Q J Med 2001; 94: 465-470
© 2001 Association of Physicians
A case ascertainment study of septic discitis: clinical, microbiological and radiological features
From the Departments of Rheumatology and 1 Radiology, Royal Bournemouth and Christchurch Hospitals, Bournemouth, UK
Received 18 May 2001 and in revised form 4 June 2001
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
We studied the spectrum of septic discitis presenting to two busy district general hospitals over 2.5 years (November 1996 to April 1999), surveying the case notes of all patients attending Royal Bournemouth and Poole Hospitals with probable septic discitis on magnetic resonance imaging (MRI). Twenty-two cases of septic discitis were identified, suggesting an annual incidence of 2/100 000/year. Seventy-three percent of patients were aged
65 years. In 91% of patients, back pain was the presenting symptom, with neurological signs evident in 45% of patients. Fever >37.5 °C was present in 68% of patients, and a marked elevation of erythrocyte sedimentation rate (ESR) in 91%. Diagnosis was originally by MRI in 86% of patients, with plain radiographs not diagnostic of discitis in the early stages of the infection. Staphylococcus aureus was the commonest pathogen (41%), but in 18% of patients, no organism was identified. The major predisposing factors to septic discitis were invasive procedures (41%), underlying cancer (25%) and diabetes (18%). Pre-existing degenerative spinal disease was found in 50% of patients. Four patients whose causative organism was not isolated had a poorer outcome: one death and three with increased morbidity. Our estimated incidence rate (2/100 000/year) is higher than that in previous studies and may be due to a higher detection rate with MRI and/or a genuine increase in the number of cases. Septic discitis should be considered in any patient who has severe localized pain at any spinal level, especially if accompanied by fever and elevated ESR, or in the immunosuppressed. |
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Septic discitis is an inflammatory process of the intervertebral disc which usually involves the discovertebral junction, and may extend in to the epidural space, posterior vertebral elements and paraspinal soft tissues.1 Spondylodiscitis accounts for 2% of all osteomyelitis, and may occur spontaneously, following surgery (especially following epidural anaesthesia), in the immunosuppressed, and following systemic infections.2
Discitis often presents with localized back pain, but additional neurological signs can develop, with radicular, meningeal or spinal cord involvement.1 The course of the discitis may be acute or chronic, and lack of specific symptoms may delay diagnosis. Although plain radiographs may suggest the diagnosis, magnetic resonance imaging (MRI) has become the investigation of choice (Figure 1
).1–6
|
This study was undertaken to determine the spectrum of septic discitis, presenting to two busy district general hospitals over a 2.5-year period.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
In the Royal Bournemouth and Poole Hospitals (serving a population of
500 000) MRI scanning is performed in all patients with possible septic discitis. Appropriate computer software allowed information retrieval of those patients with probable septic discitis on MRI imaging. Between November 1996 and April 1999, all such patients were identified. A diagnosis of definite septic discitis was based on MRI imaging and positive organism culture from biopsy, blood or urine. A diagnosis of septic discitis was thought highly likely in patients already on antibiotics, who remained culture-negative.
Note survey
Permission was obtained from the supervising doctors, and all the patients’ notes were examined according to a standard pro forma. The information recorded in each case comprised: (i) age; (ii) sex; (iii) delay in presentation (defined as the interval between the onset of symptoms attributable to septic discitis and presentation of the patient to the medical services); (iv) source of referral; (v) speciality referred to; (vi) delay in diagnosis; (vii) previous medical history; (viii) any invasive procedure or precipitating event; (ix) underlying immunosuppression, either due to disease or to cytotoxic drugs; (x) spinal level and extension of discitis; (xi) presenting signs and symptoms; (xii) white blood cell count (WBC), erythrocyte sedimentation rate (ESR), C-reactive Protein (CRP), liver function tests (LFT) and urea and electrolytes (U&E); (xiii) results of microbiological testing; (xiv) mode of diagnosis; (xv) results of radiological assessment by plain radiograph, MRI, CT, ultrasound (US) and isotope bone scan; (xvi) details of both intravenous and oral antibiotics; and (xvii) duration of hospital stay and outcome.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Twenty-two cases (10 male, 12 female) of septic discitis were identified during the study period. Of these 22, 18 (82%) were definite cases and four (18%) were highly likely. Their ages ranged from 42 to 83 years; 16 (73%) were aged
65 years.
