Primary Infections of the Spine

Clinical Symptoms

Axial back/neck pain and radicular pain are the most common presenting symptoms. There is no characteristic pain in SD, and therefore during the early stages it may be mistaken for degenerative pathology. During the initial stages, the pain symptoms are mild to moderate, continuous (often worse at night), and worse with change of posture and activities. Instability pain may also develop once spinal stability is compromised. Constitutional symptoms are not uncommon and include fever, malaise, night sweats, and weight/appetite loss. Neurological deficit may develop secondary to mechanical compression or spinal instability and has been reported in 3% to 32% of patients.

Clinical Signs

The most common physical signs include localized tenderness, sustained paraspinal spasm, and painful limitation of spinal movements. Palpable abscesses are relatively uncommon. Neurological signs may be present, based on the level and severity of involvement. Very young or very old patients and immunosuppressed patients may have relatively minimal clinical findings.

The clinical presentation is more delayed in children as compared with their adult counterparts. In children, fever is an uncommon symptom (40% incidence). Irritability (63%), refusal to bear weight (38%) and limping (38%) are common symptoms in children younger than 3 years. Children between 3 and 10 years typically present with back pain (89%), fever (59%), and abdominal pain (29%).

Elderly patients are prone to develop chronic infections and have a strong predisposition to develop neurological paralysis as compared with the younger individuals.

Differential Diagnosis

The differential diagnoses include primary and metastatic malignant spinal tumors, fractures, ankylosing spondylitis or inflammatory arthritis, polymyalgia rheumatica, infections in adjacent structures like the hips, abdominal or genitourinary structures, or psoas musculature, activated osteochondrosis, metabolic bone disease, and vertebral hemangioma.

Blood Investigations

Initial blood investigations include assessment of white blood cell (WBC) count, erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). ^, Among them, CRP is the more sensitive parameter, and elevated CRP is noticed in 90% of patients. It helps us to understand the response to therapy, as there is a favorable reduction in CRP value within 5 to 7 days after the initiation of medications. ESR is less sensitive and less specific than CRP in diagnosing PS. Persistently elevated ESR beyond 1 month of medical therapy may indicate poorer response. Elevated WBC is noticed in two-thirds of patients. Procalcitonin assessment plays a minor role because it is more cost intensive and has no prognostic advantage. Blood cultures may be positive in two-thirds of patients and should be obtained in all patients before starting antibiotics. ^, Additionally, urine cultures may be helpful in identifying a possible underlying infectious focus, especially in elderly patients with previous history of urinary tract infection and evidence of current urinary symptoms (like burning micturition, incontinence, or prostatism).

Radiological Imaging

Plain radiograph is the first radiological investigation, with a low sensitivity and specificity of 82% and 57%, respectively. The earliest changes are evident between 2 and 8 weeks after symptom onset. During the early stages, vertebral body haziness, disc height reduction, and end plate irregularity are the most typical findings. As bone and disc involvement progressively increase, variable degrees of bony destruction, progressive spinal instability, and loss of alignment become evident. ^{, ,} Griffiths et al. classified bone destruction radiographically into three stages: early (disc space narrowing), destructive (vertebral collapse), and osteosclerotic (new bone formation).

Computed tomography (CT) is a more sensitive modality than plain radiographs for evaluating bony changes. The involvement of anterior (vertebral body sclerosis/lysis/scalloping secondary to mechanical pressure from abscess, end plate changes, etc.) and posterior (spinous process, pedicle, and laminar lesions) spinal elements are clearly evident on CT. ^{, ,} Additionally, soft tissue lesions can be indirectly noted on radiographs and CT as abnormal psoas shadow, obliteration of paravertebral fat, or mediastinal and retropharyngeal widening. CT also offers a significant advantage in the planning of surgical intervention (either biopsy or definitive surgery), as the true bony involvement, site, pattern, and extent of vertebral destruction are best delineated by this modality.

