A comparison of hospital episode statistics and traditional methods to identify outcomes in a randomized trial; a sub-study of HEAT-PPCI.

BACKGROUND: This study aims to compare information from hospital episode statistics (HES) and traditional direct patient contact to identify readmission and clinical events in the follow-up of a randomized controlled trial (RCT). METHODS: The study followed 1812 patients for 28 days using direct contact (DC). In addition, we obtained HES for this period. We examined medical records for all suspected readmissions and determined confirmed events by adjudication. We compared the ability of the individual DC and HES methods to determine readmission and the occurrence of trial-specific events, confirmed at adjudication. RESULTS: In the ascertainment of readmission, compared to DC, HES demonstrated a trend towards better sensitivity (identifying 153/166 = 92.2% versus 144/166 = 86.7%; difference = 5.4%, 95% CI: 0.1-11.5%) and better specificity (1492/1492 = 100% versus 1426/1492 = 95.5%; difference = 4.4%, 95% CI: 4.2-5.6%).An examination of HES coding does not identify rates for specific events that match those from adjudication, with limitations in both sensitivity and specificity. CONCLUSION: HES is effective in the ascertainment of readmission and is a useful tool in follow-up. Information from HES provides a reflection of a patient's course and associated cost, as perceived by the healthcare system. Future studies could modify outcome definitions to reflect episode coding.

Radial versus femoral vascular access in ST-elevation myocardial infarction: Are the results of femoral operators unfairly represented in observational research?

BACKGROUND: Recent randomized controlled trials comparing femoral and radial access in primary percutaneous coronary intervention (PPCI) have shown conflicting results regarding the incidence of major adverse cardiovascular events (MACE) and major bleeding. METHODS: Using data from the HEAT-PPCI trial, we compared the primary efficacy (all-cause mortality, stroke, new myocardial infarction or unplanned repeat revascularization) and safety (major bleeding BARC 3-5) outcomes at 28 days, by final access site used (radial or femoral) and by default operator type. We then assessed outcomes in femoral cases performed by both operator types. RESULTS: Radial access (RA) was associated with fewer MACE (91/1472 = 6.2% vs. 36/332 = 10.8% P = .003) and major bleeding events (38/1472 = 2.6% vs 22/332 = 6.6% P = .001) when compared to femoral access (FA). When analyzing outcomes by default operator type, there was a similar incidence of MACE (111/1575 = 7% vs 16/229 = 7% P = .97) and major bleeding events (49/1575 = 3.1% vs 11/229 = 4.8% P = .18). In cases where FA was performed by default radial operators, there was a higher rate of MACE (22/122 = 18% vs 14/210 = 6.7% P = .003) and major bleeding events (11/122 = 9% vs 11/210 = 5.2% P < .001), potentially explained by a higher risk profile in these cases. CONCLUSION: Default femoral operators achieved comparable outcomes when compared to default radial operators. The less favorable outcomes observed in FA cases may result from its selective use by radial operators in high risk cases.

Deploying artificial intelligence software in an NHS trust: a how-to guide for clinicians.

Over the past 10 years, artificial intelligence (AI) has become one of the fastest-growing sectors in healthcare. There are now numerous new technologies designed to cut costs and improve diagnoses and treatment pathways. However, there is significant scepticism amongst National Health Service (NHS) staff regarding the usefulness of AI and it's cost to the NHS. This has likely resulted in underuse and slow adoption of software that may revolutionize our healthcare system and ensure its continued survival and effectiveness. Several governing bodies have put forward guidance on the safe and effective adoption of AI tools, but this rarely covers the reality of selecting and deploying new software. This article set out clear guidance on the practicalities and pitfalls of deploying digital solutions in healthcare, using the example of a deep learning algorithm designed to improve the accuracy of chest X-ray (CXR) interpretation in the emergency department.

Using Artificial Intelligence to Stratify Normal versus Abnormal Chest X-rays: External Validation of a Deep Learning Algorithm at East Kent Hospitals University NHS Foundation Trust.

