Diagnostic Performance of a telephone questionnaire for fracture-related infections (FRIs) in open tibia fracture patients in Tanzania.

INTRODUCTION

Fracture-related infections (FRIs) are a major cause of trauma-associated morbidity worldwide. In 2018, an expert group supported by the AO Foundation, European Bone and Joint Infection Society developed a consensus definition of FRI. Still, there is limited knowledge on the applicability of this definition in low- and middle-income countries (LMICs). Given the unique barriers that cause low follow-up rates for orthopaedic trauma patients in LMICs, this study aims to evaluate the diagnostic performance of a telephone questionnaire in identifying patients with FRIs after open tibia fracture fixation in Tanzania.

MATERIALS AND METHODS

Patients from a randomized controlled trial investigating the infection prevention benefit of locally applied gentamycin for open tibial fractures were included. Patients completed FRI based telephone questionnaires 7-10 days prior to scheduled follow-ups at 6 weeks, 3 months, 6 months, 9 months, and 1 year. The questionnaire included two "confirmatory" criteria questions for FRI (i.e., open wound and purulent drainage) and three "suggestive" criteria questions (i.e., wound drainage, fever, and warmth). Contingency tests were performed to identify the sensitivity and specificity between answers and adjudicated FRI diagnoses at the corresponding in-person follow-up. Data was analysed using STATA version 15.0 and MedCalc's online diagnostic test calculator.

RESULTS

There were a total of 234 complete questionnaires and 85 unique patients included. The sensitivity and specificity of having any positive answer in the questionnaire was highest at 6 months (100 % and 92.5 %, respectively). For all time-points pooled, sensitivity was 71.4 % and specificity was 93.0 %. Drainage had the highest sensitivity (71.4 %) while fever had the highest specificity (99.6 %). For confirmatory criteria, sensitivity was 14.3 % and specificity was 96.0 %. Contrastingly, the sensitivity for suggestive criteria was higher (71.4 %), with a similar specificity (93.8 %).

CONCLUSION

Our study indicates that telephone questionnaires have adequate diagnostic performance when assessing FRIs. The presence of drainage identified the majority of patients with FRI, and specificities were high across confirmatory and suggestive criteria. Our study is one of the first to evaluate telephone questionnaires as a diagnostic tool for FRIs in patients with open tibia fractures in a LMIC hospital and validates the FRI consensus definition criteria.