Background: Osteomyelitis is an infection of the bone. Medical imaging tests, such as radiography, ultrasound, magnetic resonance imaging (MRI), single-photon emission computed tomography (SPECT) and positron emission tomography (PET), are often used to diagnose osteomyelitis.
Objectives: To systematically review the evidence on the diagnostic accuracy, inter-rater reliability and implementation of imaging tests to diagnose osteomyelitis.
Data sources: We conducted a systematic review of imaging tests to diagnose osteomyelitis. We searched MEDLINE and other databases from inception to July 2018.
Review methods: Risk of bias was assessed with QUADAS-2 [quality assessment of diagnostic accuracy studies (version 2)]. Diagnostic accuracy was assessed using bivariate regression models. Imaging tests were compared. Subgroup analyses were performed based on the location and nature of the suspected osteomyelitis. Studies of children, inter-rater reliability and implementation outcomes were synthesised narratively.
Results: Eighty-one studies were included (diagnostic accuracy: 77 studies; inter-rater reliability: 11 studies; implementation: one study; some studies were included in two reviews). One-quarter of diagnostic accuracy studies were rated as being at a high risk of bias. In adults, MRI had high diagnostic accuracy [95.6% sensitivity, 95% confidence interval (CI) 92.4% to 97.5%; 80.7% specificity, 95% CI 70.8% to 87.8%]. PET also had high accuracy (85.1% sensitivity, 95% CI 71.5% to 92.9%; 92.8% specificity, 95% CI 83.0% to 97.1%), as did SPECT (95.1% sensitivity, 95% CI 87.8% to 98.1%; 82.0% specificity, 95% CI 61.5% to 92.8%). There was similar diagnostic performance with MRI, PET and SPECT. Scintigraphy (83.6% sensitivity, 95% CI 71.8% to 91.1%; 70.6% specificity, 57.7% to 80.8%), computed tomography (69.7% sensitivity, 95% CI 40.1% to 88.7%; 90.2% specificity, 95% CI 57.6% to 98.4%) and radiography (70.4% sensitivity, 95% CI 61.6% to 77.8%; 81.5% specificity, 95% CI 69.6% to 89.5%) all had generally inferior diagnostic accuracy. Technetium-99m hexamethylpropyleneamine oxime white blood cell scintigraphy (87.3% sensitivity, 95% CI 75.1% to 94.0%; 94.7% specificity, 95% CI 84.9% to 98.3%) had higher diagnostic accuracy, similar to that of PET or MRI. There was no evidence that diagnostic accuracy varied by scan location or cause of osteomyelitis, although data on many scan locations were limited. Diagnostic accuracy in diabetic foot patients was similar to the overall results. Only three studies in children were identified; results were too limited to draw any conclusions. Eleven studies evaluated inter-rater reliability. MRI had acceptable inter-rater reliability. We found only one study on test implementation and no evidence on patient preferences or cost-effectiveness of imaging tests for osteomyelitis.
Limitations: Most studies included < 50 participants and were poorly reported. There was limited evidence for children, ultrasonography and on clinical factors other than diagnostic accuracy.
Conclusions: Osteomyelitis is reliably diagnosed by MRI, PET and SPECT. No clear reason to prefer one test over the other in terms of diagnostic accuracy was identified. The wider availability of MRI machines, and the fact that MRI does not expose patients to harmful ionising radiation, may mean that MRI is preferable in most cases. Diagnostic accuracy does not appear to vary with the potential cause of osteomyelitis or with the body part scanned. Considerable uncertainty remains over the diagnostic accuracy of imaging tests in children. Studies of diagnostic accuracy in children, particularly using MRI and ultrasound, are needed.
Study registration: This study is registered as PROSPERO CRD42017068511.
Funding: This project was funded by the National Institute for Health Research Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 23, No. 61. See the NIHR Journals Library website for further project information.
Keywords: DIAGNOSTIC ACCURACY; MAGNETIC RESONANCE IMAGING; OSTEOMYELITIS; POSITRON EMISSION TOMOGRAPHY; SINGLE PHOTON EMISSION COMPUTED TOMOGRAPHY; SYSTEMATIC REVIEW.
Osteomyelitis is an infection of the bone and is treated with antibiotics. Left untreated, it can cause permanent damage and can lead to amputation. The best method to diagnose osteomyelitis is to take a bone sample (bone biopsy) but this is invasive and painful. Imaging may help target the best locations for biopsies or remove the need for a biopsy entirely. Several methods are available, including radiography, ultrasound, magnetic resonance imaging (MRI), single-photon emission computed tomography (SPECT) and positron emission tomography (PET). This project systematically reviewed the relevant literature to determine which tests are the most accurate and relevant for clinical practice. All types of patients and all types of osteomyelitis were reviewed. Studies were pooled using statistical methods (meta-analyses) to estimate the overall accuracy of the imaging tests. The review identified 81 studies and concluded that MRI, PET and SPECT all had similar accuracy, correctly identifying over 85% of people who did have osteomyelitis and over 80% of people who did not have osteomyelitis. Radiography and computed tomography were less accurate. Modern forms of scintigraphy have accuracy similar to PET or MRI. There was no evidence that the accuracy of the imaging tests was different depending on the cause of osteomyelitis or which body part was affected. In particular, diagnostic accuracy in people with diabetic foot ulcers was similar to other types of osteomyelitis in adults. There was not enough evidence about which tests are most accurate in children, so further studies in children are needed.