Scottish Society of Physicians – 52nd Annual Meeting

Immunity and disease – unwelcome complexity or opportunity for clinicians?

Iain B Mclnnes

Institute of Infection, Immunity and Inflammation, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK

Immune-mediated pathways are now recognized to play a significant role in many common pathologies. The immune system comprises a complex series of interactions between cellular and molecular networks. The complexity arises mainly from the requirement to offer host defence against an almost infinite microbial pathogenic attach, together with a requirement to regulate matters such that host tissue damage is uncommon or at least limited. Failure of these complex regulatory pathways often leads to induction or amplification of pathology. Using rheumatoid arthritis as an exemplar, this lecture will consider the opportunities available for clinical intervention now arising as the molecular regulatory pathways in immune responses are unraveled. The clinical success of biological agents that target cytokines e.g. tumour necrosis factor and interleukin-6 receptor provide definitive proof of concept that there are tractable checkpoints in the immune network. In future, as more information emerges as to how the immune system is organized in health and disease, it will become possible to intervene with greater discretion eventually to allow the re establishment of immune homeostasis – that is the creation of immune tolerance. Such approaches offer the prospects for inflammatory disease anticipation, abrogation and even prevention.

DOI: 10.1258/SMJ.2011.011079

The growing case for vitamin D supplementation

Professor Simon Pearce, Professor of Endocrinology

Institute of Human Genetics, International Centre for Life Central Parkway, Newcastle Upon Tyne, UK

Vitamin D is the precursor of 1,25-dihydroxycolecalciferol: a hormone with an important role in bone mineralization, calcium homeostasis and muscle function, as well as in immune system, metabolic and vascular function, and in neoplasia. UVB solar irradiation is the major natural source of vitamin D for man, but restricted availability of sunlight means that much of the UK population is vitamin D deficient or insufficient for much of the year. Thus, passive sunlight synthesis must be augmented by either dietary means, including supplements or cod liver oil, or by active sun-seeking behaviour. Sections of the population at high risk of vitamin D deficiency include those with non-white skin, exlusively breast fed infants, pregnant women, vegetarians and the elderly/institutionalized. The public health policy of universal supplementation with vitamin drops from birth to five years, which is recommended for children in the north of the UK, vegetarians and those with pigmented skin is patchily implemented at the moment, leading to a resurgence in rickets. Statutory supplementation of cows’ milk, orange juice and breakfast cereals with vitamin D would prevent rickets and likely lead to improvements in health for many people in north of the UK.

DOI: 10.1258/SMJ.2011.011080

Clinical research in Scotland

Sir John Savill, Vice Principal & Head

College of Medicine & Veterinary Medicine, University of Edinburgh, Queen's Medical Research Institute, Edinburgh, UK

This talk will highlight Scotland's strengths in clinical research, as evidenced by the 2006 UKCRC analysis of health research spend across the UK, and in RAE 2008 results. Scotland has very considerable advantages, in particular our electronic health record system and our track record in collaboration. The latter has recently been formalized by the creation in June 2009 of the Scottish Academic Health Sciences collaboration. Scottish clinical research is well placed to drive forward health and wealth gains over the next decade.

DOI: 10.1258/SMJ.2011.011081

The obesity epidemic

Professor Mike Lean, Professor of Human Nutrition

Department of Human Nutrition, University of Glasgow

Obesity has been declared an international epidemic by WHO. This presents a major challenge for physicians and healthcare planners, many of whom until recently still denied the disease status of obesity, and the need for medical management. Obesity is defined (e.g. by SIGN 2010) as the disease process of excess fat accumulation, resulting (as are all diseases) from gene-environment interactions, with multiple organ-specific pathological consequences. For epidemiological purposes, although high waist circumference is now recognized to be a better indicator of total body fat and predictor of ill-health, BMI > 30 is most widely used to mark obesity prevalence. A prevalence above 15% is used by WHO to define a nutritional epidemic. This point represents the critical threshold for intervention, when the 1:1 medical model is unsustainable, and government must institute effective preventive measures. The prevalence of obesity in UK is now 26%, and rising. Many government actions have theoretical value for obesity prevention, but evidence is lacking. Most are not directed specifically against obesity or coordinated for this purpose. The belief that obesity originates in childhood has tended to reduce efforts among young adults, where most of the weight gain occurs. While effective preventive interventions must be sought urgently, as with any epidemic it is essential to implement the best available and affordable treatment for affected individuals. A guideline-directed target of 5–10 kg loss is achieved by one in six of all the 7000 patients who have so far entered the highly cost-effective Counterweight programme, funded by the Scottish Health Department. Antiobesity drugs are safe and reliably improve results, with particular benefits for people with Type 2 diabetes. However, more aggressive treatments are now required for the rapidly growing sector of Severe and Complicated Obesity, with BMI > 35 kg/m², e.g. to generate > 15 kg loss necessary to reverse the diagnosis of diabetes. The Scottish Health Survey 2008 shows 9.4% of adults have BMI > 35, with a diabetes prevalence of 14%. The increasing BMIs, and waists, of older people point to a doubling of this figure over the next generation. Much of its costs could be avoided by more effective prevention and wider use of existing treatments.