Magnetic resonance imaging (MRI) is the most sensitive (92% for noncontrast sequences and 95.4% for gadolinium-enhanced sequences) and specific (96%) imaging modality for evaluating PS. The most useful MRI sequences include T1-weighted imaging (T1), T2-weighted imaging (T2), short tau inversion recovery (STIR) imaging, and contrast-enhanced imaging. On T1-weighted imaging, bony and soft tissue lesions are visible as hypointense regions or masses over the vertebral body, end plate, IVD, spinal canal, or paraspinal regions. On T2-weighted imaging, the lesions are evident as hyperintense regions. Fat suppression sequences (especially STIR images) have the highest sensitivity for increased water content within the infected bone and surrounding soft tissues. The bony and soft tissue lesions demonstrate increased tissue enhancement on contrast-enhanced imaging, whereas abscesses demonstrate mass lesions with peripheral enhancement. Additionally, pre- and postcontrast fat-suppressed T1-weighted imaging aid in differentiating between vascularized or hyperemic tissues and avascular/necrotic or sequestrated lesions. On MRI, Uchida et al. proposed a five-stage classification system: stage 1—bruise with localized radiolucency in the vertebral end plate, stage 2—marrow edema and fluid collection confined to the vertebral body, with poor demarcation of the lesion, stage 3—irregularly increased signal intensity with confinement of the lesion to the subligamentous (posterior longitudinal ligament) compartment, stage 4—evident fluid collection in the IVD with extensive end plate destruction and high-signal lesions in the vertebral body, along with transligamentous abscess extension into the epidural space at multiple levels, and stage 5—obvious disappearance of disc space, vertebral collapse with nonhomogenous enhancement, and extension of the lesion into the epidural space, posterior vertebral elements, and paravertebral ligaments, as well as the musculature. Broadly, stages 1 to 3 were classified as confined or contained lesions, and stages 4 and 5 were noncontained lesions. MRI is extremely useful in differentiating PS from tuberculous spondylodiscitis (TBS). Although more significant discal involvement, disc hyperintensity, and small abscesses with irregular, thick rim enhancement favor the diagnosis of PS, greater vertebral destruction, multilevel involvement, and larger, thin, smooth-rimmed abscesses with calcification are seen in TBS ( Table 44.1 ). Figs. 44.1 and 44.2 demonstrate the major imaging findings during the early and later stages of PS, respectively.

Nuclear Scans

In an ideal, current-day scenario, nuclear imaging has been almost completely replaced by MRI in the diagnosis of PS. It is a good alternative in patients with poor or delayed access to MRI, those with contraindications to contrast-enhanced CT/MRI (e.g., renal failure), and those with non–MRI-compatible pacemakers/cochlear implants or metallic implants causing serious artifacts. Nuclear scans, including gallium-67 citrate and technicium-99m scans, are more sensitive than plain radiographs during early SD. ^, Currently, among the available scintigraphy scans, fluoro-18 deoxyglucose positron emission tomography (18F-FDG-PET-CT) is probably the most useful adjunct to imaging in acute and chronic PS. It demonstrates a relative lack of specificity in differentiating neoplasia, posttraumatic bony edema, and infection, and therefore has been recommended in combination with MRI.

Culture/Sensitivity and Molecular Diagnostics

Obtaining tissue diagnosis is fundamental before initiation of antibiotics unless the clinical picture is complicated by the presence of ongoing septicemia. Percutaneous biopsy may be performed under CT or fluoroscopy guidance. If core biopsy fails to identify the organism, an open biopsy is necessary. In scenarios involving vertebral instability necessitating spinal stabilization, impending/existing or progressive neurological deficits, or any progressive deformities, biopsy may be obtained at the time of definitive surgery.

All specimens must be sent for gram stain, aerobic, anaerobic, fungal, and mycobacterial cultures, and molecular evaluation (geneXpert or polymerase chain reaction [PCR]). As the age-old dictum goes: “all tissue samples requiring histopathological evaluation must also be cultured; and all samples which require culture and sensitivity testing should also undergo histopathological assessment.” Thus, both microbiological and histopathological evaluations must be carried out for all patients. Iwata et al. reported the greatest diagnostic accuracy when both microbiology and histology were combined in the tissue diagnosis of PS. Molecular biological investigations (tissue PCR) can be helpful in situations where cultures are negative even after 48 hours or incubation. Species-specific tissue PCR for S. aureus infections provides a much better yield (with 46.7%–62.9% detection rates) than conventional PCR (26.7% detection rate). If there is suspicion of any other infectious foci (e.g., urological or pulmonary infection), additional cultures (say, urine or sputum cultures) may need to be procured.

