Cột sống ngực thắt lưng ppt

Childhood Diskitis Sean D. Early, MD, Robert M. Kay, MD, and Vernon T. Tolo, MD Abstract Childhood diskitis represents one end of a continuum of spinal infections, from diskitis to vertebral osteomyeli- tis with soft-tissue abscess. It has been recognized for decades in the pedi- atric population, with consistent dis- agreement as to the etiology and op- timal treatment. 1-14 Diskitis is now generally accepted as a bacterial in- fection involving the disk space and adjacent vertebral end plates. Opti- mal management requires the use of intravenous antibiotics. Bracing does not seem to improve the clinical course but may be indicated with ver- tebral involvement to decrease pain and minimize deformity. Biopsy is not necessary for patients who exhib- it all of the clinical characteristics, but it is indicated for those whose symp- toms do not resolve rapidly with treatment or who have atypical pre- sentation. The unique anatomy of the pedi- atric spine has implications for the etiology and pathophysiology of dis- kitis. To effectively diagnose and treat patients with diskitis, familiarity with its unique presentation is necessary to differentiate it from classic osteo- myelitis and other entities with sim- ilar symptoms. Pathogenesis As with other pediatric musculo- skeletal infections, the most common cause of diskitis is the hematogenous spread of infection. Preceding or con- comitant infections, such as otitis me- dia, urinary tract infections, and re- spiratory infections, are common. Wiley and Trueta 15 showed that nu- trient arteries are a more direct route for the spread of disease to the ver- tebral column than is the paraverte- bral venous system. They proposed that pyogenic infections of the spine result from septic emboli in the arte- rial system. This is similar to the patho- physiologic development of meta- physeal osteomyelitis in pediatric long bone infections. Although rare, direct inoculation (after trauma or sur- gery) or contiguous spread from an adjacent infection can cause diskitis in children. As in adults, the intervertebral disk is avascular in children. However, unique anatomic features of the de- veloping motion segment predispose the pediatric spine to disk space in- fections rather than to vertebral os- teomyelitis. The disk is located be- tween the hyaline cartilage end plates covering the adjacent immature ver- tebral bodies. The cartilaginous plates contain numerous canals that act as channels through which small vessels travel and terminate adjacent to the intervertebral disk. 16-18 These vascu- lar channels appear before the 16th week of gestation and persist until the ring apophyses fuse in the third de- Dr. Early is Assistant Professor, Division of Or- thopaedic Surgery, Childrens Hospital Los Ange- les and University of Southern California–Keck School of Medicine, Los Angeles, CA. Dr. Kay is Assistant Professor, Division of Orthopaedic Sur- gery, Childrens Hospital Los Angeles and Uni- versity of Southern California–Keck School of Medicine. Dr. Tolo is Professor and John C. Wil- son Chair of Orthopaedic Surgery, Childrens Hos- pital Los Angeles and University of Southern Cal- ifornia–Keck School of Medicine. None of the following authors or the departments with whichthey are affiliated has received anything of value from or owns stock in a commercial com- pany or institution related directly or indirectly to the subject of this article: Dr. Early, Dr. Kay, and Dr. Tolo. Copyright 2003 by the American Academy of Orthopaedic Surgeons. Childhood diskitis may occur in the thoracic, lumbar, or sacral spine and can affect children of all ages, but it is most common in the lumbar region in children younger than 5 years. Physical examination, laboratory tests, and radiologic studies all aid in the diagnosis of this clinical syndrome, and proper use can prevent unnecessary invasive intervention. Presentation varies with age; the child may refuse to bear weight on the lower extremities or may present with back pain, abdominal pain, a limp, or, if an infant or toddler, with irritability. The etiology appears to be a bac- terial infection, usually caused by Staphylococcus aureus. Most children improve rapidly with a 4- to 