MAY 2012 VOL 23 NO 4

AFRICA

www.cvja.co.za CardioVascular Journal of Africa (official journal for PASCAR)

• Endothelial dysfunction

• Cardiac surgery and haematological malignancies

• Cardiovascular risk factors in

• Changes in NT-proBNP

• Effusive–constrictive pericarditis in Ibadan

• Pneumonia in post cardiac surgery

• Cardiovascular disease in the Seychelles, Tanzania and Mauritius

PUBLISHED ONLINE: • Mobile atheromatous plaque of the aortic arch Cardiovascular Journal of Africa . Vol 23, No 4, May 2012 • Images in carcinoid disease Printed by Durbanville Commercial Printers Tel: 021 946 4074 Tel: Printed by Durbanville Commercial Printers “THE CURRENT AHA DIETARY GUIDELINES RECOMMEND COMBINED EPA & DHA IN A DOSE OF APPROXIMATELY 1000 mg/DAY IN PATIENTS WITH CHD “ Lavie et al, Omega - 3 Polyunsaturated Fatty Acids and Vol. 54,No. 7 2009, August 11, 2009 585– 594 TRIMEGA™

* HIGH STRENGTH EPA & DHA 300 MG EPA AND 200 MG DHA IN A SINGLE CAPSULE

* ENSURED PATIENT COMPLIANCEWITH CONVENIENTBD DOSING

* SABS TESTED FOR DECLARED CONTENT OF ACTIVE MOEITIES.

* FORMULATION ALLOWS QUICK AND EASY DISSOLUTION.

* NO ERUCTATION

For further information contact

pH PHARRM AF ICA HEALTHCARE (PTY) LTD Reg. No. 1993/003911/07 PRIVATE BAG X8, ROSETTENVILLE 2130

TOLL FREE 0800 601 098 ISSN 1995-1892 (print) ISSN 1680-0745 (online)

AFRICA

Cardiovascular Journal of Africa www.cvja.co.za

Editorial 184 Endothelial dysfunction: are we ready to heed the vasculature’s early-warning signal? H Strijdom

Cardiovascular Topics 186 association between troponin T and ICU mortality, a changing trend S Hajsadeghi • S Gholami • G Gohardehi • NS Moghadam • AS Sabet • SRJ Kerman •

CO N T EN TS M Moradi • R Mollahoseini

191 Nebivolol therapy improves QTc and QTcd parameters in heart failure patients SM Aksoy • S Cay • G Cagirci • N Sen 2012 Y 194 can cardiac surgery be performed safely on patients with haematological malignancies A Guler • MA Sahin • F Cingoz • E Ozal • U Demirkilic • M Arslan

197 the impact of cardiovascular risk factors on the site and extent of coronary artery disease AF Zand Parsa • H Ziai • L Haghighi

OL 23, NO 4. MA V OL 200 the time-course changes of NT-proBNP and tissue Doppler indices in patients undergoing mitral valve replacement DR Prakaschandra • T Esterhuizen • DP Naidoo

206 presentation pattern and management of effusive–constrictive pericarditis in Ibadan MA Salami • PO Adeoye • VO Adegboye • OA Adebo

212 risk factors for the development of pneumonia post cardiac surgery AE Topal • MN Eren

216 an audit of pregnant women with prosthetic heart valves at a tertiary hospital in South Africa: a five-year experience B Mazibuko • H Ramnarain • J Moodley

INDEXED AT SCISEARCH (SCI), PUBMED AND SABINET

Editors Topic Editors Editorial Board International Advisory Editor-in-Chief (South Africa) Nuclear Medicine and Imaging prof PA Brink PROF A LOCHNER Board PROF AJ BRINK DR MM SATHEKGE Experimental & Laboratory Biochemistry/Laboratory PROF DAVID CELEMAJER Cardiology Science Australia (Clinical Cardiology) Heart Failure Assistant Editor PROF KEITH COPELIN FERDINAND Dr g visagie PROF R DELPORT PROF BM MAYOSI Prof JAMES KER (JUN) Chemical Pathology Chronic Rheumatic Heart USA (General Cardiology) Paediatric Disease DR SAMUEL KINGUE Regional Editor dr s brown PROF MR ESSOP Cameroon (General Cardiology) DR A Dzudie Haemodynamics, Heart Failure DR MT MPE Renal Hypertension & Valvular Heart Disease Cardiomyopathy DR GEORGE A MENSAH USA (General Cardiology) Regional Editor (Kenya) dr brian rayner DR OB FAMILONI PROF DP NAIDOO Dr F Bukachi PROF WILLIAM NELSON Surgical Clinical Cardiology USA (Electrocardiology) dr f aziz PROF B RAYNER Regional Editor (South Africa) DR V GRIGOROV DR ULRICH VON OPPEL Hypertension/Society PROF R DELPORT Adult Surgery Invasive Cardiology & Heart Wales (Cardiovascular Surgery) dr j rossouw Failure PROF MM SATHEKGE PROF PETER SCHWARTZ Nuclear Medicine/Society Italy (Dysrhythmias) Electrophysiology and Pacing PROF J KER (SEN) PROF ERNST VON SCHWARZ dr a okreglicki Hypertension, Cardiomyopathy, PROF YK SEEDAT USA (Interventional Cardiology) Cardiovascular Physiology Diabetes & Hypertension Epidemiology and Preventionist DR J LAWRENSON PROF H DU T THERON Publishing Consultant dr ap kengne Paediatric Heart Disease Invasive Cardiology Mike Gibbs Letter to the Editor 205 cohort studies of cardiovascular disease in the Seychelles, Tanzania and Mauritius P Bovet • C Shamlaye

Review Article 222 Endothelial dysfunction: the early predictor of atherosclerosis M Mudau • A Genis • A Lochner • H Strijdom

Conference Report 232 Expert report on the 22nd European meeting on hypertension and cardiovascular protection, London, 26–29 April 2012 N Rapeport • S Middlemost

PUBLISHED ONLINE (Available on www.cvja.co.za and in Pubmed)

Case Reports e1 Non-obstructive membranes of the left atrial appendage T Bordonali • A Saporetti • E Vizzardi • A D’Aloia • E Chiari • L dei Cas CO N T EN TS

e3 Mobile atheromatous plaque of the aortic arch diagnosed by transthoracic echocardiography prior to coronary artery bypass surgery Which one would you choose: scepticism or wishful thinking? 2012

Y AC Hatemi • O Omay • M Başkurt • S Kücükoğlu • B Öz • K Süzer

e6 successful stenting of catheter-induced unprotected left main coronary artery dissection G Ertaş • E Ural • WJ van der Giessen

e8 New images in carcinoid heart disease M Klobučić • MH Paar • RŠ Padovan • J Vincelj • B Fila

OL 23, NO 4. MA V OL e11 aortic dissection, a complication during successful angioplasty of chronic total occlusion of the right coronary artery, was treated conservatively S Chunlai • PR Stella • A Belkacemi • P Agostoni

managing editor Production Copyright: Electronic abstracts available on Pubmed julia aalbers Co-ordinator Clinics Cardive Publishing (Pty) Ltd. Tel: 021 976 4378 WENDY WEGENER Audited circulation Fax: 086 610 3395 Layout: Tel: (021) 976-4378 Martingraphix e-mail: [email protected] e-mail: [email protected] Full text articles available on: www.cvja. Printer: co.za or via www.sabinet.co.za; for access Production Editor GAUTENG CONTRIBUTOR Durbanville Commercial Printers codes contact [email protected] SHAUNA GERMISHUIZEN PETER WAGENAAR Tel: 021 785 7178 ONLINE SERVICES: Cell 082 413 9954 Design Connection For subscription enquiries contact Fax: 086 628 1197 e-mail: [email protected] Wendy Wegener on e-mail: [email protected] All submissions to CVJA are to be [email protected] CONTENT MANAGER made online via www.cvja.co.za Editorial Assistant & Michael Meadon (Design Connection) Electronic submission by means of an The views and opinions expressed in the Circulation Tel: 021 975 3785 e-mail attachment may be considered articles and reviews published are those ELSABÉ BURMEISTER Fax: 0866 557 149 under exceptional circumstances. of the authors and do not necessarily Tel: 021 976 8129 e-mail: [email protected] reflect those of the editors of the Journal e-mail: [email protected] Postal address: PO Box 1013, or its sponsors. In all clinical instances, Durbanville, 7551 The Cardiovascular Journal of Africa, medical practitioners are referred to the development editor incorporating the Cardiovascular product insert documentation as approved Tel/Fax: 021 976 8129 by the relevant control authorities. GLENDA HARDY Journal of South Africa, is published 10 Int.: +27 21 976 8129 Cell: 071 8196 425 times a year, the publication date being e-mail: [email protected] the third week of the designated month. e-mail: [email protected] Help your patients love themselves a little more.

CRESTOR® 5 mg is suitable for select patients who need less aggressive lipid lowering1 CRESTOR® is the more effective statin at lowering LDL-C and raising HDL-C2 CRESTOR® 10 mg will get most patients to LDL-C goal1,3 CRESTOR® is well-tolerated and has a favourable benefi t-risk profi le4,5

S4 CRESTOR® 5 (Tablet) Each CRESTOR® 5 tablet contains 5 mg of rosuvastatin as rosuvastatin calcium. S4 CRESTOR® 10 (Tablet) Each CRESTOR® 10 tablet contains 10 mg of rosuvastatin as rosuvastatin calcium. S4 CRESTOR® 20 (Tablet) Each CRESTOR® 20 tablet contains 20 mg of rosuvastatin as rosuvastatin calcium. S4 CRESTOR® 40 (Tablet) Each CRESTOR® 40 tablet contains 40 mg of rosuvastatin as rosuvastatin calcium. PHARMACOLOGICAL CLASSIFICATION: A. 7.5 Serum-cholesterol reducers INDICATIONS: Primary hypercholesterolaemia, mixed dyslipidaemia and isolated hypertriglyceridaemia (including Fredrickson Type IIa, IIb and IV; and heterozygous familial hypercholesterolaemia) as an adjunct to diet when response to diet and exercise is inadequate. Indicated in patients with homozygous familial hypercholesterolaemia, either alone or as an adjunct to diet and other lipid lowering treatments. CRESTOR® 40 mg should only be considered in patients with severe hypercholesterolaemia and high cardiovascular risk who do not achieve their treatment goal on 20 mg of CRESTOR® or alternative therapy. Specialist supervision is recommended when the 40 mg dose is initiated. REGISTRATION NUMBERS: CRESTOR® 5: 41/7.5/0298, CRESTOR® 10: 36/7.5/0349, CRESTOR® 20: 36/7.5/0350, CRESTOR® 40: 36/7.5/0351. DETAILS OF THE REGISTERED LICENCE HOLDER: AstraZeneca Pharmaceuticals (Pty) Ltd Reg No. 1992/005854/07. No. 5 Leeuwkop Road, Sunninghill, 2157, South Africa. Tel: 011 797 6000. Fax: 011 797 6001. www.astrazeneca.co.za. For full details relating to any information mentioned above please refer to the package insert of CRESTOR® 5 mg, 10 mg, 20 mg and 40 mg. CRESTOR® is a registered trademark of AstraZeneca group. Licensed from Shionogi & Co Ltd, Osaka, Japan. EPI Date: 13/05/2008. Date compiled: March 2011.

References: 1. CRESTOR® package insert 2. Jones P, Davidson MH, Stein EA, et al. Comparison of the Effi cacy and Safety of Rosuvastatin Versus Atorvastatin, Simvastatin, and Pravastatin Across Doses (STELLAR* Trial). Am J Cardiol 2003;92:152-160. 3. Schuster H, Barter PJ, Stender S, et al. Effects of switching statins on achievement of lipid goals. Measuring Effective Reduction in Cholesterol Using Rosuvastatin Therapy (MERCURY I) study. Am Heart J 2004;147:705-712. 4. Rosenson RS. Statins: can the new generation make an impresssion? Expert Opin Emerg Drugs 2004;9(2):269-279. 5. Shepherd J, Hunninghake DB, Stein EA, et al. Safety of rosuvaststin. Am J Cardiol 2004;94:882-888.

16162 184 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Editorial

Endothelial dysfunction: are we ready to heed the vasculature’s early-warning signal? HANS STRIJDOM

Endothelial dysfunction (ED) refers to a spectrum of As Mudau and co-workers report, there are several pathophysiological changes in the vascular endothelium that biomarkers that could be used to assess endothelial function. ultimately results in a loss of vascular homeostasis. Traditional However, currently elevated ADMA levels appear to be the cardiovascular risk factors (e.g. diabetes mellitus, smoking, most promising biomarker of ED in terms of prognostic value dyslipidaemia and hypertension) are all associated with the and positive correlation with cardiovascular risk. Promising development of ED via sustained and harmful effects, mediated data have also been obtained from studies investigating ex-vivo by circulating stimuli such as pro-inflammatory tumour necrosis functional endothelial assessment techniques (current gold factor-alpha (TNF-alpha), oxidised low-density lipoprotein standard: flow-mediated dilatation) with regard to their ability to (ox-LDL), asymmetrical dimethyl-arginine (ADMA), predict cardiovascular risk. II and hyperglycaemia.1 Despite many promising studies showing that biomarkers The underlying cellular mechanisms of ED are directly of ED and ex-vivo endothelial function assessment are strongly or indirectly related to the development of oxidative stress correlated with cardiovascular risk, a consensus for their universal (particularly increased superoxide anion production via NADPH- clinical usefulness remains elusive.6 Reaching consensus is oxidase and xanthine oxidase), which reduces the bioavailability hampered by factors such as the availability of special equipment of the main endothelial-derived vasodilator, nitric oxide (NO) in everyday practice, the lack of data on the predictive value of via the reaction of superoxide with NO (thereby scavenging NO) more recent endothelial function assessment techniques, and the to form peroxynitrite (ONOO-), a highly reactive molecule. The lack of studies investigating which patient groups will benefit latter has the ability to uncouple endothelial NO synthase (eNOS), most from ex-vivo endothelial function measurement.6 Finally, which further reduces NO production and simultaneously more studies demonstrating that putative anti-ED therapies increases superoxide anion generation.2 As a result, vascular actually achieve their clinical benefit by the improvement of endothelial function becomes compromised, manifesting as endothelial function are required. a loss of endothelium-dependent vasorelaxation, increased In the South African/African context, research into ED, its thrombosis, the development of a generalised pro-inflammatory underlying mechanisms and potential clinical application is state (increased expression of vascular adhesion molecules) and equally relevant. Indications are pointing to significant increases increased vascular permeability.3 Ultimately, ED can develop in the incidence of cardiovascular disease among people of into atherosclerosis.2 African heritage in South Africa as a result of epidemiological The importance of ED as a potential predictor of long-term transition and chronic diseases of lifestyle.7 development of atherosclerosis and cardiovascular event rate2 Data from the Heart of Soweto (HOS) study in 2006 indicated is evident by the high number of research articles on this topic a high prevalence of traditional cardiovascular risk factors (PubMed search with keywords: endothelial + dysfunction such as hypertension and obesity, with almost two-thirds of revealed 51 600 hits and approximately 3 000–4000 articles per the patients in the study cohort presenting with multiple risk year on this topic since 2005). Herein, however, lies both the factors.7 Approximately 10% of the patients in the HOS study greatest potential and challenge of current research into ED, were diagnosed with coronary artery disease. In addition, recent namely translating the wealth of data obtained with laboratory- research has drawn attention to the interaction between HIV based research into scientifically validated predictive, diagnostic infection, anti-retroviral treatment and the increased rate of and even therapeutic tools in the clinical setting. coronary artery disease in HIV-infected patients.8 Due to the As Mudau et al. explain in their comprehensive review article fact that HIV-infected patients on treatment live longer, they are on the cellular mechanisms and clinical applications of ED in the subjected to traditional cardiovascular risk factors similar to the current issue of this journal, there is no doubt that ED serves as a non-infected population. However, of concern is the additional crucial pathophysiological link between traditional cardiovascular burden of non-traditional risk factors, namely the direct risk factors and the eventual development of atherosclerosis and vascular effects of the HI virus and the anti-retroviral drugs.7,8 ischaemic heart disease (IHD).4 Particularly helpful in this regard Consequently, HIV-infected patients on treatment now face the is the fact that ED is an early, reversible event.5 This presents risk of developing premature ED and accelerated atherosclerosis. researchers and clinicians with a golden opportunity to not only In a recent study in South Africa, significantly higher levels predict the development of atherosclerosis, IHD and possibly of inflammatory biomarkers associated with ED were detected in other cardiovascular events, but also prevent the development newly diagnosed, untreated HIV-infected participants of African of these conditions by therapeutically reversing early vascular descent compared to non-infected participants, which was dysfunction. accompanied by age-related increases in arterial stiffness.9 In view AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 185

of potentially unprecedented increases in ED and atherosclerosis HANS STRIJDOM, [email protected] in particularly the sub-Saharan African region, and the Division of Medical Physiology, Department of Biomedical subsequent burden of health this could introduce, it is imperative Sciences, Faculty of Health Sciences, Stellenbosch that researchers and research funding institutions prioritise University, South Africa ED, with respect to both basic scientific and clinical research. In conclusion, ED is an early and potentially reversible event References in the development of atherosclerosis and can therefore be 1. Esper RJ, Nordaby RA, Vilarino JO, Paragano A, Cacharron JL, regarded as a vascular early-warning signal. As cardiovascular Machado RA. Endothelial dysfunction: a comprehensive appraisal. researchers and clinicians, we should not ignore the importance Cardiovasc Diabetol 2006; 5: 4. DOI: 10.1186/1475-2840-5-4. of this window of opportunity offered to us by ED, since proper 2. Münzel T, Sinning C, Post F, Warnholtz A, Schulz E. Pathophysiology, diagnosis and prognostic implications of endothelial dysfunction. Ann detection followed by therapeutic or lifestyle interventions could Med 2008; 40: 180–196. prevent potentially catastrophic cardiovascular events later in the 3. Hsueh WA, Lyon CJ, Quinones MJ. Insulin resistance and the endothe- lives of affected patients. lium. Am J Med 2004; 117: 109–117. As explained above, this has become particularly relevant in 4. Mudau M, Genis A, Lochner A, Strijdom H. Endothelial dysfunction the sub-Saharan African region, as we are entering unchartered – the early predictor of atherosclerosis. Cardiovasc J Afr 2012; 23(4): waters with regard to a predicted surge in the incidence of 222–231. cardiovascular diseases previously unheard of in patients of 5. Chhabra N. Endothelial dysfunction – A predictor of atherosclerosis. Internet J Med Update 2009; 4(1): 33–41. African descent. This is due to a double burden of both traditional 6. Lekakis J, Abraham P, Balbarini A, et al. Methods for evaluating risk factors introduced by epidemiological transition and chronic endothelial function: a position statement from the European Society of diseases of lifestyle, but recently also due to non-traditional Cardiology Working Group on Peripheral Circulation. Eur J Cardiovasc risk factors introduced by the HI virus and anti-retroviral drugs, Prev Rehabil 2011; 18(6): 775–789. particularly with regard to their targeted pro-ED effects. As basic 7. Pretorius S, Stewart S, Sliwa K. Lessons from the Heart of Soweto research findings in the context of ED are moving ever closer to Study and future directions. SA Heart 2011; 8: 104–113. effective and potentially standardised clinical applications (i.e. 8. Fedele F, Bruno N, Mancone M. Cardiovascular risk factors and HIV disease. AIDS Rev 2011; 13(2): 119–129. early detection of ED followed by effective and targeted reversal 9. Fourie C, Van Rooyen J, Pieters M, Conradie K, Hoekstra T, Schutte A. of ED), it is crucial that researchers and clinicians in South Is HIV-1 infection associated with endothelial dysfunction in a popula- Africa and Africa remain abreast of the latest developments. Are tion of African ancestry in South Africa? Cardiovasc J Afr 2011; 22(3): we up to the challenge? 134–140.

