A Case Report of an Uncommon Cause of Cauda Equina Symptoms

CASE REPORTS A CASE REPORT OF AN UNCOMMON CAUSE CAUDA EQUINA SYMPTOMS OF Annabel Kier, DC,a Martin D. Timchur, DC,a and Peter W. McCarthy, PhDb ABSTRACT Objective: This case report discusses a patient who presented with right-sided buttock pain of apparently uncomplicated mechanical origin that was eventually diagnosed as a primary Ewing sarcoma/primitive neuroectodermal tumor of the sacrum. Clinical Features: A 32-year-old male full-time student presented for care with right-sided buttock pain. Intervention and Outcome: After examination, the patient was referred to his general practitioner for urgent magnetic resonance imaging, the report revealed no explanation for the presenting symptoms. After further imaging and biopsy, an eventual diagnosis of Ewing sarcoma/primitive neuroectodermal tumor was reached. The patient died 12 months later. Conclusion: This case highlights a nondiscal cause for cauda equina symptoms. It emphasizes potential diagnostic complexities that may present due to preconceptions based upon the probability of symptoms being related to a specific disease process. (J Manipulative Physiol Ther 2007;30:459-465) Key Indexing Terms: Chiropractic; Sarcoma; Ewing; Neuroectodermal Tumors; Primitive; Peripheral; Urinary Bladder; Neurogenic; Cauda Equina L ow back pain with or without associated buttock and/or leg symptoms is one of the most common symptom presentations observed in chiropractic clinics.1,2 The case presented here follows the progression of what initially appeared to be a sacroiliac joint syndrome through to the development of cauda equina compression signs and the eventual diagnosis of Ewing sarcoma/ primitive neuroectodermal tumor (PNET). CASE REPORT Presentation In late November 2004, a 32-year-old full-time student presented to the Welsh Institute of Chiropractic clinic complaining of a 6-week history of right-sided buttock pain a Senior Lecturer, Welsh Institute of Chiropractic (WIOC), University of Glamorgan, Pontypridd CF37 1DL, Wales, UK. b Reader, Welsh Institute of Chiropractic (WIOC), University of Glamorgan, Pontypridd CF37 1DL, Wales, UK. Submit requests for reprints to: Annabel Kier, DC, Senior Lecturer, Welsh Institute of Chiropractic (WIOC), University of Glamorgan, Pontypridd CF37 1DL, Wales, UK (e-mail: [email protected]). Paper submitted December 13, 2006; in revised form March 9, 2007; accepted April 16, 2007. 0161-4754/$32.00 Copyright D 2007 by National University of Health Sciences. doi:10.1016/j.jmpt.2007.04.012 of insidious onset. The pain was initially of a dull achy nature and progressed to a burning sensation (5/10, visual analogue scale). The intensity of the pain had apparently not increased; however, associated symptoms had developed, including progressive nocturnal pain, daily sharp pain, and numbness in the right testicle (2 weeks before consultation) and intermittent sharp shooting pain from the right buttock to the sole of the ipsilateral foot (3 weeks before consultation). Since the onset, numbness had been noted in the right buttock but later developed perianally. During the 6-week duration of symptoms, urinary and bowel retention had manifested progressively. In particular, micturition had been affected to the degree that no sense of bladder fullness was noted, and the need to urinate only registered once the bladder became so distended it was painful. In addition, urination could only be initiated and maintained while sitting and applying manual pressure over the bladder. Furthermore, penile numbness and occasional impotence were reported to have occurred over the previous 5 weeks. Prolonged static postures, notably sitting, were reported to be aggravating factors. Non-steroidal anti-inflammatory drugs and cryotherapy achieved short-term relief only. There was no loss of appetite or weight, no feeling of general malaise, no indication of night sweats, and no fever, but he reported feeling tired and attributed this to a disturbed sleep pattern and fatherhood. The patient mentioned that he was under some stress due to his studies and being a father of 3 459 460 Kier et al Non-Classic Cauda Equina Journal of Manipulative and Physiological Therapeutics July/August 2007 Fig 1. Axial T1-weighted (A and B) and axial FSE proton density fat saturated contrast (C and D) MRI sequences of the upper and lower pelvis demonstrating a large infiltrative lesion of the sacrum (arrows). small children. There was a family history of malignant tumors; two of his grandparents had been diagnosed with cancer (bone/breast). Previous History of Complaint The patient had intermittent episodic right buttock pain over the previous 4 years with a variable pain intensity of 