Diagnosis and presentation
Delay in presentation varied, with nine (41%) patients presenting within 2 weeks, seven (32%) at between 2 and 6 weeks, and six (27%) patients after up to 6 months.
Fourteen patients were referred directly to the acute hospital admissions unit by their general practitioner, two were admitted via the accident and emergency department following self-referral, three were referred following routine rheumatology or orthopaedic out-patient attendance, and three more were identified following surgical procedures. Several hospital specialities were involved in their initial management (Table 1).
|
Most patients (91%) presented with back pain, although the patient with cervical discitis exhibited shoulder pain, and one patient had no pain at all. The pain ranged from mild discomfort to very severe pain (in five). Eleven patients showed anorexia, and three had significant weight loss. Radiating pain of nerve root origin occurred in 12 (54%) patients. Of these patients, eight (36%) had sensory disturbance, segmental weakness and/or a reduction in tendon reflexes. Urinary retention was seen in two patients.
Fever >37.5 °C was documented in 68% of cases, with ten (45%) suffering from night sweats. Significant haematological abnormalities included a raised WBC in 45% of the patients (based on normal range 4–11.5x109/l). A feature of septic discitis was a marked elevation of ESR, 50–99 mm/first h in six patients (27%) and >100 mm/h in 14 (64%). Abnormal LFT (27% of cases), derangement of urea and electrolytes (27% of cases) and normocytic, normochromic anaemia (73% of cases) were also noted.
The delay in diagnosis was variable, ranging from within one week in 10 (46%), within one month in six (27%), within 2 months in four (18%) and up to 6 months in two patients (9%).
Imaging
Diagnosis was originally made using MRI in 19 (86%) patients, isotope bone scan in one patient (later shown on MRI), and by CT in two patients (also confirmed by MRI).
Of 19 patients who had plain spinal radiographs, four (18%) results were completely normal; nine (41%) had abnormal but age-related findings only. Six radiographs were suggestive of discitis.
CT scanning was generally unhelpful: of 12 cases, only two were diagnostic of discitis.
Isotope scanning was suggestive of discitis in only two of six scans undertaken, one of which had already been shown on MRI. In the others, diagnoses of degenerative spinal disease and Paget's disease were made.
Causative organisms
Material was obtained by CT-guided biopsy in seven patients. Three results were negative, two showed inflammatory cells only, and two positive cultures (Streptococcus viridans and haemolytic Streptococcus) were isolated.
Twelve patients had abdominal ultrasound scanning, six of which were normal. Four patients had pelvic infection, with a psoas abscess, an infected haematoma, a wound abscess and systemic infection due to septicaemia being amongst the findings.
Pathogens and spinal level involvement
Causative organisms were identified in 82% of cases overall. Of 21 patients who had blood cultures, five were negative, and in seven it was the sole source of diagnosis. Urine cultures were positive in 4/18 patients, two of whom also had positive blood cultures (both Staphylococcus aureus). In two patients, Escherichia coli was cultured from urine, one of whom had Salmonella enteritidis on blood culture.
Other sources of positive bacterial culture included: CT-guided biopsy of disc 2, epidural line tip 1, nasal swab 1, wound 3, tip of myocardial infarct pacing wire 1.
Staphylococcus aureus was the commonest pathogen, comprising 41% of the total number of cases (Table 2). Gram-negative infections were uncommon, and three of the four patients thus affected were aged >65 years.
|
The regional distribution and anatomic extension of the spondylodiscitis in these patients is seen in Table 3
. The most commonly affected spinal level was the lumbar region. Six individuals had an epidural abscess, with one located in the cervical spine, two in the thoracic and three in the lumbar spine. Three patients had extra-spinal sites of infection, one psoas abscess, one presacral abscess and one bacterial endocarditis.
|
Risk factors
The major predisposing factor to septic discitis was ‘invasive procedures’ (41%), which included six epidurals (three as anaesthesia for surgical procedures, two caudal and one lumbar for sciatica) and three resections of rectal cancer. Other factors included underlying carcinoma (23%), diabetes mellitus (18%), bacterial endocarditis (9%) and one after septicaemia that accompanied a urinary tract infection (UTI). Pre-existing degenerative spinal disease was found in 50% of the patients. Immunosuppression due to rheumatoid arthritis and cytotoxic drug therapy was identified in one case.