6-week course of antibiotics. Although not routinely necessary, immobilization decreases symptoms and, in the case of osseous destruction, prevents progression of spinal deformity. Biopsy of the infected disk space is reserved for chil- dren refractory to intravenous antibiotics. Follow-up should include plain radio- graphs at regular intervals for 12 to 18 months to ensure resolution of the destruc- tive process. J Am Acad Orthop Surg 2003;11:413-420 Vol 11, No 6, November/December 2003 413 cade of life. The vascular channels ap- pear to be an important source of nu- trition for the developing disk but also can deliver blood-borne bacteria to the disk during periods of bacter- emia. 19 Because of the relative avas- cularity of the intervertebral disk, bac- teria that reach the disk are relatively free from the immune system and are more likely to flourish. Differences in the vascular anato- my of the vertebral bodies in children and adults 20 result in differences in spinal infections between pediatric and adult populations. Infants and small children have widespread anas- tomoses between intraosseous arter- ies of the vertebral body; the anasto- moses decrease notably by age 15 years and disappear by adulthood. These intraosseous anastomoses de- crease the risk of vertebral body in- farction and simultaneously enhance the clearance of bacteria after a sep- tic embolus in a pediatric metaphys- eal artery. This enhanced collateral circulation in the young child there- by decreases the risk of pyogenic ver- tebral body infection. Bacteria that reach the immature motion segment through hematoge- nous spread can be deposited in the vertebral body, hyaline cartilage plate, or disk. 19 Because of the redun- dant vascular supply in the childhood vertebral body, infarction and subse- quent vertebral osteomyelitis are less likely than they would be in adults. Instead, bacteria are more likely to be deposited at the superior or inferior metaphyseal equivalent, which is ad- jacent to the vertebral end plate. 18 These bacteria then may travel through the arterioles and sinusoidal capillaries of the cartilage canals, to be deposited at either the avascular disk or the adjacent vertebral body. Should septic emboli result in oc- clusion of these cartilage canals, bac- teria may reside relatively free from host defense mechanisms in the avascular disk, and infection may develop. The adjacent vertebral bod- ies often are spared because of their hyaline cartilage–capped end plates and abundant intraosseous blood supply. With infection of the disk space, the normal stiffness of the an- ulus fibrosus may be altered by deg- radation from bacterial and host de- fense enzymes, resulting in the characteristic disk space narrowing seen on plain radiographs (Fig. 1). As the infection progresses over a period of weeks, the hyaline carti- lage covering the end plates may be eroded, thus allowing the infection to reach the bony vertebral end plates. A sawtooth pattern of end plate destruction is then evident on radiographs (Fig. 2). At this point, greater blood flow reaches the site of infection through the redundant in- traosseous network of the vertebral body, resulting in either resolution of the infection by the body’s de- fense mechanisms or progression to classic vertebral osteomyelitis and possibly soft-tissue abscess. Evaluation History The history and clinical presenta- tion of childhood diskitis are variable. Frequently, the family reports an an- tecedent or concurrent illness (viral or bacterial) and a confounding his- tory of trauma. The duration of symp- toms is quite variable, from hours to weeks. Although involvement of the lumbar spine is the most common clinical scenario in children younger than 5 years, back pain is a present- ing complaint in only 50% of children with diskitis. 6 Children younger than 3 years often present acutely with a limp or refusal to bear weight on the lower extremities. 