopathologic diagnosis of myositis.
Figure 5.
Transverse T1-weighted magnetic resonance image of the thoracic region of Th10 of case 2 post contrast administration. Both the muscles and the mass show moderate contrast enhancement (arrowhead and star). The moderate compression of the spinal cord is obvious (arrow).
The owners elected euthanasia due to the guarded prognosis and declined necropsy. However, 3 tru-cut biopsies were permitted and sampled from the mass to confirm the diagnosis. The histopathological examination revealed well-differentiated skeletal muscle as well as adipocytes infiltrating the skeletal muscle fibers, accompanied by focal lymphohistiocytic cell infiltration (Figures 6a, 6b). An infiltrative lipoma causing secondary muscle degeneration and myositis in the surrounding muscles was highly suspected.
Figure 6a.
Histopathology of the tru-cut biopsy of the dog in case no. 2 showing well-differentiated adipocytes (arrows) infiltrating skeletal muscle fibers (star, H&E, bar 60 μm).
Figure 6b.
Histopathology of the tru-cut biopsy of the dog in case 2 at higher magnification showing infiltration of the skeletal muscle with well-differentiated adipocytes resulting in dissection of the fibers (H&E, bar = 30 μm).
Discussion
Infiltrative lipomas invading and compressing the spinal cord in dogs are rarely described (9–11). In both dogs described herein, infiltrative masses compressed the spinal cord in the thoracic region causing nonambulatory paraparesis. Both, MR scans and histopathology, identified the masses as infiltrative lipomas.
Lipomas are benign neoplasms that consist of localized nodules of fat and originate from adipocytes of the subcutaneous tissue. Infiltrative lipomas are locally invasive and may cause pain and/or clinical signs due to the compression of neighboring tissues (11,12). Unlike lipomas, infiltrative lipomas may cause pain upon palpation (1). In humans, pain occurs when the infiltrative lipoma compresses or involves adjacent neurovascular and/or muscular structures (13). Histologically, infiltrative lipomas and lipomas both consist of well-differentiated adipocytes (14). The dog in Case 2 showed a mild hyperesthesia when palpated at the thoracolumbar junction of the spine. This could have been caused by compression of the adjacent neurovascular structures and secondary inflammation of the muscles in this region. Both dogs were large-breed dogs, and, although infiltrative lipomas can occur in any breed, there seems to be a higher incidence in Labradors, doberman pinschers, and mixed breed dogs (11,12). Whereas 1 report stated that older female dogs may also be overrepresented (12) another claimed that infiltrative lipomas occur commonly in middle-aged dogs (15). Females are affected 3 times as often as males (12,15). The 2 dogs in this report did not have similar signalments: case 1 involved a middle-aged (5 y) female Bernese mountain dog, whereas case 2 was a relatively old (10 y) large male mixed-breed dog. Both dogs were not overweight, which is consistent with the thesis of Kramek that obesity does not appear to be a prerequisite for the disease (15).
The 2 patients reported here clearly demonstrate the strength of MR imaging as a diagnostic tool for soft tissue masses compressing the spinal cord such as infiltrative lipomas. Radiographs led to suspicion of a neoplastic lesion in 1 dog, but did not provide definitive findings. Computed tomography (CT) was not performed. In 1 report, an infiltrative lipoma invading the pelvic canal and tendon sheath in a dog was diagnosed by CT scan (9,5,16). In both cases described here the MR imaging showed high signal intensity (hyperintensity) of the masses in T1- and T2-weighted sequences. The muscles and the mass lesions were both characterized by high signal intensity. Noticeably, both neoplasms were in the distal thoracic region and without any indication, such as a lump or swelling, at the surface of the body. In case 1, the neoplasia was in the region of Th9/10 with severe extradural compression of the spinal cord to the left side over a length of 2.0 cm. The muscles involved showed a mild contrast enhancement. The dog in Case 2 had extradural compression of the spinal cord in the region of Th10/11. The adjacent paraspinal muscles were characterized by high signal intensities in T2-weighted images. In both dogs, these muscles showed a moderate contrast enhancement which could indicate local inflammation.
To corroborate these findings a fat saturation sequence (SPAIR) was chosen. The SPAIR is a powerful technique for fat suppression which offers advantages over conventional fat suppression techniques. This technique suppresses/inverts only fat spins and uses a spectrally selective adiabatic inversion pulse to invert the fat spins in the imaging volume. Due to this suppression the fat spins do not contribute to the MR signal. Thus, fat appears hypointense (17). In this SPAIR sequence both neoplasms had the same appearance as fat, supporting the tentative diagnosis of an infiltrative lipoma.
Unfortunately, the owners of both dogs elected euthanasia due to the guarded prognosis. There are 2 reports, however, of large-breed dogs with infiltrative lipomas compressing the spinal cord that underwent surgery and had satisfactory results (9,10). Those dogs did not show severe neurological deficits. Thus, surgery might be a therapeutic option for infiltrative lipomas compressing the spinal cord because it can be curative if the infiltrative lipoma can be completely excised. In a retrospective study of infiltrative lipomas in different parts of the body, Bergman et al (2) showed that only 5 dogs out of 13 had recurrence after aggressive surgical excision, while the other 8 dogs were cured. Among 16 tumors, 5 were located in the hind limb, 3 in the forelimb, 2 in the perianal area, 1 in the periocular region, 1 adjacent to the coxofemoral joint, 1 in the lateral thoracic area, 1 lateral to the mandible, 1 adjacent to the caudal thoracic vertebrae, and 1 in the ventrolateral abdominal wall (2). Despite the good prognosis, adjuvant therapy is advisable. Ogilvie et al (7) reported a dog treated with doxorubicin for an infiltrative lipoma with a documented partial response within 6 wk. Survival time of dogs treated with radiation alone, before and after surgery is reported by McEntee et al (8). The patient survival time ranged from 6 to 94 mo (median of 40 mo) and 92% of the dogs benefited from radiation therapy (8).