MARK YOUR 17-22 FEBRUARY 2013 CALENDAR CAPE TOWN, SOUTH AFRICA

6th COME TO CAPE TOWN... Paediatric Cardiology & Cardiac Surgery A trendy, sophisticated, multi-cultural city at the foot of Africa in a diverse and beautiful natural environment. Cape Town is a destination with irresistable appeal. South Africa has a compelling history and with its abundance of game reserves Conference Secretariat o ers visitors a uniquely di erent cultural and tourist experience. Contact the conference secretariat or visit www.pccs2013.co.za The most distinguished international faculty available makes the “6th World Congress” an attractive, interactive and unique meeting place for clinicians, Event Dynamics scientists health care managers and policy developers from all across our world. P.O Box 6761, Roggebaai, 8012, South Africa Telephone: +27(0) 21 408 9796

Fax: +27(0) 21 408 9954 PROGRAMME TRACKS E-mail: [email protected] •Surgery, anaesthesia and intensive care •Catheter interventions from fetus to adult Hosted by: •Health systems and heart disease •Adults with congenital and acquired heart disease •Cardiology and the imaging revolution

SOUTH AFRICAN HEART ASSOCIATION www.pccs2013.co.za 186 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Cardiovascular Topics

Association between troponin T and ICU mortality, a changing trend S HAJSADEGHI, S GHOLAMI, G GOHARDEHI, NS MOGHADAM, AS SABET, SRJ KERMAN, M MORADI, R MOLLAHOSEINI

Abstract Keywords: troponin T, ICU, mortality, serial measurement, Background: Initially elevated levels of troponin predict changing trend adverse outcomes in patients admitted to the intensive care Submitted 23/8/10, accepted 7/6/11 unit (ICU). No research team has investigated the changes in Cardiovasc J Afr 2012; 23: 186–190 www.cvja.co.za concentration of cardiac troponin T (cTnT) during ICU stay and their association with patient outcome. DOI: 10.5830/CVJA-2011-034 Objective: We investigated whether the change in cTnT levels during ICU stay could predict outcomes (death or survival). Dysfunction of, or injury to the myocardium was recently Methods: In this cohort study, all patients admitted to the found to be an important complication in non-cardiac critically medical ICU (10 beds) from January to July 2008 were ill patients, including those with sepsis, pulmonary embolism, enrolled. Troponin levels were evaluated within the first 24 renal insufficiency and acute stroke, resulting in increased hours of ICU admission and on the fourth, seventh and 10th concentrations of cardiac troponin T (cTnT).1-5 The pathophysiology days after admission. of myocardial injury in critically ill patients is believed to Results: The study population (135 patients) had a mean age be multifactorial, including the underlying disease process, of 60.9 ± 21.5 years. The outcome was significantly different hypoxaemia and acidosis, as well as therapeutic activities.6,7 with regard to normal or elevated cTnT concentrations on Over the past decade, several studies have indicated that the first and seventh days of follow up (p = 0.03 and 0.023, cardiac dysfunction is a frequent and important factor in respectively). This difference was non-significant for cTnT determining the outcome of critically ill patients.8,9 Damage to levels on the fourth and 10th days after admission (p = 0.69 the myocardial cells results in the release of contractile regulating and 0.78, respectively). The change in cTnT levels was not proteins such as cTnT, which is a highly sensitive and specific significantly different between the deceased and discharged marker of myocardial injury.10,11 More importantly, elevated patients (p = 0.4). levels of troponin predict a poor prognosis in patients with acute Conclusion: Changes in cTnT levels during ICU stay did not coronary syndromes12-16 and may also predict adverse outcomes show a significant trend (power: 0.26). Patients whose cTnT in other patients admitted to the intensive care unit (ICU). levels were increased on the first and seventh days of ICU In a medical ICU, patients with elevated levels of troponin stay had a worse survival, which could be associated with T or troponin I admitted without a diagnosis of acute coronary cardiac events on admission or at specific times during the syndrome exhibited a fourfold higher mortality rate.17 In patients stay in ICU. in a surgical ICU, moderate elevations in troponin I were associated with higher mortality rates, and a longer stay in hospital and in the ICU.16-18 Some studies19,20 have shown that Department of Cardiology, Rasoul-e-Akram Hospital, Tehran increased cTnT levels are associated with increased mortality, University of Medical Sciences, Tehran, Iran but others did not confirm this association.21,22 The association S HAJSADEGHI, MD between elevated troponin level and adverse outcome was uncertain, so the interpretation of elevated troponin levels during Firoozgar Clinical Research Development Centre, Firoozgar Hospital, Tehran University of Medical Sciences, Tehran, Iran critical illness remains unclear. S GHOLAMI, [email protected] Cardiac events resulting in elevated biomarkers may occur during ICU stay and have an important effect on patient outcome. Medical Student Research Committee, Tehran University of There is no research evaluating the relationship between changing Medical Sciences, Tehran, Iran levels of any cardiac biomarkers during ICU stay and patients’ G GOHARDEHI NS MOGHADAM outcome. We therefore hypothesised that patients admitted to AS SABET the ICU with increased cTnT levels were at increased risk of SRJ KERMAN death, and determined whether elevated troponin levels were M MORADI related to outcome and length of stay and mortality in the Department of Neurosurgery, Firoozgar Hospital, Tehran ICU. We also investigated whether the change in cTnT levels University of Medical Sciences, Tehran, Iran during ICU stay could predict outcomes (death or survival). R MOLLAHOSEINI, MD Correlation of troponin levels with the most popular disease AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 187

severity classification system, the Acute Physiology and Chronic and 62 (46%) women, with a mean age of 60.9 ± 21.5 years, Health Evaluation II (APACHE II) score, was also evaluated. ranging from 15 to 100 years old. The most common diagnosis was infectious disease, including sepsis (n = 21, 15.5%), stroke (n = 11, 8.1%), and pulmonary Methods disease, including chronic obstructive pulmonary disease and In this cohort study, all 135 patients admitted to the medical pneumonia (n = 11, 8.1%). There was no significant difference in ICU (10 beds) from January to July 2008 were enrolled measured cTnT levels on the first, fourth, seventh and 10th days consecutively. The patients mostly suffered from sepsis (n = 21, of ICU stay between patients with different diseases (p > 0.05). 15.5%), stroke (n = 11, 8.1%) and pulmonary disease, including History of cardiovascular disease, including MI, hypertension chronic obstructive pulmonary disease (COPD) and pneumonia and heart failure was positive in 51 (37%) patients. The (n = 11, 8.1%). This investigation was approved by the ethics frequency of patients with elevated cTnT levels was not committee of the Iran University of Medical Sciences. Verbal significantly different in patients with and without a history of consent was obtained from all patients (or from their next of kin) cardiac disease. after detailed explanations and a letter of explanation was given. On admission, 83 patients (70.3%) had normal cTnT levels, On admission to the medical ICU, demographic and baseline whereas 35 (29.7%) had elevated levels > 0.01 ng/ml, mean clinical characteristics, including age, gender and APACHE = 0.28 ± 0.5 ng/ml (range 0.03–2.00 ng/ml). The clinical and II score and the diagnoses of all patients were recorded. The laboratory characteristics of the two groups are shown in APACHE II system incorporates acute physiological variables Table 1. There was no significant difference between the two and chronic health evaluation into a measurement of the risk groups with normal and elevated cTnT levels in some baseline 23 of in-hospital mortality. Levels of cardiac biomarkers such as characteristics, including Glasgow coma score (GCS) (p = creatine kinase (CK) and creatine kinase isoenzyme (CK-MB), 0.223) and partial arterial oxygen tension (PaO2; p = 0.607) on and history of cardiovascular disease were included. All patients admission. A significant difference was found in the APACHE II in the ICU had (ECG) within 24 hours of score (p = 0.003) and serum creatinine levels (p = 0.003) between admission. The length of stay in the ICU was also recorded. The the two groups with different troponin results. clinical endpoint was death or discharge from the ICU at any Patients with negative baseline troponin T levels had a time during hospitalisation. significantly lower APACHE II score (17.75 ± 7.76 vs 24.3 ± Troponin measurements were collected within the first 24 8.8, p = 0.003). The APACHE II score on the fourth day after hours of ICU admission and on the fourth, seventh and 10th ICU admission was significantly higher in patients with positive day after admission. This was suggested by the ICU team and was based on their clinical judgment. Troponin T was measured TABLE 1. BASELINE LABORATORY PARAMETERS OF using a radio-immunoassay. The analytical sensitivity (lower 118 CRITICALLY ILL PATIENTS ON ADMISSION TO ICU detection limit) of this assay is 0.01 µg/l. Troponin concentration Negative Positive of 0.1 µg/l was considered normal. Because of a laboratory cTnT level cTnT level mistake, 16 patients did not have troponin T levels measured on Baseline parameters (n = 83) (n = 35) p-value admission, but troponins were measured on days four, seven and Age (years) 59.8 ± 21.63 61.3 ± 20 0.732 10; these data were included in the analysis. Of all patients, 74 APACHE II score 17.75 ± 7.76 24.3 ± 8.8 0.003 patients had the first two, 47 had the first three, and 25 had four (mmHg) 93.47 ± 13.43 95.17 ± 16.24 0.569 serial measurements for cTnT concentrations. Patients did not (beats/min) 91.56 ± 19 94.74 ± 22.69 0.446 have cTnT levels measured on discharge or on death. Respiratory rate (per min) 21.41 ± 4.52 20.48 ± 5.7 0.363

PaO2 (mmHg) 97.5 ± 56.6 91.2 ± 46.7 0.607

Statistical analysis FiO2 (%) 61.9 ± 27.5 61 ± 26.4 0.887 HCO (mmol/l) 25.1 ± 8.34 25.11 ± 12.67 0.995 Baseline characteristics of the negative (normal levels) and 3 positive (elevated levels) cTnT groups were compared using the Sodium (mmol/l) 141 ± 7.05 140.5 ± 6.69 0.713 Pearson’s chi-square test. Continuous variables were compared Potassium (mmol/l) 4.11 ± 0.6 4.11 ± 0.7 0.998 using the Student’s t-test for normally distributed variables, and Creatinine (mg/dl) 1.19 ± 0.96 2.58 ± 2.53 0.003 the Mann–Whitney U test if either of these conditions were Haematocrit (%) 35.55 ± 9.33 32.5 ± 8.33 0.114 not met. Linear regression analysis was done to evaluate the White blood cell count (per ml) 12.22 ± 6.49 12.22 ± 4.97 0.999 independent association between APACHE score and cTnT levels. pH 7.36 ± 0.09 7.34 ± 0.1 0.422 For evaluation of the changes in cTnT levels during ICU Systolic (mmHg) 127.4 ± 18.6 129.7 ± 18.7 0.545 stay, repeat-measurement ANOVA analysis was carried out. Diastolic blood pressure (mmHg) 76.49 ± 12.58 77 ± 17.8 0.854 The Breslow method was used to compare patient survival in GCS (median) 9 10 0.223 the Kaplan–Meier analysis. Data are presented as mean ± SD. CK (U/l) 236.12 ± 255 420 ± 735 0.262 A p-value less than 0.05 was considered significant. Receiver CK-MB (ng/ml) 28.63 ± 21 128.3 ± 190 0.72 operating characteristics (ROC) were used for the detection of Male/female (ratio) 45/38 17/18 0.531 cut-off points for APACHE to predict elevated cTnT levels. History of cardiac disease 29/54 12/23 0.959 (yes/no) ST change on ECG (yes/no) 36/47 18/15 0.29

Results - PaO2: arterial O2 tension, FiO2: fraction of inspired oxygen, HCO3 : bicar- The study population consisted of 135 patients admitted to the bonate, GCS: Glasgow coma scale, CK: creatine kinase, CK-MB: creatine ICU over a period of seven months. There were 73 (54%) men kinase-MB. 188 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

TABLE 2. THE RESULTS OF ROC ANALYSIS TABLE 3. MORTALITY RATE BASED ON NORMAL OR ELEVATED cTnT CONCENTRATION IN DIFFERENT APACHE score Sensitivity Specificity MEASUREMENTS 18.5 0.684 0.552 cTnT concentration Deceased p-value 19.5 0.684 0.603 (µg/l) n (%) 20.5 0.684 0.621 cTnT-1 < 0.1 33 (40) 0.03 21.5 0.632 0.655 on the 1st day > 0.1 19 (63) 22.5 0.526 0.69 cTnT-2 < 0.1 29 (53) 0.69 23.5 0.526 0.741 on the 4th day > 0.1 11 (58) 24.5 0.526 0.793 cTnT-3 < 0.1 18 (44) 0.024 25.5 0.526 0.828 on the 7th day > 0.1 13 (65) 27 0.474 0.879 cTnT-4 < 0.1 16 (57) 0.78 28.5 0.368 0.897 on the 10th day > 0.1 5 (62) The italicised rows are suggested as the best points to predict cTnT.

TABLE 4. cTnT CONCENTRATION IN DECEASED OR cTnT levels (25.07 ± 8.01 vs 19.06 ± 8.57, p = 0.026). According DISCHARGED PATIENTS to ROC analysis, an APACHE score of more than 20.5 indicated Deceased Discharged p-value a positive cTnT with 66.7% sensitivity and 62.1% specificity. (µg/l, mean ± SD) (µg/l , mean ± SD) An APACHE II score higher than 25.5 indicated a positive cTnT 1st day 0.042 ± 0.061 0.03 ± 0.084 0.4 with 50% sensitivity and 83% specificity (p = 0.006, area under 4th day 0.103 ± 0.277 0.028 ± 0.066 0.12 the curve = 0.712 and standard error = 0.072) (Table 2, Fig. 1). 7th day 0.094 ± 0.25 0.852 ± 0.175 0.86 Serial blood sampling for cTnT levels was done in the first 10th day 0.121 ± 0.365 0.053 ± 0.119 0.42 24 hours, and on the fourth, seventh and 10th day after ICU admission (Fig. 2). No significant difference was seen in the changes in troponin levels between deceased and discharged level from the fourth to the seventh day died. patients (p = 0.4). This difference was also non-significant when The survival rate of patients on the fourth, seventh and 10th different diseases were evaluated separately. The difference days after admission was 87, 74 and 72% in normal patients, between troponin levels was not significant between deceased and 83, 61 and 56% in patients with elevated cTnT levels, and discharged patients (p > 0.05). respectively. The estimated mean survival time of patients with The outcome was significantly different with regard to elevated cTnT levels on admission was 16.77 (SE = 3.38) days, normal or elevated cTnT concentrations on the first and seventh and it was significantly lower than the group with normal levels days after admission (Table 3) but there was a non-significant on admission, with a mean survival time of 26.08 (SE = 3.15) difference between patients with different levels of cTnT on days (p = 0.027). This difference was also significant for cTnT the fourth and 10th days of admission (Table 3). The cTnT concentration on the seventh day after admission [mean survival concentration was not significantly different between deceased 31.6 (SE = 3.6) days in patients with normal cTnT levels vs 19.7 and discharged patients (Table 4). (SE = 2.5) days in patients with elevated cTnT levels, p = 0.012] There was a significant association between patients’ outcome (Fig. 3A, B). and change in cTnT levels in the first and fourth days after admission using chi-square analysis (p = 0.029). All patients Discussion with a reduced cTnT level died, whereas all patients with an The present study shows that in critically ill patients, elevated increased cTnT level survived. Also a significant relationship cTnT levels were significantly associated with increased was found between the outcome of patients and the level of mortality, and the survival time was significantly shorter in cTnT on the fourth and the seventh days of ICU stay (p = 0.02). patients with elevated cTnT than in those with normal cTnT All patients with an elevated cTnT level from the fourth to the levels. seventh day survived, whereas all patients with a decreased cTnT

0.12 Outcome ROC curve died 1.0 0.10 discharged 0.8 0.8

0.6 0.6

0.4

0.4 Means of c T nt S ensitivity 0.2 0.2

0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1st 4th 7th 10th Specificity Day of measurment Fig. 1. The results of ROC analysis (area under the curve Fig. 2. Trend in cTnT levels over four time periods for 25 = 0.712 and standard error = 0.072). patients (± SD). AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 189

A B 1.0 < 0.1 1.0 cTnT < 0.1 > 0.1 cTnT > 0.1 0.8 < 0.1-censored 0.8 censored > 0.1-censored 0.6 0.6

0.4 0.4 C um survival C um survival 0.2 0.2

0.0 0.0 0 30 40 60 0 20 40 60 Length of hospital stay (days) Length of hospital stay (days) Fig. 3. Patient survival and comparison between patients with high and low cTnT levels on (A) first day of admission (p = 0.027), (B) seventh day of ICU stay (p = 0.012).

Cardiac troponin I and troponin T are the most specific and Ammann et al.17 showed that troponin positivity on admission sensitive laboratory markers of myocardial cell injury. However, was associated with a fourfold increased risk of mortality in 58 they may be elevated in patients presenting with many conditions critically ill patients without acute coronary syndrome. In their other than acute coronary syndrome.24 Elevated levels of cTnT study, a significant difference in survival between troponin- were previously considered a specific marker for the diagnosis positive and troponin-negative patients was found mainly in of MI. Several recent studies have reported unexpectedly high a subgroup of patients without volume-refractory shock. This cTnT levels in the serum of ICU patients who did not have implied that the analysis of troponin level could predict mortality underlying coronary syndrome,9,10,25–28 indicating unrecognised in the early but not the late stage of the disease. In another myocardial injury during their disease process. study by Vlad et al., high levels of cTnT on admission had an It is reasonable to suggest that critically ill patients are at independent association with in-hospital, short- and long-term increased risk of myocardial cell injury. This is due to exposure mortality in 2 078 patients with acute respiratory distress.31 to many stresses that increase myocardial oxygen demand, A limitation of the study was that we did not measure renal whereas the myocardial oxygen supply can be limited by function in any patient. Renal function is an important factor shock, anaemia, , hypoxaemia and impaired tissue in cTnT concentrations because the kidney filters cTnT from .9 These events can result in the release of troponin from the blood. An investigation with a larger sample size is also cardiomyocytes into the serum. In addition, tumour necrosis necessary. An inevitable limitation of the sample size was factor (TNF), produced by inflammatory cells, can depress the decreasing number of patients over time because of the myocardial function and induce cardiomyocyte apoptosis, which death of some between admission and the 10th day. Another results in low coronary artery flow and decreased ejection limitation was the assay method, which could not detect a cTnT fraction, which may lead to necrosis and cTnT release from concentration less than 0.01 µg/l, so we did not have exact cardiomyocytes.29,30 It was found that patients with higher cTnT concentrations for many patients, who were reported to have zero levels had a shorter survival rate. cTnT concentrations. The APACHE II scoring system predominately evaluates haemodynamic changes rather than heart function, but cTnT levels could provide a direct marker for cardiac injury, even Conclusion if clinically unrecognised. In the present study, the APACHE We found that levels of cTnT could predict outcome in critically II score on ICU admission was significantly different between ill patients at specific times. There was a significant association patients with elevated and normal cTnT levels. This could have between outcome and cTnT level on the first and seventh days been due to a more serious baseline condition. of ICU stay, and a non-significant association with cTnT level on Levels of cTnT on admission and on the seventh day could the fourth and 10th days. The level of cTnT had an association predict an adverse prognosis in critically ill patients, whereas with outcome and survival, and was shown to be a predictor of troponin levels on the fourth and 10th days could not. The outcome. Cardiac TnT levels during ICU stay did not show a peak concentration of cTnT was different in these two groups significant trend overall, which may have been due to the small (fourth day in surviving patients and seventh day in patients who sample size (power: 0.26) but changes in cTnT levels at specific died), which in addition to the other findings, represents some times could be a useful predictor. A study focusing on defining undefined importance in the time of assessment of cTnT levels. the best time for measurement of cTnT levels would also provide Although serial measurements did not provide additional crucial data. statistical value for risk stratification, the different peaks of troponin levels in deceased and discharged patients are shown in References Fig. 2. Different outcomes during ICU stay related to decreasing 1. Robert E., Fromm Jr. Cardiac troponins in the intensive care unit: or increasing cTnT levels have not been reported before. We Common causes of increased levels and interpretation. Crit Care Med found that whenever cTnT levels begin to decrease, an adverse 2007; 35(2): 584–588. 2. Wu AH. Increased troponin in patients with sepsis and septic shock: outcome could be expected, and increasing cTnT level was a myocardial necrosis or reversible myocardial depression? Intensive predictor of a favourable outcome. This indicates the significant Care Med 2001; 27: 959–961. role that daily changes in cTnT levels play, independent of 3. Parker MM. Myocardial dysfunction in sepsis: injury or depression? baseline cTnT levels. Crit Care Med 1999; 27: 2035–2036. 190 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

4. Guest TM, Ramanathan AV, Tuteur PG, et al. Myocardial injury in troponin concentrations are associated with increased morbidity and critically ill patients. A frequently unrecognized complication. J Am mortality rates in surgical intensive care unit patients. Crit Care Med Med Assoc 1995; 273: 1945–1949. 2003; 31: 2598–2603. 5. Kollef MH, Ladenson JH, Eisenberg PR. Clinically recognized cardiac 18. Jensen JK. Kristensen SR, Bak S, et al. Frequency and significance of dysfunction: an independent determinant of mortality among critically troponin T elevation in acute ischemic stroke. Am J Cardiol 2007; 99: ill patients. Is there a role for serial measurement of cardiac troponin I? 108–112. Chest 1997; 111: 1340–1347. 19. Di AE, Fiorelli M, Toni D, et al. Prognostic significance of admission 6. Hamm CW, Ravkilde J, Gerhardt W, et al. The prognostic value of levels of troponin I in patients with acute ischaemic stroke. J Neurol serum troponin in unstable angina. N Engl J Med 1992; 327: 146–150. Neurosurg Psychiatry 2005; 76: 76–81. 7. Ohmann EM, Armstrong PW, Christenson RH, et al. Risk stratification 20. Etgen T, Baum H, Sander K, et al. Cardiac troponins an N-terminal with admission cardiac troponin T levels in acute myocardial ischemia. pro-brain natriuretic peptide in acute ischemic stroke do not relate to N Engl J Med 1996; 335: 1333–1334. clinical prognosis. Stroke 2005; 36: 270–275. 8. Christensen RH, Duh SH, Newby LK, et al. Cardiac troponin T and 21. Barber M, Morton JJ, Macfarlane PW, et al. Elevated troponin levels cardiac troponin I: relative values in short-term risk stratification of are associated with sympathoadrenal activation in acute ischaemic patients with acute coronary syndromes. Clin Chem 1998; 44: 494–501. stroke. Cerebrovasc Dis 2007; 23: 260–266. 9. Jaffe AS, Ravkilde J, Roberts R, et al. It’s time for a change to a 22. Ammann P, Pfisterer M, Fehr T, et al. Raised cardiac troponins. Br Med troponin standard. Circulation 2000; 102: 1216–1220. J 2004; 328: 1028–1029. 10. Turner A, Tsamitros M, Bellomo R. Myocardial cell injury in septic 23. Knaus WA, Draper EA, Wagner DP, et al. APACHE II: a severity of shock. Crit Care Med 1999; 27: 1775–1780. disease classification system. Crit Care Med 1985; 13(10): 818–829. 11. Cunnion RE, Schaer GL, Parker MM, et al. The coronary circulation in 24. Hamm CW. Risk stratifying acute coronary syndromes: gradient of risk human septic shock. Circulation 1986; 73: 637–644. and benefit. Am Heart J 1999; 138: 6–11. 12. Stubbs P, Collinson P, Moseley D, et al. Prospective study of the role 25. Willging S, Keller F, Steinbach G. Specificity of cardiac troponin I and of cardiac troponin T in patients admitted with unstable angina. Br Med T in renal disease. Clin Chem Lab Med 1998; 36: 87–92. J 1996; 313: 262–264. 26. Spies C, Haude V, Fitzner R, et al. Serum cardiac troponin-T as a prog- 13. Lindahl B, Venge P, Wallentin L. Relation between troponin T and the nostic marker in early sepsis. Chest 1998; 113: 1055–1063. risk of subsequent cardiac events in unstable coronary artery disease. 27. Arlati S, Brenna S, Prencipe L, et al. Myocardial necrosis in ICU The FRISC study group. Circulation 1996; 93: 1651–1657. patients with acute non-cardiac disease: a prospective study. Intensive 14. Galvani M, Ottani F, Ferrini D, et al. Prognostic influence of Care Med 2000; 36: 31–37. elevated values of cardiac troponin I in patients with unstable angina. 28. Paterson D, Beyne P, Laperche T, et al. Elevated circulating troponin I Circulation 1997; 95: 2053–2059. in patients with cirrhosis. Hepatology 1999; 29: 640–643. 15. Newby LK, Christenson RH, Ohman EM, et al. Value of serial troponin 29. Meldrum DR: Tumor necrosis factor in the heart. Am J Physiol 1998; T measures for early and late risk stratification in patients with acute 274: 577–595. coronary syndromes. The GUSTO-IIa Investigators. Circulation 1998; 30. Shames BD, Barton HH, Reznikov LL, et al. Ischemic alone is suffi- 98: 1853–1859. cient to induce TNF-mRNA and peptide in the myocardium. Shock 16. Ammann P, Maggiorini M, Bertel O, et al. Troponin as a risk factor for 2002; 17: 114–119. mortality in critically ill patients without acute coronary syndromes. J 31. Vasile VC, Chai HS, Khambatta S, et al. Significance of elevated Am Coll Cardiol 2003; 41: 2004–2009. cardiac troponin T levels in critically ill patients with acute respiratory 17. Relos RP, Hasinoff IK, Beilman GJ. Moderately elevated serum disease. Am J Med 2010; 123: 1049–1058.