DOI: 10.1258/SMJ.2011.011082

New developments in device therapy in heart failure

Professor Henry Dargie, Consultant Cardiologist

Golden jubilee National Hospital, Glasgow, UK

The modern era of Devices was ushered in with the extension of the use of pacemakers not only to regulate the rate of the heart beat but also to enhance cardiac function by facilitating atrio-ventricular transport. The concept of ‘resynchronizing’ ventricular contraction by the addition of left ventricular pacing to the already established ‘physiological’ pacing techniques fully translated into routine practice following several ‘landmark’ clinical trials, among the most notable being CARE-HF, coordinated by the Scottish cardiologist John Cleland. Cardiac resynchronization therapy, now universally referred to as ‘CRT’, is no longer ‘new’ but recent trials suggest that its indications may soon extend to a spectrum much wider than that of patients with advanced heart failure and more often to include an intra-cardiac cardioverter-defibrillator. Also not ‘new’ but only recently available in Scotland is the ventricular assist device, or ‘VAD’. The only NHS funded indication is as a ‘bridge to transplantation’, though it is clear that both the scope and scale of their implantation are expanding. The extent of investment in VADs and their increasing success in rescuing patients from catastrophic acute and chronic heart failure and affording an acceptable quality of life suggests that they may become an acceptable alternative, as well as an adjunct, to heart transplantation. Valvular heart disease as a frequent cause of heart failure is now receiving more attention due to the introduction of new therapeutic technologies. Aortic stenosis can be treated by transcutaneous aortic valve implantation of which there have been >1000 procedures in England, Wales and Northern Ireland. Similarly, mitral incompetence can be treated by a percutaneous version of the Alfieri stitch’ operation known as ‘MitraClip’ or by a simulation of surgical mitral annuloplasty by a ring placed round the mitral annulus delivered via the coronary sinus. Neither of these technologies is available in Scotland at the present time, although some patients have been treated in English centres. Definitive clinical trials are required before these techniques will enter routine clinical practice. Clinical trials show that small ultrafiltration devices can remove fluid more effectively than intravenous diuretics with less perturbation of electrolyte composition. Since diuretic resistance, hyponatraemia and hypo or hyperkalaemia are common problems in advanced heart failure such devices could prove very useful. Validation in larger clinical trials is awaited. Several ‘telemonitoring’ devices are now in use or undergoing trials to record the clinical or physiological status of patients with heart failure. Meta-analyses of these telemonitoring techniques reveal mixed but some favourable results. Invasive monitoring by utilizing pacemakers with enhanced function to record cardiac function can detect impending failure in previously stable patients. Recent data using the CARDIOMEMS system whereby a microchip is permanently implanted in a small pulmonary arteriole and interrogated by an external ‘wand’ has been shown to reduce hospitalization rates. Finally, new technologies to measure natriuretic peptides to monitor patient status by a ‘fingerstick’ test performed by patients in their own homes has been developed in Scotland and is undergoing clinical assessment to determine efficacy in improving clinical status and reducing hospital admission rates. To say that there are now ‘myriad’ devices to diagnose, assess and treat heart failure would still be an exaggeration but a vibrant biotechnology industry encouraged by the extensive needs of a complex and very common condition such as heart failure suggests that we may be seeing only the tip of an iceberg of ‘Devices’ for its future management. Clearly, these technologies will require persuasive evidence not only of efficacy but also cost-effectiveness before they can be adopted in routine clinical practice. For appropriate deployment of any of these techniques an adequate clinical infrastructure that can identify patients most likely to benefit is mandatory. The recent national audits across the UK indicate that this, presently, is not the case.