Treatment
Of the 22 patients, 17 were initially started on intravenous (i.v.) antibiotics, which continued for between 2 days and 6 months. The patient with tuberculosis (TB) was treated with anti-tuberculosis drugs and the other four patients were started on oral antibiotics. A total of 14 different antibiotic combinations were used in the initial intravenous treatment, the most common component being flucloxacillin (41%). Three patients had 2–3 days i.v. treatment, seven patients had up to 2 weeks, six had between 3 weeks and 2 months, and one patient, whose discitis was caused by methicillin-resistant Staph. aureus (MRSA) had 6 months treatment.
Five patients with epidural collection of infection had surgical drainage; three of these required a laminectomy. The patient with a psoas abscess had it drained under ultrasound control.
All patients treated initially with i.v. antibiotics were then switched to oral treatment. The duration of treatment ranged from 4 days to 3 years (and still continuing). The median duration of treatment was 19 weeks. There were 14 different combinations of oral antibiotics used, most commonly including flucloxacillin (11 patients) and ciprofloxacin (9 patients).
All the patients were hospitalized, except for one, and the hospital stays lasted between 1 week and 5 months. Two patients were hospitalized more than once for treatment of their discitis.
Outcome
One patient, with discitis and epidural abscess following hemicolectomy, died during her in-patient stay, and six remain on oral antibiotics. Five patients did not regain premorbid mobility and function, and four patients continue to have neurological symptoms.
Of the four patients where the causative pathogen was not established, one died, two remain on antibiotics, and the fourth is dependent on residential care. None of the patients with Gram-negative bacilli are still on antibiotics, but two of each of those with Staph. aureus and streptococcal infections remain on antibiotics.
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Septic discitis is considered to be a rare condition. In a review of the literature by Sapico and Montgomerie,7 two cases/year were diagnosed in an active tertiary care hospital (serving a population of
500 000). This compares to an incidence rate in this study of two cases per 100 000 per year. Kapeller et al.1 described 41 cases within 7 years in a district general hospital which, like our two district general hospitals, served 500 000 inhabitants, giving an incidence rate of 1/100 000. The higher detection rate in our study may be due to a genuine increase in the number of cases, but other factors include improved diagnostic modalities, or demographic differences in the local population—30% aged >65 years compared with 18.7% nationally.8 In 86% of cases, diagnosis was established using MRI. In one case the diagnosis was missed on initial MRI report; two cases had CT scanning prior to their MRI which was diagnostic. These results are consistent with previous studies demonstrating MRI to be the method of choice for the diagnosis of septic discitis. It is non-invasive3,5,6 and of high sensitivity in the early stages, including better definition of the paravertebral and epidural space.2 Plain radiographs of the spine are not diagnostic of discitis in the early stages of infection, and alteration is rarely seen before 2–4 weeks.2,9 This concurs with our study, in that the six radiographs showing discitis were taken at a mean duration of 10 weeks following presentation. Maiuri et al.2 suggests that MRI differentiates tuberculosis infections from other bacterial spondylodiscitis. In fact, MRI imaging of our patient with TB was characteristic, showing relative preservation of the disc.
It can be seen from this study that a high index of suspicion is necessary, because when the delay in presentation and diagnosis was >3 months, all cases continued on long-term antibiotic treatment. Some 91% of the patients had both back pain and an ESR >50 mm/first h. In 63%, ESR was >100 mm/first h. In addition, a precipitating factor such as an invasive procedure and/or immunosuppressive disease should alert the clinician to this diagnosis. Our study confirms the reports of others1,10 that diabetes mellitus is a predisposing factor for septic discitis.
