Peter Sharpe

Novel biomarkers in early diagnosis of acute myocardial infarction compared with cardiac troponin T

AIMS To evaluate the role of novel biomarkers in early detection of acute myocardial infarction (MI) in patients admitted with acute chest pain. METHODS AND RESULTS A prospective study of 664 patients presenting to two coronary care units with chest pain was conducted over 3 years from 2003. Patients were assessed on admission: clinical characteristics, ECG (electrocardiogram), renal function, cardiac troponin T (cTnT), heart fatty acid binding protein (H-FABP), glycogen phosphorylase-BB, NT-pro-brain natriuretic peptide, D-dimer, hsCRP (high sensitivity C-reactive protein), myeloperoxidase, matrix metalloproteinase-9, pregnancy associated plasma protein-A, soluble CD40 ligand. A > or = 12 h cTnT sample was also obtained. MI was defined as cTnT > or = 0.03 microg/L. In patients presenting <4 h of symptom onset, sensitivity of H-FABP for MI was significantly higher than admission cTnT (73 vs. 55%; P = 0.043). Specificity of H-FABP was 71%. None of the other biomarkers challenged cTnT. Combined use of H-FABP and cTnT (either one elevated initially) significantly improved the sensitivities of H-FABP or cTnT (85%; P < or = 0.004). This combined approach also improved the negative predictive value, negative likelihood ratio, and the risk ratio. CONCLUSION Assessment of H-FABP within the first 4 h of symptoms is superior to cTnT for detection of MI, and is a useful additional biomarker for patients with acute chest pain.

Biochemical detection and monitoring of alcohol abuse and abstinence

The merits and limitations of traditional and new markers for alcohol abuse (and abstinence) are critically examined for detection and monitoring of alcoholics, hazardous drinkers and binge drinkers. The traditional markers discussed include γ-glutamyltransferase (GGT), aspartate and alanine aminotransaminases (AST, ALT) and mean corpuscular volume (MCV); new markers include mitochondrial AST, carbohydrate-deficient transferrin (CDT), serum/urine 5- hydroxytryptophol, β-hexosaminidase and acetaldehyde adducts. The strengths and weaknesses of several of the self-reporting screening questionnaires are also explored. No laboratory test is reliable enough on its own to support a diagnosis of alcoholism. Sensitivities and specificities vary considerably and depend on the population concerned. GGT continues to remain the test that combines greatest convenience and sensitivity; its diagnostic accuracy can be enhanced by combination with other traditional markers (AST, ALT, MCV). None of the newer markers offers significant advantage, although CDT seems to be better at monitoring patients for increased alcohol consumption or progress towards abstinence.

Dietary Intake of Fruits and Vegetables Improves Microvascular Function in Hypertensive Subjects in a Dose-Dependent Manner

Background— Observational evidence has consistently linked increased fruit and vegetable consumption with reduced cardiovascular morbidity; however, there is little direct trial evidence to support the concept that fruit and vegetable consumption improves vascular function. This study assessed the dose-dependent effects of a fruit and vegetable intervention on arterial health in subjects with hypertension. Methods and Results— After a 4-week run-in period during which fruit and vegetable intake was limited to 1 portion per day, participants were randomized to consume either 1, 3, or 6 portions daily for the next 8 weeks. Endothelium-dependent and -independent arterial vasodilator responses were assessed by venous occlusion plethysmography in the brachial circulation before and after intervention. Compliance was monitored with serial contemporaneous 4-day food records and by measuring concentrations of circulating dietary biomarkers. A total of 117 volunteers completed the 12-week study. Participants in the 1-, 3-, and 6-portions/d groups reported consuming on average 1.1, 3.2, and 5.6 portions of fruit and vegetables, respectively, and serum concentrations of lutein and &bgr;-cryptoxanthin increased across the groups in a dose-dependent manner. For each 1-portion increase in reported fruit and vegetable consumption, there was a 6.2% improvement in forearm blood flow responses to intra-arterial administration of the endothelium-dependent vasodilator acetylcholine (P=0.03). There was no association between increased fruit and vegetable consumption and vasodilator responses to sodium nitroprusside, an endothelium-independent vasodilator. Conclusions— The present study illustrates that among hypertensive volunteers, increased fruit and vegetable consumption produces significant improvements in an established marker of endothelial function and cardiovascular prognosis.