6 If diskitis is not promptly recognized and treated, it may progress to the point that the child is uncomfortable in all positions except when lying supine. The child between the ages of 3 and 8 years also frequently avoids activity but com- Figure 1 Posteroanterior (A) and lateral (B) radiographs of a 3-year-old girl with a 2-week history of irritability and refusal to walk for 2 days. The disk space narrowing at L3-4 is con- sistent with diskitis. Childhood Diskitis 414 Journal of the American Academy of Orthopaedic Surgeons plains of vague abdominal or back pain more often than would a young- er child. Although older children and teenagers occasionally present with diskitis, vertebral osteomyelitis is more common. This age group also may complain of abdominal symp- toms but usually can localize the pain to the thoracolumbar or lumbar spine. They may complain of buttock and leg pain as a result of nerve root ir- ritation. Physical Examination Low-grade fever may be present; however, most children with diskitis do not appear to be acutely ill. Young- er children may refuse to ambulate and may demonstrate irritability of the hip, but not to the degree that would be compatible with septic ar- thritis. Assessment of all limping chil- dren must include spinal examina- tion. Local tenderness is common, and the thoracic, lumbar, and sacral regions must be palpated. Paraverte- bral muscle spasm, decreased range of motion of the spine, and hamstring tightness are common. A positive straight leg raise may be seen; how- ever, most children do not have find- ings on neurologic examination. Chil- dren who will walk may show loss of normal, fluid spinal motion and may keep the back stiff and rigid. When asked to pick up an object from the floor, the young child may bend the knees and squat while keeping the back straight. Laboratory Tests In cases of suspected diskitis, as with other pediatric musculoskeletal infections, laboratory examination should include a complete blood count with differential blood cell count, blood cultures, erythrocyte sedimentation rate (ESR), and C-reactive protein level (if avail- able). 21 The white blood cell count is often in the high normal range, al- though the child may have a left shift and/or mild leukocytosis. 11 Blood cultures should be obtained, partic- ularly if the child is febrile. 8 The ESR will be elevated in nearly all patients, consistent with a chronic inflamma- tory or infectious process. 7,8,22 The C-reactive protein level is helpful, es- pecially in early infections, and can be used to follow the effect of initial antibiotic therapy. The purified pro- tein derivative skin test for tubercu- losis also should be done with con- trols, especially for children with disk space narrowing or vertebral end plate erosions on plain radio- graphs. 23,24 Radiography The initial radiologic workup should consist of plain radiographs. In the patient with acute onset of dis- kitis, initial radiographs will not dem- onstrate osseous changes, although there may be a loss of normal lum- bar lordosis. When symptoms have been present for 1 week or more,plain radiographs also may reveal isolated intervertebral disk space narrowing, and by 3 to 4 weeks may show saw- tooth erosion of adjacent vertebral end plates. 25,26 Scalloping of the su- perior and inferior vertebral body may be seen with long-standing in- fections. Vertebra magna with resul- tant canal narrowing, permanent loss of disk height, or block vertebra caused by spontaneous disk space fu- sion may be noted in patients with re- solved infections (Fig. 3). For continued clinical suspicion in the absence of changes on plain ra- diographs, technetium Tc 99m ( 99m Tc)– labeled bone scan can isolate the area of pathology to a specific motion seg- ment and help confirm the diagno- sis. This is especially useful for young children in whom localization of the involved area can be extremely dif- ficult by physical examination alone. Scintigraphic changes may appear as early as 3 to 5 days after the onset of clinical symptoms, well before any changes would be seen on plain ra- diographs. No additional radiologic Figure 2 Posteroanterior (A) and lateral (B) radiographs of a 19-month-old boy with a 4-week history of refusal to walk. Note the L4-5 end plate erosion (arrow), early vertebral body scal- loping (arrowheads), and disk space narrowing. Sean D. Early, MD, et al Vol 11, No 6, November/December 2003 415 studies except plain radiographs are necessary in patients with character- istic bone scan findings. 