ABSTRACT
Purpose: An MRI-based scoring tool assessing surgical repair after proximal hamstring avulsions may provide benefit in the context of research, while serial post-operative MRI will provide insight of what to expect in the clinical context of early re-injury requiring imaging. This study developed and assessed the reliability of a Proximal Hamstring Objective Magnetic Resonance Imaging Score (PHOMRIS), further assessing MRI-based repair status and its correlation with patient-reported outcome. Methods: 15 patients that underwent proximal hamstring surgical repair underwent MRI and clinical review pre-operatively and at 3-, 6- and 12-months. Clinical scores included the Lower Extremity Functional Scale (LEFS), the Perth Hamstring Assessment Tool (PHAT) and Tegner Activity Scale (TAS). The MRI-based tool assessed the conjoint (semitendinosus & biceps femoris) and semimembranosus insertion components based on bone-tendon healing, signal and retraction. Inter- and intra-observer reliability of the tool was assessed. Results: Inter-observer reliability indicated a strong correlation for the semimembranosus (rho = 0.827, p < 0.0001) and conjoint (rho = 0.851, p < 0.0001) components. Intra-observer reliability indicated a strong correlation for the semimembranosus (rho = 0.852, p < 0.0001) and conjoint (rho = 0.996, p < 0.0001) components. All clinical scores and the semimembranosus hamstrings component MRI score significantly improved (p < 0.05) over time, though the conjoint component did not (p = 0.219). At 12 months, a higher LEFS was significantly associated with a better semimembranosus MRI score (r = -0.57, p = 0.042), though no other significant correlations (p > 0.05) were observed between clinical and MRI measures. Conclusions: Excellent reliability was observed for the MRI-based scoring tool, which may prove useful in both a research and clinical setting.
ABSTRACT
INTRODUCTION: To evaluate the rates of inferior vena cava (IVC) filter retrieval and the number of patient's lost to follow-up, before and after the establishment of an IVC filter database and interventional radiology (inserting physician) led follow-up. METHODS: On the 1st of June 2012, an electronic interventional radiology database was established at our Institution. In addition, the interventional radiology team took responsibility for follow-up of IVC filters. Data were prospectively collected from the database for all patients who had an IVC filter inserted between the 1st June 2012 and the 31st May 2014. Data on patients who had an IVC filter inserted between the 1st of June 2009 to the 31st of May 2012 were retrospectively reviewed. Patient demographics, insertion indications, filter types, retrieval status, documented retrieval decisions, time in situ, trackable events and complications were obtained in the pre-database (n = 136) and post-database (n = 118) cohorts. RESULTS: Attempted IVC filter retrieval rates were improved from 52.9% to 72.9% (P = 0.001) following the establishment of the database. The number of patients with no documented decision (lost to follow-up) regarding their IVC filter reduced from 31 of 136 (23%) to 0 of 118 patients (P = < 0.001). There was a non-significant reduction in IVC filter dwell time in the post-database group (113 as compared to 137 days, P = 0.129). CONCLUSION: Following the establishment of an IVC filter database and interventional radiology led follow-up, we demonstrate a significant improvement in the attempted retrieval rates of IVC filters and the number of patient's lost to follow-up.
