An appropriateness review of urgent in-hours nontrauma CT brain scans at a single tertiary referral centre in South Africa - are we scanning rationally?

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
Aim: To determine if urgent, non-trauma in-hours CT head scan requests in the GSH department of radiology are being appropriately requested and completed, as well as determining the radiation dose for each study. Secondly, to make recommendations to improve local practice based on the findings of this study. Methods: A retrospective study was undertaken of 100 qualifying in-hours urgent, nontrauma CT head scans completed at the GSH department of radiology between 01/10/2015 and 31/03/2016. All qualifying CT request data and dose records were collected and anonymised, after ethical and institutional approval. Three radiologists at GSH were enlisted to review the request information. Each request was reviewed and categorised by both the researcher and each consultant individually to determine the indication and appropriateness. The researcher used previously published, objective criteria (Rothrock Criteria) to review requests, while the radiologists used their own interpretation of accepted local practice. The researcher recorded positive and negative scan outcomes, radiation doses and calculated the Effective Dose (ED) for each study. Results were recorded in Excel and statistical analysis using weighted Kappa analysis was undertaken. Results: Study cohort CT scans made up 15.6% of the total emergency head CT scans over the study period. The mean patient age was 52.3 years (range: 18.8-87.4 years). One-third (34; 34%) were older than 60 years with 33 (97%) having at least 2 positive Rothrock criteria. Most CT scans (86%) consisted of a single study, while the remaining comprised two or more. Average ED was 3.27 mSv (range: 1.03-4.33 mSv). 52 (52%) participants had abnormal CT findings, independent of age-group. Discrepancy in assigning study indication and appropriateness between the researcher and consultants was present, with at best moderate agreement (weighted-Kappa range 0.09-0.52). The researcher showed slight to fair agreement between scan outcome and request appropriateness using the Rothrock criteria (weighted kappa 0.20; 95% CI: 0.06-0.35, p=.00861), while the consultant consensus performed slightly worse (0.10; 95% CI: -0.05-0.26, p=.19728). Conclusion: Retrospective application of the Rothrock criteria to patient referrals produced better correlation with outcome than current departmental practice. However, there is at best moderate agreement between consultants with regard to classification of referrals which could negatively affect the application of Rothrock criteria in practice. Incorporating the Rothrock criteria into published departmental guidelines, in conjunction with other interventions to improve clinician requesting practices, is recommended. The formation of a Quality Assurance team and the use of existing dose-reducing techniques may assist in reducing radiation doses further.