23,27 In patients whose symptoms and findings clearly localize to a specific area of the spine, bone scan can be omitted in favor of more advanced imaging. Computed tomography (CT) scans may show bony end plate erosion but otherwise do not add much value to management deci- sions. 14 Magnetic resonance imaging (MRI) (Fig. 4) is more useful because it allows excellent visualization of the disk, neural tissues, surrounding soft tissues, and pathophysiologic changes in the vertebral body. 25,28-34 MRI is more sensitive than nuclear studies or CT scan in detecting dis- kitis, helps differentiate diskitis from classic vertebral osteomyelitis, and can detect paraspinal or epidural ab- scesses. This may help reduce diag- nostic delay while also guiding the extent and duration of treatment 18,24 (Fig. 5). Fernandez et al 13 studied 50 patients with spinal infection, 36 of whom had diskitis. Of the 33 patients with diskitis who had radiographs of the spine, 25 (76%) had plain radio- graphic abnormalities at initial eval- uation. Of the 30 children who un- derwent 99m Tc bone scan, 27 (90%) demonstrated increased marker up- take at the affected area. Nine of 10 children also showed MRI abnormal- ities consistent with diskitis. The au- thors recommended nuclear bone scanning in very young children in whom localization of the inflamma- tory process is uncertain by physical examination and MRI in children with suspected diskitis who do not display characteristic radiographic changes. Although MRI is more sensitive than other radiologic modalities at detecting diskitis, its use rarely alters treatment. MRI is used to evaluate children who present with either a sagittal or coronal spinal deformity, who have no plain radiographic or bone scan abnormality, or who do not improve after 2 to 3 days of parenter- al antibiotic therapy. In such patients, MRI allows detection of conditions that may require surgical interven- tion, such as soft-tissue abscess, bone destruction, and involvement of neu- rologic elements. MRI also can be use- ful in the thoracic spine, where inter- pretation of plain radiographs may prove to be difficult. Differential Diagnosis The differential diagnosis of a child with irritability and back pain in- cludes a number of noninfectious eti- ologies. Scheuermann’s kyphosis may manifest as back pain in adoles- cents. Schmorl’s nodes are often present on plain radiographs, espe- cially in children with lumbar spine involvement. Patients with metastatic tumor and leukemia tend to have ver- tebral body involvement rather than end plate erosion on plain radio- graphs. These changes often occur at multiple levels throughout the spine. Vertebra plana, secondary to eosino- philic granuloma, is characteristic on radiographs. Osteoid osteoma and osteoblastoma most commonly in- volve the posterior elements of the spine. In a child with fever and elevated laboratory test results, other etiologies of infection must be considered. Sep- tic arthritis of the sacroiliac joint may clinically mimic diskitis. It can be dif- ferentiated with careful palpation, use of the flexion, abduction, and exter- nal rotation (FABER) maneuver, and characteristic bone scan findings. Destructive pyogenic osteomyelitis, which may be the sequela of long- standing untreated diskitis, may lead to vertebral body destruction and seg- mental collapse. In acutely ill chil- dren, or those with neurologic or meningeal signs, paraspinal or epidu- ral abscess must be ruled out by MRI evaluation. Infections secondary to brucellosis, fungal species, or tuber- culosis are indolent and rarely cause acute symptoms. Steplike vertebral Figure 3 Posteroanterior (A) and lateral (B) radiographs of the patient in Figure 2, 5 years after presentation. Note the extreme narrowing of the disk space with near formation of a block vertebra. The patient was asymptomatic. Childhood Diskitis 416 Journal of the American Academy of Orthopaedic Surgeons body erosions and elevated Brucella titers (>1:80) are characteristic of bru- cellosis. Fungal infections secondary to species of Aspergillus or Cryptococ- cus generally affect immunocompro- mised patients. Tuberculosis more commonly affects the thoracic and lumbar spine