Outcomes after inter-hospital transfer of intensive care patients with haemorrhagic stroke: a 5-year retrospective review.

BACKGROUND: Inter-hospital transfer (IHT) of intensive care patients is a limited resource. We assessed the outcomes of patients with haemorrhagic stroke requiring IHT and intensive care and aimed to identify early prognostic factors of poor neurological outcome. METHODS: We conducted a retrospective observational cohort study of patients admitted to a single tertiary intensive care unit (ICU) with haemorrhagic stroke after IHT between January 2014 and December 2018. Primary outcome was poor neurological outcome (modified Rankin Scale ≥4 at time of discharge from hospital or rehabilitation unit). Secondary outcomes were mortality rate, rate of intervention, rate of organ donation surgery (ODS) and potentially avoidable transfer (PAT). PAT was defined as transfer where the patient did not receive an intervention and had a poor neurological outcome. RESULTS: Ninety patients were included in this study, 48 with intracerebral haemorrhage (ICH) and 42 with subarachnoid haemorrhage (SAH). Fifty-one (56.7%) patients had a poor neurological outcome, including 30 (33%) who died. Factors significantly associated with poor neurological outcome included age > 80 years, lower presenting Glasgow Coma Score (GCS) and bilaterally fixed and dilated pupils. Stepwise logistic regression demonstrated history of hypertension as significantly associated with poor neurological outcome in patients with ICH (P = 0.021). Seven (7.8%) patients had ODS. Sixty-four (71.1%) patients received intervention and 20 (22.2%) transfers were potentially avoidable. CONCLUSIONS: Patients in this cohort are at high risk of poor neurological outcome. Prognostic factors identified in this study may help referring, retrieval and receiving clinicians to discuss futility prior to pursuing IHT.

Five-year analysis of magnetic resonance imaging as a screening tool in women at hereditary risk of breast cancer.

INTRODUCTION: Women at very high risk of breast cancer are recommended to undertake enhanced surveillance with annual MRI in addition to mammography. We aimed to review the performance of breast MRI as a screening modality over its first 5 years at our institution. METHODS: The study used a retrospective review using prospectively collected data from a consecutive series of women at high risk of developing breast cancer undergoing surveillance MRI. RESULTS: Two hundred twenty-three women had at least one screening MRI. The median age was 42 years old. Sixty-nine (30.9%) were confirmed genetic mutation carriers. The remaining 154 (69.1%) women were classified as high risk based on family history, without a confirmed genetic mutation. Three hundred forty screening MRI studies were performed. Of these, 69 patients (20.3%) were recalled for further assessment. There was a significant reduction in the recall rate throughout the study for prevalent screens, from 50% (17/34) in 2008 to 14% (9/54) in 2011 (P = 0.004). The overall biopsy rate was 39 in 340 screens (11.5%). Four cancers were identified. Three were in confirmed BRCA1/BRCA2 mutation carriers, and one was found to be a carrier after the cancer was diagnosed. All four were identified as suspicious on MRI, with two having normal mammography. The cancer detection rate of MRI was 1.2% (4/340 screens). The overall positive predictive value was 7.0%, 6.7% for prevalent screens and 7.1% for subsequent screens. CONCLUSIONS: Breast MRI as a screening modality for malignant lesions in women with high hereditary risk is valuable. The recall rate, especially in the prevalent round, improved with radiologist experience.