3-7/ 10. No paraesthesia, anesthesia, muscle weakness, or bowel and bladder changes had been noted previously. Each episode of stiffness and/or pain that he had sought treatment for had apparently responded well to spinal manipulative therapy. Examination The patient’s blood pressure was 130/90 mm Hg bilaterally, pulse rate was 70 beats/min, and respiration rate was 14 breaths/min. Axillary temperature (36.38C) suggested a normal core temperature. No abnormality was detected during the full lymph node screen. The right patella, posterior tibialis, and Achilles reflexes were absent, whereas left-sided reflexes were +2 (normal). Babinski and Clonus responses were absent (normal). Although myotomes were +5 (normal), the patient perceived weakness of the right gluteal muscles. Complete loss of sensation to sharp and soft touch was detected in S2 and S3 dermatomes and severely reduced in S4. All orthopedic screening tests were negative except for straight leg raise, which reproduced the right-sided buttock pain at 408. Based upon bowel and bladder involvement and sensory deficit, a diagnosis of cauda equina syndrome was established. Follow-Up The patient was referred to his general medical practitioner (GP) for further evaluation. Magnetic resonance imaging (MRI) of the lumbar spine performed within 24 hours revealed some degeneration of the L5/S1 disk with a very small right paracentral bulge just touching the right S1 nerve root. No evidence of cauda equina compression was reported wherefore the hospital placed the patient on the waiting list for a consultant neurologist appointment. Further degeneration of the patient’s condition prompted the patient to obtain a referral for an urgent private urologic consultation, which identified the need for urinary ultrasound scan with flow test and residual volume testing, for which he was placed upon the waiting list. During December 2004, the patient reported development of night sweats and increased pain level and was prescribed an oral solution of morphine sulphate, metoclopramide for nausea, and a stimulant laxative for constipation. Journal of Manipulative and Physiological Therapeutics Volume 30, Number 6 Kier et al Non-Classic Cauda Equina necrotic material failing to produce a firm diagnosis. A subsequent laparotomy and pelvic biopsy were then performed (February 2005), and the patient was referred to a local oncology hospital. The macrodescription from the biopsy revealed multiple pieces of firm gray and brown tissue. Microscopy showed fibrous and adipose tissue, with fragments of cellular tumor showing extensive necrosis. The tumor consists of small cells, with scant cytoplasm and only occasional nuclear grooves and mitoses. The nuclei were mildly pleomorphic with fine granular chromatin. Immunocytochemistry showed that a range of tissue markers were negative, including S100 and synaptophysin (potential markers for neural tissue) and desmin (a marker for muscle). However, the immunocytochemistry showed focal, strong, positive staining of lesional cells for CD99 and Bcl-2. This profile is suggestive of Ewing sarcoma/PNET and was supported by the finding of ESWR1 (22q12) gene rearrangement as identified by fluorescence in situ hybridization.3,4 Over the course of the following months, metastatic dissemination in the cervical spine, femur, and lungs ensued, and despite chemotherapy and radiotherapy treatment, the patient died in March 2006, 12 months after the diagnosis. Fig 2. Isotope bone scan demonstrating a photopoenic area in the right side of the sacrum with peripheral accumulation of the radiopharmaceutical agent (white arrow). Several additional areas of radiotracer activity are identified (black arrows). In mid January 2005, a neurological consultation concluded a sacral radiculopathy with need for cerebrospinal fluid testing, neurophysiologic examination, and a wide range of blood tests. Blood tests revealed a hypochromic microcytic anemia, indicating an iron deficiency. In February 2005, the patient represented to the our clinic with a tender swelling over the right sacroiliac and upper posterolateral gluteal area that had appeared and enlarged over the previous 2 to 3 days. The mass measured 3.5  4.5 in, was tender, fixed, firm, and had palpable margins. The patient was urgently referred to his GP and readmitted to the department of neurology where a digital x-ray and MRI (Fig 1) of the pelvis, plus bone scintigraphy (Fig 2), were performed. The x-ray report stated the film to be normal for the patient’s sex and age. In particular, it was emphasized that the right sacroiliac joint (site of complaint) was not destroyed, and therefore, a diagnosis of infection was considered unlikely. The MRI revealed an infiltrative