Diabetes congress diary Date Conference Venue Contact details to register MAY 2012 24–27 May 27th Annual Clinical Conference on Diabetes Florida, USA [email protected] JUNE 2012 8–12 June 72nd American Diabetes Association Scientific Sessions Pennsylvania, USA http://professional.diabetes.org JULY 2012 9–12 July The 3rd International Congress on Abnormal Obesity (ICAO) Quebec City, Canada Icao2012.myhealthwaist.org 24–26 July CDE Centres for Diabetes and Endocrinology Johannesburg, South Africa 25–28 July 1st African Diabetes Congress Arusha, Tanzania www.africadiabetescongress.org OCTOBER 2012 1–5 October 48th EASD Annual Meeting Berlin, Germany www.easd.org 10–13 October ISPAD 2012 – 38th Annual Meeting of the International Society for Istanbul, Turkey Pediatric and Adolescent Diabetes 20–22 October 8th Asian Pacific Society of Atherosclerosis and Vascular Disease Phuket, Thailand www.apsavd2012.com DECEMBER 2012 4–6 December 1st American Diabetes Association Middle East Congress: Diabetes Dubai, UAE Prevention and Treatment To advertise your conference/meeting, e-mail details and half-page pdf advert to [email protected] AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 191

Nebivolol therapy improves QTc and QTcd parameters in heart failure patients SM AKSOY, S CAY, G CAGIRCI, N SEN

Abstract effect of nebivolol on QTc and QTcd in patients with heart Aim: It has previously been shown that β-blocker therapy failure with systolic dysfunction. reduces QT dynamics in heart failure patients. The aim of this study was to demonstrate this improvement with the Methods β third-generation -blocker, nebivolol. Consecutive β-blocker-naïve (no use during the four weeks Methods: A total of 72 heart failure patients with systolic before study entry) systolic heart failure patients were selected dysfunction were included in the study. Corrected QT (QTc) for the study. A total of 75 patients were possible candidates and and QT dispersion (QTcd) were measured manually by two were included in the study. However, three were subsequently independent observers at baseline and after nebivolol use (5 excluded because of side effects of the drug. At the end of the mg/day) in the first and third months of follow up. follow-up period, 72 patients taking the drug remained. Results: Both QTc and QTcd were found to be significantly All the participants had depressed left ventricular ejection ± ± p < reduced in the first (455.3 26.7 vs 441.2 25.7 ms, 0.001 fraction (< 40%) as per the study protocol. Patients were p for QTc, and 65.6 ± 5.3 vs 58.2 ± 5.6 ms, = 0.001 for QTcd) interviewed about any cardiovascular drug use, and NYHA p and third months (455.3 ± 26.7 vs 436.0 ± 28.7 ms, < 0.001 functional class was determined. Systolic and diastolic blood p for QTc, and 65.6 ± 5.3 vs 56.0 ± 6.2 ms, < 0.001 for QTcd) pressure and heart rates were measured at the beginning of the compared with baseline values. study and in the third month. Laboratory parameters including Conclusion: Nebivolol was associated with improved QT serum electrolytes were assayed. Nebivolol 5 mg/day was started dynamics in heart failure patients with systolic dysfunction. in all patients and the same dose was continued throughout the Keywords: improvement, nebivolol, QT dynamics study period. Patients with β-blocker use in the previous four weeks, Submitted 18/2/10, accepted 30/8/11 electrolyte abnormalities, any drug use prolonging QT Cardiovasc J Afr 2012; 23: 191–193 www.cvja.co.za interval, coronary artery disease or acute coronary syndromes, DOI: 10.5830/CVJA-2011-046 pacemakers, atrial fibrillation, and significant hepatic and renal dysfunction were excluded from the study. Heart failure, both systolic and diastolic, is a clinical syndrome The electrocardiograms (ECG) were recorded after 10 minutes with different treatment modalities and a poor prognosis, of rest in the supine position, using 12 simultaneously recording especially in the advanced stages. β-blocker therapy, including leads: three standard (DI, DII, DIII), three unipolar (aVR, aVL, nebivolol, improves survival in these patients by modulating aVF) and six precordial (V1–V6), at a paper speed of 25 mm/s. several neurohormonal mechanisms.1-4 Readings were done manually by the same two cardiologists blinded to the study. There is evidence that manual measurement Nebivolol is a selective β1-blocker with vasodilatory properties due to modulation of nitric oxide release, which is superior to automatic measurement of QTd.9,10 decreases peripheral .5,6 In addition, it has The QT interval was measured in each lead from the onset of been demonstrated that in patients with hypertension, nebivolol the QRS complex: the beginning of the Q wave to the terminal reduces QT dispersion and corrected QT interval (QTc) and inscription of the in the lead with clearly identified corrected QT dispersion (QTcd), which are indicators of the T-wave termination, or from the beginning of the R wave if the heterogeneity of myocardial repolarisation and electrical Q wave was absent. The terminal inscription of the T wave was instability.7 Abnormality of these parameters has been also found determined as the return to the TP baseline. When a U wave to be associated with adverse cardiac events and mortality.8 was present, the QT interval was measured to the nadir of the In the current study, we aimed to demonstrate the beneficial curve between the T and U waves. The QTcd was defined as the difference between the maximum and minimum of the QTc intervals, measured in milliseconds in any of the measured ECG leads. Department of Anesthesiology, Diskapi Yildirim Beyazit Education and Research Hospital, Ankara, Turkey SM AKSOY, MD Statistical analysis Department of Cardiology, Yuksek Ihtisas Heart Education Data were analysed with SPSS software version 15.0 for and Research Hospital, Ankara, Turkey Windows (SPSS Inc, Chicago, Illinois, USA). Continuous S CAY, MD, [email protected] variables are presented as mean ± SD and categorical variables N SEN, MD as frequency and percentage. The Kolmogorov–Smirnov test Department of Cardiology, Diskapi Yildirim Beyazit was applied to assess the distribution of continuous variables. Education and Research Hospital, Ankara, Turkey The Student’s t-test was used to compare normally distributed G CAGIRCI, MD continuous variables and the Mann–Whitney U-test for variables 192 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

without normal distribution. Pearson correlation was used for population, both QTc and QTcd were significantly decreased the correlation analysis. The mean inter-observer difference in the first and third months compared to baseline values (Table and inter-class coefficients were used to evaluate inter-observer 2). QTc and QTcd values correlated significantly between the variability. A two-tailed p-value < 0.05 was considered two observers. Inter-class coefficients and mean inter-observer statistically significant. differences at baseline and in the first and third months are presented in Table 3. Results In male subjects, both QTc and QTcd were significantly Baseline characteristics of patients including demographic, decreased in the first and third months compared to baseline clinical and laboratory parameters are outlined in Table 1. The values. In female subjects, both QTc and QTcd were significantly study population was mostly elderly patients and 69.4% were decreased in the third month compared to baseline values. over 65 years. Approximately two-thirds of the patients were However, in the first month of follow up, no significant decrease white males. All patients were on a drug affecting the renin– was detected in QTc or QTcd compared to baseline values. In angiotensin–aldosteron system. NYHA functional class II was addition, both QTc and QTcd were significantly decreased in the most common category in the population. No discontinuation the third month compared to the first month (data not presented). of the study drug was observed during the study period. There was a significant positive correlation between age and At the end of the study, vital parameters of patients including baseline QTcd in the whole study population (β = 0.567, p < systolic (110 ± 15 vs 100 ± 15 mmHg, p < 0.001) and diastolic 0.001). There was no significant difference between patients (72 ± 10 vs 68 ± 10 mmHg, p < 0.001) blood pressure and heart in NHYA class I–II and III–IV, according to QTc and QTcd at rates (84 ± 11 vs 74 ± 11 bpm, p < 0.001) were significantly baseline and in the first and third months (p > 0.05 for all). decreased, as expected. However, there were no significant changes in the left ventricular and NYHA Discussion functional class of the patients between the baseline and follow- We found that nebivolol therapy significantly improved both up values at one and three months (32.3 ± 5.0 vs 32.4 ± 4.9%, QTc and QTcd parameters in patients with systolic heart failure, p = 0.327 and 2.3 ± 0.6 vs 2.2 ± 0.5%, p = 0.103), respectively. which has not been reported in the literature before. Sympathetic QTc and QTcd were measured at baseline and in the first and tone, excitation–contraction coupling and myocardial fibrosis third month of the study. The measurements and calculations may be the reasons for impaired QT dynamics in heart failure. were performed by two independent observers blinded to the Both QTc and QTcd are the indicators of the heterogeneity of study protocol. The mean values of the QT dynamics from myocardial repolarisation and electrical instability.8,11 The action two independent observers are given in Table 2. In the whole potential is prolonged and repolarisation is delayed in heart failure patients. TABLE 1. BASELINE DEMOGRAPHIC, CLINICAL The QT interval on the surface ECG is a readily measurable AND LABORATORY CHARACTERISTICS OF reflection of cardiac repolarisation. The QT interval is an THE STUDY POPULATION index of ventricular repolarisation that is directly influenced Characteristic Value (n = 72) by myocardial health and autonomic nervous system activity. Age (years) 71.0 ± 10.4 Patients with heart failure and prolonged action potential Male (%) 45 (62.5) durations have abnormalities of the QT interval. In a small group Diabetes mellitus (%) 6 (8.3) Smoking (%) 13 (18.1) Hypertension (%) 23 (31.9) TABLE 2. QT DYNAMICS IN THE WHOLE POPULATION CCB use (%) 5 (6.9) Characteristic Baseline First month Third month ACE inhibitor use (%) 62 (86.1) QTc (ms) 455.3 ± 26.7 441.2 ± 25.7* 436.0 ± 28.7* ARB use (%) 10 (13.9) QTcd (ms) 65.6 ± 5.3 58.2 ± 5.6** 56.0 ± 6.2* Statin use (%) 10 (13.9) The mean values of two independent observers are presented. *p < 0.001 for baseline to first month and baseline to third month. Aldosterone antagonist use (%) 21 (29.2) **p = 0.001 for baseline to first month. Diuretic use (%) 68 (94.4) Digoxin use (%) 16 (22.2) NYHA class TABLE 3. INTER-OBSERVER DIFFERENCES AND INTER-CLASS COEFFICIENTS BETWEEN TWO I (%) 4 (5.6) OBSERVERS FOR QT DYNAMICS II (%) 47 (65.3) Mean III (%) 19 (26.4) inter- Inter- IV (%) 2 (2.8) observer class LVEF (%) 32.3 ± 5.0 difference coeffi- Characteristics (ms) Range (ms) cient 95% CI p-value Creatinine (mg/dl) 0.96 ± 0.27 QTc baseline –2.8 –15 to +20 0.986 0.978–0.991 < 0.001 Na (mmol/l) 136.3 ± 4.2 QTcd baseline –0.5 –2.5 to +3.5 0.999 0.999–1.000 < 0.001 K (mmol/l) 4.2 ± 0.6 QTc month 1 –3.0 –20 to +20 0.995 0.993–0.997 < 0.001 Ca (mmol/l) 2.37 ± 0.12 QTcd month 1 –0.5 –2.5 to +3.5 0.999 0.999–1.000 < 0.001 ACE: angiotensin converting enzyme; ARB: angiotensin receptor blocker; CCB: calcium channel blocker; LVEF: left ventricular ejection fraction; QTc month 3 –2.6 –15 to +20 0.993 0.988–0.995 < 0.001 NYHA: New York Heart Association. QTcd month 3 –0.5 –3.0 to +3.5 0.999 0.999–1.000 < 0.001 AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 193

of heart failure patients, QT dispersion has been shown to be patients in our study can partly be attributed to improvement in a marker of electrical instability and increased risk of sudden their QT dynamics. death.12 In patients with known repolarisation abnormalities, the QTcd has been demonstrated to be a better prognostic indicator References 13 of arrhythmic risk than the QT itself. Unstable ventricular 1. The MERIT-HF study group. Effects of metoprolol CR/XL in chronic repolarisation may contribute to the development of ventricular heart failure: metoprolol CR/XL randomised intervention trial in tachycardia and fibrillation. congestive heart failure (MERIT-HF). Lancet 1999; 353: 2001–2007. It is well known that β-blockers are the only anti-arrhythmic 2. The CIBIS-II investigators and committees. The Cardiac Insufficiency drug class effectively reducing mortality and arrhythmic sudden Bisoprolol Study II (CIBIS II): a randomised trial. Lancet 1999; 353: death in patients with heart failure.14 These drugs have been also 9–13. 3. Packer M, Coats AJ, Fowler MB, Katus HA, Krum H, Mohacsi P, et al. shown to improve QT dynamics.15-17 for the Carvedilol Prospective Randomized Cumulative Survival study It has been demonstrated that nebivolol therapy reduced group. Effect of carvedilol on survival in severe chronic heart failure. N the composite risk of all-cause mortality or cardiovascular Engl J Med 2001; 344: 1651–1658. hospital admission compared with placebo in patients with 4. Flather MD, Shibata MC, Coats AJ, van Veldhuisen DJ, Parkhomenko heart failure.4 Nebivolol therapy as an antihypertensive drug A, Borbola J, et al.; SENIORS investigators. Randomized trial to deter- has been extensively studied and approved. Its effect on QT mine the effect of nebivolol on mortality and cardiovascular hospital dynamics in hypertensive patients has shown that nebivolol admission in elderly patients with heart failure (SENIORS). Eur Heart J 2005; 26: 215–225. significantly reduced QTcd in hypertensive subjects without 5. Zanchetti A. Clinical pharmacodynamics of nebivolol: new evidence of 7 affecting left ventricular mass. In our study, no statistically nitric-oxide mediated vasodilating activity and peculiar haemodynamic significant difference was found between hypertensive and properties in hypertensive patients. Blood Pressure 2004; 13: 18–33. non-hypertensive patients according to QTc and QTcd at 6. Kuroedov A, Cosentino F, Luscher TF. Pharmacological mechanisms of baseline and in the first and third months of follow up (p > 0.05 clinically favorable properties of a selective beta1-adrenoceptor antago- for all). nist, nebivolol. Cardiovasc Drug Rev 2004; 22: 155–168. Statin therapy has also been found to be associated with 7. Galetta F, Franzoni F, Magagna A, Femia FR, Pentimone F, Santoro G, 18 Carpi A. Effect of nebivolol on QT dispersion in hypertensive patients improved QT dynamics in heart failure patients. In our study with left ventricular hypertrophy. Biomed Pharmacother 2005; 59: there was no significant difference between patients on statin 15–19. therapy and those without statins, according to QTc and QTcd 8. Vrtovec B, Delgado R, Zewail A, Thomas CD, Richartz BM, parameters at baseline (p > 0.05 for all). We found that in female Radovancevic B. Prolonged QTc interval and high B-type natriuretic subjects, neither QT dynamic significantly decreased in the peptide levels together predict mortality in patients with advanced heart first month compared to baseline. The reason for this could be failure. Circulation 2003; 107: 1764–1769. the small number of females, which did not reach statistical 9. Glancy JM, Weston PJ, Bhullar HK, et al. Reproducibility and automat- ic measurement of QT dispersion. Eur Heart J 1996; 17: 1035–1039. significance. 10. Murray A, McLaughlin NB, Campbell RWF. Measuring QT dispersion: Although ACE inhibitor therapy has been shown to decrease man versus machine. Heart 1997; 77: 539–542. QTd,19 in our study no significant difference was found between 11. Haigney MC, Zareba W, Gentlesk PJ, Gentlesk PJ, Goldstein RE, patients on ACE inhibitors and those not on them at baseline (p Illovsky M, et al., Multicenter Automatic Defibrillator Implantation > 0.05). Also, no significant difference was found between those Trial II investigators. QT interval variability and spontaneous ventricu- on digoxin and those not on digoxin at baseline (p > 0.05). lar tachycardia or fibrillation in the Multicenter Automatic Implantation Trial (MADIT) II patients. J Am Coll Cardiol 2004; 44: 1481–1487. Neither QTc nor QTcd were different among NHYA functional 12. Barr CS, Naas A, Freeman M, Lang CC, Struthers AD. QT dispersion classes at baseline and in the first and third months of follow up. and sudden unexpected death in chronic heart failure. Lancet 1994; This demonstrates that QT dynamics were impaired regardless of 343: 327–329. NYHA functional class. Since there were no significant changes 13. Day CP, McComb JM, Matthews J, Campbell RWF. Reduction in QT in the left ventricular ejection fraction and NYHA functional dispersion by sotalol following myocardial infarction. Eur Heart J class, this indicates that the QT changes were the result of the 1991; 12: 423–427. drug’s effect on the heart rate and blood pressure and not due to 14. Singh BN. CIBIS, MERIT-HF and COPERNICUS trial outcomes: do they complete the chapter on beta-adrenergic blockers as antiarrhyth- improvement in the heart failure. mic and antifibrillatory drugs? J Cardiovasc Pharmacol Ther 2001; A limitation of this study was that this was a case–control 6: 107–110. observational study without primary end-points such as 15. Singh JP, Musialek P, Sleight P, Davey P, Marinho M, Hart G. Effect of hospitalisation for heart failure, cardiac and total mortality. atenolol or metoprolol on waking hour dynamics of the QT interval in Therefore, randomised blinded studies with longer follow-up myocardial infarction. Am J Cardiol 1998; 81: 924–926. periods are needed. In addition, we only studied heart failure 16. Hintze U, Mickley H, Moller M. Effects of beta-blockers on the relation patients with systolic dysfunction and normal . between QT interval and heart rate in survivors of acute myocardial infarction. Ann Noninvas Electrocardiol 1998; 3: 319–326. Similar studies should be performed in heart failure patients with 17. Bonnar CE, Davie AP, Caruana L, Fenn L, Ogston SA, McMurray JJ, preserved systolic function and coronary heart disease. Struthers AD. QT dispersion in patients with chronic heart failure: beta blockers are associated with a reduction in QT dispersion. Heart 1999; 81: 297–302. Conclusion 18. Xie RQ, Cui W, Liu F, Yang C, Pei WN, Lu JC. Statin therapy shortens QTc, QTcd, and improves cardiac function in patients with chronic Similar to other older-generation β-blockers, nebivolol is a heart failure. Int J Cardiol 2008 Nov 28. [Epub ahead of print]. selective β1-blocker with vasodilator properties that have the 19. Barr CS, Naas AA, Fenwick M, et al. Enalapril reduces QT dispersion potential to improve QT dynamics in patients with heart failure. in patients with mild congestive heart failure secondary to coronary Therefore the good clinical results obtained in the heart failure artery disease. Am J Cardiol 1997; 79: 328–333. 194 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Can cardiac surgery be performed safely on patients with haematological malignancies A GULER, MA SAHIN, F CINGOZ, E OZAL, U DEMIRKILIC, M ARSLAN

