17 Spinal Infections in Geriatric Patients
Introduction
The spine is a common site of musculoskeletal infections. Spinal infection encompasses pyogenic spondylodiskitis, vertebral osteomyelitis, epidural abscess, and iliopsoas abscess. The disease causes not only local symptoms, such as back pain, leg pain, and neurologic deficit, but also deterioration of the patient’s general condition, which is evident in such symptoms as general malaise, fever, and vital organ failure due to sepsis. Spinal infection is a life-threatening disease. Several studies 1 – 7 reviewed spinal infection and discussed the management of spinal infection, as it is an important disease entity worldwide. This chapter focuses on the clinical characteristics, diagnosis, and treatment of spinal infection in geriatric patients.
Epidemiology
The estimated incidence of spinal infection is reported to range from 1 in 40,000 to 1 in 250,000 population per year. 8 The Japanese Survey for Vertebral Osteomyelitis, using the Japanese Diagnosis Procedure Combination Data base, reported that the estimated incidence of vertebral osteomyelitis has gradually increased in recent years. 8 This survey reported an incidence of 5.3 in 100,000 population per year in 2007, increasing to 6.3 in 2008, 7.1 in 2009, and 7.4 in 2010. The same increasing tendency was evident from the data from 1995 to 2008 in Denmark. 9 It has also been evident in Japanese studies that the incidence markedly increases with age among people 60 to 80 years of age. 8 , 9 The incidence per 100,000 population per year in patients younger than 59 was 1.7, whereas in those aged 60 to 69 years it was 10.9, in those aged 70 to 79 years it was 21.6, and in those 80 or older it was 25.1. 8 The incidence was 9.8 in 100,000 per year in New Zealand for those aged > 65 years, 10 and it was the highest in Denmark in those aged > 70 years. 11 Geriatric patients commonly have comorbidity related to their general condition. In the survey of Carragee, 6 the patients with pyogenic vertebral osteomyelitis were older and more likely to have a comorbid condition than was previously reported.
As for surgical site infection (SSI) in spine surgery, the Japanese Society for Spine Surgery–Related Research (JSSR) reported that the incidence of deep wound infection was 1.4% in 8,033 patients who had spine surgery. 12 Another multicenter study in Japan reported that the incidence of complications was significantly higher in patients older than 65 years, including SSI. 13 Several studies have shown that geriatric patients are more prone to SSI in spine surgery. 14 , 15 This might be caused by immune deficiency in geriatric patients.
Pathology
Primarily, bacteremia causes hematogenous infection of the spine. There are two major routes for hematogenous dissemination of bacteria: venous and arterial. Wiley and Trueta 16 suggested that bacteria can become lodged in the arteriolar network at the vertebral end plate. Batson has demonstrated retrograde flow of veins from the pelvic venous plexus to the perivertebral plexus which has been confirmed in other studies as well. 17 The infection in the pelvis or urinary tract may be spread via Batson plexus routes. The main focus of spinal infection is the intervertebral disk (IVD) in adults. The IVD is avascular, and bacteria invades the end-arterial arcades in the metaphyseal region adjacent to the IVD. The infection then spreads by direct extension, with rupture of the infective lesion through the end plate into the IVD. It may extend from the vertebral body to the subligamentous paravertebral area, the epidural space and contiguous vertebral bodies. 18 The other route is direct invasion from the needle at diskography. Diskitis after diskography is due to bacterial penetration into the IVD by a contaminated needle. The incidence of diskitis from diskography was reported to be 1 to 4%. 19 One case report found a pyogenic spondylodiskitis due to Pseudomonas aeruginosa after computed tomography (CT)-guided facet joint steroid injection in a 78-year-old man. 20
Hematogenous vertebral osteomyelitis is generally a single bacterial infection, 21 although the pathogenic bacteria may not always be detected. This is especially common when antibiotics are used before examinations for the detection of bacteria. Many studies demonstrated that Staphylococcus aureus is the most commonly isolated bacteria. 1 , 22 , 23 Although S. aureus accounts for more than 50% of isolates, the diversity of causative bacteria is wide. 1 In a study on vertebral osteomyelitis in geriatric patients, a shift in the frequency distribution of pathogens to gram-negative bacteria, such as Escherichia coli, has been observed. Belzunegui et al 24 reported that E. coli was isolated in seven of 34 elderly patients and in none of 38 young patients. This might be characteristic of elderly men who have a history of urinary tract infection. However, Sobottke et al 23 stated that the most frequent pathogen identified in 50% of cases was S. aureus. They did not see a shift to gram-negative pathogens in their study, which was focused on spondylodiskitis in elderly patients.