and is usually accom- panied by slowly progressive consti- tutional symptoms. The infection characteristically tunnels below the anterior and posterior longitudinal ligaments, initially sparing the disk spaces.Apositive purified protein de- rivative skin test or characteristic chest radiograph findings may be seen early in the disease process be- fore spinal deformity occurs. Back pain usually occurs late in tubercu- lous infections, after bony collapse be- comes evident on plain radiographs (Fig. 6). Biopsy Positive cultures have been report- ed in approximately 60% of patients undergoing biopsy for suspected dis- kitis, with Staphylococcus aureus the most commonly isolated organ- ism. 2,4,6,7,14,29 These low yields are compatible with biopsy results in pe- diatric acute hematogenous osteomy- elitis and septic arthritis. They are most likely the result of confounding factors, such as inadequate sampling, culture technique, previous antibiot- ic treatment, or brisk host response to a pathogen of low virulence, rath- er than a noninfectious etiolo- gy. 19,24,35,36 Chew and Kline 37 report- ed positive cultures in 39 of 43 patients who underwent CT-guided percutaneous disk space aspiration. However, the mean age of their pa- tients was 57 years, with no patients younger than 13 years. In five recent reports of pediatric disk space infec- tions, biopsy was not done in most patients, and only 37% (29/79) of the children who underwent biopsy had positive culture results. 10,11,13,14,24 Because of the low yield, potential morbidity, and need for conscious se- dation or general anesthesia in the young child, biopsy is not routinely recommended for the evaluation of the child with diskitis. 12,19,21,24 Biop- sy should be reserved for patients who do not respond to empiric intra- venous antistaphylococcal antibiotics because an unusual or highly virulent organism may be present. 8,19,24 In such patients, CT-guided aspiration or core needle biopsy under general anesthe- sia with immediate Gram stain and frozen section analysis is preferred. If the specimen is nondiagnostic, open surgical biopsy is indicated to rule out neoplasm, fungal infection, brucellosis, tuberculosis, and non- staphylococcal pyogenic infections. Treatment The goals of treatment are eradication of infection and minimization of mor- bidity. The variable responses to a myriad of treatment protocols in pre- vious studies of diskitis in the pedi- atric population have resulted in con- tinued confusion and discussion as to Figure 4 A, Sagittal postcontrast T1-weighted MRI scan of the patient in Figure 1. Note the increased uptake of contrast at the L3-4 disk space and adjacent vertebral bodies. There is no evidence of soft-tissue abscess or canal impingement. B, Sagittal T2-weighted MRI scan. Note the loss of normal signal intensity at the L3-4 disk and mild signal increase at the ad- jacent vertebral bodies. These findings are consistent with diskitis. Figure 5 Sagittal T2-weighted MRI scan of the patient in Figure 2, 4 months after pre- sentation. Note the decreased signal inten- sity at the L4-5 disk, consistent with resolved diskitis. The child now walks normally. Sean D. Early, MD, et al Vol 11, No 6, November/December 2003 417 its underlying etiology. Bed rest or various modes of immobilization have been almost universally recom- mended, but debate existed regard- ing the need for antibiotic therapy. In some series, no difference was found between patients who received anti- biotics and those treated with only rest and immobilization. 1,2,4,5,8,10,26 However, those series had few pa- tients and sometimes followed incon- sistent treatment algorithms. Most patients who did not improve after weeks of rest and immobilization were started on antistaphylococcal antibiotics, at which time there was often rapid clinical resolution. Al- though antibiotic therapy was neces- sary for improvement in some chil- dren, authors often concluded that no apparent difference was seen between patients who received antibiotics and those who did not. In other series, nearly all patients were treated with antibiotic therapy in addition to rest and immobilization because the au- thors believed diskitis to be of an in- fectious etiology. 