DOI: 10.1258/SMJ.2011.011083

Hospital-acquired infection: from doctors to patients, with complements

Professor John Simpson, Professor of Respiratory Medicine

Institute of Cellular Medicine, University of Newcastle

The talk will be in two parts. The first part will briefly discuss aspects of hospital-acquired infection (HAI) such as epidemiology, pathogenesis, diagnostic issues and issues of antibiotic stewardship. The main examples will be drawn from hospital-acquired pneumonia (HAP) and infections acquired in the intensive care unit (ICU), as HAP in the ICU is associated with higher mortality than any other HAI.

The second part will consider recent evidence suggesting that activated complement may play an important role in the innate immune dysfunction linked to HAL The implications of these data for diagnosing HAI, assessing risk of HAI and identifying new therapeutic targets will be discussed.

DOI: 10.1258/SMJ.2011.011084

Aldosterone, hypertension and cardiovascular risk – rediscovery of an old hormone

Professor John Connell, Dean of Medical School and Professor of Endocrinology

College of Medicine, Dentistry and Nursing, University of Dundee

Excess aldosterone production has long been associated with hypertension through the entity of primary aldosteronism. In recent years, the true frequency of this disorder has been debated – some studies suggest that approximately 10% of patients with high blood pressure may have primary aldosterone excess, although the definition of the disorder, and the ability to distinguish it clearly from low renin essential hypertension remains problematic. Nevertheless, it is clear that a very significant proportion of patients with high blood pressure have excessive production of aldosterone relative to renin. Aldosterone exerts a range of deleterious effects on the cardiovascular system, including cardiac hypertrophy, renal damage and pro-inflammatory effects that lead to significant tissue damage. In a recent study, we have demonstrated, for example, a significant excess of superoxide generation in samples from patients with primary aldosteronism compared with control hypertensives. It is now clear that the effects of aldosterone are exercised across the blood pressure range. In a population of middle-aged adults, we have shown that aldosterone levels correlate well with levels of blood pressure, so that subjects with the highest values have, on average, a level of blood pressure 10 mmHg greater than those with the lowest levels. In very recent work, we have demonstrated that this is likely to reflect a genetic influence on the regulation of aldosterone and blood pressure. This population influence of aldosterone also has consequences for other cardiovascular risk. The effect of aldosterone excess on the risk of development and progression of heart failure has been well recognized. More recently, however, evidence has been produced to show that levels of aldosterone have a major influence on outcomes following acute myocardial infarction. Taken together, therefore, these data from patients with high blood pressure, heart failure and ischaemic heart disease identify aldosterone as a major risk factor and one that is amenable to therapy. Mineralocorticoid receptor antagonists, such as spironolactone, have long been known to be effective antihypertensive agents. More recently, however, this influence was illustrated in the ASCOT study, where addition of spironolactone produced a substantial fall in blood pressure in patients resistant to other treatments. It remains unclear, however, whether spironolactone is effective as a single-line agent in lower doses. We have recently compared the effect of spironolactone with a standard thiazide diuretic in a study in which patients were stratified by levels of aldosterone and renin. While pretreatment levels of these hormones did not influence blood pressure response to spironolactone, it was clear that this drug was at least as effective in lowering blood pressure as a standard thiazide diuretic. In summary, aldosterone has now moved from its association with a relatively rare cause of secondary hypertension to a mainstream position as a key regulator of cardiovascular pathophysiology. In turn, this identifies aldosterone blockade as an important therapeutic opportunity in the treatment of patients with cardiovascular disease – the challenge is now to identify drugs with a more acceptable profile of side-effects to improve tolerability.

DOI: 10.1258/SMJ.2011.011085

Oral Abstracts

Allopurinol reduces both left ventricular hypertrophy and endothelial dysfunction in patients with chronic kidney disease

Michelle P Kao, Donald S Ang and Allan D Struthers

Division of Cardiovascular and Lung Biology, University of Dundee, Ninewells Hospital and Medical School, Dundee, DD1 9SY