Abstract Several procedures are performed nowadays on patients with Introduction: Surgical strategy in patients with haematologi- co-morbidity factors, with acceptable morbidity and mortality 1 cal malignancies must be planned and carried out with the rates. specific aim of decreasing postoperative complications. The Haematological malignancies are diagnosed in all age aim of this study was to present our experience on patients groups but the chronic forms are predominantly seen in elderly 2 previously diagnosed with haematological malignancies who populations. Great strides have been made to improve the subsequently underwent cardiac surgery. We include data to quality of life of these patients and many clinicians are now assist other surgeons predict factors affecting postoperative focusing on finding solutions to other symptoms these patients morbidity and mortality in this group of patients. may have. In an era when atherosclerotic heart disease shows an Methods: Fifteen patients diagnosed with haematological increasing prevalence, cardiac surgeons are encountering this malignancies who had cardiac surgery were retrospectively population more frequently. analysed. Eight patients had chronic lymphocytic leukaemia, The operative risk of patients with malignant haematological six had non-Hodgkin’s lymphoma and the rest had chronic disorders is increased, as this may include coagulation defects, myelocytic leukaemia. Coronary artery bypass graft surgery changes in blood viscosity, immune suppression and bone marrow 3 was performed on all of them. insufficiency. When surgically treating these patients, one must Results: There were no hospital mortalities. The average be concerned about postoperative infection, haemorrhage and follow-up period was 35 ± 11 (23–56) months. Three patients leukaemic transformation. Surgical trauma and cardiopulmonary required early postoperative re-operation because of exces- bypass (CPB), because of their immune-depressing effects, have sive bleeding. No mortalities were seen in the early post- the potential risk of increasing the haematological problems, 4 operative period. There were five (33%) deaths during the leading to fatal or morbid complications. late follow-up period. Three patients were lost due to intrac- There are few reports on how to deal with these patients, and ranial bleeding (confirmed by autopsy) in the 16th, 23rd also little knowledge on their progress after cardiac surgery. Due and 38th months after surgery. The remaining two patients to these concerns, the aim of our study was to detail our clinical had sudden death in the eighth and 55th months from non- experience and data on the postoperative period. detectable causes. Conclusion: Cardiac surgery can be performed with accept- Methods able early postoperative outcomes in patients with haemato- We retrospectively reviewed hospital records of 15 patients with logical malignancies. Intracranial bleeding is an important haematological malignancies who underwent cardiac surgery factor contributing to late mortality and patient selection at the Cardiovascular Surgery Department of our institution and risk stratification are crucial to improving surgical between 2003 and 2009. Eight patients suffered from chronic benefits. lymphocytic leukemia (CLL), six had non-Hodgkin’s lymphoma Keywords: haematological malignancy, cardiac surgery, (NHL), and the rest had chronic myelocytic leukaemia. Coronary intracranial bleeding artery bypass graft (CABG) surgery was performed on all patients. Submitted 4/9/10, accepted 6/9/11 The diagnosis of haematological malignancy was assigned Cardiovasc J Afr 2012; 23: 194–196 www.cvja.co.za based on the international ICD-10 code. Patients received DOI: 10.5830/CVJA-2011-053 routine intravenous antibiotic prophylaxis (cephazolin Na 1 g) for three days, beginning the night before surgery. CPB was Increased surgical experience and technological advances in performed after the standard heparinisation (ACT > 400 s). All cardiac surgery have encouraged surgeons to perform complex the patients were given antegrade cold crystalloid cardioplegia cardiac operations in patients with unrelated complications. with moderate hypothermia, and topical cold slush solution was used. The left internal mammary artery (LIMA) and saphenous vein grafts were prepared for CABG surgery. Distal anastomoses were performed with 7/0 polypropylene sutures and a continuous Department of Cardiovascular Surgery, Gulhane Military suture technique. Proximal anastomoses were performed with Medical Academy, Etlik, Ankara, Turkey 6/0 polypropylene sutures and a continuous suture technique A GULER, MD with a side clamp. Salicylic acid 100 mg was started on the first MA SAHIN, MD, [email protected] F CINGOZ, MD postoperative day in all patients. E OZAL, MD Patients’ hospital charts, demographics, peri-operative data U DEMIRKILIC, MD and complications were reviewed. Follow-up data were obtained M ARSLAN, MD by review of subsequent hospital admissions and telephone AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 195

interviews. The average follow-up period was 35 ± 11 (23–56) A 27-year-old male patient diagnosed with CLL was months. hospitalised one month after the first operation due to In statistical analysis, values were expressed as mean ± mediastinitis. This patient was treated with antibiotics according standard deviation for continuous variables, and number and to bacterial culture and he then underwent sternal dehiscence percentage for dichotomous variables. In comparisons, the revision surgery. The hospital course was uneventful after Student’s t-test was used for continuous variables and the Mann- the second operation and he soon returned to work. Clinical Whitney U-test for dichotomous variables. Comparison of and peri-operative variables of patients with CLL and NHL gender was done using the Chi-square test. A p-value below 0.05 were similar (Tables 1, 2). No mortality was seen in the early was set as significant. postoperative period. The three-year survival rate was 80%. There were five (33%) late deaths during the follow-up period. All deaths were of a Results non-cardiac nature. Three patients were lost due to intracranial There were a total of 15 patients, 11 were male and four female. bleeding in the 16th, 23rd and 38th months after surgery. The Their pre-operative data are summarised in Table 1. Mean age other two patients had sudden death in the eighth and 55th was 65 ± 14 (range 27–76) years. The mean time interval from months and the reason for death could not be determined. One the diagnosis of the haematological malignancy to the CABG of these patients was brought to the emergency department with surgery was 4.6 years (7 months to 9 years). All were in remission cardiopulmonary arrest. Resuscitation was performed but the and under supervision of the Haematology Department. patient died and the family refused an autopsy. An average of 2.75 ± 1.2 grafts was placed in the 15 patients undergoing isolated CABG. LIMA grafts were used in 14 patients. All were operated on with an arrested heart under CPB. Discussion Two patients required intra-aortic balloon pump during the peri- Haematological malignancies, particularly the lymphocytic operative period. Haematology reports reflected the expected types, affect mainly elderly patients. Survival of these patients ranges in all patients during the postoperative period. can range from one year to several decades.2,5 Over the past few In patients with CLL and NHL, the average length of stay in years, treatment options have continued to improve survival rates. ICU was 1.2 ± 0.3 and 1.5 ± 0.5 days, respectively. The average Cardiac surgical experience in patients with haematological of packed red blood cells required was 2.1 ± 0.6 and 1.5 ± malignancies is limited and detailed investigation is mandatory 0.6 units, respectively, and the average of fresh frozen plasma in decision making.2 required was 1.3 ± 0.5 and 1.5 ± 0.6 units, respectively. There It is obvious that CPB affects all systems, including the was no apparent difference in the postoperative course and mean haematopoietic system. CPB, which aggravates cell damage, postoperative stay between the two groups (Table 2). also has immune-depressant properties, resulting in an Four postoperative complications occurred in four patients increased incidence of infection.6,7 Furthermore, haematological (26%). Three of them required early re-operation because of malignancies may lead to antibody deficiency, leucopenia or bleeding. They were respectively, 55-, 63- and 69-year-old male impaired T-cell function. Previous reports demonstrate infection patients, diagnosed with NHL. More bleeding was seen in the as the primary cause of morbidity.2,3,8 Some studies show patients with NHL than in those with CLL, and the difference morbidity rates of between 23 and 57%. Samuels et al. reported was statistically significant (p < 0.05). The hospital stay was the incidence of infectious complications as 42%, underlining the uneventful after the second operation. major role they play in the hospital stay.2,5 Some investigations emphasise the use of additional intravenous immunoglobulin, as broad-spectrum antibiotic prophylaxis was found to be TABLE 1. CLINICAL DATA insufficient to prevent or control infection in these patients.2 Non- Hodkgin’s Diagnosis CLL lymphoma p-value CML TABLE 2. PERI-OPERATIVE DATA Number 8 6 1 Gender (male/female) 6/2 4/2 0.733 0.733** M Non- Hodkgin’s Age (years) 66 ± 16 60 ± 12 0.128* 78 Diagnosis CLL lymphoma p-value CML Body mass index 24 ± 4.2 26.5 ± 4.9 0.096* 22 CPB time (min) 83 ± 21 88 ± 16 0.215* 95 Pre-operative EF (%) 49 ± 13 52 ± 18 0.226* 44 Cross-clamp time (min) 64 ± 19 69 ± 24 0.083* 72 Pre-operative haemoglobin (g/dl) 13.3 ± 3.6 12.8 ± 2.9 0.258* 15.4 Packed red blood cells (units) 2.1 ± 0.6 2.3 ± 0.8 0.573** 3 Post-operative haemoglobin (g/dl) 11.9 ± 7.2 10.8 ± 8.6 0.086* 12.3 Fresh frozen plasma (units) 1.3 ± 0.5 1.5 ± 0.6 0.755** 1 Pre-operative platelets (× 109 /dl) 206 ± 74 254 ± 89 0.068* 139 Post-operative IABP 1 1 0.950** – Post-operative platelets (× 109 /dl) 174 ± 51 152 ± 73 0.070* 128 ICU stay (days) 1.2 ± 0.3 1.5 ± 0.5 0.491* 1 Pre-operative prothrombin time (s) 12.8 ± 4.1 12.5 ± 2.8 0.825* 13.3 Mortality early/late –/3 –/2 0.950** – Post-operative prothrombin time (s) 13.2 ± 7.2 12.9 ± 9.4 0.652* 13.1 Bleeding complications – 3 0.030** Pre-operative WBC (× 109 /dl) 33.2 ± 9.1 38.5 ± 10.8 0.075* 20.0 Infection 1 – 0.755** Post-operative WBC (× 109 /dl) 36.1 ± 1.4 34.8 ± 5.3 0.348* 17 CLL: chronic lymphocytic leukaemia, CML: chronic myelocytic leukae- CLL: chronic lymphocytic leukemia, CML: chronic myelocytic leukemia, EF: mia, CPB: cardiopulmonary bypass, IABP: intra-aortic balloon pump, ejection fraction, WBC: white blood cells, Postoperative: 14th day. *Student’s ICU: intensive care unit, *Student’s t-test, **Mann-Whitney U-test, t-test, **Chi-square test; p-value represents the comparison between CLL and p-value represents the comparison between CLL and non-Hodkgin’s non-Hodkgin’s lymphoma patients. lymphoma patients. 196 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

According to the advice of our infections committee, we We surmise that the small number of patients seriously used only cephazolin Na for prophylactic antibiotic therapy. affected the late-period outcomes and no predictive factors Mediastinitis was observed in one patient and it was the only for identifying such high-risk patients were found. Although infection-related complication encountered in our cases. This intracranial bleeding was a significant mortality factor, was an acceptable infection rate and comparable with that of prothrombin time, bleeding time and thrombocyte level were open-heart surgery in the general population. in the normal range in all patients. Aspirin 100 mg was the White blood cell count generally increases after open-heart anti-aggregant therapy and it may have been responsible for surgery, and CPB may stimulate a leucomoid reaction, which can intracranial bleeding. However, there is no evidence and future lead to a relapse in an otherwise quiescent illness.6,7 In our study, guidelines should describe a detailed approach and treatment. white blood cell count increased between acceptable ranges, as It is reasonable to use less-invasive techniques (percutaneous seen in Table 1. The general opinion on cardiac surgery is that coronary interventions) in high-risk patients when they meet it does not exacerbate haematological malignancies in low-risk the operative indications. However, these procedures require patients or those with low-grade disease (in the remission heavy post-intervention anti-agregant therapy, and the risk for period). However patients with an intermediate to high risk or intracranial bleeding would therefore remain high. grade of disease may show progression and this may be the cause Predictive scores such as the Euroscore or the Society of of late death.5 Our patients were all in complete remission and Thoracic Surgeons’ score have no predictive values for mortality there was no leucomoid reaction or relapse of the disease in the and morbidity in patients with haematological malignancies follow-up period, as confirmed by the Haematology Department. undergoing cardiac surgery. Therefore, future studies giving Bleeding is another potential complication in this group of accurate rates of post-operative morbidity or mortality would patients. Fecher et al.5 reported a 16.6% rate of bleeding and it be of great use to provide a general approach for these patients. was the main postoperative complication they observed. They had only one mediastinal bleeding that led to re-operation, Conclusion and two cases of gastrointestinal and one of femoral artery Cardiac surgery can be undergone with acceptable early haemorrhage. They were not group related. postoperative outcomes in patients with haematological In our series, three patients (20%) had excessive bleeding in malignancies. The expectations of the patient and surgeon the early postoperative period, showing a similar rate to that of should be appropriately discussed and the medical team should Fecher et al.5 However, all were of mediastinal origin and led to be focused on potential complications such as bleeding, infection re-operation and all were patients with NHL. One of the cases and prolonged hospital stay. Long-term multidisciplinary follow of haemorrhage was surgery related while a general leakage up and adequate medical treatment are essential in prolonging was observed in the other two cases. We did not detect any survival. significantly low platelet count or elevated INR peri-operatively. Bleeding was the only statistically significant difference References observed between the patients with CLL and NHL. We did not 1. Avery GJ, Ley SJ, Hill JD, et al. Cardiac surgery in the octogenar- find any reports in the literature regarding NHL and tendency to ian: evaluation of risk, cost, and outcome. Ann Thorac Surg 2001; 71: haemorrhage and we felt that statistical analysis was not feasible 591–596. for some factors because of the small size of our study group. 2. Samuels LE, Kaufman MS, Morris RJ, et al. Open heart surgery in However, it may be valuable in future clinical decisions. patients with chronic lymphocytic leukemia. Leukemia Res 1999; 23: This article has some limitations. It was retrospective and 71–75. 3. Christiansen S, Schmid C, Löher A, et al. Impact of malignant hemato- lacked a control group of standard patients with CABG. The logical disorders on cardiac surgery. Cardiovasc Surg 2000; 8: 149–152. number of patients and types of malignancies were different 4. Haiston P, Manos JP, Graber CD, et al. Depression of immuno- among different institutions, therefore it was difficult to find a logic surveillance by pump oxygenator perfusion. J Surg Res 1969; 9: large series of patients to determine the morbidity and mortality 587–593. rates in a homogeneous group. 5. Fecher AM, Birdas TJ, Haybron D, et al. Cardiac operations in patients Incomplete follow-up data was another shortcoming. In with hematologic malignancies. Eur J Cardiothorac Surg 2004; 25: previous reports, 30-day mortality rates ranged from zero to 537–540. 6. Drury N, Ali A, Mussa S, et al. Acute leukaemoid reaction following 2,5,9 17%. In our series, there was no early mortality, but late cardiac surgery. J Cardiothorac Surg 2007; 9: 2–3. mortality was somewhat higher (33%). Three patients died 7. Ghosh P, Carroll I, Kanhere A, et al. Cardiac operations in patients with because of intracranial bleeding and two others due to unknown low-grade small lymphocytic malignancies. J Thorac Cardiovasc Surg reasons. 1999; 118: 1033–1037. Samuels et al. pointed out that these patients’ long-term 8. Finck SJ, Cockerill KJ, Jeter JE, et al. Coronary artery bypass grafting outcomes are variable and non-cardiac related.2 The high late- in patients with chronic lymphocytic leukemia. Ann Thorac Surg 1993; 55: 1192–1196. mortality rate of our patients may be partially related to our 9. Sommer SP, Lange V, Yildirim C, Schimmer C, Aleksic I, Wagner C, limited experience in this specific group of patients. Appropriate et al. Cardiac surgery and hematologic malignancies: a retrospective long-term anti-aggregant therapy should be designed in single-centeranalysis of 56 consecutive patients. Eur J Cardiothorac collaboration with haematology departments. Surg 2010; Dec 9 (Epub ahead of print). AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 197

The impact of cardiovascular risk factors on the site and extent of coronary artery disease AF ZAND PARSA, H ZIAI, L HAGHIGHI

Abstract a heterogeneous disease in terms of severity, extent and site of Background: In patients with coronary artery disease (CAD), involvement, these are the most important predictors of outcome the site and extent of coronary artery involvement in terms of of patients with coronary artery disease. The main question is proximal versus distal stenosis and multi- versus single-vessel whether or not these heterogeneities have any relationship with disease have a crucial effect on patients’ outcome. This study cardiovascular risk factors, and if so, which is responsible for was designed to evaluate the relationship between cardiovas- which kind of lesion. cular risk factors and the site and extent of coronary artery Although in some studies a relationship between diabetes 1-5 4-6 involvement. mellitus (DM) and hyperlipidaemia, and severity of CAD Methods: In this study of patients who had undergone coro- has been reported, these studies were focused on the severity of nary angiography in our hospital, 125 with proximal lesions lesions according to the scoring system used and not to the site of were enrolled as the case group (group 1) and an equal age- lesion in terms of proximal versus distal stenosis. In our study we and gender-matched number of patients with non-proximal tried to evaluate the impact of major cardiovascular risk factors lesions were selected as the control group (group 2). The two such as DM, hypercholesterolaemia, hypertriglyceridaemia, groups were compared based on the presence or absence of hypertension (HTN) and cigarette smoking on the site and extent diabetes mellitus (DM), hypercholesterolaemia, hypertrigly- of coronary artery involvement in terms of proximal versus distal ceridaemia, hypertension (HTN) and cigarette smoking. and multi- versus single-vessel disease. Results: The frequency of DM was 33.6 and 10.4% in the case and control groups, respectively, which was statistically Methods significant (p < 0.0001). However, the frequency of hyper- This study was a prospective case–control study that included cholesterolaemia in the case and control groups was 30.4 patients who had undergone coronary angiography in our and 29.6% (p = 0.89), respectively; for hypertriglyceridaemia hospital. Patients with normal coronary arteries were excluded it was 19.2 and 16.8% (p = 0.062), respectively; for HTN it from the study. The case group included 125 patients with was 33.6 and 28.8% (p = 0.4), respectively; and for cigarette significant proximal coronary artery stenosis (≥ 50% luminal smoking it was 28.8 and 39.2% (p = 0.08), respectively, which narrowing) and the control group included 125 patients with were not statistically significant. Diabetic patients compared significant non-proximal stenosis, and matched with case group to non-diabetics had more multi-vessel disease (89.1 vs 61%, regarding gender and age. p < 0.0001, respectively), which was statistically significant. Coronary angiography of all patients was re-evaluated by There was no relationship between hypercholesterolaemia, an expert cardiologist and in the case of controversy, by two hypertriglyceridaemia, HTN and cigarette smoking and experts who were not aware of the patients’ risk factors or other extent (multi-vessel involvement) of CAD (p = NS). clinical conditions. The two groups were compared for major Conclusion: Proximal and multi-vessel involvement of the cardiovascular risk factors such as hyperlipidaemia, HTN, DM coronary arteries in patients with CAD was related to a and cigarette smoking. history of DM but not of hypercholesterolaemia, HTN, ciga- rette smoking and hypertiglyceridaemia. Statistical analysis Keywords: coronary artery disease, coronary angiography, SPSS version 11.5 was used for analysing the data. The Student’s proximal versus distal stenosis, cardiovascular risk factors t-student test and Chi-square test were used for numerical and Submitted 15/10/10, accepted 6/9/11 continuous variables, respectively. For evaluating data, odds ratio (OR) with 95% confidence interval (CI) were used and p > Cardiovasc J Afr 2012; 23: 197–199 www.cvja.co.za 0.05 was considered significant. DOI: 10.5830/CVJA-2011-052 Results Coronary artery disease (CAD) is one of the leading causes of In the case group 87 patients (69.6%) and in the control group 95 morbidity and mortality worldwide. As atherosclerotic CAD is patients (76%) were male. The mean age in the case and control groups was 59.6 ± 10.8 and 58.8 ± 10.9 years, respectively. Division of Cardiology, Imam Khomeini Medical Center, Regarding gender and age, there was no significant difference Tehran University of Medical Sciences, Tehran, Iran between the two groups (p = 0.556 and p = 0.256, respectively). AF ZAND PARSA, MD, [email protected] Clinical and demographic characteristics of patients are presented H ZIAI, MD in Table 1. Shahid Akbari Hospital, Iran University of Medical Sciences, Diabetes mellitus was more prevalent in the case group than Tehran, Iran the control (33.6 vs 10.4%, p < 0.0001, OR = 4.36; 95% CI: L HAGHIGHI, MD 2.2–8.6, respectively) and it was statistically significant. There 198 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

was no difference between the two groups regarding HTN (OR burden and extension score have been reported.1,2,4,5,7 However = 1.25; 95% CI: 0.732–2.14, p = 0.41), hypercholesterolaemia Pajumen et al. in their study did not find any relationship (OR = 1.04; 95% CI: 0.60–1.78, p = 0.86), hypertriglyceridaemia between diabetes and extent of CAD compared to non-diabetics.8 (OR = 1.18; 95% CI: 0.62–2.25, p = 0.62) and cigarette smoking Uddin et al.1 and Synkija et al.2 studied site of coronary artery (OR = 0.63; 95% CI: 0.37–1.06, p = 0.08). The frequency of involvement (proximal versus distal involvement) in diabetic cardiovascular risk factors in both groups is presented in Fig. 1. patients versus non-diabetics. Although the trend was towards Regarding the extent of CAD, multi-vessel disease was proximal involvement in diabetic patients, it was not statistically more frequent than single-vessel disease in the diabetics significant (p > 0.05). than the non-diabetics (89.1 and 10.9% vs 61 and 39%, Synkija reported more multi-vessel disease in hypertensive respectively, p < 0.0001). In hypercholestrolaemic compared to patients than in non-hypertensives (p < 0.0003).2 Hong et al.9 non-hypercholestrolaemic patients, the trend was in favour of also reported more multi-vessel than single-vessel disease in more multi-vessel disease but the difference was not statistically hypertensive patients (p < 0.01). However, Sposito et al.10 did significant (76 and 24% vs 63.4 and 36.6%, respectively, p = not find any relationship between hypertension and extent of 0.052). Hypertriglyceridaemia, HTN and cigarette smoking had coronary artery involvement (multi-vessel disease), which was no impact on the extent of coronary artery involvement in terms similar to what we found in our study. of multi-vessel versus single-vessel disease. The relationship of Synkija et al.2 and Sposito et al.10 found no relationship these cardiovascular risk factors to the extent of coronary artery between hypercholesterolaemia and extent of CAD. Syvanne et involvement is presented in Table 2. al.,5 Kosuge et al.6 and Hong et al.9 reported a relationship between low-density lipoprotein cholesterol and total cholesterol:high- density lipoprotein cholesterol ratio and extent of coronary Discussion artery involvement, based on ABI or index of extent of artery Although the relationship between cardiovascular risk factors involvement (p = 0.027, p = 0.01 and p < 0.01, respectively), but and CAD has held investigators’ attention for a long time, there not to the site of coronary artery involvement (proximal versus are no clear data regarding the impact of risk factors on the site, distal). extent and complexity of coronary artery involvement in terms In the study of Sposito et al.,10 post-menopausal women with of proximal or distal and diffuse or segmental involvement. hypertrygliceridaemia had more extensive CAD compared to Most studies that have been conducted in this regard were based those without hypertrygliceridaemia (p = 0.0013). Wilson et al.11 on index of atheroma burden and extension score of CAD in reported more extensive coronary artery involvement in smokers patients with DM. In the majority of these studies, a strong compared to non-smokers based on score of extent of artery relationship between DM and increased index of atheroma involvement (p < 0.005), which was opposite to what we found in our study. 60 None of these studies assessed relationship between 49 39.2% hypertriglyceridaemia or cigarette smoking and site of coronary 50 42 42 artery stenosis (proximal versus distal stenosis). In our study 33.6% 33.6% 38 37 36 30.4% 36 40 28.8% 29.6% 28.8% the trend was towards distal stenosis in smokers compared to non-smokers but it was not statistically significant (p = 0.08). 30 24 19.2% 21 Also we did not find a relationship between hypertriglyceridaemia 16.8% 20 13 and hypertension and the site of coronary involvement (p = NS). 10.4% Uddin et al.1 and Synkija et al.2 were the only investigators 10 who considered relative frequency of proximal versus distal involvement in their patients. According to their findings, 0 DM HTN HCL HTG C/S DM = diabetes mellitus HTG = hypertriglyceridaemia HTN = hypertension C/S = cigarette smoking TABLE 2. RELATIONSHIP BETWEEN CARDIOVASCULAR HCL = hypercholesterolaemia RISK FACTORS AND EXTENT OF CORONARY case group control group ARTERY DISEASE Fig. 1. Frequency of cardiac risk factors in the two groups. Extent of CAD Single-vessel Multi-vessel disease disease TABLE 1. CLINICAL AND DEMOGRAPHIC Subgroups n (%) n (%) p-value CHARACTERISTICS OF PATIENTS Diabetic 6 (10.9) 49 (89.1) Case group Control group < 0.0001 n = 125 n = 125 Non-diabetic 76 (39) 119 (61) Demographic characteristics (100%) (100%) p-value Hypertensive 20 (25.6) 58 (74.4) NS Males, n (%) 87 (69.6) 95 (76) NS Non-hypertensive 62 (36) 110 (64) Age mean ± SD (years) 59.6 ± 10.8 58.8 ± 10.9 NS Hypercholesteraemic 18 (24) 57 (76) 0.052 Diabetes mellitus, n (%) 42 (33.6) 13 (10.4) < 0.0001 Non-hypercholestraemic 64 (36.6) 111 (63.4) Hypertension, n (%) 42 (33.6) 36 (28.8) NS Hypertriglyceridaemic 12 (26.7) 33 (73.3) NS Hypercholesterolaemia, n (%) 38 (30.4) 37 (29.6) NS Non-hypertriglyceridaemic 70 (34.1) 135 (65.9) Hypertriglyceridaemia, n (%) 24 (19.2) 21 (16.8) NS Cigarette smoker 31 (36.5) 54 (63.5) NS Cigarette smoking, n (%) 36 (28.8) 49 (39.2) 0.08 Non-cigarette smoker 51 (30.9) 114 (69.1) AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 199