3,6,7,9,14,18,24,38 Diskitis is now accepted to be a bacterial infection of the disk space and adjacent vertebral end plates. 12,21 The severity of the infec- tion likely depends on a dynamic in- terplay between the virulence of the infecting organism and the patient’s immunologic defenses. 23,24 Relative rest or immobilization without anti- biotic therapy may be sufficient for se- lect immunocompetent children without virulent infection. However, in other children a delay in antibiotic therapy may lead to prolonged hos- pitalization, recurrent symptoms, and worsening infections. Boston et al 4 recommended antibiotics for any pa- tient with pain, compensatory scoli- osis, or persistently elevated ESR. Scoles and Quinn 7 recommended 7 to 10 days of intravenous antibiotics for all patients with high fever, extreme pain, or positive blood, sputum, throat, or urine cultures. In most re- cently published reports, intravenous antibiotics are recommended as ini- tial management. 11,14,18,23,24,36 Ring et al 11 analyzed a series of 47 patients. Of the 38 nonimmobilized patients, 22 were treated with intra- venous antibiotics, 10 received oral antibiotics, and 6 received no antibi- otics. Prolonged or recurrent symp- toms occurred in 9 of the 16 patients who did not receive intravenous an- tibiotics but in only 4 of the 22 patients treated with intravenous antibiotics. Seven patients were treated with im- mobilization. Only one of five pa- tients treated with immobilization and intravenous antibiotics had a re- currence. Both patients treated with immobilization alone developed re- current symptoms after immobiliza- tion was removed. Thus, intravenous antibiotics were believed more likely to lead to rapid resolution of symp- toms over 2 to 4 days without recur- rence than were oral antibiotics or no antibiotic therapy. (The remaining 2 of the 47 patients required surgical drainage of paraspinal abscesses.) It is difficult to initially separate children infected with organisms of low virulence from those infected with highly virulent bacteria. The pre- sentation of disk space infection of- ten is benign and does not allow differentiation based on clinical char- acteristics. The need for general an- esthesia, the low incidence of positive cultures from obtained specimens, and the potential risks of biopsy pre- clude the use of direct biopsy in ev- ery patient. 11,14,24 Thus, to allow rap- id healing and prevent complications from worsening infection, all patients should receive intravenous antibiot- ics once the diagnosis of diskitis is es- tablished. Empiric coverage is direct- ed against S aureus because it has been the most common organism isolated from culture-positive biopsy speci- mens. There are no absolute recommen- dations to guide duration of antibi- otic treatment. Initial treatment is with parenteral antibiotics, with a change to oral antibiotics when there is clin- ical and laboratory evidence of patient response to antibiotic therapy. The rec- ommended duration of parenteral an- tibiotics ranges from 1 to 8 weeks, while oral antibiotics may be continued for as long as 3 to 6 months. 11,13,18,19,24 The C-reactive protein level improves more rapidly than the ESR in response to appropriate initial antibiotic therapy and can help guide the duration of antibiotic treatment. Typical duration of antibiotic therapy should be between 4 and 6 weeks, with parenteral anti- biotics administered for 1 to 2 weeks, followed by oral antibiotics for a to- tal course of 4 to 6 weeks. If compli- ance with oral antibiotics is unlikely, parenteral antibiotics may need to be administered for the duration of an- tibiotic treatment. Relative rest and immobilization with a lumbosacral corset or thora- columbosacral orthosis may be used in conjunction with antibiotics to im- prove comfort. A thoracolumbosacral Figure 6 Lateral radiograph of a 15-year-old girl with a 2-year history of tuberculous in- fection of the spine and progressive kyphotic deformity. She had no pain despite near com- plete destruction of the L1 vertebral body and 76° of focal kyphosis. Childhood Diskitis 418 Journal of the American Academy of Orthopaedic Surgeons orthosis is recommended if, on imag- ing studies, sagittal or coronal defor- mity of the spine exists or if extensive bony destruction and soft-tissue in- volvement are evident. Surgical débridement should be considered only in the rare patient with a documented abscess who is systemically ill or has an evolving neurologic deficit. In light of clinical improvement with antibiotic therapy, MRI findings of paraspinal fluid col- lection do not necessitate surgical débridement. However, drainage may be necessary if clinical improve- ment does not occur. 