lesion of the right side of the sacrum with a large soft tissue mass extending anteriorly through the sciatic notch along the sciatic nerve and out through the sacrum posteriorly (Fig 1). The bone scan (Fig 2) identified a photopoenic mass with peripheral uptake corresponding with the lesion in the sacrum; in addition, other lesions were present, which suggested metastasis. A soft tissue core biopsy was performed for identification. Histology from this initial biopsy showed inflammatory and DISCUSSION Cauda equina syndrome is an acute neuropathy that, by affecting the smaller nerve fibers or autonomic fibers, leads to varying degrees of urinary and fecal incontinence, localized sensory loss in the perineal area, and varying degrees of leg weakness.5,6 Although motor and/or sensory symptoms caused by the compression of exiting nerve roots are common symptoms associated with cauda equina syndrome, Chang et al7 emphasize the importance of weakness of the gluteal muscles, saddle-type hypoesthesia or anesthesia, and urinary retention. Bladder symptoms can occur without motor weakness in the lower extremities, and classic bilateral saddle-type hypoesthesia in cauda equina syndrome suggests bilateral involvement of the lower sacral roots, indicating impairment of the normal reflex arc of the sacral spine. This, when considered in the context of the bulbocavernosus reflex (S2-S3), implies that the latter reflex can be used in an attempt to differentiate between strictly urological and neurogenic disturbances of micturition.7,8 However, because the imaging report in early December 2005 indicated no evidence of cauda equina compression, other differential diagnoses needed to be considered. With respect to the symptoms, 2 possibilities had to be contemplated: (a) all symptoms were related, and one underlying cause was being sought; or (b) all symptoms were not related and appeared simultaneously; therefore, more than one cause had to be sought. 461 462 Kier et al Non-Classic Cauda Equina Journal of Manipulative and Physiological Therapeutics July/August 2007 3. The denervated bladder (autonomic bladder). Here, both components of the reflex arc are interfered with either by a peripheral lesion or by damage to the spinal bladder center in the conus terminalis. A flaccid, distended bladder develops. This condition may develop after trauma, infection, spinal arachnoiditis, or it may accompany spina bifida as well as tumors in the spinal canal. 4. Reflex neurogenic bladder (automatic bladder). The sacral centers and efferent and afferent fibers are intact. The suprasegmental reflex arc is interrupted anatomically or functionally. This can be seen in spinal cord lesions above the level of the conus, in multiple sclerosis, pernicious anemia, and in normal infants. The emptying of the bladder occurs as a reflex and in spite of not being initiated or interrupted voluntarily, it can be started by certain maneuvers. Fig 3. Bladder function and control centers. As the symptom of primary concern to the patient at the time of initial presentation was the lack of urinary control, the following discussion will focus on this topic. Symptoms of leg pain, reflex changes, and associated saddle paraesthesia or anesthesia are well documented in the literature.9-11 Micturition in a normally developed adult is a spinal reflex that can be released or inhibited by higher (cerebral) centers, exerting action through the autonomic pathways in the lateral columns of the spinal cord and is therefore subject to voluntary control. Bladder function and control is a multifactorial finely tuned neurological process (Fig 3). It is essential, should bladder dysfunction occur, that all neurological aspects mentioned below are considered, in addition to kidney pathology and obstruction local to the detrusor muscle and the urethra. The following 4 possibilities should be considered8: 1. The deafferentated bladder (sensory paralytic bladder) will have partial or complete sensory loss where the lesion is located in the dorsal nerve roots. The bladder overfills, becomes hypotonic, and develops thinning of its walls. 