AAHS Annual Meeting
Back to main AAHS site
Annual Meeting Home
Preliminary Program
Past & Future Meetings


Back to 2015 Annual Meeting Program


Sensitivity and Specificity of Radiographs in the Diagnosis of Carpometacarpal Joint Injuries
S. Hassan, MB, ChB, MRCS; A. Aziz, MRCS; T. R. C. Davis, MBChB, FRCS, MD; ND Downing, MA, BM, BCh, FRCS; R. W. Trickett, MBBCh, MSc, FRCS
Department of Hand Surgery, Queen's Medical Centre, Nottingham, United Kingdom

Introduction: Injuries to the 4th and/or 5th carpometacarpal joints (CMCJs) are often described as rare injuries. These “fractures of frustration” can be missed at initial presentation due to misinterpretation of the radiographs, perhaps leading to an underestimation of the incidence. We aim to determine the sensitivity and specificity of standard hand radiographs in these injuries.

Materials & Methods: Four patients with confirmed 4th/5th CMCJ injuries were identified from our trauma theatre records. Four patients with normal hand radiographs were used as controls. All had undergone 4 radiographic views – anteroposterior, lateral, pronated oblique and supinated oblique. Radiographs were anonymised, duplicated and randomly ordered.

Radiographs were shown to 2 cohorts of higher orthopaedic trainees (n=1 and 11) in a timed “test”. Test 1 showed the radiographs only and no clinical history; Test 2 described a history and clinical photograph consistent with the injury. Emergency Department Nurse Practitioners (ENP) viewed the same radiographs (as 8 individual cases) in an untimed test without clinical history (Test 3). Diagnoses for each radiograph/case were recorded. Sensitivity, specificity, positive predictive value and negative predictive value were calculated for each radiographic view.
Results: The pronated oblique view had the highest sensitivity and specificity as well as yielding the most correctly interpreted radiographs. The supinated oblique view performed least well throughout. In test 3, the diagnosis was missed completely despite reviewing all 4 radiographs in 22% of cases. In no cases did the supinated oblique view correctly identify an injury when the other 3 views had been interpreted as normal.
Table 1: The calculated sensitivity (Sens), specificity (Spec), positive predictive value (PPV), negative predictive value (NPV) and total percentage correct for each of the radiographic views.

Conclusions: The pronated oblique view offers the best chance of a successful diagnosis. A succinct history and examination aids diagnosis and suspected injuries should be referred on this basis. We recommend 3 views: anteroposterior, lateral and pronated oblique. Education is essential for correct interpretation.

Conclusions: The pronated oblique view offers the best chance of a successful diagnosis. A succinct history and examination aids diagnosis and suspected injuries should be referred on this basis. We recommend 3 views: anteroposterior, lateral and pronated oblique. Education is essential for correct interpretation.



Back to 2015 Annual Meeting Program
© 2017 American Association for Hand Surgery. Privacy Policy.