Anshul Agarwal MD, PhD, Presenter: Nothing to Disclose

Ji Young Buethe MD, Abstract Co-Author: Nothing to Disclose

Hythem Adnan Omar MD, Abstract Co-Author: Nothing to Disclose

Jon A. Anderson PhD, Abstract Co-Author: Nothing to Disclose

Orhan Kemal Oz MD, Abstract Co-Author: Nothing to Disclose

William A. Erdman MD, Abstract Co-Author: Nothing to Disclose

To evaluate if 99m-TcWBC-SPECT/CT (WBC/CT) imaging can aid in assessing severity of lower extremity infections by predicting outcome in patients where MRI scans were either clinically incongruent (MRci) or inconclusive (MRin).

A retrospective chart review was conducted in 36 cases in which MRI and WBC/CT were both performed to evaluate for osteomyelitis. MRci was defined as clinical impression inconsistent with MRI interpretation. Cases were considered MRin when the presence or absence of osteomyelitis could not be established by MRI criteria, as recorded in the impression. MRI and WBC/CT interpretations (no osteomyelitis, osteomyelitis, or inconclusive) were correlated with clinical outcomes. MRI or WBC/CT interpreted as no osteomyelitis with successful resolution of infection after <6 weeks of antibiotic therapy were considered true negative (TN); those with osteomyelitis and failure after 6 weeks antibiotic therapy (readmission, amputation or additional antibiotic therapy) were considered true positive (TP). Cases interpreted as no osteomyelitis who failed <6 weeks antibiotic therapy were considered false negative (FN) while those interpreted as osteomyelitis with successful response to < 6 weeks antibiotics were considered false positive (FP).  

In 19 MRci cases, analysis of each modality’s ability to predict clinical outcome was performed. MRI had 8 TP, 3 TN, 7 FP, 1 FN yielding a sensitivity of 89%, specificity 30%, PPV 53%, NPV 75%, and accuracy 58%. In the same group, WBC/CT had 8 TP, 6 TN, 4 FP, 1 FN yielding the same sensitivity of 89% but a higher specificity 60%, PPV 67%, NPV 86%, and accuracy 74%. In 17 MRin cases, WBC/CT had 8 TP, 4 TN, 0 FP, 5 FN yielding a sensitivity of 62%, specificity 100%, PPV 100%, NPV 44%, and accuracy 71% for predicting outcome. Overall, WBC/CT had sensitivity of 73%, specificity 71%, PPV 80%, NPV 63%, and accuracy 72% in clinical outcome prediction.  

When MRI is clinically incongruent WBC/CT is superior for predicting response to therapy. When MRI findings are inconclusive, WBC/CT scanning offers a high specificity and positive predictive value for identifying disease with likelihood of failing therapy.

Although MRI is effective in diagnosing osteomyelitis, it may be inconclusive or clinically incongruent. In such cases, WBC/CT is useful in assessing disease severity and predicting therapy outcome.

Agarwal, A, Buethe, J, Omar, H, Anderson, J, Oz, O, Erdman, W, Utility of 99m-Tc WBC SPECT/CT in Predicting Outcome of Diabetic Lower Extremity Infections with Clinically Incongruent or Inconclusive MRI.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11006169.html