Outcome of septic discitis is influenced by age, as the average hospital stay for patients aged 70 and over was 13 weeks, whereas the mean stay for all age groups was 8 weeks. Pre-existing spinal disease also affects outcome, as patients may be less likely to complain of further back pain and doctors may be less likely to investigate it. Delay in diagnosis occurred because MRI was not thought warranted in the initial diagnostic tests. Outcome was poorer in our patients whose antibiotic treatment was started before the causative organism was detected. O'Grady et al.11 stated that the initial choice for treatment of skeletal infection or spinal epidural abscess should consist of flucloxacillin, together with an appropriate cephalosporin. This agrees with the predominant use of flucloxacillin in our study patients.
The duration of antimicrobial therapy was evaluated individually, depending on time taken for resolution of systemic and local signs and the restoration of a normal ESR. If clinical response was slow and ESR remained elevated, antibiotic therapy was continued.
In conclusion, our data suggest that septic discitis should be considered in any patient who has severe localized pain at any spinal level, especially if it is accompanied by fever and an elevated ESR, and they are elderly and/or a precipitating factor is identified. A definite diagnosis is essential for appropriate antimicrobial therapy of discitis and MRI should be accompanied by blood and urine cultures, and disc culture if necessary.
|
Notes |
|---|
Address correspondence to Dr N. Hopkinson, Department of Rheumatology, Royal Bournemouth and Christchurch Hospitals, Castle Lane East, Bournemouth BH7 7DW
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Kapeller P, Fazekas F, Krametter D, Koch M, Roob G, Schmidt R, Offenbacher H. Pyogenic infectious spondylitis: clinical, laboratory and MRI features. Eur Neurol1997; 38:94–8.[Web of Science][Medline]
2. Maiuri F, Iaconetta G, Gallicchio B, Manto A, Briganti F. Spondylodiscitis: clinical and magnetic resonance diagnosis. Spine1997; 22:1741–6.[Web of Science][Medline]
3. Kuker W, Mull M, Mayfrank L, Topper R, Thron A. Epidural spinal infection: variability of clinical and magnetic resonance imaging findings. Spine1997; 22:544–50.[Web of Science][Medline]
4. Honan M, White GW, Eisenberg GM. Spontaneous infectious discitis in adults. Am J Med1996; 100:85–9.[Web of Science][Medline]
5.
Sharif HS. Role of MR imaging in the management of spinal infections. Am J Roentgenol1992; 158:1333–45.
6. Thrush A, Enzmann D. MR imaging of infectious spondylitis. AJNR1990; 11:1171–80.[Abstract]
7. Sapico FL, Montgomerie JZ. Pyogenic vertebral osteomyelitis: report of nine cases and review of the literature. Rev Infect Dis1979; 1:754–76.[Web of Science][Medline]
8. Morris S, Bell H. A population atlas for Dorset: social statistics from the census. Dorset County Council, 1996.
9. Schindler OS, Wilson-MacDonald J. Acute MRI changes in infectious discitis: report on two cases. Eur Spine J1995; 4:360–1.[Medline]
10. Perronne C, Saba J, Behloul Z, Salmon-Ceron D, Leport C, Vilde JL, Kahn MF. Pyogenic and tuberculous spondylodiscitis (vertebral osteomyelitis) in 80 adult patients. Clin Infect Dis1994; 19:746–50.[Web of Science][Medline]
11. O'Grady F, Lambert HP, Finch RG, Greenwood D. Antibiotic and chemotherapy: Anti-infective agents and their use in therapy, 7th edn. Edinburgh, Churchill Livingstone, 1997.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D A Enoch, J S Cargill, R Laing, S Herbert, T W Corrah, and N M Brown Value of CT-guided biopsy in the diagnosis of septic discitis J. Clin. Pathol., June 1, 2008; 61(6): 750 - 753. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. A. Michel, C. W. A. Pfirrmann, N. Boos, and J. Hodler CT-Guided Core Biopsy of Subchondral Bone and Intervertebral Space in Suspected Spondylodiskitis. Am. J. Roentgenol., April 1, 2006; 186(4): 977 - 980. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Hatton, M. Gupta, P. Balint, and M. Field Septic discitis presenting following intravenous cannulation QJM, March 1, 2002; 95(3): 189 - 191. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||