Musculoskeletal systemCase Type
Julian Pohlan 1, Vaclav Cink 2, Friederike Schömig 2, Matthias Pumberger 2, Torsten Diekhoff 1Patient
80 years, male
An 80-year-old man was admitted with systemic signs of infection, two weeks after decompression of L4-S1 for symptomatic spinal canal stenosis. He was discharged asymptomatically one week post-surgery.
Upon admission, he reported mild back pain, and the surgical wound was healing well. Antibiotic therapy was initiated, without clinical improvement.
Search for the infectious focus was then repeated including clinical examination, unremarkable chest X-ray and microbiological probing of blood and urine. Thereafter, magnetic resonance imaging (MRI) suggested spondylodiscitis at the L2-3 level. After transfer from an outside hospital, a CT-guided biopsy was scheduled.
The postoperative MRI indicates pyogenic spondylodiscitis new to the preoperative MRI (Figs. 1 and 2). During CT-guided biopsy, an unenhanced DECT scan with acquisition at 135 kVp and 80 kVp was obtained. Collagen maps were reconstructed using a collagen-specific gradient of 1.1 and virtual non-calcium images for detection of bone marrow lesions with a gradient of 0.69. Collagen mapping shows evidence of an altered connective tissue structure with complete loss of collagen integrity in the disc (Fig. 3). These DECT findings are consistent with the MRI performed by the outside hospital (Fig. 2).
For microbiology, a biopsy was sampled from the L2-3 disc with CT guidance (Fig. 4). With CT-guided probes remaining sterile initially, additional transpedicular biopsy was performed in the operating room under general anaesthesia using fluoroscopic guidance (Fig. 4).
The incidence of pyogenic osteomyelitis has recently increased, resulting in significant morbidity.  Pyogenic, or bacterial spondylodiscitis – often caused by Staphylococcus aureus – typically develops through haematogenous inoculation as endplate-near vertebral osteomyelitis and secondly spreads into the adjacent avascular disc.  It can also occur as a complication of spinal procedures, e.g. infiltration or surgery. 
Spondylodiscitis may be difficult to diagnose due to misleading clinical presentation, e.g. with pain suggestive of degenerative disease. Pyogenic spondylodiscitis will require at least six weeks of antibiotic therapy. Surgery, i.e. posterior stabilisation, is indicated in patients with neurological deficits due to compression, spinal instability or deformity and after failure of conservative treatment. 
Imaging is usually required to prove the extent of tissue destruction in spondylodiscitis. [5, 6] Magnetic resonance imaging (MRI) accurately detects inflammatory bone marrow lesions, destruction of the endplate and disc, and soft-tissue abscesses. Soft-tissue inflammation of the disc is the hallmark of the disease in imaging, i.e. fluid-in-disc sign. Also, T1 typically reveals a loss of fat-signal and a loss of endplate definition. Administration of gadolinium will further increase diagnostic accuracy with contrast-enhancement of inflammatory lesions. Still, early imaging may be non-specific for spondylodiscitis due to similarities with Modic I subchondral reactive oedema. 
While computed tomography (CT) allows detection of endplate destruction, it is not well suited to identify soft-tissue involvement. However, dual-energy computed tomography (DECT) visualises connective tissue components by exploiting collagen’s spectral properties. 
In this patient, DECT collagen mapping indicated severely altered connective tissue components with complete destruction of the disc. Thus, DECT has added value compared with conventional CT for diagnosing spondylodiscitis, especially when MRI is unavailable or contraindicated, e.g. in claustrophobia.
DECT visualises the structure of collagen-rich connective tissue, potentially useful in diagnosing a range of diseases.  Recently, DECT collagen maps were shown to diagnose vertebral disc injuries in elderly patients with vertebral fractures. 
CT-guided biopsy and microbiology may be needed to isolate a causative pathogen. The diagnostic yield can be low, especially with antibiotic pretreatment.
In our patient, Dual-energy computed tomography accurately shows lumbar spondylodiscitis as established using MRI. Blood culture as well as microbiology from CT-guided and transpedicular biopsy revealed Staphylococcus epidermidis as the culprit germ. Antibiotic therapy with vancomycin and fosfomycin was initiated based on the resistogram. The patient showed clinical improvement with normalisation of previously elevated C-reactive protein (Fig. 5).