proximal involvement was more frequent in diabetic patients, 4. Fallow GD, Singhj. The prevalence, type and severity of cardiovascular but it was not statistically significant (p > 0.05). In our study, not disease in diabetics and non diabetic patients: amathced – paired retro- only frequency of proximal stenosis but also frequency of multi- spective analysis using CAG as the diagnosis tools. Mol cell Biochem 2004; 261(1–2): 263–269. vessel involvement was significantly higher in diabetics than in 5. Syvanne M, Pajunen P, Kahri J. Determination of the severity and extent non-diabetics (p < 0.0001). of CAD in patients with type 2 diabetes and non-diabetes subjects. Coron Arter Dis 2001; 12(2): 99–106. 6. Kosuge M, Kimmurak, Ishikawa T, et al. Different clinical and coronary Conclusion angiographic findings according to ratios of total cholesterol to HDL From our findings, proximal involvement of the coronary during the acute phase of MI. J Cardiol 2004; 43(6): 251–258. arteries and more extensive coronary artery disease (multi-vessel 7. Pajunen P, Tashinen MR, Nieminen MS, Syvanne M. Angiographic disease) were strongly related to a history of DM, but less so severity and extent of coronary artery disease in patients with type I diabetes mellitus. Am J Cardiol 2000; 86(10): 1080–1085. to a history of hypercholesterolaemia, and not to a history of 8. Pajunen P, Nieminen MS, Taskinen MR. Quantitative comparison of hypertriglyceridaemia, hypertension and cigarette smoking. angiographic characteristics of coronary artery disease in patients with NIDDM compared with matched non-diabetic control subjects. Am J Cardiol 1997; 80: 550–556. References 9. Hong MK, Romm PA, Peagan K, Green CE Rackleyce. Usefulness of 1. Uddin SN, Malik F, Bari MA, et al. Angiographic severity and extent the total cholesterol to HDL ratio in predicting angiographic coronary of coronary artery diseases in patients with type 2 diabetes mellitus. artery disease in woman. Am J Cardiol 1991; 68(17): 1646–1650. Mymentsingh Med J 2005; 14(1): 32–37. 10. Sposito AC, Mansur AP, Maranhao RC, et al. Triglyceride and LPL (a) 2. Synkija R, Aronow WS, Najak D, et al. Increased fasting plasma insulin are markers of coronary artery disease severity among post-menopausal consentration are associated with the severity of angiographic CAD. women. Maturitas 2001; 39(3): 203–208. Angiology 2005; 56(3): 249–251. 11. Wilson SH, Celermajer DS, Nakagomi A, Wydhom RN, Janu MR. 3. Ammann P, Brunner- La, Rocca H, Fehr T, et al. Coronary anatomy Benfreedoms. Vascular risk factors correlate to the extent as well as and LVEF in patients with type 2 diabetes admitted for elective coro- severity of coronary atherosclerosis. Coron Artery Dis 1999; 10(7): nary angiography. Catheter Cardiovasc Inter V 2004; 62(4): 432–438. 449–453.

Cardiovascular congress diary Date Conference Venue Contact details to register MAY 2012 17–20 May Congress on Cardiac Problems in Pregnancy (CPP 2012) Berlin, Germany www.cppcongress.com 18/20 May 1st Annual Congress of the Faculty of Consulting Physicians of CTICC, Cape Town, South Africa www.physician.co.za South Africa Internal Medicine SA JUNE 2012 8–9 June CCC 2012 – Cardiovascular Complications Conference Frankfurt, Germany www.complications2012.org 27 June ICI 2012 – Imaging in Cardiovascular Interventions Frankfurt, Germany www.ici-congress.org 28–30 June CSI 2012 – Catheter Interventions in Congenital & Structural Frankfurt, Germany www.csi-congress.org Heart Disease JULY 2012 9–12 July 18th World Congress of the International Society for the Study Geneva, Switzerland www.isshp2012.com of Hypertension in Pregnancy 13–15 July ASEAN Federation of Cardiology Congress (AFCC) Singapore www.afcc2012.com 19–22 Jul y 13th Annual SA Heart Congress Sun City, South Africa www.saheart.org AUGUST 2012 25–29 August 2012 ESC, European Society of Cardiology Congress Munich, Germany www.escardio.org SEPTEMBER 2012 29 September Trend 2012 Asia–Pacific Hong Kong www.csi-trend.org OCTOBER 2012 5 October New Horizons in Echocardiography Sandton, South Africa [email protected] 10–13 October 8th World Stroke Congress Brasilia, Brazil www.2.kenes.com/stroke/pages/home.aspx 20 October The Many Faces of AF symposium Cape Town, South Africa [email protected] 20–22 October Acute Cardiac Care Istanbul, Turkey www.escardio.org 24 October The Many Faces of AF symposium Durban, South Africa [email protected] 27 October The Many Faces of AF symposium Johannesburg, South Africa [email protected] NOVEMBER 2012 3–7 November American Heart Association Scientific Sessions Los Angeles, US www.americanheart.org 16–17 November LAA 2012 Frankfurt, Germany www.csi-laa.org DECEMBER 2012 5–8 December The 16th Annual EUROECHO and other imaging modalities Athens, Greece www.euroecho.org To advertise your conference/meeting, e-mail details and half page pdf advert to [email protected] 200 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

The time-course changes of NT-proBNP and tissue Doppler indices in patients undergoing mitral valve replacement DR PRAKASCHANDRA, T ESTERHUIZEN, DP NAIDOO

Abstract Submitted 20/9/10, accepted 15/9/11 Background: In severe mitral regurgitation, a subset of Cardiovasc J Afr 2012; 23: 200–205 www.cvja.co.za patients who are asymptomatic may develop left ventricu- DOI: 10.5830/CVJA-2011-057 lar decompensation before changes in echocardiographic parameters become evident. Since N-terminal brain natriu- Chronic organic mitral regurgitation (MR) has a variable course retic peptide (NT-proBNP) is used to detect early heart fail- and requires careful monitoring by the clinician. Symptoms often ure, we hypothesised that NT-proBNP would be activated in occur late due to the compliance properties of the left atrium patients with mitral regurgitation. that allow it to accommodate large volumes of blood without Methods: Patients submitted to surgery were prospectively a significant rise in pressure. As the regurgitation becomes evaluated over eight months in the Department of Cardiology more severe, contractile dysfunction may precede the onset of at Inkosi Albert Luthuli Central Hospital. Control patients symptoms associated with disease progression as the ejection with severe mitral regurgitation were obtained from the fraction (EF) declines but it may still remain in the normal outpatient clinic. In order to define their value in identi- range. An EF less than 60% has been shown to be associated fying left ventricular decompensation, NT-proBNP levels with poorer survival rates after corrective surgery and is likely to and tissue Doppler imaging (TDI) indices were simultane- indicate covert contractile dysfunction in MR patients.1 ously measured and compared with conventional echocardio- Although numerous qualitative and quantitative echo- graphic indices at baseline and this was repeated at one week cardiographic modes have been developed, previous studies have and at six weeks after valve replacement. demonstrated that existing measures of severity of MR correlate Results: Mean NT-proBNP levels were markedly elevated poorly with clinical signs and symptoms.2 There are few data on pre-operatively in all surgical cases compared to controls the newer echocardiographic modalities, notably tissue Doppler (p = 0.0001). The diastolic E-mitral/E-annulus ratio, meas- imaging (TDI), the predictive values of which have not been ured using TDI, was higher in the study group, indicating determined. While the mitral inflow measurements are - higher left ventricular filling pressure present in the study dependent, diastolic tissue velocities measured using TDI are far group. NT-proBNP levels increased further at one week less influenced by these parameters and may be more reliable in after surgery and subsided at the six-week follow-up visit to assessing contractile changes. levels similar to the control group. The TDI diastolic ratio The development of contractile dysfunction and its relation to also decreased at one week, and increased slightly again at the severity of in MR is not clearly understood. the six-week follow up. These changes were accompanied Prolonged contractile dysfunction eventually becomes by significant reduction in left atrium and left ventricular irreversible even after the MR is corrected and is predictive chamber dimensions with an increase in the ejection fraction of both congestive heart failure and death.1 In chronic organic from one to six weeks. MR, there are as yet no clear definitions of BNP physiological Conclusion: Marked differences in mean NT-proBNP levels determinants and outcome implications. We hypothesised that and TDI ratios between the study and control groups suggest N-terminal brain natriuretic peptide (NT-proBNP) would be that using TDI and NT-proBNP assays may detect covert left activated in MR and, because this is a validated diagnostic test ventricular decompensation. in heart failure, it could prove to be an early marker for the Keywords: mitral regurgitation, tissue Doppler imaging, development of left ventricular (LV) dysfunction. NT-proBNP, mitral valve replacement In this study we evaluated tissue Doppler imaging and NT-proBNP levels in patients with severe chronic MR and determined their time-course patterns following mitral valve Department of Biomedical and Clinical Technology, Durban replacement. University of Technology; and Department of Cardiology, University of KwaZulu-Natal, Durban, South Africa Methods DR PRAKASCHANDRA, MMed Sci (Cardiology), The study population was selected from Inkosi Albert Luthuli [email protected] Central Hospital, Durban. Informed consent was obtained from Department of Cardiology, University of KwaZulu-Natal, each patient, and in the case of minors, from the parent or legal Durban, South Africa guardian. The study protocol conforms to the ethical guidelines DP NAIDOO, MD, FRCP, FESC, [email protected] of the 1975 Declaration of Helsinki. Ethics approval was given Department of Biostatistics, University of KwaZulu-Natal, by the Biomedical Research Ethics Committee at the University Durban, South Africa of KwaZulu-Natal, Nelson R Mandela School of Medicine (Ref T ESTERHUIZEN, MSc (Biostatistics) No. H112/06). AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 201

The study enrolled patients prospectively from February the cases was evaluated using paired t-tests. The discriminating to September 2007. Patients with severe, chronic isolated MR capacity of the NT-proBNP for separating surgical cases from underwent comprehensive quantitative Doppler echocardiography controls was assessed by the construction of ROC curves. performed by one trained echocardiographer (RP). Patients were Bivariate analysis was performed to assess the ability of the excluded if they had acute MR, MR due to ischaemic heart different parameters in predicting a favourable outcome using an disease or cardiomyopathy, previous valve surgery and associated NT-proBNP level of 50 pmol/l. aortic or congenital valve disease. Patients with associated mitral 2 stenosis were excluded if the valve area was less than 2.0 cm . Results Clinical evaluation and management of the patients were A total of 54 patients with severe rheumatic mitral regurgitation conducted by their independent clinicians. Assessment of were enrolled in the study and surgical cases were followed up symptoms was determined clinically by the New York Heart for six weeks. Their baseline characteristics are shown in Table Association (NYHA) classification, and atrial fibrillation was 1. All but one of the patients in our sample population were evaluated by electrocardiography. Doppler echocardiographic of African descent (98%). There were 27 control patients with recordings and blood samples were collected simultaneously severe chronic rheumatic mitral regurgitation who were recruited and estimation of the NT-proBNP levels were processed from the cardiology follow-up clinic. independently. Only one patient in the control group had markedly elevated Control subjects with severe MR were selected from the systolic pressure (63 mmHg) together with NYHA class III cardiology outpatient follow-up clinic where they were assessed symptoms. This patient subsequently had a valve replacement as not requiring surgery in the short term and were receiving seven months after the echocardiographic and NT-proBNP medical therapy. Controls were selected as severe MR, but assessment. without evidence of ischaemic heart disease, as it is known that The orifice area was markedly increased in 12 patients in the 3 NT-proBNP is activated in the presence of ischaemia. study group (Table 2). Group comparisons revealed that early Colour Doppler echocardiography was performed on all diastolic filling ratios (E-mitral/E-annulus) was higher in the patients using a Siemens Sequoia machine (Acuson, Germany). study group (p = 0.04). A similar pattern was observed with Dimensions and wall thickness were measured according to the NT-proBNP level, which was elevated in both groups, but was American Society of Echocardiography guidelines using the markedly higher (p < 0.001) in the study group (Table 3). leading-edge method,4 and EF was measured from the apical The immediate post-operative period showed a rise in four-chamber view using the modified Simpson’s method. The NT-proBNP levels from pre-operative levels (262 ± 224 pmol/l) rate of rise of pressure (dP/dT) was also calculated as a measure to a mean of 395 pmol/l at one week, and subsiding thereafter to of ventricular systolic function. The six frequently applied 94 pmol/l at six weeks (see Fig. 1). These changes were mirrored echocardiographic variables described by Thomas et al.5 were by a significant reduction in the left atrium (LA) size and used to evaluate MR. Mitral regurgitation was quantified by volume, as well as in the LV chamber dimension to levels similar measurement of the regurgitant volume and fraction. to the control population group (Table 4). It is noteworthy that TDI was performed in the apical four-chamber view. The inasmuch as there was a slight reduction in the end-diastolic myocardial systolic wave (Sm) velocity, the diastolic indices, volume (EDV) and end-systolic volume (ESV), these changes namely early myocardial (Em) and atrial contraction (Am) were not statistically significant. Although the TDI systolic wave peak velocities, Em/Am ratio, and early diastolic filling ratios indices were unchanged between the two time points, there was (transmitral/annular) (Em/Ea ratio) were measured. The ratio a significant increase noted in the early diastolic filling ratios, of the E wave across the mitral valve to the annulus E wave suggesting a further rise in the left ventricular filling pressures on tissue Doppler (E/Ea) was used as an estimate of the LV at six weeks. filling pressure. All echocardiographic measurements were Four patients exhibited persistently elevated NT-proBNP performed by a single echocardiographer (RP) blinded to all other levels at six weeks compared to the other study cases. The ejection measurements. The images and measurements were reviewed fraction fell in three of these patients post-operatively, and was off-line by a trained cardiologist (DPN). The intra-observer accompanied by little change in the end-systolic dimension variability for the measurements of proximal isovelocity surface (ESD) and NT-proBNP values compared to pre-operative levels. area (PISA) and TDI was < 5%. For measurement of NT-proBNP, venous blood samples were TABLE 1. BASELINE (PRE-OPERATIVE) taken in gel-filled tubes with the patient resting quietly at the CHARACTERISTICS IN SEVERE MR time of echocardiography. Additional samples were taken from Variables Controls (n = 27) Study (n = 27) p-value subjects undergoing mitral valve replacement at one and six Age 23 ± 13 20 ± 11 0.08 weeks post surgery. Males/females 7/20 7/20 1.000 Statistical analysis NYHA I–II 24 10 0.001 NYHA III–IV 3 17 SPSS for Windows version 15.0 was used for the statistical Heart failure 1 5 analysis. Clinical variables were normally distributed and EF (mean ± SD) 67 ± 6 67 ± 9 1.000 expressed as mean ± SD. NT-proBNP levels were log-transformed for statistical analysis. Group comparisons were performed with Diuretics 24 20 0.307 ANOVA, t-test or chi-square tests. Associations of baseline ACE inhibitors 25 25 1.000 NT-proBNP were tested with linear and non-parametric regression Atrial fibrillation 6 10 0.372 (categorical variables). The time course of NT-proBNP within NYHA = New York Heart Association class; EF = ejection fraction. 202 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

TABLE 2. BASELINE QUANTIFICATION TABLE 4. POST-OPERATIVE EVALUATION AT OF THE SEVERITY OF MR ONE AND SIX WEEKS’ FOLLOW UP Variables Controls (n = 27) Study (n = 27) p-value 1 week 6 weeks CI p-value RF < 70% 6 2 LA size (mm) 64 ± 16 58 ± 16 2.0; 11.2 0.009 70–80% 6 3 0.192 LA volume (ml) 234 ± 152 182 ± 140 21.2; 82.9 0.003 > 80% 15 19 EDV (ml) 139 ± 52 125 ± 55 –15.7; 43.8 0.326 RV > 60 ml 27 27 1.000 ESV (ml) 86 ± 47 73 ± 47 –10.7; 37.5 0.250 EOA < 0.35 3 1 EDD (mm) 59 ± 9 54 ± 8 2.5; 10.2 0.004 0.36–0.40 2 4 ESD (mm) 45.5 ± 13 38.7 ± 9 12.7; 1.0 0.03 0.058 0.41–0.90 17 10 EF 42 ± 13 51 ± 13 –17.9; –1.0 0.0032 > 0.90 5 12 TDI syst (m/s) 0.07 ± 0.016 0.07 ± 0.015 –0.01; -0.2 0.821 PAS < 30 mmHg 0 0 TDI Em/Ea 12 ± 4 15 ± 3 –6.7; –0.2 0.418 30–39 10 6 NT-proBNP (pmol/l) 395 ± 460 94 ± 161 87.5; 514.1 0.009 0.192 40–59 16 18 LA = left atrium; EDV = end-diastolic volume; ESV = end-systolic > 60 mmHg 1 4 volume; EDD = end-diastolic dimension; ESD = end-systolic dimen- sion; EF = ejection fraction, TDI = tissue Doppler imaging; TDI Em/Ea = LA size < 40 mm 4 0 diastolic filling ratio. 40–49 mm 5 2 0.074 > 49 mm 20 25 heart failure (NT-proBNP > 53 pmol/l), the specificity for EF < 50% 0 1 NT-proBNP improved to 74%, and the positive predictive value 51–60% 1 5 0.123 to 84%. Of interest, eight patients in the control group had BNP > 60% 26 21 levels > 53 pmol/l, and of these, six underwent surgery within RF = regurgitant fraction; RV = regurgitant volume; LA = left atrium; RV the ensuing six months. = right ; PAS = systolic pressure; EOA = effec- tive orifice area; EF = ejection fraction. A second ROC curve (Fig. 3) was constructed using all the variables to separate NYHA classes I–II from III–IV. Once again, NT-proBNP emerged with the highest area under the There was an increase in the diastolic filling ratios (Em/Ea) and a curve, followed closely by the ESD. Bivariate analysis was decrease in the TDI systolic wave. The fourth patient had a large performed to assess the ability of the different parameters to drop in the NT-proBNP level, as well as an increase in the TDI predict a favourable outcome, defined arbitrarily as NT-pro BNP systolic wave, in keeping with the preserved LV contractility. level < 50 pmol/l. These were an effective orifice area (EOA) (p There was a small decrease in the ESD post-operatively (Fig. = 0.014), tissue Doppler S wave (p = 0.049) and left atrium (LA) 1). These four were identified by NT-proBNP cut-off levels size (p = 0.027), all pointing to the interrelationship between set by Januzzi’s rule-in criteria (450 pg/ml, ~53 pmol/l) for the severe regurgitation, systolic function and NT-proBNP levels. diagnosis of heart failure in the PRIDE study6 at six weeks. A ROC curve was constructed for each variable to assess its discriminating capacity to distinguish between surgical Discussion cases and controls. The area under the curve was highest for This is one of the first studies to use new modalities of NT-proBNP; this was the only parameter that approached 90% measuring left ventricular function with TDI and NT-proBNP (Fig. 2). However, at established cut-off levels for normality, assays. NT-proBNP level has been shown to be a marker of left NT-proBNP (12 pmol/l = 125 pg/ml) yielded the highest ventricular dysfunction and has been used to predict diastolic sensitivity of 96% but had a low specificity of 45%. Using abnormalities in patients with normal systolic function,7 limiting cut-off criteria established by Januzzi6 for the detection of the need for expensive cardiac imaging modalities.8,9 In this study, NT-proBNP level yielded the highest predictive value for TABLE 3. PRE-OPERATIVE QUANTIFICATION discriminating between cases selected for surgery and controls OF LV FUNCTION followed up at the cardiology outpatients’ clinic (sensitivity of Controls Study p-value 96%). The ESD at the established cut-off point of 45 mm that LA size (mm) 59.6 ± 13 76 ± 16 < 0.001 defines the need for surgery had a higher specificity of 81%. LA volume (ml) 177 ± 119 309 ± 183 0.003 EDV (ml) 177 ± 69 165 ± 48 0.472 3.00 case ESV (ml) 58 ± 25 57 ± 25 0.898 control 2.50 EDD (mm) 67 ± 6 67 ± 9 0.06 (pmol/l) ESD (mm) 38 ± 9 43 ± 8 0.043 2.00 EF 65 ± 10 67 ± 7 1.000 TDI (m/s) 0.09 ± 0.02 0.1 ± 0.07 0.377 1.50 TDI Em/Ea 14 ± 8 20 ± 8 0.004 1.00

NT-proBNP (pmol/l) 57 ± 88 262 ± 224 < 0.001 95% CI log BN P LA = left atrium; EDV = end-diastolic volume; ESV = end-systolic 0.50 volume; EDD = end-diastolic dimension; ESD = end-systolic dimen- Baseline 1st week post op 6 weeks post op sion; EF = ejection fraction, TDI = tissue Doppler imaging; TDI Em/Ea = Time points diastolic filling ratio. Fig. 1. NT-pro BNP time-course changes. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 203

ROC curve Source of the Curve 1.0 LV Esd EF (%) BNP pmol/l 0.8 s (l) E/Ea (l) EOA 0.6 dP/dT Regurg. Fraction Reference Line

S ensitivity 0.4 LV Esd 0.704 EF (%) 0.529 BNP pmol/l 0.882 0.2 s(l) 0.545 dP/dT E/Ea(l) 0.614 EOA 0.673 0.0 dP/dT 0.364 0.0 0.2 0.4 0.6 0.8 1.0 Regurg. fraction 0.711 1 – Specificity Digital segments are produced by tiers. Area under the curve

Fig. 2. Receiver operating characteristics curve: surgical cases and controls.