Left ventricular hypertrophy (LVH) and vascular dysfunction are key pathophysiological processes that lead to cardiovascular (CV) events in patients with chronic kidney disease (CKD). As oxidative stress is thought to contribute to both LVH and endothelial dysfunction in CKD, we hypothesized that allopurinol might regress left ventricular (LV) mass and improve endothelial dysfunction and arterial stiffness in CKD patients. A randomized, double-blind, placebo-controlled, parallel study was conducted in patients with CKD stage 3 and LVH. Subjects received 300 mg/day Allopurinol or placebo for nine months. Cardiac magnetic resonance (CMR) was performed to measure LV mass index (LVMI). Endothelial function was assessed by flow-mediated dilation (FMD) of the brachial artery, while central arterial stiffness was assessed by pulse wave analysis (PWA) and pulse wave velocity (PWV). Fifty-three patients completed the study (27 active, 26 placebo). Allopurinol significantly reduced LVH ($DLLVMI: −1.42 ± 4.67 g/m², allopurinol versus + 1.28 ± 4.45 g/m², placebo [P = 0.036]). Allopurinol also significantly improved brachial artery FMD ($DLFMD: +1.26 ± 3.06%, allopurinol versus −1.05 ± 2.84%, placebo [P = 0.009]). Central augmentation index (AIx) also improved significantly on allopurinol ($DLAIx: −4.70 ± 9.30%, allopurinol versus +0.77 ± 6.06%, placebo [P = 0.015]). There was also a trend towards improvement in PWV in patients on allopurinol ($DLPWV was —0.39 ± 1.13 m/s, allopurinol versus and +0.20 ± 1.28 m/s, placebo [P = 0.086]). In conclusion, this is the first study to demonstrate that allopurinol can regress LV mass in any population and the first to demonstrate that it improves endothelial dysfunction in CKD patients. As LVH and endothelial dysfunction are both important surrogate markers for prognosis, this study should prompt future trials to examine whether allopurinol reduces CV events in CKD patients with LVH.

DOI: 10.1258/SMJ.2011.011086

The use of implantable loop recorders (ILRs) in the investigation of patients with unexplained syncope in the Royal Alexandra Hospital (RAH), Paisley

Craig McCallum^* and Stuart Hood^†

^* Faculty of Medicine, University of Glasgow; ^†Royal Alexandra Hospital, Paisley

Syncope is a transient loss of consciousness due to transient global cerebral hypoperfusion characterized by rapid onset, short duration and spontaneous complete recovery.¹ In the presence of underlying structural heart disease, syncope is a marker of an adverse prognosis. In 2009 the European Society of Cardiology (ESC) guidelines highlighted the role of a diagnostic strategy based on prolonged monitoring of patients with syncope.¹ Implantable loop recorders (ILRs) are subcutaneous devices, which are inserted in the pectoral area under local anaesthesia. The patient activates the device after a syncopal episode and an ECG recording of the previous 40 minutes is stored. The device can also automatically detect changes in rhythm disturbance should the patient be unable or fail to make a recording. We performed a follow-up audit of patients who underwent ILR device implantation at the RAH to assess usage and diagnostic yield of ILRs in our hospital. Between 2000 and 2010, a total of 55 patients had ILRs implanted at the RAH. A review of these cases was undertaken by analysis of their case sheets. Data were collected on patient demographics, symptoms, investigations performed before implantation of the ILR and whether the ILR aided or confirmed diagnosis. Forty-eight of 55 (87%) of case sheets were available for review. Eighteen (37%) were male with a mean age of 65 years (37–87) and 30 (63%) were female with a mean age of 60 years (22–87). Seventeen patients (35%) had their device implanted since 2008, suggesting increasing use of ILRs. Forty-one (85%) had experienced syncope, one (2%) complained of only presyncope and six (13%) complained of presyncope and palpitations without syncope. Thirteen (27%) patients have had the cause of syncope diagnosed by the ILR (all presented with syncope). Twenty-seven (56%) patients remained without a diagnosis and eight (17%) have been asymptomatic since implant. Of those in whom a diagnosis was confirmed by ILR, six patients (45%) had sick sinus syndrome, four patients (31%) had narrow complex tachycardia which correlated with symptoms, one patient (8%) had torsades de pointes, one patient (8%) had complete heart block and one patient (8%) had paroxysmal ventricular tachycardia. ILR revealed the cause of syncope/presyncope in 27% of patients in whom conventional monitoring revealed no cause. This diagnostic yield is broadly in keeping with current literature, which reports yields of 30–50%.² This study demonstrates that ILRs can identify malignant rhythm disturbances that may otherwise be missed if only non-invasive monitoring is performed. All patients who received a diagnosis as a result of monitoring with the ILR presented with syncope (with or without associated symptoms). Reserving ILR implantation for patients with syncope, rather than presyncope or palpitations will increase the diagnostic yield and reduce costs.