Prognostic value of a multimarker approach for patients presenting to hospital with acute chest pain.

To evaluate the prognostic role of novel biomarkers for the risk stratification of patients admitted with ischemic-type chest pain, a prospective study of 664 patients presenting to 2 coronary care units with ischemic-type chest pain was conducted over 3 years beginning in 2003. Patients were assessed on admission for clinical characteristics, electrocardiographic findings, renal function, cardiac troponin T (cTnT), markers of myocyte injury (heart fatty acid-binding protein [H-FABP] and glycogen phosphorylase BB), neurohormonal activation (N-terminal-pro-brain natriuretic peptide [NT-pro-BNP]), hemostatic activity (fibrinogen and D-dimer), and vascular inflammation (high-sensitivity C-reactive protein, myeloperoxidase, matrix metalloproteinase-9, pregnancy-associated plasma protein-A, and soluble CD40 ligand). A >or=12-hour cTnT sample was also obtained. Myocardial infarction (MI) was defined as peak cTnT >or=0.03 microg/L. Patients were followed for 1 year from the time of admission. The primary end point was death or MI. Elevated fibrinogen, D-dimer, H-FABP, NT-pro-BNP, and peak cTnT were predictive of death or MI within 1 year (unadjusted odds ratios 2.5, 3.1, 5.4, 5.4, and 6.9, respectively). On multivariate analysis, H-FABP and NT-pro-BNP were selected, in addition to age, peak cTnT, and left ventricular hypertrophy on initial electrocardiography, as significant independent predictors of death or MI within 1 year. Patients without elevations of H-FABP, NT-pro-BNP, or peak cTnT formed a very low risk group in terms of death or MI within 1 year. A very high risk group had elevations of all 3 biomarkers. In conclusion, the measurement of H-FABP and NT-pro-BNP at the time of hospital admission for patients with ischemic-type chest pain adds useful prognostic information to that provided by the measurement of baseline and 12-hour cTnT.

The relationship between microvascular endothelial function and carotid radial pulse wave velocity in patients with mild hypertension

Carotid-radial pulse wave velocity (CRPWV) can be measured rapidly using applanation tonometry and significantly higher values have been reported among patients with risk factors for vascular disease. Forearm blood flow responses to intrabrachial infusion of acetylcholine independently predict cardiovascular morbidity among hypertensive patients. We aimed to examine the relationship between CRPWV, a potentially informative, noninvasive measure and this more established parameter of arterial health. One hundred and fifteen mildly hypertensive individuals (67% men, mean (± SD) age 54 ± 8 years, mean (± SD) blood pressure (BP) 143 ± 16/83 ± 12 mmHg) were recruited from a weekly medical outpatient clinic. Each volunteer had CRPWV measured using sequential tonometry before forearm blood flow responses to intra-arterial, endothelium-dependent (acetylcholine) and independent (sodium nitroprusside) vasodilators were assessed. There was a significant negative correlation between CRPWV and maximum forearm blood flow response to acetylcholine (r = −0.225, p = 0.016). This association remained significant in a multiple regression analysis (β = −0.213, p = 0.034). Mean arterial pressure and weight were additional independent predictors of CRPWV in this model. There was no such relationship between CRPWV and response to sodium nitroprusside (r = 0.088, p = 0.349). In patients with mild hypertension, a poor forearm blood flow response to acetylcholine independently predicted faster CRPWV, thus linking an established measure of microvascular endothelial function with a noninvasive index of conduit vessel stiffness.

Paradoxical decrease in serum high-density lipoprotein cholesterol with Tredaptive® (m/r nicotinic acid 1 g and laropiprant 20 mg).