In a study similar to ours, Pillai et al.10 assessed pre-operative The six-week correlation between NT-proBNP levels and NT-proBNP levels in a group of 20 patients with rheumatic chamber dimensions suggests that in MR, changes in volume heart disease. They showed that elevated pre-operative load may be paralleled by changes in the NT-proBNP level and NT-proBNP levels were an indicator of underlying myocardial that the fall in NT-proBNP was related to corrections in volume dysfunction, which was not evident by routine two-dimensional and removal of the diastolic run-off into the left atrium. These echocardiography, and advocated pre-operative assessment of results are also in keeping with those found by the Mayo Clinic,13 NT-proBNP levels to detect underlying myocardial dysfunction. which showed that the severity of mitral regurgitation, although Two other studies have examined patients with varying univariately associated with NT-proBNP concentrations, was degrees of MR, and showed that changes in ventricular function not an independent predictor of high NT-proBNP levels. They occur early in the disease process, even before they could be examined a group of 124 patients with varying degrees of detected echocardiographically.11,12 The diastolic filling ratios organic mitral regurgitation and showed that NT-proBNP level were higher in the study group, indicating LV decompensation was independently predictive of mortality/heart failure over a and a rise in the filling pressures. However, the diastolic ratios 4.4-year follow-up period. were also elevated in the control group, suggesting that LV Increased Em/Ea (> 12) ratios and elevated NT-proBNP decompensation with elevated LV filling pressure was already (> 170 pg/ml) have been shown to be useful parameters to present in patients deemed by the clinician to be stable enough identify patients at increased risk of developing paroxysmal to be followed up at the clinic. atrial fibrillation (AF) as well as to reflect early left ventricular

ROC curve Source of the Curve 1.0 LV Esd EF (%) BNP pmol/l 0.8 s (l) E/Ea (l) EOA 0.6 dP/dT Regurg. Fraction Reference Line

S ensitivity 0.4 LV Esd 0.769 EF (%) 0.556 BNP pmol/l 0.797 0.2 s(l) 0.373 E/Ea(l) 0.623 EOA 0.602 0.0 dP/dT 0.326 0.0 0.2 0.4 0.6 0.8 1.0 Regurg. fraction 0.577 1 – Specificity Digital segments are produced by tiers. Area under the curve

Fig. 3. Receiver operating characteristics curve: NYHA all classes. 204 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

dysfunction.14 We found that NT-proBNP and tissue Doppler pg/ml (12.4 pmol/l for NT-proBNP) identified asymptomatic levels in patients with AF were elevated, both in the study and in patients with severe MR who were at higher risk. the control groups, indicating that symptomatology was not an Most of the subjects in our control group had values above early marker of ventricular decompensation and that our patients this level, indicating that all our patients were at risk while being needed to be evaluated and referred for surgical intervention at followed up at the clinic. We attributed this to the subtle nature of an earlier stage in the course of their illness. symptoms, or particularly in our study, lack of awareness (on the The challenge in evaluating mitral regurgitation is a part of patients or patients’ caregivers) of worsening symptoms determination of what really constitutes normal ventricular and poor referral guidelines from peripheral healthcare centres. function in these patients. The limitations of using the ejection This low rate of intervention has also been reported in other centres fraction in the timing of surgery become clear in subjects with as well.23 The Euro Heart Survey suggests that 31% of patients apparently normal ejection fraction and minimal symptoms. with severe valve disease and symptoms were not operated on.24 In a study of 84 asymptomatic patients who underwent In our study, no parameter could discriminate pre-operatively surgical correction for MR, Agricola et al.15 demonstrated that between the four patients who exhibited persistently elevated TDI systolic indices could predict postoperative left ventricular NT-proBNP levels at six weeks and the rest of the sample. function. In contrast, our study has shown that the TDI systolic All four of these patients had atrial fibrillation pre-operatively, wave cut-off point of 0.06 m/s, which has been previously used which persisted at six weeks and at six months. There was no to rule out systolic dysfunction,16 had very low sensitivities in indication of difficulties with myocardial preservation to suggest separating surgical cases from controls (see Fig. 2). Using a this as the cause for the decline in postoperative EF by almost higher cut-off point of 0.085 m/s only marginally improved 30% from normal pre-operative levels. the specificity. It is possible that different cut-off points in One of the limitations of this study was that it was performed conjunction with strain measurements17 may be more sensitive only in patients with severe MR. The study needs to be repeated in determining impaired LV contractile function in MR, which in different subgroups with varying degrees of MR in order to can only be established in future studies with serial evaluations determine the time course of NT-proBNP levels in the different at different time intervals. stages of MR as well as in patients with ischaemic MR. Finally, Left ventricular contractile dysfunction is present in many we need to determine the association of NT-pro BNP levels in a patients with severe MR despite a normal ejection fraction and longer follow-up study with hard end-points such as heart failure returns to normal after corrective mitral valve surgery in most and death. but not all patients.18 This was very apparent in our patients. In fact, all our patients experienced more than 10% decline in the Conclusion EF immediately post surgery. This was improved at the six-week We have shown that decision making regarding the timing follow-up visit in all but three of our patients. Despite symptomatic of surgery in this cohort of rheumatic heart disease patients improvement, postoperative left ventricular dysfunction (EF was determined largely by advanced symptomatology, so that < 50%) has been shown to occur frequently, occurring in patients are referred to surgery late in the course of severe MR, close to a third of the patients successfully operated on.19 with a risk of permanent LV decompensation. We propose that In our study, 15 (15/27) of the patients referred for surgery NT-proBNP level be an additional marker, particularly in less had ejection fractions above 60% and ESD values below 45 mm, symptomatic patients, even if the EF is normal. In time, it may indicating that the reason for surgery in these patients was the prove to be the composite marker for the assessment of LV presence of significant symptoms while on medical treatment. decompensation. These findings support those of Detaint et There is an inherent subjectivity in defining significant valve- al.25 in that the BNP level reflects the severe haemodymamic, related symptoms. Indeed, when the second ROC curve was ventricular and atrial consequences of MR. constructed using NYHA class as the determinant, NT-proBNP again emerged as the strongest discriminator of advanced NYHA References class. The low specificity of 45% for NT-proBNP indicates a 1. Griffin BP. Timing of surgical intervention in chronic mitral regurgita- tion: is vigilance enough? Circulation 2006; 113(18): 2169–2172. high number of false positive cases, i.e. patients selected 2. Rosenhek R, Rader F, Klaar U, Gabriel H, Krejc M, Kalbeck D, et al. by NT-proBNP level as requiring surgery when in fact they Outcome of watchful waiting in asymptomatic severe mitral regurgita- were still being followed up in the clinic on medical therapy. tion. Circulation 2006; 113: 2238–2244. We believe that what is considered the high false-positive 3. Weber M, Dill T, Arnold R, et al. N-terminal B-type natriuretic peptide rate with NT-proBNP actually indicates that many of these predicts extent of coronary artery disease and ischemia in patients with control subjects with or without minimal symptoms actually stable angina pectoris. Am Heart J 2004; 148: 612–620. 4. Zoghbi WA, Enriquez-Sarano M, Fosrer E, et al. American Society of required surgery, rendering them true positives. Waiting for Echocardiography. Recommendations for evaluation of the severity of more advanced symptoms or change in dimensions increases the native valvular regurgitation with two-dimensional and Doppler echo- risk of left ventricular dysfunction postoperatively. Two studies, cardiography. J Am Soc Echocardiogr 2003; 16: 777–802. both from the Mayo Clinic, have highlighted the poor outcomes 5. Thomas L, Foster E, Hoffman JIE, Schiller NB. The mitral regurgita- in patients with severe MR who were managed conservatively tion index: an echocardiographic guide to severity. J Am Coll Cardiol rather than surgically.20,21 1999; 3320: 16–2022. In another study, Pizzaro et al.22 reported that NT-proBNP 6. Januzzi JL, Carlos AC, Saif A, et al. The N-terminal pro-BNP Investigation of Dyspnoea in the Emergency Department (PRIDE) level was a stronger prognostic marker than ESD or EOA and Study. Am J Cardiol 2005; 95: 948–954. contributed independent prognostic information additional to 7. Cowie MR, Jourdain P, Maisel J, Dahlstrom U. Clinical applica- other echo parameters. They showed that the BNP cut-off of 105 tions of B-type natriuretic peptide testing. Eur Heart J 2003; 24(19): AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 205

1710–1718. Am J Cardiol 1991; 67: 222. 8. Lubien E, DeMaria A, Krishnaswamy P, et al. Utility of B-natriuretic 17. Carlhall CJ, Nguyen TC, Itoh A, et al. alterations in transmural myocar- peptide in detecting diastolic dysfunction: comparison with Doppler dial strain. Circulation 2008; 118: S256–S262. velocity recordings. Circulation 2002; 105: 595–601. 18. Enriquez-Sarano M, Avierinos JF, Messika-Zeitoun D, Detaint D, 9. Mair J, Hammerer-Lercher A, Puschendorf B. The impact of cardiac Capps M, Nkomo V, et al. Quantitative determinants of the outcome of natriuretic peptide determination on the diagnosis and management of asymptomatic mitral regurgitation. N Engl J Med 2005; 352: 875–883. heart failure. Clin Chem Lab Med 2001; 39: 571–588. 19. Kouris N, Ikonomidis I B, Kontogianni D. Mitral valve repair versus 10. Pillai S, Hemalatha R, Saibabu Meena Rani, Mullasari Ajit S. replacement for isolated non-ischemic mitral regurgitation in patients Assessment of preoperative pro-BNP levels in patients with rheu- with preoperative left ventricular dysfunction. A long-term follow-up matic heart disease with normal left ventricular function; Institute of echocardiography study. Eur J Echocardiogr 2005; 6(6): 435–442. Cardiovascular Diseases. Indian Heart J 2003; 55(5): 450. 20. Kernis SJ, Nkomo VT, Messika-Zeitoun D, Gersh BJ, Sundt TM III, 11. Shimamoto K, Kusumoto M, Sakai R, Watanabe H, Ihara S, Koike N, Ballman KV, et al. Atrial fibrillation after surgical correction of mitral et al. Usefulness of the brain natriuretic peptide to atrial natriuretic regurgitation in sinus rhythm: incidence, outcome, and determinants. peptide ratio in determining the severity of mitral regurgitation. Can J Circulation 2004; 110: 2320–2325. Cardiol 2007; 24(4): 295–300. 21. Enriquez-Sarano M. Timing of mitral valve surgery. Heart 2002; 87: 12. Brookes CI, Kemp MW, Hooper J, Oldershaw PJ, Moat NE. Plasma 79–85. brain natriuretic peptide concentrations in patients with chronic mitral 22. Pizaro R, Bazzino O, Oberti P, et al. Prospective validation of the prog- regurgitation. J Heart Valve Dis 1997; 6: 608–612. nostic usefulness of brain natriuretic peptide in asymptomatic patients 13. Enriquez-Sarano M, Schaff H, Orszulak T, et al. Congestive heart fail- with chronic severe mitral regurgitation. J Am Coll Cardiol 2009; ure after surgical correction of mitral regurgitation. A long-term study. 54(12): 1099–1106. Circulation 1995; 92: 2496–2450. 23. Bach DS, Awais M, Gurm HS, Kohnstamm S. Failure of guideline 14. Nakamura S, Naruse M, Naruse K, et al. Atrial natriuretic peptide and adherence for intervention in patients with severe MR. Am Coll Cardiol brain natriuretic peptide coexist in the secretory granules of human 2009; 54(9): 860–865. cardiac myocytes. Am J Hypertens 1991; 4: 909–912. 24. Iung B, Baron G, Butchart EG, et al. A prospective survey of patients 15. Agricola E, Galderisi M, Oppizzi M, et al. Pulsed tissue Doppler imag- with valvular heart disease in Europe: The Euro Heart Survey on ing detects early myocardial dysfunction in asymptomatic patients with Valvular Heart Disease. Eur Heart J 2003; 24: 1231–1243. severe mitral regurgitation. Heart 2004; 90: 406–410. 25. Detaint D, Suitton M, Avierinos JF, et al. B-type natriuretic peptide 16. Pai RG, Bodenheimer MM, Pai SM, et al. Usefulness of systolic excur- in organic mitral regurgitation: determinants and impact on outcome. sion of the mitral anulus as an index of left ventricular systolic function. Circulation 2005; 111: 2391–2397.

Letter to the Editor

Cohort studies of cardiovascular disease in the Seychelles, Tanzania and Mauritius

We read with interest the review by Kengne et al. on cohort antihypertensive treatment was low in 50 hypertensive patients studies of cardiovascular disease in sub-Sahara Africa.1 We followed for 12 months, despite free healthcare.3 In this study agree with the authors that cohort studies are important tools to adherence was measured with electronic pill containers, the gold advance our knowledge of cardiovascular disease in the region standard for assessment of therapeutic adherence. In a cohort and inform appropriate clinical and public health responses. study of 153 smokers followed for six months, smokers who We recognise the difficult challenge of identifying all cohort were shown pictures of their own atherosclerotic plaques in their studies in the region. We wish however to mention several carotid arteries (B-mode ultrasonography) had improved rates cohort studies in the Seychelles, Tanzania and Mauritius, which of smoking cessation.4 A cohort study among 644 Seychelles were published in leading medical journals but were not included children enrolled at birth showed no overall effect of pre-natal in the review, although they met inclusion criteria set by the exposure to organic mercury on blood pressure (BP) levels at age authors of the review. 12 and 15 years.5 The Republic of Seychelles, which lies in the Indian Ocean In Tanzania, 653 participants with BP ≥ 160/95 mmHg around 1 000 km east of Kenya, belongs to south Saharan Africa. and 653 with BP < 160/95 mmHg from a population survey Seychelles is part of WHO AFRO, is a member of the South of 9 254 subjects in Dar es Salaam had BP readings on three African Development Community (SADEC) and contributes additional visits over an eight-week follow-up period. Their epidemiological data to the Global Burden of Disease project BP decreased markedly over subsequent visits, irrespective of for estimates of the east Africa region. The majority of the baseline BP levels, and the prevalence of hypertension dropped population of Seychelles is of African descent. by approximately 50% based on BP values on the second, third In a cohort study of 5 514 Seychelles children, there was a or fourth visits, compared to BP values on the first visit.6 strong association between weight gain during the first year of life and overweight/obesity at age five to 17 years.2 Adherence to continued on page 215 … 206 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Presentation pattern and management of effusive– constrictive pericarditis in Ibadan MA SALAMI, PO ADEOYE, VO ADEGBOYE, OA ADEBO

Abstract Keywords: pericarditis, effusive, constrictive, Ibadan, African Background: Effusive–constrictive pericarditis is a syndrome Submitted 24/5/10, accepted 5/10/11 in which constriction by the visceral pericardium occurs in Cardiovasc J Afr 2012; 23: 206–211 www.cvja.co.za the presence of a dense effusion in a free pericardial space. Treatment of this disease is problematic because pericardio- DOI: 10.5830/CVJA-2011-066 centesis does not relieve the impaired filling of the heart and surgical removal of the visceral pericardium is challenging. Effusive–constrictive pericarditis is a clinical syndrome We sought to provide further information by addressing characterised by concurrent pericardial effusion and pericardial the evolution and clinico-pathological pattern, and optimal constriction where constrictive haemodynamics are persistent surgical management of this disease. after the pericardial effusion is removed. The treatment of Methods: We conducted a prospective review of a consecutive effusive–constrictive pericarditis is problematic because series of five patients managed in the cardiothoracic surgery pericardiocentesis does not relieve the impaired filling of the unit of University College Hospital, Ibadan, in the previous heart, and surgical removal of the fibrinous exudate coating 1 year, along with a general overview of other cases managed the visceral pericardium may not be possible. Pericardiectomy over a seven-year period. This was followed by an extensive following development of a pericardial skin that is amenable to literature review with a special focus on Africa. surgical stripping is usually the most successful treatment option. Results: The diagnosis of effusive–constrictive pericarditis The objectives of this case series were to document the evolution was established on the basis of clinical findings of features of and clinico-pathological pattern of this disease in Nigerians. pericardial disease with evidence of pericardial effusion, and echocardiographic finding of constrictive physiology with or Methods without radiological evidence of pericardial calcification. A We conducted a prospective review of a consecutive series review of our surgical records over the previous seven years of five patients managed in the cardiothoracic surgery unit revealed a prevalence of 13% among patients with pericar- of University College Hospital, Ibadan in the previous year, dial disease of any type (11/86), 22% of patients presenting along with a general overview of other cases managed over a with effusive pericardial disease (11/50) and 35% who had seven-year period. This was followed by an extensive literature had pericardiectomy for constrictive pericarditis (11/31). All review with a special focus on Africa. The diagnosis of five cases in this series were confirmed by a clinical scenario effusive–constrictive pericarditis was established on the basis of non-resolving cardiac impairment despite adequate open of clinical findings of features of pericardial disease with pericardial drainage. They all improved following pericar- evidence of pericardial effusion, and echocardiographic finding diectomy. of constrictive physiology with or without radiological evidence Conclusion: Effusive–constrictive pericarditis as a subset of of pericardial calcification. pericardial disease deserves closer study and individualisa- tion of treatment. Evaluating patients suspected of having the disease affords clinicians the opportunity to integrate Results clinical features and non-invasive investigations with or A review of our surgical records over the previous seven without findings at pericardiostomy, to derive a management years revealed a prevalence of 13% among patients with plan tailored to each patient. The limited number of patients pericardial disease of any type (11/86), 22% of patients in this series called for caution in generalisation. Hence our presenting with effusive pericardial disease (11/50) and 35% aim was to increase the sensitivity of others to issues raised who had pericardiectomy for constrictive pericarditis (11/31). and help spur on further collaborative studies to lay down The present subset was chosen for the prospective follow up due guidelines with an African perspective. to the unusual consecutive presentation and a dearth of studies specifically on this subset of patients from Africa. All five cases in this series were confirmed by a clinical Department of Surgery, Cardiovascular and Thoracic scenario of non-resolving cardiac impairment despite adequate Surgery Division, University College Hospital and College of open pericardial drainage. All five patients were prospectively Medicine, University of Ibadan, Ibadan, Nigeria followed up. One patient, who we treated for effusive–contrictive MA SALAMI, MBBS, MRCS (Glasgow), FWACS, drmudathirsalami pericarditis, is described in detail and four other cases are @yahoo.com summarised in tabular form (Table 1). PO ADEOYE, MBBS, FWACS VO ADEGBOYE, MBBS, FMCS College of Health Sciences, Bowen University, Iwo, Nigeria Case studies OA ADEBO, FRCSC A 20-year-old, HIV sero-negative lady presented to the AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 207

TABLE 1. SUMMARY OF CASES OF EFFUSIVE–CONSTRICTIVE PERICARDITIS Age HIV Post-op Patient (years) Gender Comorbid conditions status Initial procedure Pericardial histology NYHA 1. SB 46 M Superficial thigh wound from Negative Pericardial window and Tuberculous pericarditis II Pre-op NYHA III gunshot biopsy 2. DS 19 M Haemoglobin AS – Pericardial window and Non-specific calcific I Pre-op NYHA III biopsy pericarditis 3. AO 20 F – Negative Pericardial window and Non-specific chronic I Pre-op NYHA IV biopsy pericarditis 4. MN 19 F Endomyocardial fibrosis Negative Pericardial window and Pericardial fibrosis I Pre-op NYHA IV Tricuspid regurgitation biopsy 5. OS 20 M Fournier’s gangrene Negative Pericardial window Non-specific chronic I Pre-op NYHA IV Upper gastrointestinal bleeding pericarditis cardiothoracic unit of the University College Hospital, Ibadan The pericardiostomy tube was removed one week post operation. with a three-year history of easy fatigability, exertional dyspnoea A subsequent radiograph revealed evidence of re-accumulation and weight loss. There was a history of cough productive of of pericardial fluid. The patient and her relatives still declined whitish sputum. There was an associated history of orthopnoea, surgery and asked for a discharge. chest discomfort and bulging chest, but no history of leg swelling. She represented about 48 hours later with evidence of The patient was wasted and afebrile with a respiratory rate of 32 massive pericardial effusion and cardiac tamponade. She then breaths/min. Her blood pressure and were, respectively, had an emergency pericardiocentesis under echocardiographic 105/80 mmHg and 102 per min. Her neck veins were distended guidance, during which 1 940 ml of haemorrhagic effusion was and she had a bulging anterior chest and hepatomegaly. aspirated and another 2 250 ml four days later. She improved The patient’s packed cell volume was 40%. Her blood following this and then had a pericardiectomy. chemistry findings were normal. The chest radiograph showed Findings at surgery included a thickened parietal and visceral a globular heart shadow (Fig. 1). The ECG revealed low-voltage pericardium, about 1.5 l of serosanguinous fluid in the pericardial waves. An echocardiogram revealed a large pericardial effusion space, and an area of calcification particularly over the right with echo speckles within it and a thickened pericardium. atrium (Fig. 3A, B). Both the parietal and visceral pericardium There was septal bounce and a dilated inferior vena cava were stripped. The patient had an uneventful postoperative with blunted respiratory fluctuations in diameter. A diagnostic recovery period and was discharged home 10 days after surgery. pericardiocentesis yielded serosanguinous fluid. She has been seen twice since discharge, the last visit eight The patient underwent a subxiphoid tube pericardiostomy months post operation, with remarkable recovery, and NYHA with pericardial biopsy. A postoperative chest radiograph class I status. showed evidence of pericardial calcification (Fig. 2). She was SM had a pre-operative (pericardial window) echo, which scheduled for an elective pericardiectomy, which was declined. showed effusion with constrictive physiology. He had modest

Fig. 1. Radiograph showing massive globular heart Fig. 2. Radiograph showing evidence of pericardial calci- shadow. fication. 208 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