2. The deefferentated bladder (motor paralytic bladder) where the motor component of the reflex arc is interrupted. At the time of presentation, the patient had lost both bladder sensation and the ability to void voluntarily. As a consequence, eliciting the exact progression of this aspect of his symptoms proved to be unachievable. Although the symptoms were of serious concern and thought to be caused by pathology either in the area of the sacrum or from an unusual presentation of multiple sclerosis, arriving at a final diagnosis regarding the urinary difficulties was considered as being outside of the scope of chiropractic practice. This resulted in a series of discussions with the patient’s GP, who in early December 2004 requested urgent referral to a neurologist, an appointment that could not be achieved until January of 2005. Magnetic resonance imaging of the brain was requested privately to reduce the possibility of there being any obvious higher center involvement. This investigation revealed no findings capable of explaining the symptoms. A private consultant urologist appointment was established in mid December 2004, which revealed the need for further investigation. Unfortunately, there was a waiting list for the required tests. The patient was urged to attend the accident and emergency department to attempt to hasten the investigation, a path this patient never chose to take. Records have been obtained from other practitioners consulted by this patient dating back to 2000 (Table 1). After his demise, conversations with the patient’s spouse revealed that at times the patient was not completely forthcoming regarding the pain intensity. She stated, bHe could be in tears the night before, yet during a consultation the following day claimed that the pain had not been too severe.Q In addition, night sweats had started in October 2005, although documentation indicates that the patient did not reveal such symptoms. Details like these highlight the reliance upon a patient’s ability and willingness to honestly answer questions that he or she either finds embarrassing or know could indicate more serious and complicated conditions. This is an issue all primary practitioners may face and identifies the value of Journal of Manipulative and Physiological Therapeutics Volume 30, Number 6 Kier et al Non-Classic Cauda Equina Table 1. Overview of symptoms and treatment before presentation to authors Episode Symptom presentation Examination findings Treatment Outcome October 2000 Lumbar stiffness NAD diagnosis: musculoskeletal SMT 2 treatments Patients stated he is fine November 2003 1-d history of right-sided buttock pain. Onset: noticed upon waking NAD diagnosis: musculoskeletal SMT 2 treatments Pain improving. Patient did not return for further treatment. November 2003 8-d history of sharp right-sided buttock pain NAD diagnosis: musculoskeletal SMT 1 treatment Patient stated he is fine. March 2004 10-d history of sharp right-sided buttock pain. Subjective numbness noted in area first 4 days since subsided NAD SLR 758 bilaterally, right Achilles reflex considered mildly less brisk than left, diagnosis: musculoskeletal SMT 2 treatments Patient stated no pain, stiffness, or numbness October 2004 7-d history of intermittent 3/10 right-sided buttock pain with irregular sleep disturbance NAD except for reproduction of buttock pain on right SLR 508, diagnosis: musculoskeletal SMT 4 treatments over the following 3 weeks Patient is urged to seek consultation with general medical practitioner as urinary retention develops. November 2004 6-wk history of right-sided sciatica Signs and symptoms of cauda equina None Patient referred. NAD indicates no abnormality detected; SMT, spinal manipulative therapy; SLR, straight leg raise. requesting the presence of a partner, spouse, or family member to confirm certain details where appropriate.16 It is considered that few low back cases (1%-2%) have significant underlying pathology of this nature.17,18 In reality, this equates to 1 significant pathology per 5 to 10 weeks of practice in a clinic that has 10 new patients with low back pain per week. Most significantly, however, is that (as in this case) serious pathology can mimic nonserious conditions until the disease process has progressed, and symptoms alert the patient and/or the practitioner. Multiple Ewing sarcoma family tumor (ESFT)19,20 cases are documented in the literature, where the initial symptom presentation was insidious onset of buttock pain mimicking sacroiliac joint irritation. Symptoms subsequently progressed gradually over time to include sciatica with or without cauda equina compression signs, remarkably comparable to this case.19-24 Likely, primary symptom generator(s) can, in most cases, be determined by a thorough case history, physical examination, and appropriate imaging.17 It is not the remit of this case report to elaborate on the differential diagnosis of low back pain, but Figure 4 outlines potential red flags. Ewing sarcoma/PNETs are small round cell tumors that have been termed peripheral PNET/ESFTs and include Ewing sarcoma of the bone, extraosseous Ewing sarcoma, pPNET (peripheral neuroepithelioma), and Askin tumor.19 Ewing sarcoma and PNETs have a number of characteristics in common and, for diagnostic purposes, are now considered to be part of the same family of tumors, the