Background: The chest radiograph (CXR) is the most frequently performed radiological examination worldwide. The increasing volume of CXRs performed in hospitals causes reporting backlogs and increased waiting times for patients, potentially compromising timely clinical intervention and patient safety. Implementing computer-aided detection (CAD) artificial intelligence (AI) algorithms capable of accurate and rapid CXR reporting could help address such limitations. A novel use for AI reporting is the classification of CXRs as 'abnormal' or 'normal'. This classification could help optimize resource allocation and aid radiologists in managing their time efficiently. Methods: qXR is a CE-marked computer-aided detection (CAD) software trained on over 4.4 million CXRs. In this retrospective cross-sectional pre-deployment study, we evaluated the performance of qXR in stratifying normal and abnormal CXRs. We analyzed 1040 CXRs from various referral sources, including general practices (GP), Accident and Emergency (A&E) departments, and inpatient (IP) and outpatient (OP) settings at East Kent Hospitals University NHS Foundation Trust. The ground truth for the CXRs was established by assessing the agreement between two senior radiologists. Results: The CAD software had a sensitivity of 99.7% and a specificity of 67.4%. The sub-group analysis showed no statistically significant difference in performance across healthcare settings, age, gender, and X-ray manufacturer. Conclusions: The study showed that qXR can accurately stratify CXRs as normal versus abnormal, potentially reducing reporting backlogs and resulting in early patient intervention, which may result in better patient outcomes.

Misdiagnosis of Pulmonary embolism in a Fontan's patient: When standard protocol CT Pulmonary angiogram is inadequate.

Patients who have undergone a Fontan's procedure have an abnormal circulation that presents a unique challenge when performing computed tomography pulmonary angiograms. In a standard imaging protocol, contrast is injected into the upper limb veins that feed into the superior vena cava. In Fontan's patients the Computed tomography pulmonary angiograms bypasses the heart and preferentially fills the right lung, with only a small amount of mixture of contrast and noncontrast blood in the pulmonary arteries. In this article, we present the case of a 35-year-old female complaining of chest and abdominal pain with oxygen saturations of 85% on room air. Computed tomography pulmonary angiograms showed suboptimal imaging of the left lung and apparent filling defects in the right lung suggesting a radiological diagnosis of a pulmonary embolism. The abnormal flow and distribution of contrast in the pulmonary arteries can result in a false positive diagnosis of pulmonary embolism. To overcome this, experts advise using a dual-injection of contrast via upper and lower limb central veins to achieve optimal imaging.

Twelve-month mortality from the "How Effective are Antithrombotic Therapies in Primary Percutaneous Coronary Intervention (HEAT-PPCI) Trial".

BACKGROUND: There is ongoing uncertainty regarding the safety and efficacy of unfractionated heparin and bivalirudin when used for systemic anticoagulation in patients undergoing primary percutaneous coronary intervention (PPCI). This paper reports 12-month mortality from the HEAT-PPCI randomised trial. METHODS: In this open-label, randomised controlled trial (RCT) we enrolled consecutive adults with suspected ST-elevation myocardial infarction (STEMI). Patients were randomised to heparin (bolus 70 U/kg) or bivalirudin (bolus 0.75 mg/kg followed by an infusion 1.75 mg/kg/h for the duration of the procedure). We report the pre-specified secondary outcome of all-cause mortality at 12 months. Mortality was classified as cardiovascular or not, blinded to treatment allocation. Deaths in the first 28 days were classified by formal event adjudication and later events classified from death certificates. RESULTS: Mortality status at 12 months was obtained for 1805/1812 = 99.6% of participants. Overall mortality was 160/1812 = 8.9%. There were more deaths in those randomised to bivalirudin (95/902 = 10.5% vs 65/903 = 7.2%; HR 1.48; 95% CI 1.08 to 2.03; p = 0.015). Most deaths were classified as cardiovascular (71/902 = 7.9% in the bivalirudin group and 53/904 = 5.9% in the heparin group). The difference between the rates of cardiovascular deaths in each treatment group did not reach statistical significance: HR 1.35; 95% CI 0.95 to 1.93; p = 0.095. CONCLUSIONS: At 12 months, treatment with bivalirudin, rather than heparin, was associated with a higher rate of all-cause mortality. Cardiovascular mortality was higher with bivalirudin although this difference was not statistically significant.

A vulnerable plaque identified on CT coronary angiography: when should we act in stable coronary artery disease?

CT coronary angiography (CTCA) is increasingly being used to diagnose coronary artery disease (CAD). Recent technological advancements, including dual energy CT and improved gantry times, have led to the ability to image coronary arteries with excellent spatial resolution at low radiation doses. Atheromatous plaques can be identified using CTCA and assessed to establish the risk of acute coronary syndrome from each individual plaque. If CTCA identifies CAD, it should then be used in conjunction with functional testing or invasive angiography with physiological assessment to establish the significance of coronary disease in an individual patient. In this case, the patient was diagnosed with an acute coronary syndrome originating from an atheromatous plaque that had been identified on CTCA 15 months before the acute event. The patient had positive ischaemic testing on myocardial perfusion scan but no symptoms of angina prior to the acute event. This case highlights the increasing difficulties clinicians face when deciding on management for patients with high-risk plaques, as there are little guidelines beyond aggressive secondary prevention.