Nonoperative Treatment

The mainstay of treatment of spinal infections is administration of appropriate antibiotic therapy. Patients with intact or maintained neurological status, stable spinal alignment or absence of any evidence of instability, absence of significant spinal canal compromise or impending neurological deficit, small or resolving abscess cavities, and those who are systemically unable to tolerate surgical interventions are ideal candidates for nonoperative treatment. Yashimoto et al. reported resolution of neurological symptoms in 72.7% of SD patients within 10 months following antibiotic therapy alone. de Graeff et al. reported that weekly assessment of clinical status and inflammatory parameters is necessary to promptly identify treatment failure in high-risk patients (diabetes mellitus, concomitant abscess, or osteomyelitis). They recommended repeat MRI in patients with suspected treatment failure, which was defined as absent clinical improvement or any additional radiological deterioration after 4 weeks of treatment.

Empirical Antibiotic Therapy. The initial, empirical anti-microbial therapy should involve broad-spectrum antibiotics and account for common organisms present in the community (especially S. aureus ). The initiation of antibacterial treatment should not be performed before obtaining appropriate tissue samples for biopsy, unless florid septicemia complicates the clinical picture, necessitating immediate control of infections. In patients with sepsis, initial coverage is essential for both S. aureus and gram-negative bacilli.

Definitive Antibiotic Therapy. It is of utmost importance to obtain tissue samples to identify the offending pathogen. Once the etiological agent is identified, the appropriate antibacterial therapy is administered based on the sensitivity analysis. In patients with culture-negative SD, the antibiotic administered should actively cover the most common pathogens, including S. aureus , streptococci, and Escherichia coli . The response to therapy is monitored by evaluating the clinical response (including resolution of pain or neurological symptoms, improvement in the general systemic status like fever resolution, weight gain, improvement in appetite, general wellbeing, etc.), complete blood count (CBC), and recovery of acute-phase reactants like ESR and CRP.

Duration of Antibiotic Therapy. The general consensus regarding the duration of medical treatment has been to administer intravenous antibiotics for 6 to 8 weeks, followed by oral antibiotics, until disease eradication is confirmed by complete clinical recovery and normalization of laboratory parameters, including ESR and CRP. In situations where the patient is toxic/septicemic or when the tissue diagnosis is broadly unyielding of definitive results, long-term antibiotic therapy must cover both gram-positive and gram-negative organisms. Any failure of resolution of symptoms or persistent elevation of laboratory parameters during the course of treatment should be taken seriously, and necessitate an immediate and detailed reevaluation of the patient, including procurement of repeat tissue samples (percutaneous or open), as required. However, more recently there have been growing concerns with regard to long-term antibiotic therapy, including adverse reactions, health care–related infections, treatment costs, and antibiotic resistance. ^{, ,}

With these concerns in the background, there has been a paradigm shift toward shorter duration of parenteral antibiotic therapy and earlier shift to oral medications over the past few years. The Infectious Diseases Society of America proposed guidelines for antibiotic treatment of SD. They state that 6 weeks of therapy is adequate and that oral therapy (drugs with good bioavailability, including quinolones, clindamycin, and cotrimoxazole) is a reasonable alternative to intravenous treatment. Recently, Bernard et al. published the results of a multicenter, randomized controlled trial designed to determine the ideal duration of antibiotic therapy. Based on these results, they also recommended a 6-week total duration of antibiotic therapy in PS (where culture sensitivity is known), with an initial parenteral course of antibiotic treatment for no more than 2 weeks. A longer course of antibiotic treatment (12 weeks’ duration) may be necessary in patients with MRSA infections, older patients (age >75 years), nonpyogenic infections, and those without microbiological identification. Closely-monitored, home-administered, ambulant antibiotic therapy, with additional brace support as needed, is the most common nonoperative strategy employed by most experts.

Recurrent Infections after Conservative Management of PS. Park et al. performed a retrospective study of risk factors for disease recurrence following antibiotic treatment for SD. The mean time to relapse was around 5 weeks, and the major risk factors for recurrence included MRSA, persistent paravertebral or psoas abscesses, and renal failure. Box 44.1 lists the major factors that may predict failure of response to conservative management in patients with PS.