A systematic review and meta-analysis were conducted per PRISMA guidelines. Multiple databases were searched for studies examining ultrasound, MRI, or CT for primary (unrepaired) hand flexor tendon injuries in adults or cadavers. Studies with 3 or more cases were included. Two reviewers screened titles, abstracts, and full papers independently. Then, study details were extracted including imaging techniques, sensitivity/specificity data, and other noteworthy findings. Meta-analysis used random-effects models to pool diagnostic accuracy measures with 95% confidence intervals. Study heterogeneity was measured with I² statistics and forest plots were created for visualization. A combined sensitivity-specificity plot allowed comparison between imaging types.
RESULTS
Sixteen studies qualified for analysis: 12 ultrasound, 6 MRI, and 1 CT study. Three studies examined both ultrasound and MRI. Ultrasound studies had pooled sensitivity of 91% (95% CI: 0.78-0.99) and specificity of 97% (95% CI: 0.88-1.00). MRI studies showed pooled sensitivity of 88% (95% CI: 0.55-1.00) and specificity of 100% (95% CI: 0.95-1.00). The CT study had 92% sensitivity and 100% specificity. Heterogeneity was high for sensitivity in ultrasound (I² = 88.74%) and MRI (I² = 80.86%) studies. Specificity heterogeneity was moderate for ultrasound (I² = 62.35%) and absent for MRI (I² = 0%).
CONCLUSIONS
This analysis demonstrates that both ultrasound and MRI yield high sensitivity and specificity for detecting primary flexor tendon injuries of the hand. Ultrasound has clear practical advantages due to lower cost, ability to visualize tendon movement in real time, and greater availability in clinics and emergency departments. Though more expensive and less accessible, MRI may be more specific for ruling in flexor tendon injuries. We therefore recommend the use of ultrasound, and secondarily MRI, as effective diagnostic tools in cases of suspected tendon injuries when clinical exam is indeterminate.
