Imperial College London


Faculty of MedicineNational Heart & Lung Institute

Professor of Cardiology



+44 (0)20 7594 3381d.francis Website




Miss Juliet Holmes +44 (0)20 7594 5735




Block B Hammersmith HospitalHammersmith Campus





Publication Type

579 results found

Francis DP, Davies LC, Thorne S, Wilson K, Coats AJSet al., 1999, Oxygenation in a functionally univentricular circulation with complete mixing: Should saturation and flow be considered interchangeable?, Heart, Vol: 81, ISSN: 1355-6037

Background: Perioperative management with patients with complete pulmonary-systemic arterial mixing is delicate, and centres on balancing pulmonary (Qp) and systemic (Qs) blood flows. Clinical experience has suggested Qp/Qs ∼ 1 is optimal. Yet theoretical studies of maximisation of flow and saturation suggest Qp/Qs ratios well below 1. We hypothesised that this was due to an erroneous choice of combination low-saturation parameter: total "oxygen delivery", DO2 =Qsx[O2]art. With this, a fall in saturation from 90% to 45% could be fully compensated for by doubling flow: this is clinically invalid. Methods:. Oxygen delivered at low saturations is less metabolically useful. We thus defined "useful" DO2 as flow x arterial oxygen content above an arbitrary threshold in saturation (SatThresh): DUO2=QSx([O2]art-[O 2]thresh). We assessed the impact of this new flow-saturation parameter on optimal Qp:Qs balance. Results: "Useful" DO2 peaks at higher Qp/Qs ratios than does total DO2, more so as we increase SatThresh away from the 0% assumed by DO2. The figure shows DO2 (thick line labelled 0%), and DUO2 for SatThresh values of 10-50% (labelled). With SatThresh >37%, the maximum "useful" O2 delivery occurs at Qp/Qs above 1. Conclusion: Blood flow and oxygen saturation are not interchangeable. Clinical experience tells us that a fall in saturation is more harmful than the same proportionate fall in flow. If our theoretical models are made to reflect this, they yield results closer to clinical reality. In practical terms, maximising DUO2 rather than DO2 can lead to a 10% increase in tissue pO2. (Graph Presented).

Journal article

Francis DP, Shamim W, Davies LC, Chambers JS, Anker SD, Coats AJSet al., 1999, VE/VCO<inf>2</inf> slope and peak VO<inf>2</inf> interact and define a non-linear prognostic "risk surface" in chronic heart failure, Heart, Vol: 81, ISSN: 1355-6037

Background: Reduced peak oxygen consumption (peak VO2) during exercise is a known adverse prognostic factor in chronic heart failure (CHF). However, another feature, increased exercise hyperpnea (VE/VCO2 slope), may also be an adverse marker. Methods: 301 CHF patients underwent treadmill cardiopulmonary exercise testing, using a modified Bruce protocol. Peak VO2 and VE/VCO2 slope were determined using a pneumotachograph and mass spectrometer (Amis, DK). Results: During a median follow-up of 33 months (range 24-68), 87 patients died. Univariate Cox proportional-hazards analysis revealed powerful predictive value in both peak VO2 and VE/VCO2 slope (p<0.0001 for each). More importantly, bivariate mortality analysis identified independent prognostic value peak VO2 (p=0.0003) and VE/VCO2 slope (p<0.0001). In this graph, we demonstrate the relationship of 2-year mortality to peak VO2 and VE/VCO2 slope, using a model which does not assume linearity of risk or of interaction between peak VO2 and VE/VCO2 slope. Conclusion: During cardiopulmonary exercise testing, it is important to take note of the VE/VCO2 slope, as it gives valuable prognostic information independent of peak VO2. In the worst prognostic group by peak VO2 (<10 ml/kg/min) patients are classifiable by VE/VCO2 slope into a range of 2-year mortalities from 20% to 60%. Even in the best prognostic group by peak VO2, enhanced VE/VCO2 slope identifies a subgroup at elevated risk. (Graph Presented).

Journal article

Davies LC, Francis DP, Coats AJS, 1999, Baroreflex sensitivity and irregular breathing, Heart, Vol: 81, ISSN: 1355-6037

Attenuation of Baroreflex sensitivity (BRS) in patients with chronic heart failure (CHF) and post myocardial infarction carries important prognostic information. Yet measurement of BRS suffers from technical difficulties at low values of BRS and with irregular breathing, both of which are common in CHF. Irregular breathing causes transient disturbances in blood pressure (BP) and R-R interval, which have different time courses, making direct correlation between the two unreliable. The sizes of the fluctuations caused by respiration are significant when compared to the changes induced experimentally and can result in falsely negative values for BRS. Assessing changes in BP and RR separately may eliminate this. METHODS: We used the standard bolus phenylephrine method of elevating BP and continuously measured non-invasive BP, RR interval and respiration by plethysmography. BRSConv was defined as the conventional regression slope between RR and BP. Our proposed index, BRSInd, was defined as the mean rate of change in RR interval divided by the mean rate of change in BP. Both were determined on identical time intervals. RESULTS: In the example shown (see Fig.), respiratory artefact resulted in a BRSConv value of -0.04ms/mmHg. However, BRSInd gave a value of +0.6ms/mmHg as the overall trend of both BP and RR interval was upwards. In controls, where the relative size of respiratory artefact was small, the two methods of calculating BRS agreed closely. (Standard deviation of the percentage difference in value between BRSInd and BRSConv=10.7%: n=66) (Graph Presented) CONCLUSION: Respiratory measurement and double regression analysis allow minimisation of respiratory artefact improving the accuracy of BRS measurement in patients with a low BRS and irregular breathing.

Journal article

Francis D, Moriarty AJ, Lim PO, 1999, ECG interpretation [1] (multiple letters), British Journal of Cardiology, Vol: 6, Pages: 165-166, ISSN: 0969-6113

Journal article

Francis DP, Seydnejad SR, Kitney RI, Coats AJSet al., 1999, Dynamic Chemoreceptor Response Using Laguerre Expansion Technique, IEEE Proc Engineering in Medicine and Biology Conference

Conference paper

Davies LC, Francis DP, Coats AJS, 1998, Paramedic direct admission of heart-attack patients to coronary care unit, LANCET, Vol: 352, Pages: 2017-2018, ISSN: 0140-6736

Journal article

Edhouse J, Wardrope J, Morris FP, 1998, Paramedic direct admission of heart-attack patients to coronary care unit., Lancet, Vol: 352, Pages: 2018-2019, ISSN: 0140-6736

Journal article

Rauchhaus M, Francis DP, Schmidt H, Teichmann W, Glaeser Cet al., 1998, Not the ACE gene polymorphism, but the E-selectin gene polymorphism helps to predict restenosis after coronary angioplasty, CIRCULATION, Vol: 98, Pages: 393-394, ISSN: 0009-7322

Journal article

Keell SD, Chambers JS, Francis DP, Edwards DF, Stables RHet al., 1998, Shuttle-walk test to assess chronic heart failure, LANCET, Vol: 352, Pages: 705-705, ISSN: 0140-6736

Journal article

This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.

Request URL: Request URI: /respub/WEB-INF/jsp/search-html.jsp Query String: id=00167319&limit=30&person=true&page=20&respub-action=search.html