I read with interest the recent case report describing paradoxical decreases in high-density lipoprotein cholesterol (HDL-C) with simvastatin and atorvastatin in a patient with type 2 diabetes mellitus. The accompanying editorial highlighted papers reporting paradoxical decreases in HDL-C by certain fibrates. I had previously reported that paradoxical decreases in HDL-C were actually quite common (in 43 out of 94 patients, 46%, on fenofibrate). The editorial also stated that there were no reports to the author’s knowledge of a paradoxical decrease in HDL-C with nicotinic acid use. However, I now wish to highlight a recent case where I believe that nicotinic acid, in the form of Tredaptive (m/r nicotinic acid 1 g and laropiprant 20 mg; Merck Sharp & Dohme Ltd, Hertfordshire, UK), was associated with a paradoxical reduction in HDL-C which was reversed when the agent was stopped. I believe this to be the first reported case in the literature. A 40-year-old man of south-east Asian ethnicity was referred to the lipid clinic in 2005. He had recently been an inpatient with acute pancreatitis and was noted to have a mixed hyperlipidaemia, predominantly hypertriglyceridaemia (fasting cholesterol 8.36 mmol/L, triglycerides 31.0 mmol/L and HDL-C 1.34 mmol/L). There was no evidence of gallstones and alcohol consumption was less than 15 units per week. He admitted to a very-high-fat diet but had a normal body mass index of 22.7 kg/m. He smoked 15 cigarettes per day. He had been commenced on Fenofibrate 200 mg daily (Lipantil Micro; Abbott Healthcare Products Ltd, Southampton, UK) and at the clinic his fasting lipids demonstrated cholesterol 4.48 mmol/L, triglycerides 7.1 mmol/L and HDL-C 0.87 mmol/L. His fenofibrate was increased to 267 mg daily (Lipantil Micro). Apo E genotyping was not consistent with Type III hyperlipidaemia. All other biochemical analyses, including plasma glucose, renal function, liver profile and thyroid profile, were normal and remained so. Total cholesterol, HDL-C and triglycerides were measured by standard methods for Roche Modular (HDL-C plus third generation; Roche Diagnostics Ltd, West Sussex, UK). The coefficients of variation (CV%) for total cholesterol, HDL-C and triglycerides were 2.7%, 1.8% and 2.4%, respectively. In March 2006, his lipid profile was as follows: cholesterol 3.89 mmol/L, triglycerides 3.55 mmol/L and HDL-C 0.69 mmol/L. Reduction in HDL-C is a phenomenon that I frequently see with fenofibrate and as I was satisfied with his cholesterol and triglycerides, he remained on fenofibrate 267 mg daily (Lipantil Micro). During the next few years his lipid results remained relatively stable. In February 2009, his triglycerides had risen to 5.3 mmol/L and HDL-C had fallen further to 0.59 mmol/L. Omacor (omega-3-acid ethyl esters; Abbott Healthcare Products Ltd) 4 g daily was commenced. In May 2010, his cholesterol was 3.64 mmol/L, triglycerides 4.1 mmol/L and HDL-C 0.48 mmol/L. In view of the falling HDL-C, I replaced his Lipantil Micro 267 mg daily with Tredaptive one tablet daily. In December 2010, his fasting lipids were as follows: cholesterol 5.88 mmol/L, triglycerides 19.7 mmol/L and HDL-C 0.27 mmol/L. He denied any non-compliance with medications. He was having some generalized mild itch which he related to his Tredaptive. The HDL-C result was verified by ultracentrifugation (0.25 mmol/L). His Tredaptive was stopped and he went back onto Lipantil Micro 267 mg daily. A lipid profile in March 2011 demonstrated cholesterol 4.98 mmol/L, triglycerides 4.97 mmol/L and HDL-C 0.74 mmol/L. I believe that this man experienced a paradoxical reduction in HDL-C with fenofibrate which is a phenomenon that has been previously described in almost 50% of our patients (also seen with other fibrates). There was a further significant 44% reduction in HDL-C when he was taking Tredaptive and this recovered fully when the Tredaptive was stopped and fenofibrate recommenced. There is some uncertainty regarding his compliance with Tredaptive as his triglycerides were also high (19.7 mmol/L) alongside the low HDL-C (0.27 mmol/L), but in the past when he was not taking any lipid-lowering medication, his baseline HDL-C was 1.34 mmol/L. He has always denied any compliance issue. In summary, I believe that this case demonstrates paradoxical lowering of HDL-C with both fenofibrate (a common phenomenon in my experience) and a further significant lowering with nicotinic acid (Tredaptive). The actual mechanism(s) by which this decrease occurs is unclear. I am unaware of any previous reports in the literature describing a paradoxical decrease in HDL-C with nicotinic acid use, but clinicians should remain alert to this possibility.