Clinical Features
The diagnosis of pyogenic spinal infection is based on clinical symptoms, radiological findings, and microbiological analyses. Local pain, usually back pain, is the most common symptom, with a prevalence of more than 90% of the patients with pyogenic spinal infection. This might be due to destruction of the vertebral structure or pressure rendered by the abscess. However, as this is not a specific clinical symptom, a delay in diagnosis is not uncommon. Especially in geriatric patients, it might be erroneously assumed that the pain is caused by degenerative spinal change. In addition, the presence of comorbidities such as diabetes can result in less severe constitutional symptoms. Thus, it takes from 2 to 12 weeks and occasionally more than 3 months before the diagnosis is made. 23 Fever is thought to be associated with infection, but it is typically not present and it accounts for less than 20% of infections. This is usually true even in geriatric patients. Other symptoms such as anorexia, general malaise, and weight loss are also not specific to spinal infection. Neurologic deficit might be caused when an epidural abscess compresses the spinal cord, cauda equina, or spinal nerves. However, the incidence of motor palsy by spinal infection is less common. Butler et al 25 reported that 27% of patients with infective spondylitis had neurologic involvement requiring surgical intervention. Most of these patients had an incomplete neurologic deficit with mild extremity motor weakness; this might be one reason for the delay in the diagnosis, as physicians might assume that the weakness is caused by general malaise or disuse in geriatric patients with spinal infection, even though spinal infection associated with epidural abscess causes a neurologic deficit. However, in a retrospective study, Eismont et al 26 reported that old age as well as the associated diseases of diabetes mellitus and rheumatoid arthritis and a more cephalad level of infection were the predisposing factors to paralysis.
Geriatric patients tend to have comorbidities. In some cases, geriatric patients develop spondylodiskitis from a primary focus infection elsewhere, such as urinary tract infection or pneumonia. Symptoms related to spondylodiskitis should be sought. Further, preexisting neurologic conditions such as sensorimotor neuropathy, motor neuron disease, and Parkinson’s disease can mask typical neurologic symptoms, and, contrarily, the neurologic symptoms of neural compression can be mistaken for diabetic neuropathy. Therefore, special attention should be paid to patients with these conditions.
Diagnosis
Laboratory Data
Laboratory markers, such as white blood cell (WBC) count, C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR), are well-known inflammatory biomarkers for infection, although they are not specific for spinal infection. 1 , 3 An elevation in these markers should not be taken as pathognomonic for an infection, but these markers serve as good screening and surveillance examinations in the diagnosis and treatment of spinal infection. Among the three biomarkers, Yee et al 27 assessed the relationship between both presentation and peak value and age in patients with pyogenic spondylosis who were intravenous drug users. In their results, the WBC count on presentation for patients 65 years of age or older was significantly higher than in those younger than 65 years of age; in contrast, there was no significant difference in ESR and CRP. Further, there was no significant difference in the peak values of these markers between the patients > 65 years old and the younger patients.
Recently, procalcitonin (PCT) has been used to distinguish bacterial infection from non-bacterial infection. Jeong et al 28 examined the clinical value of PCT in patients with spinal infection and found that the sensitivity of PCT was lower than that of CRP. Based on these results, they concluded that patients with spinal infection who showed an elevated serum PCT level should be examined for combined infection, and that antibiotics targeted to spinal infection and to the other combined infection should be prescribed. Yoon et al 29 reported that a combination of clinical data and biomarkers had better predictive value for differential diagnosis compared with biomarkers alone in their multicenter study for infectious spondylodiskitis.
It is important to perform urine and blood cultures to identify the organism when infection is highly probable, although the rate of identification of the organism is not very high. Prophylactic use of antibiotics usually masks the presence of the organism. It is also essential to check for anemia, albumin (malnutrition), and liver function status in the management of spinal infection in geriatric patients.
Imaging Studies
Imaging studies are very important for the diagnosis of spinal infection. They include plain radiographs, CT, magnetic resonance imaging (MRI), 18F-fluorodeoxyglucose–positron emission tomography (FDG-PET), scintigraphy, and single photon emission computed tomography (SPECT). There are no typical image findings in geriatric patients with spinal infection. The findings are relatively dependent on the stage of the disease.