14,24 Follow-up After resolution of symptoms with antibiotic treatment, patients are fol- lowed for 12 to 18 months. Recur- rence of symptoms, although rare af- ter a 4- to 6-week course of antibiotics, should prompt the treating physician to obtain new radiographs, a com- plete blood count, C-reactive protein level, and ESR. In general, C-reactive protein level and ESR should grad- ually return to normal predisease lev- els; a rise in these levels may indicate recurrent infection. Periodic plain radiographs will show continued residual disk space narrowing. 18 With disease resolution, partial reconstitution of the disk height may gradually occur, but reconstitution is rarely complete 18 (Fig. 7). Vertebra magna, with resul- tant canal narrowing, or block verte- bra, from spontaneous fusion, also may be seen. 9 MRI is rarely needed but can be done if extensive bony destruction or soft-tissue involvement occurred in conjunction with the disk space infec- tion. MRI findings are not acutely af- fected by antibiotic therapy. Howev- er, after complete resolution, there is a return to normal signal with a de- creased disk width on T1-weighted images.Anormal vertebral body with low disk signal will be seen with T2- weighted imaging. 28,29 Summary Despite previous controversy regard- ing etiology and treatment, childhood diskitis is now accepted to be a pyo- genic infection of the disk space and bony vertebral end plates. Clinical presentation varies depending on the age of the child, but characteristic findings on physical examination, laboratory studies, plain radiographs, and bone scan make early diagnosis possible. Early diagnosis can prevent unnecessary testing and invasive in- tervention. Intravenous antibiotic therapy, empirically directed against S aureus, is the mainstay of treatment. With clinical resolution, a switch to oral antibiotics is appropriate, for a total antibiotic course of 4 to 6 weeks. Immobilization and rest may be add- ed for symptomatic relief. Follow-up should continue for 12 to 18 months after healing. Biopsy and surgery are reserved for patients refractory to in- travenous antibiotic therapy. References 1. Menelaus MB: Discitis: An inflamma- tion affecting the intervertebral discs in children. J Bone Joint Surg Br 1964;46: 16-23. 2. Spiegel PG, Kengla KW, Isaacson AS, Wilson JC Jr: Intervertebral disc-space inflammation in children. J Bone Joint Surg Am 1972;54:284-296. 3. Kemp HBS, Jackson JW, Jeremiah JD, Hall AJ: Pyogenic infections occurring primarily in intervertebral discs. J Bone Joint Surg Br 1973;55:698-714. 4. Boston HC Jr, Bianco AJ Jr, Rhodes KH: Disk space infections in chil- dren. Orthop Clin North Am 1975;6: 953-964. 5. Fischer GW, Popich GA, Sullivan DE, Mayfield G, Mazat BA, Patterson PH: Figure 7 Posteroanterior (A) and lateral (B) radiographs of the patient in Figure 1, 3 years after resolution of diskitis. Although some reconstitution of the disk height at L3-4 has oc- curred (arrow), it is not complete. Sean D. Early, MD, et al Vol 11, No 6, November/December 2003 419 Diskitis: A prospective diagnostic anal- ysis. Pediatrics 1978;62:543-548. 6. Wenger DR, Bobechko WP, Gilday DL: The spectrum of intervertebral disc- space infection in children. J Bone Joint Surg Am 1978;60:100-108. 7. ScolesPV, Quinn TP: Intervertebral dis- citis in children and adolescents. Clin Orthop 1982;162:31-36. 8. Crawford AH, Kucharzyk DW, Ruda R, Smitherman HC Jr: Diskitis in children. Clin Orthop 1991;266:70-79. 9. Jansen BRH, Hart W, Schreuder O: Dis- citis in childhood: 12-35-year follow-up of 35 patients. Acta Orthop Scand 1993; 64:33-36. 10. Ryoppy S, Jaaskelainen J, Rapola J, Al- berty A: Nonspecific diskitis in chil- dren: A nonmicrobial disease? Clin Orthop 1993;297:95-99. 