SLR - March 2020 - Megan L. Herring
Performance of A Rapid Two-Sequence Screening Protocol for Osteomyelitis of the Foot
Reference: Singer AD, Umpierrez M, Kakarala A, Schechter MC, Maceroli M, Sharma GB, Rajani RR. Performance of A Rapid Two-Sequence Screening Protocol for Osteomyelitis of the Foot. Skeletal Radiol. 2020 Jan 14. doi: 10.1007/s00256-019-03367-x.
Scientific Literature Review
Reviewed By: Megan L. Herring, DPM
Residency Program: Christus St. Patrick Hospital – Lake Charles, LA
Podiatric Relevance: Diabetic foot ulcerations (DFUs) are a condition that is prevalent in the podiatric field and can not only be costly and time-consuming to the patient but also to the medical team managing the ulcers. It is known that DFUs can lead to increased complications including but not limited to osteomyelitis. While bone biopsy remains the gold standard for diagnosing osteomyelitis, many physicians utilize advanced imaging to aid in the diagnosis and to assist in surgical planning. This study is a retrospective cohort that looks to determine whether a rapid two-sequence MRI protocol is sufficient and accurate in diagnosing osteomyelitis when compared to a full MRI protocol.
Methods: This study looked into 132 foot MRIs (18 hindfoot, 114 forefoot in 109 patients with 121 unique feet). Four musculoskeletal board certified radiologists reviewed the 132 full protocol MRIs. Two musculoskeletal radiologists blinded to clinical information read two sequences from the 132 full protocol MRIs. The two sequences included a T1 non-fat-suppressed sequence and one fluid sensitive fat-suppressed sequence. The majority was sagittal images; however, in 28 of the cases, coronal fluid sensitive images were used instead of sagittal. The two radiologists recorded the presence of osteomyelitis, reactive osteitis/early osteomyelitis, or an abscess. Those were then compared with the full protocol MRI reads.
Results: The rapid two-sequence radiologists and the full protocol MRI radiologists were classified in agreement with one of four classifications, fair (0.21-0.40), moderate (0.41-0.60), substantial (0.61-0.80), and almost perfect (0.81-1.0). Screening for normal versus abnormal MRI was highly sensitive and was in the almost perfect classification. Osteomyelitis agreement between the groups was substantial. Agreement for abscess was fair. The inclusion of IV contrast in the full protocol did not prove to be statistically significant in the diagnosis of osteomyelitis; however, it was significant in the diagnosis of an abscess.
Conclusions: The data collected and analyzed in this study suggests that this rapid two-sequence protocol is moderate to highly sensitive in ruling out osteomyelitis. Furthermore, it suggests that if osteomyelitis is not seen in the rapid two-sequence scan, then it is unlikely to be seen with additional imaging. This is an important consideration as it could help reduce costs and scan time. Many of our diabetic patients have renal compromise, so this rapid protocol could also help reduce their exposure to IV contrast. A major limitation of the rapid two-sequence protocol is real time reads. Not all facilities have a radiologist available in real time to read the two sequences to determine whether a full protocol is warranted. Another limitation is this study compared the rapid protocol to a full protocol, and it was not compared to the gold standard in diagnosing, bone biopsy. A higher level study could be performed comparing these results with bone biopsy. Despite these limitations, this study could help to create a way of triaging patients to allow medical resources to be used more efficiently.