DOI: 10.1258/SMJ.2011.011088

Cyclophilin A is a damage-associated molecular pattern that mediates paracetamol-induced liver injury

James W Dear^*‡, Kenneth Simpson^#, Melianthe P J Nicolai^*, James H Catterson^*, Jonathan Street^*, Tineke Huizinga^*, Darren Craig^#, Kevin Dhaliwal^#, Sheila Webb^§, D Nicholas Bateman^‡ and David J Webb*

*Univeristy/BHF Centre for Cardiovascular Science, Edinburgh University; ^#Centre for Inflammation Research, Edinburgh University; ^‡National Poisons Information Service, Edinburgh; ^§MRC Human Genetics Unit, Edinburgh University

The innate immune system is alerted to cell injury by damage-associated molecular patterns (DAMPs)– intracellular molecules that are released following damage to the cell membrane. Cyclophilin A (CypA) is a novel inflammatory mediator we previously identified as important in the pathophysiology of sepsis.¹ As massive cell death is typical of severe paracetamol poisoning, we hypothesized that CypA may be an important mediator of liver injury by virtue of it being a novel DAMP. This translational study used both mouse models and human samples. We generated a mouse lacking the gene for CypA (CypA KO) to investigate its role in paracetamol poisoning (induced by 350 mg/kg intraperitoneal injection). Furthermore, we determined the effect on paracetamol poisoning of blockade of the extracellular receptor for CypA (CD147) with a specific antibody. We generated necrotic cells from the liver of wild-type (WT) and knockout (KO) mice and quantified the inflammatory response when necrotic cells were injected into WT mice. With ethics approval, urine was collected from patients attending the Royal Infirmary of Edinburgh with varying degrees of paracetamol-induced liver injury and the CypA concentration was measured. CypA KO mice were resistant to paracetamol poisoning (ALT values 24 hours postparacetamol injection: KO 1096 ± 501 μ/L n = 24; WT 4244 ± 1266 μ/L n = 15, P = 0.01). Inhibition of CD147 also significantly reduced paracetamol-induced liver injury. Six hours after injection into WT mice, CypA KO necrotic liver cells induced less of an inflammatory response than WT cells. Inhibition of CD147 also significantly reduced the host inflammatory response to necrotic liver cells. Urine from patients with paracetamol-induced liver injury had significantly elevated CypA compared with patients without evidence of liver injury. In conclusion, CypA is a critical mediator of paracetamol-induced liver injury. Our data support its pathophysiological mechanism being a DAMP. In humans, this protein is released into the urine of patients with liver injury and it may represent a biomarker in addition to a novel therapeutic target.

DOI: 10.1258/SMJ.2011.011089

Joint Winner of the Fitzgerald Peel Prize

Risk assessment of patients presenting with upper GI haemorrhage: a multicentre comparison of the Glasgow Blatchford and Rockall scores

E Reed*, H R Dalton^†, O Blatchford^‡, D Ashley^§, C Mowat**, D R Gaya**, A Cahill**, M Groome**, W Murray*, U Warshow^†, E Thompson^§ and A J Stanley^§

*GI Unit, Glasgow Royal Infirmary; ^†Royal Cornwall Hospital, Truro; ^‡Health Protection, Scotland; ^§North Tees University Hospital, Stockton; **Ninewells Hospital, Dundee