so-called ESFTs.19,20 However, they have been previously referred to as separate entities, which can lead to confusion. Ewing sarcomas are the fourth most common primary malignant tumor of bone.4 The characteristic clinical presentation is one of a localized palpable mass with or without pain in the involved area. Fever (remittent 388C), anemia, weight loss, leukocytosis, and increased sedimentation rate are commonly noted (which may mimic infection).26-28 The tumors are aggressive, metastasizing to lung, bone, and bone marrow.19 The primary skeletal locations common to these tumors include the long tubular bones followed by the pelvis/sacrum, ribs, and less so the skull, vertebra, scapula, and short tubular bones.27,29 Physeal plates and joint spaces generally act as a barrier to infiltration of tumors. Consequently, when joint involvement is present, infection ranks highly in the list of differential diagnoses. However, joint destruction and transarticular spread may be seen in some aggressive tumors,30 including cases of Ewing sarcoma.23 Although Ewing sarcomas usually originate in bone, primary soft tissue sarcomas can also occur. Common soft tissue locations include the chest wall and paravertebral extradural locations, but the extremities, retroperitoneum and large peripheral nerves such as the brachial plexus and sciatic nerve also include preferential sites.19 Ninety percent of ESFTs occur in the first 3 decades of life.26-28,31 There appears to be a slight male predominance in Ewing cases, and an equal distribution or slight female predilection for PNET.3,28 Ewing sarcoma has a distinct preponderance for whites (95%).3,4,28 The survival rate appears to be inversely dependant upon the presence of metastatic disease at presentation, primary extraosseous tumor, and age being 26 years or more. In 463 464 Kier et al Non-Classic Cauda Equina Journal of Manipulative and Physiological Therapeutics July/August 2007 pain relief is initially gained does not alone eliminate the possibility of sinister pathology.24,32 Progression of severity and/or frequency of symptoms should alert the practitioner and trigger reexamination, which may warrant further investigation. It may at times be pertinent to confirm various aspect of the presentation with the patient’s partner or spouse. In this case, the patient had visited various practitioners over a period of time, thus creating certain difficulties, a situation that emphasizes the need to request patient notes from colleagues or other healthcare practitioners, a task which may be problematic as the patient may not disclose the fact. Where seemingly straightforward cases repeatedly present, the physician should consider the wider differentials, including uncommon causes of symptoms. In closing, one can only conclude that vigilance must be maintained and the clinician should always be prepared to expect the unexpected. Practical Application Fig 4. Red flags: potential signs of serious spinal pathology in low back pain.17,25 particular, pelvic origin in adults is associated with higher mortality,26,29 a consideration based upon the notion that tumors can remain silent and detection only occurs once the growth is large enough to cause symptoms. The classic radiographic appearance of ESFT in a long bone resemble that of an osteolytic poorly defined area of permeative bone destruction in a diaphyseal location demonstrating cortical destruction with a laminated periosteal reaction and soft tissue mass. In some cases, a more aggressive, spiculated periosteal reaction is observed (ie, the trimmed whiskers effect).4,28 An uncommon, but classic radiographic finding is an obliquely oriented cortical erosion known as bcortical saucerization.Q A spinal location may lack the classic features of long bone Ewing sarcoma but still shows the aggressive osteolytic nature of the lesion with poorly defined margins and soft tissue extension. Magnetic resonance imaging is the best method to adequately identify the degree of soft tissue extension, marrow changes, and is used for staging the tumor.4,28 Bone scintigraphy is the most useful investigation for establishing metastatic bone disease. CONCLUSION Although uncommon, serious pathology may appear in chiropractic clinical practice. The importance of attention to case history, comorbidities, and associated signs and symptoms cannot be overemphasized. The mere fact that ! Initial pain relief may be gained from SMT—this does not eliminate the possibility of underlying serious pathology ACKNOWLEDGMENT The authors thank the following people for their assistance with this manuscript: Dr O. Tilsley and staff (Velindre Hospital), and Dr Robinson and staff (Heath Hospital). We would also like to thank S. Hajdu, MD for information on Ewing sarcoma. Our thoughts and thanks got to G. Harding for permitting his data to be utilized for the manuscript as well as to N. Harding for her review of the manuscript, assistance in data collection, and continual support. REFERENCES 1. Dagenais S, Haldeman S. Chiropractic. Prim Care 2002;29: 419-37. 