In the histopathological examination, mitotic figures were not detected in either case. These findings confirmed the diagnosis of an infiltrative lipoma as these neoplasms do not produce mitotic figures and are characterized by well-differentiated adipocytes.
These 2 cases of infiltrative lipomas compressing the spinal cord and causing severe neurological deficits show that MR imaging is a sensitive diagnostic tool for soft tissue neoplasms such as infiltrative lipomas when the spinal cord is involved (2). Spinal cord compression by infiltrative lipomas, as described here, is extremely rare. For surgical planning the MR imaging is irreplaceable, as the margins of the neoplasm can be depicted as precisely as possible by this technique. Magnetic resonance imaging can be helpful to distinguish lipomas from infiltrative lipomas and a special fat saturation sequence, such as the SPAIR sequence, can provide the evidence that the neoplasms were of fatty origin. CVJ
Footnotes
Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA office (hbroughton@cvma-acmv.org) for additional copies or permission to use this material elsewhere.
References
- 1.Thomas RC, Fox LE. Tumors of the skin and subcutis. In: Morrison WB, editor. Cancer in Dogs and Cats. Baltimore, Maryland: Lippincott, Williams and Wilkins; 1998. pp. 500–501. [Google Scholar]
- 2.Bergman PJ, Withrow SJ, Straw RC, Powers BE. Infiltrative lipoma in dogs: 16 cases (1981–1992) J Am Vet Med Assoc. 1994;205:322–324. [PubMed] [Google Scholar]
- 3.McChesney AE, Stephens LC, Lebel J, Snyder S, Ferguson HR. Infiltrative lipoma in dogs. Vet Pathol. 1980;17:316–322. doi: 10.1177/030098588001700305. [DOI] [PubMed] [Google Scholar]
- 4.Moulton JE. Tumors in Domestic Animals. 3rd ed. Berkeley, California: University of California Press; 1990. pp. 31–33.pp. 98 [Google Scholar]
- 5.McEntee MC, Thral DE. Computed tomographic imaging of infiltrative lipoma in 22 dogs. Vet Radiol Ultrasound. 2001;42:221–225. doi: 10.1111/j.1740-8261.2001.tb00928.x. [DOI] [PubMed] [Google Scholar]
- 6.Strafuss AC, Smith JE, Kennedy GA, Dennis SM. Lipomas in dogs. J Am Anim Hosp Assoc. 1973;9:555–561. [Google Scholar]
- 7.Ogilvie GK, Reynolds HA, Richardson RC, et al. Phase II evaluation of doxorubicin for treatment of various canine neoplasms. J Am Vet Med Assoc. 1989;195:1580–1583. [PubMed] [Google Scholar]
- 8.McEntee MC, Page RL, Mauldin GN, Thrall DE. Results of irradiation of infiltrative lipoma in 13 dogs. Vet Radiol Ultrasound. 2000;41:554–556. doi: 10.1111/j.1740-8261.2000.tb01889.x. [DOI] [PubMed] [Google Scholar]
- 9.Morgan LW, Toal R, Siemering G, Gavin P. Imaging diagnosis-Infiltrative lipoma causing spinal cord compression in a dog. Vet Radiol Ultrasound. 2007;48:35–37. doi: 10.1111/j.1740-8261.2007.00201.x. [DOI] [PubMed] [Google Scholar]
- 10.O’Driscoll JL, McDonnell JJ. What is your neurologic diagnosis. J Am Vet Med Assoc. 2006;229:933–935. doi: 10.2460/javma.229.6.933. [DOI] [PubMed] [Google Scholar]
- 11.Kim HJ, Chang HS, Choi CB, et al. Infiltrative lipoma in cervical bones in a dog. J Vet Med Sci. 2005;67:1043–1046. doi: 10.1292/jvms.67.1043. [DOI] [PubMed] [Google Scholar]
- 12.Goldschmidt M, Shofer F. Skin tumors of the dog and cat. Oxford, United Kingdom: Pergamon Press; 1992. [Google Scholar]
- 13.Austin RM, Mack GR, Townsend CM, et al. Infiltrating (intramuscular) lipomas and angiolipomas. Arch Surg. 1980;115:281–284. doi: 10.1001/archsurg.1980.01380030031007. [DOI] [PubMed] [Google Scholar]
- 14.Thomson MJ, Withrow SJ, Dernell WS, Powers BE. Intermuscular lipomas of the thigh region in dogs: 11 cases. J Am Anim Hosp Assoc. 1999;35:165–167. doi: 10.5326/15473317-35-2-165. [DOI] [PubMed] [Google Scholar]
- 15.Kramek BA, Spackman CJA, Hayden DW. Infiltrative lipoma in three dogs. J Am Vet Med Assoc. 1985;186:81–82. [PubMed] [Google Scholar]
- 16.Pereira JM, Sirlin CB, Pinto PS, Casaola G. CT and MR imaging of extrahepatic fatty masses of the abdomen and pelvis: Techniques, diagnosis, differential diagnosis, and pitfalls. RadioGraphics. 2005;25:69–85. doi: 10.1148/rg.251045074. [DOI] [PubMed] [Google Scholar]
- 17.Lauenstein TC. Spectral Adiabatic Inversion Recovery (SPAIR) MR imaging of the Abdomen. MAGNETOM Flash 2/2008. [Last accessed November 5, 2012]. Available from www.siemens.com/magnetom-world.