11. Ring D, Johnston CE II, Wenger DR: Pyogenic infectious spondylitis in chil- dren: The convergence of discitis and vertebral osteomyelitis. J Pediatr Orthop 1995;15:652-660. 12. Hensinger RN: Acute back pain in chil- dren. Instr Course Lect 1995;44:111-126. 13. Fernandez M, Carrol CL, Baker CJ: Dis- citis and vertebral osteomyelitis in chil- dren:An 18-year review. Pediatrics 2000; 105:1299-1304. 14. Garron E, Viehweger E, Launay F, Guillaume JM, Jouve JL, Bollini G: Non- tuberculous spondylodiscitis in chil- dren. J Pediatr Orthop 2002;22:321-328. 15. Wiley AM, Trueta J: The vascular anat- omy of the spine and its relationship to pyogenic vertebral osteomyelitis. J Bone Joint Surg Br 1959;41:796-809. 16. Coventry MB, Ghormley RK, Kernohan JW: The intervertebral disc: Its micro- scopic anatomy and pathology: I. Anat- omy, development, and physiology. J Bone Joint Surg 1945;27:105-112. 17. Whalen JL, Parke WW, Mazur JM, Stauffer ES: The intrinsic vasculature of developing vertebral end plates and its nutritive significance to the interverte- bral discs. J Pediatr Orthop 1985;5:403- 410. 18. Song KS, Ogden JA, Ganey T, Guidera KJ: Contiguous discitis and osteomyeli- tis in children. J Pediatr Orthop 1997;17: 470-477. 19. Glazer PA, Hu SS: Pediatric spinal in- fections. Orthop Clin North Am 1996;27: 111-123. 20. Ratcliffe JF: Anatomic basis for the pathogenesis and radiologic features of vertebral osteomyelitis and its differen- tiation from childhood discitis: Amicro- arteriographic investigation. Acta Radiol Diagn (Stockh) 1985;26:137-143. 21. King HA: Back pain in children. Orthop Clin North Am 1999;30:467-474. 22. Schulak DJ, Rayhack JM, Lippert FG III, Convery FR: The erythrocyte sedimen- tation rate in orthopaedic patients. Clin Orthop 1982;167:197-202. 23. Cushing AH: Diskitis in children. Clin Infect Dis 1993;17:1-6. 24. Brown R, Hussain M, McHugh K, Novelli V, Jones D: Discitis in young children. J Bone Joint Surg Br 2001;83: 106-111. 25. Szalay EA, Green NE, Heller RM, Horev G, Kirchner SG: Magnetic reso- nance imaging in the diagnosis of child- hood discitis. J Pediatr Orthop 1987;7: 164-167. 26. Grünebaum M, Horodniceanu C, Mukamel M, Varsano I, Lubin E: The imaging diagnosis of nonpyogenic dis- citis in children. Pediatr Radiol 1982;12: 133-137. 27. Maliner LI, Johnson DL: Intervertebral disc space inflammation in children. Childs Nerv Syst 1997;13:101-104. 28. Paushter DM, Modic MT, Masaryk TJ: Magnetic resonance imaging of the spine: Applications and limitations. Radiol Clin North Am 1985;23:551-562. 29. du Lac P, Panuel M, Devred P, Bollini G, Padovani J: MRI of disc space infec- tion in infants and children: Report of 12 cases. Pediatr Radiol 1990;20:175-178. 30. Szypryt EP, Hardy JG, Hinton CE, Worthington BS, Mulholland RC: A comparison between magnetic reso- nance imaging and scintigraphic bone imaging in the diagnosis of disc space infection in an animal model. Spine 1988;13:1042-1048. 31. Heller RM, Szalay EA, Green NE, Horev GD, Kirchner SD: Disc space in- fection in children: Magnetic resonance imaging. Radiol Clin North Am 1988;26: 207-209. 32. Modic MT, Pavlicek W, Weinstein MA, et al: Magnetic resonance imaging of in- tervertebral disk disease: Clinical and pulse sequence considerations. Radiolo- gy 1984;152:103-111. 33. Gabriel KR, Crawford AH: Magnetic resonance imaging in a child who had clinical signs of discitis: Report of a case. J Bone Joint Surg Am 1988;70:938- 941. 34. Mahboubi S, Morris MC: Imaging of spinal infections in children. Radiol Clin North Am 2001;39:215-222. 35. Ring D, Wenger DR: Magnetic resonance-imaging scans in discitis: Sequential studies in a child who need- ed operative drainage. A case report. J Bone Joint Surg Am 1994;76:596-601. 36. Brook I: Two cases of diskitis attribut- able to anaerobic bacteria in children. Pediatrics 2001;107:E26. 37. Chew FS, Kline MJ: Diagnostic yield of CT-guided percutaneous aspiration procedures in suspected spontaneous infectious diskitis. Radiology 2001;218: 211-214. 38. Leahy AL, Fogarty EE, Fitzgerald RJ, Regan BF: Discitis as a cause of abdom- inal pain in children. Surgery 1984;95: 412-414. Childhood Diskitis 420 Journal of the American Academy of Orthopaedic Surgeons