Plain radiographs should be the first imaging study in all patients suspected of having spinal infection. They demonstrate various findings based on the stage of spinal infection. Plain radiographs are also suitable for follow-up study. At the early stage, 2 to 8 weeks after the onset of the infection, end-plate irregularity and disk space narrowing are found. After 8 to 12 weeks, bone destruction might be observed, which results in changes in spinal alignment. Spinal infection sometimes causes local kyphotic deformity, which can rapidly progress in geriatric patients who have severe osteoporosis. The presence of preexisting spinal degeneration, diffuse idiopathic skeletal hyperostosis (DISH), osteoporosis, spondylolisthesis, and spinal stenosis should also be determined in radiographs. These conditions can help determine the type of management and the approach for surgical treatment. Plain radiographs can also demonstrate soft tissue swelling, suggesting paraspinal abscess.
Computed tomography is used for the examination of bone destruction as well as soft tissue swelling. A moth-eaten appearance at both end plates on the affected IVD is seen in spondylodiskitis ( Fig. 17.1a–c ). CT is commonly used for image-guided biopsy.
Magnetic resonance imaging (MRI) is the gold standard for diagnosing spinal infection. It is more useful for the detection of spinal infection at an early stage, as compared with plain radiographs and CT. The sensitivity, specificity, and accuracy of MRI are reported to be 96%, 92%, and 94%, respectively, which are very high. 30 However, care must be taken in the interpretation of the MRI findings, because the presence of exuberant edema in adjacent zones may seem to indicate that the spinal infection covers a larger area than it does. The typical MRI pattern for the early stage of spinal infection is low intensity on T1-weighted imaging (T1 low) and high intensity on T2-weighted imaging (T2 high) at the IVD and the end plates ( Fig. 17.1d,e ). This finding suggests a fluid signal caused by bone edema. A mirror image, which shows upper and lower end-plate damage at the affected IVD level, can be seen on T1 and T2 MRI in patients with spinal infection. T2 high is sometimes clearly observed at the IVD. This suggests degradation of the IVD by the enzyme from bacteria. On the other hand, a metastatic lesion is often seen in the bone structure, such as the vertebral body and pedicle. The metastatic lesion occurs through blood flow; thus, the metastatic lesion preserves the structure of the IVD, as the IVD is an avascular tissue. This is a characteristic finding in the differential diagnosis between infectious and metastatic lesions in the spine.
Magnetic resonance imaging is the best noninvasive imaging tool to assess the spread of infection to the epidural space. An epidural abscess causes a neurologic deficit associated with intolerable local pain. The detection of the epidural abscess is very important for predicting the symptoms and also to determine the treatment. T1-gadolinium–enhanced MRI is sometimes useful for the detection of an epidural abscess.
Degenerative end-plate changes seen in MRI are called as Modic signs, which can sometimes be confused with spinal infections ( Fig. 17.2 ). Modic signs are of three types 31 : type 1 shows T1 low and T2 high, corresponding to vertebral body edema and hypervascularity; type 2 shows T1 high and T2 high, reflecting fatty replacements of red bone marrow; type 3 shows T1 low and T2 low, suggestive of subchondral bone sclerosis. Modic type 2 is most frequently seen, followed by Modic type 1 and then Modic type 3. Interestingly, some studies suggest that Modic type 1 is associated with infection. Albert et al 32 reported that treatment with anti biotics (amoxicillin-clavulanate) is effective in Modic type 1 patients with lower back pain (LBP). Ohtori et al 33 followed 71 patients with Modic type 1 for 2 years and found that three patients had a spinal infection that was found later, which might imply that some Modic type 1 patients have a hypovirulent infection, especially at the end plate. Other studies have demonstrated that FDG-PET or diffusion MRI might be useful for the detection of infection in patients with Modic type 1. 34 , 35 More evidence should be collected to confirm these findings. Table 17.1 lists the differential diagnosis of spondylodiskitis, spinal metastasis, and osteoporotic vertebral fracture using MRI findings.
18F-fluorodeoxyglucose–positron emission tomography is a three-dimensional imaging tool. 18F-fluorodeoxyglucose has increased uptake in tissue that is active for metabolism, such as malignant tumor and infection. The usefulness of FDG-PET has been shown in musculoskeletal infection, including spinal infection. A meta-analysis regarding FDG-PET as a diagnostic tool of spinal infection reported that the sensitivity was 97% and specificity was 88%; the positive predictive value was 0.96 and negative predictive value was 0.85. 36
Scintigraphy has been frequently used as a diagnostic tool for spinal infection as well as bone metastasis. Bone scan employing technetium 99m as the radioisotope-labeled tracer is useful for the detection of infection at the early stage. Gallium 67 is also used for the analysis of infection. These are two-dimensional imaging studies. SPECT is a three-dimensional imaging study. In the comparison between bone scan and SPECT, SPECT had higher sensitivity and specificity. It has been reported that SPECT detected an additional 30% of solitary vertebral lesions in patients with LBP. 37
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