A B

Fig. 3. (A) Thickened pericardium and a large pericardial space. (B) Final phase of visceral pericardial stripping. postoperative improvement and was discharged but he the prevalence of 13% among patients with pericardial disease represented three months later with worsening of pre-operative of any type in our seven-year review (11/86). We are not aware symptoms. He then had a pericardiectomy, following which he of any specific series from Africa. improved progressively. Patients with effusive–constrictive pericarditis present with Following tube pericardiostomy, DS had very transient symptoms due to limitation of diastolic filling. These findings improvement in his symptoms. Repeat lateral chest X-ray showed are secondary not only to the pericardial effusion but also evidence of pericardial calcification while echocardiography the pericardial constriction. Symptoms and physical findings showed moderate pericardial effusion and diastolic dysfunction vary, while a moderate-to-large pericardial effusion may occur. (Fig. 4). He made a rapid recovery following pericardiectomy. Management of effusive–constrictive pericarditis is therefore MN had minimal improvement following tube pericardiostomy, fraught with challenges. remaining dyspnoeic at rest. Postoperative chest radiography The diagnosis is usually made by echocardiography, which and echocardiography showed pericardial calcification. In should demonstrate diastolic dysfunction. The diagnosis can addition, there was a markedly enlarged right atrium, grade easily be missed by an unwary clinician because of the usual III–IV tricuspid regurgitation and a small right ventricle with superimposed features of accompanying pericardial effusion endocardial thickening, suggestive of endomyocardial fibrosis. or tamponade. This may have accounted for the premature We elected to go ahead with a pericardiectomy on account of discharge and re-admission of one of our patients (SB). the pericardial thickening with calcification. She improved Pericardial effusion is seen as an echo-free space around following pericardiectomy, with NYHA class I status. the heart on echocardiography (Fig. 4). The presence of a OS had pericardiostomy with slight improvement and large pericardial effusion with frond-like projections and a was discharged home on anti-tuberculous therapy. He had a thick ‘porridge-like’ exudate is suggestive of an exudate but pericardiectomy three months later, during which he had an intra- not specific for a tuberculous aetiology.1 Patients with acute operative complication of right ventricular wall injury, which haemorrhagic effusions may have pericardial thrombus appearing was promptly repaired. He had an uneventful postoperative as an echo-dense mass.5 recovery until the 12th and 19th days postoperatively, when Small pericardial effusions are only seen posteriorly, while he developed Fournier’s gangrene and upper gastrointestinal those large enough to produce cardiac tamponade are usually bleeding, respectively. These were successfully managed and he circumferential. In large pericardial effusions, the heart may was discharged home on the 36th day postoperatively. move freely within the pericardial cavity (‘swinging heart’). In the parasternal long-axis view, pericardial fluid reflects at the posterior atrio-ventricular groove, while pleural fluid continues Discussion under the left atrium, posterior to the descending aorta. Rarely, Effusive–constrictive pericarditis is said to be an uncommon tumour masses are found within or adjacent to the pericardium pericardial syndrome.2 In a prospective study of 1 184 patients and may masquerade as tamponade.6 with pericarditis, Sagrista-Sauleda et al. reported a prevalence of Diagnostic criteria for cardiac tamponade include diastolic only 1.3% among patients with pericardial disease of any type collapse of the right atrial and ventricular anterior free wall, and (15/1 184) and 6.9% among patients with clinical tamponade left atrial and very rarely left ventricular collapse. Right atrial (15/218).3 However, a recent observational study by Mayosi collapse is more sensitive for tamponade, but right ventricular et al. reported 28 (15.1%) of 185 patients with tuberculous collapse lasting more than one-third of is a more pericarditis as belonging to that subset.4 This is quite similar to specific finding for cardiac tamponade. Doppler findings include AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 209

A Because constrictive pericarditis can be corrected surgically, it is important to distinguish chronic constrictive pericarditis from restrictive cardiomyopathy, which has a similar physiological abnormality, i.e. restriction of ventricular filling. Helpful in the differentiation of these two conditions are right ventricular trans-venous endomyocardial biopsy (by revealing myocardial infiltration or fibrosis in RCM) and echocardiography, CT scan or cardiac magnetic resonance imaging (by demonstrating a thickened pericardium in constrictive pericarditis but not in RCM).8 Our fourth patient (MN) actually presented this challenge but a convincing thickening of the pericardium at echocardiography was enough to help us clarify the diagnosis. Another important problem is the lack of placebo-controlled trials from which appropriate therapy may be selected, and of guidelines that assist in important clinical decisions. As a result, the practitioner must rely heavily on clinical judgment.9 The absence of guidelines specific to this subset of pericardial B disease may be due to its relative rarity in the Western world. The recent European Society of Cardiology guidelines on management of pericardial diseases was also silent on the subset of patients with effusive constrictive pericarditis, presumably due to a paucity of data on the subject.6 Other reasons could be difficulty in reaching a diagnosis and varied aetiopathogenesis, necessitating different evolution patterns. While there is an abundance of diagnostic armamentarium in the West, practitioners in sub-Saharan Africa largely have to cope with severe limitations in diagnostic facilities. An exception to this may be South Africa, where a recent report highlighted the value of contrast-enhanced magnetic resonance imaging (MRI) in delineating epicardial and pericardial inflammation in effusive–constrictive pericarditis.10 Cost is still an issue even if MRI becomes widely available. Clinical acumen and reasoning therefore still form the bedrock of clinical practice in most C centres. The cases managed in this series illustrate this point. In only two of the five cases was there a hint of constrictive physiology at the initial echocardiography, even though it is known there is a phase of transient sub-acute constriction, which may improve after pericardial drainage and medical treatment, especially with anti-tuberculous therapy in those arising secondary to tuberculosis. The only strong evidence of a high likelihood of need for pericardiectomy was the duration of the history in the first three patients. They all had a history longer than two years, suggestive of a chronic process. Reaching an aetiological diagnosis is a real challenge globally but more problematic in our local practices. The results of pericardial fluid culture are frequently falsely negative and pericardial biopsy has a higher yield of diagnostic specimens.11-13 One therefore has to rely on pericardial tissue biopsy microbiology Fig. 4. Echocardiography showing moderate pericardial and histology. None of our patients had positive evidence from effusion (PE). RV = right ventricle; LV = left ventricle; RA = right atrium; LA = left atrium. pericardial fluid microbiology or cytology. The histology of their pericardia is shown in Table 1. Three of the patients were distension of the inferior vena cava that does not diminish with therefore treated empirically with anti-tuberculous therapy. inspiration, which is a manifestation of the elevated venous The difficulty in establishing a bacteriological or histological pressure in tamponade.6 In addition, there can be marked diagnosis is foremost among unresolved issues in patients reciprocal respiratory variation in mitral and tricuspid flow with pericarditis.14 A definite or proven diagnosis is based on velocities. Tricuspid flow increases and mitral flow decreases demonstration of tubercle bacilli in the pericardial fluid or on during inspiration (the reverse in expiration). histological section of the pericardium. A probable or presumed A challenging differential diagnosis is endomyocardial fibrosis, diagnosis is based on proof of tuberculosis elsewhere in a a common form of restrictive cardiomyopathy (RCM) in Africa.7 patient with otherwise unexplained pericarditis, a lymphocytic 210 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

pericardial exudate with elevated biomarkers of tuberculous was performed in seven patients between 13 days and four infection, and/or appropriate response to a trial of anti- months after pericardiocentesis owing to the persistence of tuberculosis chemotherapy. severe right heart failure. The diagnoses in these seven patients The diagnostic difficulty is best demonstrated by a recent were idiopathic pericarditis in four, radiation pericarditis in one, series of patients with tuberculous pericarditis where most tuberculous pericarditis in one, and postsurgical pericarditis in patients were treated on clinical grounds, with microbiological one. evidence of tuberculosis obtained in only 13 (7.0%) patients.4 The patients in our limited series, as in others cases due to Hence, the focus currently is on indirect tests for tuberculous tuberculosis, usually had attendant pericardial calcification with infection, including ADA levels and more importantly, lysozyme no room for improvement without pericardectomy. This partly or IFN-γ assay, which appears to hold promise for reaching explains the need for pericardectomy in these patients. diagnosis of cases arising secondary to tuberculosis.14-18 Technical and financial constraints may, however, limit the diagnostic utility of IFN-γ in many developing countries.1 These tests are Decision based on timing of presentation and currently not available in our centre. response to medication The importance of recognising the haemodynamic syndrome Related to aetiology is the timing of presentation. Transient of tamponade and constriction characteristic of effusive– sub-acute effusive–constrictive pericarditis is known to resolve constrictive pericarditis lies in an acknowledgment of the after pericardiocentesis without the need for pericardiectomy.3,23,24 contribution of the visceral layer of the pericardium to the In fact in two of three patients with idiopathic pericarditis who pathogenesis of constriction and of the need to remove it surgically. had resolution of their symptoms following pericardiocentesis in However, not only is it sometimes surgically challenging to do the Sagrista-Sauleda series, the onset of their illness was stated an epicardectomy in some patients due to a flimsy, fibrinous to be very recent. The monitoring of intra-cardiac and intra- visceral pericardium with attendant risk of haemorrhage; some pericardial pressures as part of a pericardiocentesis procedure patients may recover with medical treatment alone – so-called has been suggested in patients who present with a sub-acute transient effusive–constrictive pericarditis.3,19 Three of the course of pericardial tamponade, particularly those in whom the patients in this series actually had intra-operative haemorrhage condition is idiopathic or is related to infection, neoplasm or from atrial or ventricular injury during the epicardectomy part rheumatological disease.2 of the procedure. The duration of pericardial disease in three of our patients Visceral pericardiectomy is therefore a much more difficult was more than two years, suggesting chronicity and need for and hazardous procedure than parietal pericardiectomy, but pericardectomy. Although the duration in the fourth and fifth it is necessary for a good clinical result in cases of effusive– patients was relatively short, non-resolution of their symptoms constrictive pericarditis. The clinical decision as to which and presence of pericardial calcification in the fourth patient patients need to be observed on medical treatment depends on appeared to be a predictor of need for pericardial stripping. presumed or confirmed aetiology, timing of presentation, and response to medical therapy.

Effusive–constrictive pericarditis diagnosed? Decision based on aetiology • Based on radiological finding of cardiomegaly Causes of effusive–constrictive pericarditis are varied and • Evidence of constrictive physiology on echocardiography usually practice-dependent. Tuberculosis is said to be responsible for approximately 70% of cases of large pericardial effusion and most cases of constrictive pericarditis in developing countries. Clinical or echo findings of tamponade? However, in industrialised countries, tuberculosis accounts for only 4% of cases of pericardial effusion and an even smaller Yes No proportion of instances of constrictive pericarditis.14 Series from Europe and North America report a predominance of idiopathic cases, followed by cases that occur after radiotherapy or cardiac Pericardiostomy + Duration less than 1 year surgery, or as a result of neoplasia or tuberculosis.3,11,20 pericardial biopsy The aetiological spectrum indeed reflects the general Yes No aetiological spectrum of pericardial diseases in each area and can be influenced by the changing aetiological spectrum of pericarditis in general and constrictive pericarditis in Medical treatment +/– particular.3,21,22 The varying aetiological spectrum impacts on the pericardiocentesis need for and timing of pericardiectomy.17 In the Sagrista-Sauleda series, pericardiectomy was not Tuberculous/calcification/ performed in eight of 15 patients; in five of them owing to a poor persistent symptoms/ general prognosis (four patients with neoplastic pericarditis) elevated venous pressures or a high surgical risk (one patient with radiation pericarditis), and in three patients (all with idiopathic pericarditis) because Pericardiectomy including visceral pericardium of progressive improvement and eventually resolution of the illness after pericardiocentesis. Wide anterior pericardiectomy Fig. 5. Potential algorithm for the management of effu- sive–constrictive pericarditis. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 211

Management BMC Infect Dis 2006; 6: 2. 5. Knopf WD, Talley JD, Murphy DA. An echo-dense mass in the peri- One can propose a management algorithm from the above cardial space as a sign of left ventricular free wall rupture during acute discussion (Fig. 5). We would suggest pericardiocentesis followed myocardial infarction. Am J Cardiol 1987; 59: 1202. by pericardiostomy and pericardial biopsy for bacteriology and 6. The Task Force on the Diagnosis and Management of Pericardial histology as a first step in patients with tamponade or imminent Diseases of the European Society of Cardiology. Guidelines on tamponade. Duration of illness should be the next guide in the Diagnosis and Management of Pericardial Diseases Executive those without tamponade, with those patients with duration Summary. Eur Heart J 2004; 25: 587–610. 7. Bukhman G, Ziegler J, Parry E. Endomyocardial fibrosis: still a more than one year offered pericardiostomy and biopsy. Other mystery after 60 years. PLoS Negl Trop 2008; 2: e97. doi:10.1371/ patients could be tried on medical treatment for six to eight journal.pntd.0000097. weeks and operated on when there is persistent evidence of 8. Wynne J, Braunwald E. Cardiomyopathy and myocarditis. In: Fauci constriction. Presence of pericardial thickening with calcification AS, Braunwald E, Kasper DL, Hauser SL, Longo DL, Jameson JL, following pericardiocentesis is an absolute indicator of need for Loscalzo J (eds). Harrison’s Principles of Internal Medicine, 2008, ch a pericardiectomy. This can be further confirmed on a cardiac 23: 17e. http://www.accessmedicine.com.myaccess.library.utoronto.ca/ content.aspx?aID=2902533. CT scan. 9. Hoit BD. Management of effusive and constrictive pericardial heart We believe this management algorithm is preliminary at best disease. Circulation 2002; 105: 2939–2942. and is subject to improvement with more collaborative research. 10. Russell JBW, Syed FF, Ntsekhe M, Mayosi BM, Moosa S, Tshifularo M, The current on-going multicentre study on the role of steroids in Smedema JP. Tuberculous effusive-constrictive pericarditis. Cardiovasc the prevention of constrictive pericarditis, involving centres in J Afr 2008; 19: 200–201. South Africa, Nigeria and other African countries, is one such 11. Fowler NO. Tuberculous pericarditis. J Am Med Assoc 1991; 266: study.4 Other studies could focus on influence of aetiology and 99–103. 12. Gooi HC, Smith JM. Tuberculous pericarditis in Birmingham. Thorax duration of pericardial disease on the need for pericardiectomy 1978; 33: 94–96. in other areas. 13. Trautner BW, Darouiche RO. Tuberculous pericarditis: Optimal diagno- sis and management. Clin Infect Dis. 2001; 33: 954–961. 14. Mayosi BM, Syed FF. A modern approach to tuberculous pericarditis. Conclusion Prog Cardiovasc Dis 2007; 50: 218–236. Effusive–constrictive pericarditis as a subset of pericardial 15. Aggeli C, Pitsavos C, Brili S, Hasapis D, Frogoudaki A, Stefanadis C, et disease deserves closer study and individualisation of treatment. al. Relevance of adenosine deaminase and lysozyme measurements in Evaluating patients suspected of having the disease affords the diagnosis of tuberculous pericarditis. Cardiology 2000; 94: 81–85. 16. Burgess LJ, Reuter H, Carstens ME, Taljaard JJ, Doubell AF. The use clinicians the opportunity to integrate clinical features and of adenosine deaminase and interferon-gamma as diagnostic tools for non-invasive investigations with or without findings at tuberculous pericarditis. Chest 2002; 122: 900–905. pericardiostomy to expeditiously arrive at a patient-specific 17. Reuter H, Burgess L, van Vuuren W, Doubell A. Diagnosing tubercu- management plan. The limited number of patients in this series is lous pericarditis. Q J Med 2006; 99: 827–839. a limitation, which calls for caution in generalisation. Hence our 18. Tuon FF, Litvoc MN, Lopes MI. Adenosine deaminase and tuberculous aim was to increase the sensitivity of others to issues raised and pericarditis-A systematic review with meta-analysis. Acta Tropica 2006; 99: 67–74. help spur on further collaborative studies to lay down guidelines 19. Oh JK, Hatle LK, Mulvagh SL, Tajik AJ. Transient constrictive pericar- with an African perspective. ditis: diagnosis by two-dimensional Doppler echocardiography. Mayo Clin Proc 1993; 68: 1158–1164. 20. Hancock EW. Subacute effusive-constrictive pericarditis. Circulation References 1971; 43: 183–192. 1. Mayosi BM, Burgess LJ, Doubell AF. Tuberculous pericarditis. 21. Cameron J, Oesterle SN, Baldwin JC, Hancock EW. The etiologic Circulation 2005; 112: 3608–3616. spectrum of constrictive pericarditis. Am Heart J 1987; 113: 354–360. 2. Hancock EW. A clearer view of effusive-constrictive pericarditis. 22. Ling LH, Oh JK, Schaff HV, Danielson GK, Mahoney DW, Seward Circulation 2004; 350: 435–437. JB, et al. Constrictive pericarditis in the modern era: evolving clinical 3. Sagristà-Sauleda J, Angel J, Sánchez A, Permanyer-Miralda G, Soler- spectrum and impact on outcome after pericardiectomy. Circulation Soler J. Effusive-constrictive pericarditis. N Engl J Med 2004; 350: 1999; 100: 1380–1386. 469–475. 23. Woods T, Vidarsson B, Mosher D, Stein JH. Transient effusive-constric- 4. Mayosi BM, Wiysonge CS, Ntsekhe M, Volmink JA, Gumedze F, tive pericarditis due to chemotherapy. Clin Cardiol 1999; 22: 316–318. Maartens G, et al. Clinical characteristics and initial management of 24. Tanaka K, Kawauchi M, Murota Y, et al. Reversible subacute effusive- patients with tuberculous pericarditis in the HIV era: the Investigation constrictive pericarditis after correction of double-chambered right of the Management of Pericarditis in Africa (IMPI Africa) registry. ventricle: a case report. J Cardiol 2002; 39: 267–270. 212 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Risk factors for the development of pneumonia post cardiac surgery AE TOPAL, MN EREN

Abstract the development of nosocomial pneumonia post cardiac surgery Objectives: Postoperative pneumonia is a devastating compli- and thus contribute to decreasing the incidence of pneumonia by cation after cardiac surgery that increases morbidity and identifying preventable risk factors. mortality. The objective of this study was to identify potential risk factors for the development of nosocomial pneumonia Methods post cardiac surgery by the way of logistic regression analysis. This retrospective study was performed on the last 162 patients Design: Data of the last 162 patients undergoing cardiac who underwent cardiac surgery (coronary artery bypass graft surgery before November 2009 were retrospectively collected surgery, valve-replacement surgery) at our reference centre and analysed. up to November 2009. The exclusion criteria were usage of Results: The mean age of the patients was 65.57 ± 10.48 years immunosuppressive agents and an identifiable infection prior to and 83 (51%) were male. Postoperative pneumonia was surgery. diagnosed in 21 (13%) patients. The mean remaining time All patients received standardised anaesthetic management. in the intensive care unit and mean length of hospitalisa- In the operating room, leads II and V5 on the electrocardiogram tion were longer for patients with postoperative pneumonia. (ECG) and arterial blood pressure were continuously monitored. Pre-operative heart rate, previous diabetes mellitus, previ- Anaesthesia was induced with intravenous midazolam (0.03–0.07 ous chronic obstructive pulmonary disease, postoperative mg/kg), sufentanil (1.5–3.0 mg/kg) and rocuronium bromide urea, creatinine and potassium levels, extubation time, (0.9 mg/kg), and maintained with sevoflurane (0.8–1.5%) and postoperative atrial fibrillation, and number of units of continuous infusion of sufentanil (0.5–1.5 mg/kg/h). transfused packed red blood cells (pRBC) and fresh frozen All surgical procedures were performed through a median plasma were associated with higher occurrence of postopera- sternotomy. All patients included in the study received tive pneumonia on univariate analysis. prophylactic administration of intravenous cefazolin peri- Conclusions: On logistic regression analysis, pRBC transfu- operatively (1 g intravenously 30 minutes prior to the first sion, previous chronic obstructive pulmonary disease and incision, every eight hours during surgery and postoperatively postoperative atrial fibrillation remained as independent for three days). predictors for the development of postoperative pneumonia. Pneumonia was considered clinically present as a new Keywords: cardiac surgery, pneumonia, atrial fibrillation, trans- radiographic pulmonary infiltrate, consolidate, cavitation or fusion, chronic obstructive pulmonary disease pneumatocele in the presence of the following conditions: fever (> 38°C) without other recognised causes, leucocytosis (> Submitted 29/7/10, accepted 17/1/12 12 000/μl) or leucopenia (< 4 000/μl) and new-onset purulent Cardiovasc J Afr 2012; 23: 212–215 www.cvja.co.za sputum with a Gram-positive stain finding. DOI: 10.5830/CVJA-2012-005 Possible risk factors and outcomes associated with pneumonia post cardiac surgery were analysed, including pre-operative Despite the progress made in surgery and anaesthesia, the risk of variables [age, gender, heart rate, mean blood pressure, body developing nosocomial infections remains a real threat as more surface area, urea, creatinine and potassium levels, co-morbidities, patients of greater age and with more co-morbidities are operated NYHA class, and left ventricular ejection fraction (LVEF)], on.1 Particularly cardiac surgery creates a high risk for the operative variables [on/off pump surgery, cross-clamp time, development of hospital infections and among these, pneumonia cardiopulmonary bypass (CPB) time, total operation time, and plays an important role as it increases morbidity and mortality by need for intra-operative inotropic support], and postoperative causing pulmonary dysfunction or multi-organ failure. variables [extubation time, chest tube drainage, number of Patients undergoing cardiovascular operations are currently units of transfused packed red blood cells (pRBC) and fresh older and with serious co-morbid disease. Compared to their frozen plasma (FFP), urea, creatinine and potassium levels, and younger counterparts, heart surgery in elderly patients has been postoperative atrial fibrillation (AF)]. implicated in the higher risk of mortality and recurrent pulmonary complications.2 Moreover, emergence of antibiotic-resistant Statistical analysis pathogens increases the incidence of refractory pneumonia. The aim of our study was to identify potential risk factors for The normality of the variables was analysed by Kolmogorov– Smirnov test. Continuous variables are presented as means with standard deviations and were compared among groups using Cardiovascular Surgery Department, Dicle University the Student’s t-test or Mann-Whitney U-test when appropriate Medical Faculty, Diyarbakır,Turkey (non-parametric data). Dichotomous variables are presented AE TOPAL, MD, [email protected], [email protected] as percentages and were compared among groups using a MN EREN, MD Chi-square or Fisher exact test when appropriate. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 213