The pre-endoscopic Glasgow Blatchford Score (GBS) identifies low-risk patients with upper gastrointestinal haemorrhage (UGIH) who may be suitable for outpatient management. Studies suggest the GBS has high accuracy in predicting clinically relevant endpoints. Our aim was to compare the GBS with the preendoscopy (admission) and postendoscopy (full) Rockall scores in predicting outcome. Data were collected on consecutive patients presenting to four UK hospitals (Glasgow Royal Infirmary, Royal Cornwall Hospital Truro, University Hospital of North Tees and Ninewells Hospital Dundee). Admission history, clinical and laboratory data, endoscopic findings, and treatment and clinical follow-up were recorded. Receiver Operator Curves (ROC) were used to compare the scores in the separate prediction of death, endoscopic or surgical intervention, or transfusion. A total of 1556 patients (mean age 56.7 years; 62% male) presented with UGIH to the four hospitals during the study period. Seventy-four (4.8%) died, 223 (14.3%) had endoscopic or surgical intervention and 363 (23.3%) required transfusion. The GBS was equally effective at predicting death compared with both the admission Rockall score (area under ROC curve [AUROC] 0.804 versus 0.801) and the full Rockall score (AUROC 0.741 versus 0.790). The GBS was superior to the admission Rockall score (AUROC 0.858 versus 0.705; P < 0.00005) at predicting endo/surgical intervention, but similar to the full Rockall score (AUROC 0.822 versus 0.797). In predicting need for transfusion, the GBS was superior to both the admission Rockall (AUROC 0.944 versus 0.756; P < 0.00005) and the full Rockall score (AUROC 0.935 versus 0.792; P < 0.00005). In conclusion, the GBS is as effective as the admission and full Rockall scores in predicting death after UGIH. It is superior to both the admission and full Rockall scores in predicting need for transfusion and superior to the admission Rockall score in predicting endoscopic or surgical intervention. This simple scoring tool can be used in A&E to accurately assess risk in patients presenting with UGIH.

DOI: 10.1258/SMJ.2011.011090

Metformin in insulin resistant LV dysfunction: a double-blind, placebo controlled trial (TAYSIDE trial)

Aaron K F Wong, Ruth Symon, Matlooba A AlZadjali, Donald S C Ang, AnnaMaria Choy, John R Petrie, Allan D Struthers and Chim C Lang

There is increasing appreciation of the reciprocal relationship between chronic heart failure (CHF) and insulin resistance (IR). The degree of IR has been shown to correlate with disease severity and clinical outcome in CHF. However, these association studies do not distinguish between cause and effect. Therefore, the significance of IR in the pathophysiology of CHF remains to be determined. It is not certain if IR is merely a marker reflecting the severity of disease or it is the culprit in CHF. If it is a culprit that worsens CHF, improving it may lead to better clinical outcome. The purpose of our study was to determine if reversing IR with metformin in non-diabetic CHF patients results in clinical benefits. In a double-blind, placebo-controlled trial, 62 non-diabetic insulin-resistant CHF patients (mean age, 65.2 ± 8.0 years; male, 90%; left ventricular ejection fraction, 32.6 ± 8.3%; NYHA I/II/III/IV 11/45/6/0) were randomized to receive either four months of metformin (n = 39, 2 g per day) or matching placebo (n = 23). IR was defined by fasting insulin resistance index (FIRI) of ≥ 2.7. Cardiac-pulmonary exercise testing, echocardiography, flow-mediated dilation, EndoPAT (a measurement of endothelial dysfunction), six-minute walk test, Minnesota Living with Heart Failure Questionnaire, anthropometric measurements, FIRI and biomarkers were assessed at baseline and after four months of intervention. Compared with placebo, metformin decreased FIRI (from 5.8 ± 3.8 to 4.0 ± 2.5, P < 0.001), decreased fasting glucose (from 5.6 ± 0.6 to 5.2 ± 0.4 mmol/L, P = 0.005), reduced fasting insulin (from 26.8 ± 14.3 to 20.2 ± 10.4 mU/L, P < 0.001), reduced serum HbA1c (from 5.7 ± 0.3% to 5.5 ± 0.3%, P = 0.002), decreased serum leptin (16.6 ± 23.3 to 12.1 ± 16.5 ng/mL, P < 0.05) and resulted in a weight loss of 1.9 kg (P < 0.001). Metformin treatment significantly improved NYHA functional class (from 1.89 ± 0.5 to 1.75 ± 0.5, P = 0.046). Although peak exercise parameters and endothelial function did not differ between treatment groups, submaximal parameters were significantly improved with metformin therapy, notably VE/VCO₂ slope (from 32.9 ± 15.9 to 28.1 ± 8.8, P = 0.05) and ventilatory class (from 1.9 ± 0.9 to 1.6 ± 0.9, P = 0.021). In conclusion, this novel study provides supportive evidence that metformin is a safe treatment in non-diabetic CHF patients which improves glucose homeostasis but more importantly results in significant improvement in NYHA class, VE/VCO₂ slope and ventilatory class, all of which are important prognostic markers in CHF. Therefore, reversing IR may represent a new target for treatment in CHF.

DOI: 10.1258/SMJ.2011.011091