Jessica Webb1, Jane Draper2, Tiago Rua2, Yee Yiu2, Susan Piper3, Thomas Teall2, Lauren Fovargue4, Elena Bolca2, Tom Jackson5, Simon Claridge5, Ben Sieniewicz5, Bradley Porter5, Adam McDiarmid2, Ronak Rajani5, Stamatis Kapetanakis5, Christopher A Rinaldi5, Reza Razavi5, Theresa A McDonagh6, Gerald Carr-White5. 1. Department of Cardiology, Guy's and St Thomas' NHS Foundation Trust, London SE1 7EH, United Kingdom; Division of Imaging Sciences and Biomedical Engineering, King's College London, SE1 7EH, United Kingdom. Electronic address: Jessica.webb@kcl.ac.uk. 2. Department of Cardiology, Guy's and St Thomas' NHS Foundation Trust, London SE1 7EH, United Kingdom. 3. Department of Cardiology, Guy's and St Thomas' NHS Foundation Trust, London SE1 7EH, United Kingdom; Department of Cardiology, King's College Hospital NHS Foundation Trust, London SE5 9RS. 4. Division of Imaging Sciences and Biomedical Engineering, King's College London, SE1 7EH, United Kingdom. 5. Department of Cardiology, Guy's and St Thomas' NHS Foundation Trust, London SE1 7EH, United Kingdom; Division of Imaging Sciences and Biomedical Engineering, King's College London, SE1 7EH, United Kingdom. 6. Division of Imaging Sciences and Biomedical Engineering, King's College London, SE1 7EH, United Kingdom; Department of Cardiology, King's College Hospital NHS Foundation Trust, London SE5 9RS.
Abstract
AIMS: The 2014 National Institute of Clinical Excellence (NICE) guidelines on the management of acute heart failure recommended using a plasma NT-proBNP threshold of 300pg/ml to assist in ruling out the diagnosis of heart failure (HF), updating previous guidelines recommending using a threshold of 400pg/ml. NICE based their recommendations on 6 studies performed in other countries. This study sought to determine the diagnostic and economic implications of using these thresholds in a large unselected UK population. METHODS: Patient and clinical demographics were recorded for all consecutive suspected HF patients over 12months, as well as clinical outcomes including time to HF hospitalisation and time to death (follow up 15.8months). RESULTS: Of 1995 unselected patients admitted with clinically suspected HF, 1683 (84%) had a NTproBNP over the current NICE recommended threshold, of which 35% received a final diagnosis of HF. Lowering the threshold from 400 to 300pg/ml would have involved screening an additional 61 patients and only would have identified one new patient with HF (sensitivity 0.985, NPV 0.976, area under the curve (AUC) at 300pg/ml 0.67; sensitivity 0.983, NPV 0.977, AUC 0.65 at 400pg/ml). The economic implications of lowering the threshold would have involved additional costs of £42,842.04 (£702.33 per patient screened, or £ 42,824.04 per new HF patient). CONCLUSION: Applying the recent updated NICE guidelines to an unselected real world population increases the AUC but would have a significant economic impact and only identified one new patient with heart failure.
AIMS: The 2014 National Institute of Clinical Excellence (NICE) guidelines on the management of acute heart failure recommended using a plasma NT-proBNP threshold of 300pg/ml to assist in ruling out the diagnosis of heart failure (HF), updating previous guidelines recommending using a threshold of 400pg/ml. NICE based their recommendations on 6 studies performed in other countries. This study sought to determine the diagnostic and economic implications of using these thresholds in a large unselected UK population. METHODS:Patient and clinical demographics were recorded for all consecutive suspected HF patients over 12months, as well as clinical outcomes including time to HF hospitalisation and time to death (follow up 15.8months). RESULTS: Of 1995 unselected patients admitted with clinically suspected HF, 1683 (84%) had a NTproBNP over the current NICE recommended threshold, of which 35% received a final diagnosis of HF. Lowering the threshold from 400 to 300pg/ml would have involved screening an additional 61 patients and only would have identified one new patient with HF (sensitivity 0.985, NPV 0.976, area under the curve (AUC) at 300pg/ml 0.67; sensitivity 0.983, NPV 0.977, AUC 0.65 at 400pg/ml). The economic implications of lowering the threshold would have involved additional costs of £42,842.04 (£702.33 per patient screened, or £ 42,824.04 per new HF patient). CONCLUSION: Applying the recent updated NICE guidelines to an unselected real world population increases the AUC but would have a significant economic impact and only identified one new patient with heart failure.