Take Home Message
Pyogenic spondylodiscitis can be accurately diagnosed with DECT. With excellent availability and fast acquisition time, DECT can easily be implemented in the diagnostic workup of spondylodiscitis, e.g. in emergencies. Besides, low-dose techniques significantly reduce radiation exposure. MRI is the current imaging standard in diagnosing spondylodiscitis. Though, MRI has been criticised for gadolinium deposits of unclear significance with discussions about toxicity. [12-14] In future, DECT-guided biopsies may increase the microbiological yield, possibly improving patient outcomes.
Written informed patient consent for publication has been obtained.
TD received research funds from Canon Medical Systems GmbH.
 Kehrer M, Pedersen C, Jensen TG, Lassen AT. Increasing incidence of pyogenic spondylodiscitis: a 14-year population-based study. Journal of infection. 2014;68(4):313-20 (PMID: 24296494)
 Tali ET. Spinal infections. European journal of radiology. 2004;50(2):120-33 (PMID: 15081128)
 Cheung W, Luk KD. Pyogenic spondylitis. International orthopaedics. 2012;36(2):397-404 (PMID: 22033610)
 Rutges J, Kempen D, Van Dijk M, Oner F. Outcome of conservative and surgical treatment of pyogenic spondylodiscitis: a systematic literature review. European Spine Journal. 2016;25(4):983-99 (PMID: 26585975)
 Esendagli-Yilmaz G, Uluoglu O. Pathologic basis of pyogenic, nonpyogenic, and other spondylitis and discitis. Neuroimaging Clinics. 2015;25(2):159-61 (PMID: 25952170)
 Lucio E, Adesokan A, Hadjipavlou AG, Crow WN, Adegboyega PA. Pyogenic spondylodiskitis: a radiologic/pathologic and culture correlation study. Archives of pathology & laboratory medicine. 2000;124(5):712-6 (PMID: 10782153)
 Nicolaou S, Liang T, Murphy DT, Korzan JR, Ouellette H, Munk P. Dual-energy CT: a promising new technique for assessment of the musculoskeletal system. American Journal of Roentgenology. 2012;199(5_supplement):S78-S86 (PMID: 23097171)
 Czuczman GJ, Marrero DE, Huang AJ, Mandell JC, Ghazikhanian V, Simeone FJ. Diagnostic yield of repeat CT-guided biopsy for suspected infectious spondylodiscitis. Skeletal radiology. 2018;47(10):1403-10 (PMID: 29777259)
 Pumberger M, Fuchs M, Engelhard N, Hermann KG, Putzier M, Makowski MR, et al. Disk injury in patients with vertebral fractures—a prospective diagnostic accuracy study using dual-energy computed tomography. European radiology. 2019:1-8 (PMID: 30649597)
 Peltola EK, Koskinen SK. Dual-energy computed tomography of cruciate ligament injuries in acute knee trauma. Skeletal radiology. 2015;44(9):1295-301 (PMID: 26025120)
 Jeon JY, Lee S-W, Jeong YM, Yu S. The utility of dual-energy CT collagen material decomposition technique for the visualization of tendon grafts after knee ligament reconstruction. European journal of radiology. 2019;116:225-30 (PMID: 31054789)
 Gulani V, Calamante F, Shellock FG, Kanal E, Reeder SB. Gadolinium deposition in the brain: summary of evidence and recommendations. The Lancet Neurology. 2017;16(7):564-70 (PMID: 28653648)
 Conte G, Preda L, Cocorocchio E, Raimondi S, Giannitto C, Minotti M, et al. Signal intensity change on unenhanced T1-weighted images in dentate nucleus and globus pallidus after multiple administrations of gadoxetate disodium: an intraindividual comparative study. European radiology. 2017;27(10):4372-8 (PMID: 28357495)
 Young LK, Matthew SZ, Houston JG. Absence of potential gadolinium toxicity symptoms following 22,897 gadoteric acid (Dotarem®) examinations, including 3,209 performed on renally insufficient individuals. European radiology. 2019;29(4):1922-30 (PMID: 30276674)
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.