2. Wilson FC. A survey of chiropractors in the United Kingdom. Eur J Chiropr 2003;50:185-8. 3. Hajdu SI. A note from history: the enigma of Ewing’s sarcoma. Ann Clin Lab Sci 2006;36:108-10. 4. Yochum TR, Rowe L. Yochum and Rowe’s essentials of skeletal radiology. 3rd ed. Baltimore, MD7 Williams and Wilkins; 2005. p. 1200-5. 5. Bilir A, Gulec S. Cauda equina syndrome after epidural steroid injection: a case report. J Manipulative Physiol Ther 2006;29:6. 6. Crowther ER. Slow onset cauda equina syndrome: a case report. J Can Chiropr Assoc 1993;37:203-9. 7. Chang HS, Nakagawa H, Mizuno J. Lumbar herniated disc presenting with a cauda equina syndrome. Long-term followup of four cases. Surg Neurol 2000;53:100-4. Journal of Manipulative and Physiological Therapeutics Volume 30, Number 6 8. Mumenthaler M, Mattle H, Taub E. Neurology. 4th ed. Thieme, 2004. p. 395-9. 9. Chen HJ, Liang CL, Lu K, Liliang PC, Tsia YD. Cauda equina syndrome caused by delayed traumatic spinal subdural haematoma. Injury 2001;32:505-7. 10. Casey ATH. Spinal cord and root compression. Medicine 2004;32:103-5. 11. Orendacova J, Cizkova D, Kafka J, Lukacova N, Marsala M, Sulla I, et al. Cauda equina syndrome. Prog Neurobiol 2001;64: 613-37. 12. Bharat R, Kavia C, Dasgupta R, Fowler CJ. Functional imaging and the central control of the bladder. J Comp Neurol 2005;493:27-32. 13. Blok BFM. Central pathways controlling micturition and urinary incontinence. Urology 2002;59(Suppl 5A):13-7. 14. Sugaya K, Nishijima S, Miyazato M, Ogawa Y. Central nervous control of micturition and urine storage. J Smooth Muscle Res 2005;41:117-32. 15. Fowler CJ. Bladder afferents and their role in the overactive bladder. Urology 2002;59(Suppl 5A):37-42. 16. Kier A, McCarthy PW. Cerebrovascular accident without chiropractic manipulation: a case report. J Manipulative Physiol Ther 2006;29:330-5. 17. Waddell G. Editorial: subgroups within bnon-specificQ low back pain. J Rheum 2005;32:395. 18. L&rum E, Dullerud R, Kirkesola G, et al. Acute low back pain interdisciplinary clinical guidelines. By: Norwegian Back Pain Network. Division of medicine - KGR. Ullev3l Hospital. N-0407 Oslo, Norway; 2002. 19. Mobley B, Roulston D, Shah GV, Bijward KE, McKeever PE. Peripheral primitive neuroectodermal tumour/Ewing’s sarcoma of the craniospinal vault: case reports and review. Hum Pathol 2006;37:845-53. 20. Isefuku S, Seki M, Tajino T, et al. Ewing’s sarcoma in the spinal nerve root: a case report and review of the literature. Tohoku J Exp Med 2006;209:369-77. Kier et al Non-Classic Cauda Equina 21. Hashimoto M, Akabane Y, Tate E. Ewing’s sarcoma of the sacrum. Radiat Med 1999;17:451-3. 22. Modrego PJ, Pina MA. Extraskeletal Ewing’s sarcoma presenting with neurological symptoms. Report of two cases. Neurol Sci 2001;22:257-9. 23. Drnovsek V, Zafiroski G, Byron IG, Plavsic BM. Transarticular spread of Ewing’s sarcoma across the sacroiliac joint: CT and MRI correlation. Orthopedics 1999;22:977-9. 24. Baker N, Dorfman D. Ewing’s sarcoma of the sacrum. Skeletal Radiol 1996;25:302-4. 25. Shimada Y, Miyakoshi N, Kasukawa Y, Hongo M, Ando S, Itoi E. Clinical features of cauda equina tumours requiring surgical treatment. Tohoku J Exp Med 2006;200:1-6. 26. Baldini EH, Demetri GD, Fletcher CDM, Foran J, Marcus KC, Singer S. Adults with Ewing’s sarcoma/primitive neuroectodermal tumour. Adverse effect of older age and primary extraosseous disease on outcome. Ann Surg 1999;230:79. 27. Soulard R, Claude V, Campara P, Dufau JP, Saint-Blancard P, Gros P. Primitive neuroectodermal tumour of the stomach. Arch Pathol Lab Med 2004;129:107-10. 28. Resnick D. Diagnosis of bone and joint disorders. 4th ed., vol. 4. Philadelphia, PA: WB Saunders Company; 2002. p. 4060-73. 29. Ushigome S, Machinami R, Sorensen PH. Ewing sarcoma/ primitive neuroectodermal tumour (PNET). In: Fletcher CDM, Unni KK, Mertens F, editors. World Health Organization classification of tumours. Pathology and genetics of tumours of soft tissue and bone7 IARC PressLyon; 2002. p. 298-300. 30. Abdelwahab IF, Miller TT, Hermann G, Klein MJ, Kenen S, Lewis MM. Transarticular invasion of joints by bone tumours: hypothesis. Skeletal Radiol 1991;20:279-83. 31. Ilaslan H, Sundram M, Dekutoski MB. Primary Ewing’s sarcoma of the vertebral column. Skeletal Radiol 2004;33: 506-13. 32. Webster M. Sacrococcygeal chordoma in chiropractic practice; a radiology case report. Eur J Chiropr 2003;51:173-8. 465