All variables showing an association (p ≤ 0.05) with Results pneumonia post cardiac surgery were then entered into a The study group comprised 162 patients who underwent cardiac forward stepwise multivariate logistic regression model. A surgery. The mean age of the patients was 65.57 ± 10.48 years two-sided p-value < 0.05 was considered significant in the (range 43–84 years), and 83 (51%) were male. Of 162 operations, multivariate logistic regression model. Adjusted odds ratios 140 were coronary artery bypass graft (CABG) surgery, and 22 (AORs), 95% confidence intervals (CIs), and two-tailed p-values patients underwent valve replacement surgery. Sixteen CABG were calculated for all variables retained in the multivariate operations were performed without cardiopulmonary bypass logistic regression model. Statistical analyses were carried out (CPB). using the statistical packages for SPSS 15.0 for Windows (SPSS Before surgery, 20 patients were in New York Heart Inc., Chicago, IL, USA). Association (NYHA) functional class I, 101 patients were in class II, and 41 were in class III. Pre-operative co-morbid diseases were diabetes in 106 patients, hypertension in 111 patients, chronic obstructive pulmonary disease (COPD) in 43 TABLE 1. EFFECT OF PATIENTS’ CHARACTERISTICS patients, peripheral artery disease in 10 patients and AF in 29 AND PERI-OPERATIVE VARIABLES ON DEVELOPMENT OF PNEUMONIA POST CARDIAC SURGERY patients. Ninety-one (56.2%) patients were tobacco users. The patients’ characteristics and peri-operative variables are shown Patients without Patients with in Table 1. pneumonia pneumonia Postoperative pneumonia was detected in 21 (13%) patients. (n = 141) (n = 21) p-value Mean remaining time in the intensive care unit and mean length Male, n (%) 72 (51.1) 11 (52.4) 0.911** of hospitalisation were longer for patients with postoperative Age (years) 65.3 ± 10.4 67.5 ± 11.0 0.362* pneumonia compared to the patients without postoperative Pre-operative variables pneumonia (4.5 ± 2.7 vs 3.1 ± 1.1 days, p < 0.001; 13.1 ± 9.4 NYHA class, n (%) 0.889** vs 8.8 ± 4.3 days, p = 0.001). There was no difference between I 17 (12.1) 3 (14.3) CABG and valve-replacement surgery regarding postoperative II 89 (63.1) 12 (57.1) development of pneumonia (p = 0.435). III 35 (24.8) 6 (28.6) Pre-operative heart rate was related to postoperative incidence Ejection fraction (%) 50.5 ± 8.7 47.6 ± 9.4 0.170* of pneumonia (p = 0.047). The percentage of patients with Heart rate (/min) 92.2 ± 6.3 89.2 ± 7.4 0.047* previous COPD and diabetes was greater in the group with Mean blood pressure (mmHg) 91.7 ± 9.7 88.3 ± 6.9 0.130* postoperative pneumonia. The remaining patients’ characteristics Body surface area (m2) 1.7 ± 0.1 1.7 ± 0.2 0.242* regarding pre-operative variables were similar between the Urea (mg/dl) 40.2 ± 14.3 44.7 ± 13.1 0.174* groups. Creatinin (mg/dl) 1.0 ± 0.2 1.0 ± 0.2 0.403* Whereas none of the intra-operative variables had any effect on development of pneumonia, many postoperative variables Potassium (mmol/l) 4.3 ± 0.4 4.2 ± 0.5 0.855* were significant risk factors. In patients with postoperative Hypertension, n (%) 97 (68.8) 14 (66.7) 0.845** pneumonia, intubation time was longer, postoperative urea, Hyperlipidaemia, n (%) 89 (63.1) 14 (66.7) 0.753** creatinine and potassium levels were higher, more chest tube Tobacco usage, n (%) 79 (56) 12 (57.1) 0.924** drainage was encountered, and the need for transfusion of pRBC Peripheral arterial disease, n (%) 9 (6.4) 1 (4.8) 0.774** and FFP was increased. Atrial fibrillation, n (%) 25 (17.7) 4 (19) 0.884** All variables showing an association (p ≤ 0.05) with COPD, n (%) 27 (19.1) 16 (76.2) < 0.001** occurrence of postoperative pneumonia were then entered into Diabetes mellitus, n (%) 88 (62.4) 18 (85.7) 0.037** a forward stepwise multivariate logistic regression model. The Intra-operative variables following variables were included in the multivariate model: Off pumpn, n (%) 13 (9.2) 3 (14.3) 0.469** pre-operative heart rate, previous diabetes, previous COPD, Cross-clamp time (min) 36.2 ± 10.8 37.8 ± 9.8 0.552* postoperative urea, creatinine and potassium levels, extubation CPB time (min) 60.4 ± 16.9 61.2 ± 15.0 0.859* time, number of transfused FFP units, number of transfused Total operation time (min) 114.9 ± 20.3 114.8 ± 15.0 0.971* pRBC units and postoperative AF. Upon logistic regression Need for inotropic support, n (%) 30 (21.3) 8 (38.1) 0.091** analysis of these risk factors, pRBC transfusion, previous COPD Postoperative variables and postoperative AF remained as independent predictors for the Extubation time (hour) 7.5 ± 2.8 25.0 ± 21.3 < 0.001* development of pneumonia post cardiac surgery (Table 2). Chest tube drainage (ml) 610.6 ± 733.3 ± 0.069* 286.0 287.4 Units of transfused FFP 2.9 ± 1.5 4.8 ± 3.3 < 0.001* Units of transfused pRBC 5.8 ± 1.6 10.8 ± 3.3 < 0.001* TABLE 2. THE OUTCOMES OF FORWARD STEPWISE BINARY LOGISTIC REGRESSION AND ODDS RATIO Urea (mg/dl) 45.2 ± 15.7 57.8 ± 21.6 0.001* Variables β SE Wald OR (95% CI) p-value Creatinine (mg/dl) 1.1 ± 0.4 1.3 ±.5 0.009* pRBC transfusion 0.910 0.220 17.131 2.484 (1.614–3.821) < 0.001 Potassium (mmol/l) 3.7 ± 0.7 4.1 ± 1.0 0.008* Previous COPD 3.026 0.932 10.530 20.613 (3.315–128.191) 0.001 Atrial fibrillation, n (%) 25 (17.7) 13 (61.9) < 0.001** Postoperative AF 1.732 0.855 4.100 5.653 (1.057–30.228) 0.043 *Student’s t-test, **Fisher’s exact test. COPD: chronic pulmonary obstructive disease, CPB: cardiopulmonary pRBC: packed red blood cells, COPD: chronic obstructive pulmonary bypass, FFP: fresh frozen plasma, pRBC: packed red blood cells. disease, AF: atrial fibrillation. 214 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Discussion to chemotactic stimuli and produce more prostaglandin E2, Although cardiac surgery-related mortality has substantially resulting in decreased activity of antigen-presenting cells and the 3 reduced due to advances in surgical techniques and peri- production of interleukin 2. operative care, the incidence of pneumonia post cardiac surgery There is an association between the length of storage of is still high, varying between 1.5 and 21% in most series.3-8 Hortal transfused red blood cells and the development of postoperative 3,8 et al.9 reported a 45.9% incidence of pneumonia in the sub-group pneumonia, which is seen more rarely among patients with of patients needing mechanical ventilation for longer than 48 fresh transfusions. Various immunosuppressive hours. This wide range in the incidence rates was attributed to substances are released from white blood cell granules into the difference in the characteristics of the study population and red blood cell components during blood storage, contributing 8 the diagnostic criteria used to define nosocomial pneumonia.3 to transfusion-induced immunomodulation. Furthermore, the Surgical technique plays an important role in the occurrence deleterious effect of stored blood may be due to depleted levels of nosocomial infections. For instance, inadequate haemostasis of 2,3-diphosphoglycerate and decreased deformability of stored 3 can lead to hypovolaemia, resulting in an increased need for red blood cells, both impairing oxygen delivery to the tissues. blood transfusion, inotropic support, duration of surgery, or even Recently, the involvement of inflammation in atrial fibrillation possible re-operation. However, besides the surgical technique, has been documented, and high levels of pro-inflammatory several risk factors for pneumonia post cardiac surgery have proteins, such as C-reactive protein, have been suggested been identified: age,4,9 unnecessary use of broad-spectrum to promote the persistence of atrial fibrillation by inducing antibiotics,5,10,11 duration of mechanical ventilation,4-6,9,12,13 CPB structural and/or electrical remodelling of the atria. Atrial time,3,9 re-intubation,3,4,9 emergency surgery,4,5,9 intra-operative biopsies taken from patients in AF have also demonstrated inotropic support,9 and pre-operative renal dysfunction.14 evidence of inflammatory infiltrates within the atrial tissue, with In univariate analysis of our study, duration of mechanical evidence of oxidative damage or occult myocarditis, even among 20 ventilation had a significant effect on postoperative pneumonia, persons who were thought to have had lone AF. whereas age, gender, CPB time and need for inotropic support The fact that inflammation plays an important role in the had no association with pneumonia. Postoperative but not development of both AF and pneumonia may explain the pre-operative high creatinine and urea levels, indicating renal concomitance of these complications after cardiac surgery. It dysfunction, were more common among patients with pneumonia is classic knowledge that pneumonia is one of the non-cardiac compared to those without pneumonia. However, multivariate causes of AF, predominantly in elderly patients. However, in our analysis depicted only prior COPD, transfusion of pRBC and study, AF was an independent risk factor for pneumonia post postoperative atrial fibrillation as independent risk factors for cardiac surgery. pneumonia. In patients with AF, contraction of the ventricles averages COPD was reported to cause postoperative pneumonia in 150/minute. At that rate, the ventricles may not have enough another series.5 Nosocomial pneumonia is a frequent event in the time to fill maximally with blood before the next contraction, course of acute exacerbation of COPD. There is clear evidence particularly without the normal contraction of the atria.21 that in up to 50% of stable COPD patients, the lower airways are Moreover, the contractility of the ventricle decreases after CPB. colonised by potential pulmonary pathogens. Advanced age and Therefore AF decreases the amount of blood pumped by the severity of lung disease are strongly associated with increased ventricles and the body begins to compensate by retaining fluid, risk for pneumonia. resulting in the accumulation of fluid in the lungs. Alveolar Cardiac surgery, especially CPB, aggravates COPD. Moreover oedematous fluid is a good culture medium for the development the use of inhaled corticosteroids among patients with COPD of secondary pneumonia. CPB contributes to this process by significantly increases the risk of developing pneumonia.15,16 aggravating the pulmonary oedema and inflammation.22 According to Lomas,17 inhaled corticosteroid use for at least Also, AF may play an active role in the development 24 weeks is associated with a 60 to 70% higher relative risk of of postoperative pneumonia by prolonging the postoperative pneumonia. Corticosteroid use before elective cardiac surgery intubation time. However, postoperative AF has not been may be limited or at least the dose may be decreased in order to suggested as a cause of pneumonia post cardiac surgery in any decrease the incidence of pneumonia. In addition, immunisation previous reports. Our findings must be confirmed in larger series against influenza in older patients with COPD is associated with and it must be clarified whether pneumonia is only a cause or a 52% reduction in hospitalisations for pneumonia.18 also a consequence of AF. The second independent risk factor for the occurrence The main limitations in our study were the retrospective nature of pneumonia following cardiac surgery was the need for of the analysis and the small sample size, affecting particularly blood transfusion, consistent with previous reports.3-5,8,9 Blood the subgroup with postoperative pneumonia. Therefore any claim transfusions may cause transient immune suppression, thus about an associative relationship between pre- or peri-operative increasing the susceptibility to infection. It was found to be an variables and outcomes should be viewed with caution. The independent risk factor of deep sternal wound infections.19 study group was also not treated entirely uniformly, as off-pump The mechanism of the immunomodulatory effect of allogenic CABG patients were included. However, peri-operative care blood transfusion remains elusive. The infusion of foreign was standardised, which serves to strengthen conclusions on the antigens in either soluble or cell-associated form has been shown results. to induce immune suppression, anergy and clonal deletion, most likely mediated by allogenic white blood cells.8 Another concern is the altered function of macrophages. Conclusion After transfusion macrophages lose migratory ability in response On logistic regression analysis, pRBC transfusion, previous AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 215

chronic obstructive pulmonary disease and postoperative 10. Kollef MH, Morrow LE, Niederman MS, Leeper KV, Anzueto A, atrial fibrillation remained as independent predictors for the Benz-Scott L, et al. Clinical characteristics and treatment patterns development of postoperative pneumonia. AF particularly among patients with ventilator-associated pneumonia. Chest 2006; 129: 1210–1218. should be investigated in future series as a cause of pneumonia. 11. Soo Hoo GW, Wen YE, Nguyen TV, Goetz MB. Impact of clinical guidelines in the management of severe hospital-acquired pneumonia. References Chest 2005; 128: 2778–2787. 1. Segers P, de Mol BA. Prevention of ventilator-associated pneumonia 12. Morrow BM, Argent AC, Jeena PM, Green RJ. Guideline for the after cardiac surgery: prepare and defend! Intensive Care Med 2009; diagnosis, prevention and treatment of paediatric ventilator-associated 35: 1497–1499. pneumonia. S Afr Med J 2009; 99: 253–267. 2. Cheitlin M, Ziper D. Cardiovascular disease in the elderly. In: 13. Tang CW, Liu PY, Huang YF, Pan JY, Lee SSJ, Hsieh KS, et al. Braunwald E, Zipes D, Libby P. (eds). Heart Disease. 6th edn. WB. Ventilator-associated pneumonia after pediatric cardiac surgery in Saunders Company, 2001: 2019–2037. southern Taiwan. J Microbiol Immunol Infect 2009; 42: 413–419. 3. El Solh AA, Bhora M, Pineda L, Dhillon R. Nosocomial pneumonia 14. Thakar CV, Yared JP, Worley S, Cotman K, Paganini EP. Renal dysfunc- in elderly patients following cardiac surgery. Respir Med 2006; 100: tion and serious infections after open-heart surgery. Kidney Int 2003; 729–736. 64: 239–246. 4. Bicer Y, Simsek S, Yapici N, Aydin O, Sogut F, Aykac Z. Risk factor 15. Joo MJ, Au DH, Fitzgibbon ML, Lee TA. Inhaled corticosteroids and analysis of pneumonias developing after open heart surgery. Crit Care risk of pneumonia in newly diagnosed COPD. Respir Med 2010; 104: 2005; 9: 10. 246–252. 5. Kinlin LM, Kircher C, Huiling Z, Daley J, Fisman DN. Derivation 16. Rees PJ. Review: inhaled corticosteroids do not reduce mortality but and validation of a clinical prediction rule for nosocomial pneumonia increase pneumonia in COPD. Evid Based Med 2009; 14: 74. after coronary artery bypass graft surgery. Clin Infect Dis 2010; 50: 17. Lomas C. Inhaled corticosteroids raise risk of pneumonia in patients 493–501. with COPD. Nursing Times 2009; 105: 7. 6. Puzio J, Kucewicz E, Siola M, Dworniczak A, Wojarski J, Zeglen S, 18. Nazir SA, Al-Hamed MM, Erbland ML. Chronic obstructive pulmo- et al. Atypical and opportunistic pulmonary infections after cardiac nary disease in the older patient. Clin Chest Med 2007; 28: 703–715. surgery. Anestezjol Intens Ter 2009; 41: 41–45. 19. Steingrimsson S, Gottfredsson M, Kristinsson KG, Gudbjartsson T. 7. Fukui T, Manabe S, Shimokawa T, Takanashi S. Incidence and Deep sternal wound infections following open heart surgery in Iceland. outcomes of pneumonia after isolated off-pump coronary artery bypass A population-based study. Scand Cardiovasc J 2008; 42: 208–213. grafting. Heart Surg Forum 2009; 12: 194–198. 20. Boss CJ, Lip GYH. The role of inflammation in atrial fibrillation. Int J 8. Vamvakas EC, Carven JH. Exposure to allogeneic plasma and risk of Clin Pract 2005; 59: 870–872. postoperative pneumonia and/or wound infection in coronary artery 21. Shehab A. Heart rate variability as an indicator of left ventricular bypass graft surgery. Transfusion 2002; 42: 107–113. systolic dysfunction. Cardiovasc J Afr 2009; 20: 278–283. 9. Hortal J, Giannella M, Perez MJ, Barrio JM, Desco M, Bouza E, et al. 22. Staton GW, Williams WH, Mahoney EM, Hu J, Chu H, Duke PG, et al. Incidence and risk factors for ventilator associated pneumonia after Pulmonary outcomes of off-pump vs on-pump coronary artery bypass major heart surgery. Intensive Care Med 2009; 35: 1518–1525. surgery in a randomized trial. Chest 2005; 127: 892–901.

… continued from page 205 2. Stettler N, Bovet P, Shamlaye H, Zemel BS, Stallings VA, Paccaud F. Prevalence and risk factors for overweight and obesity in children In a subsequent 12-month cohort study, 161 untreated from Seychelles, a country in rapid transition: the importance of early growth. Int J Obes 2002; 26: 214–219. Tanzanian participants who had BP ≥ 160/95 mmHg on four 3. Bovet P, Burnier M, Madeleine G, Waeber B, Paccaud P. Monitoring separate visits were advised to seek healthcare. Twelve months one-year compliance to antihypertension medication in the Seychelles. later, only 34% reported to have attended a healthcare provider Bull WHO 2002; 80: 33–39. and antihypertensive treatment was taken by only 34% at some 4. Bovet P, Perret F, Cornuz J, Quilindo J, Paccaud F. Improved smoking point during and 3% at the end of the 12-month follow up.7 cessation in smokers given ultrasound photographs of their own athero- Mauritius is also part of sub-Saharan Africa although a sclerotic plaques. Prev Med 2002; 34: 215–220. substantial proportion of the population is of Indian descent. 5. Thurston SW, Bovet P, Myers GJ, Davidson PW, Georger LA, Shamlaye C, Clarkson TW. Does prenatal methylmercury expo- Many large cohort studies have been performed there. We sure affect blood pressure in childhood? Neurotoxicology 2007; 28: 8,9 mention just two, as it is not possible to include all of them in 924–930. the context of this letter. 6. Bovet P, Gervasoni JP, Ross AG, Mkamba M, Mtasiwa DM, Lengeler C, Burnier M, Paccaud F. Assessing the prevalence of hypertension in Pascal Bovet, MD, MPH, [email protected], populations: are we doing it right? J Hypertension 2003; 21: 509–517. [email protected] 7. Bovet P, Gervasoni JP, Mkamba M, Balampama M, Lengeler C, University Institute of Social and Preventive Medicine, Paccaud F. Low utilization of health care services following screening Lausanne, Switzerland for hypertension in Dar es Salaam (Tanzania): a prospective longitudi- nal study. BMC Public Health 2008; 8: 407. Conrad Shamlaye, MD, MSc, [email protected] 8. Nyamdorj R, Qiao Q, Söderberg S, Pitkäniemi J, Zimmet P, Shaw J, Ministry of Health, Republic of Seychelles et al. Comparison of body mass index with waist circumference, waist- to-hip ratio, and waist-to-stature ratio as a predictor of hypertension incidence in Mauritius. J Hypertens 2008; 26: 866–870. References 9. Magliano DJ, Söderberg S, Zimmet PZ, Cartensen B, Balkau B, 1. Kengne AP, Ntyintyane LM, Mayosi BM. A systematic overview of Pauvaday V, et al. Mortality, all-cause and cardiovascular disease, over prospective cohort studies of cardiovascular disease in sub-Saharan 15 years in multiethnic Mauritius: impact of diabetes and intermediate Africa. Cardiovasc J Afr 2011; 22: online publication forms of glucose tolerance. Diabetes Care 2010; 33: 1983–1989. 216 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

An audit of pregnant women with prosthetic heart valves at a tertiary hospital in South Africa: a five-year experience B MAZIBUKO, H RAMNARAIN, J MOODLEY

Abstract Submitted 3/11/10, accepted 6/3/12 Background: Cardiac disease in pregnancy is a common Cardiovasc J Afr 2012; 23: 216–221 www.cvja.co.za problem in under-resourced countries and a significant cause DOI: 10.5830/CVJA-2012-022 of maternal morbidity and mortality. A large proportion of patients with cardiac disease have prosthetic mechanical Women with mechanical prosthetic heart valves (MPHV) heart valve replacements, warranting prophylactic antico- are at greater risk of developing complications than those agulation. with cardiac disease without MPHV.1 The main reason is that Aim: To evaluate obstetric outcomes in women with pros- MPHV require lifelong anticoagulation to reduce the high risk thetic heart valves in an under-resourced country. of associated thrombo-embolic complications.1,2 In addition Methods: A retrospective chart review was performed of 61 pregnancy, being a hypercoaguable state, further increases the pregnant patients with prosthetic valve prostheses referred to risk of thrombo-embolic complications and results in a 35% our tertiary hospital over a five-year period. functional deterioration of MPHV.2 It is therefore not surprising Results: Sixty-one (6%) of 1 021 pregnant women with a that complications associated with cardiac valvular disease in diagnosis of cardiac disease had prosthetic heart valves. pregnancy carry with them significant mortality and morbidity, Fifty-nine had mechanical valves and were on prophylactic particularly in under-resourced countries.3-6 anticoagulation therapy, three had stopped their medication Prophylactic anticoagulation treatment options for patients prior to pregnancy and two had bioprosthetic valves. There with MPHV in pregnancy include warfarin, unfractionated were forty-one (67%) live births, two (3%) early neonatal heparin (UH) and low-molecular weight heparin (LMWH).1,7 deaths, 12 (20%) miscarriages and six (10%) stillbirths. These agents are associated with increased maternal and foetal Maternal complications included mitral valve thrombosis complications, treatment failures, high financial costs and (n = 4), atrial fibrillation (n = 8), infective endocarditis (n = potential teratogenic effects.1-4,8 Warfarin usage in the first 6), caesarean section wound haematomas (n = 7), broad liga- trimester of pregnancy, for example is associated with high rates ment haematoma (n = 1) and warfarin embryopathy (n = 4). of congenital malformations and foetal losses.1-4 Therefore many Haemorrhagic complications occurred in five patients and authors suggest that warfarin be replaced by heparin, at least in all five required blood transfusions. the first trimester.1,4-6 Conclusion: Prophylactic anticoagulation with warfarin in High rates of treatment failure and mortality with the use patients with mechanical heart valve prostheses was associ- of UH have also been reported.6 Low-molecular weight or ated with high rates of maternal and neonatal complica- fractionated heparin does not cross the placental barrier and is tions, including significant foetal wastage in the first and not reported to have teratogenic potential. Its use for prophylactic early second trimesters of pregnancy. Health professionals anticoagulation therapy may be preferred in pregnancy.2,7 In providing care for pregnant women with prosthetic heart addition, LMWH has a longer half-life than UH. However, its valves must consistently advise on family planning matters, use in pregnancy is still controversial due to the lack of adequate adherence to anticoagulation regimes and consider the use clinical trials.1,8 of prophylactic anticoagulant regimens other than warfarin, The majority of pregnant patients with MPHV have been particularly during the first trimester of pregnancy. managed at a single tertiary/quaternary facility in Durban by Keywords: prosthetic heart valves, anticoagulation, maternal a multidisciplinary team since 2003. It is therefore likely that and foetal outcomes a single approach to prophylactic anticoagulation was used at this facility. The opportunity was therefore taken to conduct a retrospective chart review of management of patients with MPHV in pregnancy.

Department of Obstetrics and Gynaecology, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Methods Durban, South Africa This was a retrospective study of pregnant patients with MPHV, B MAZIBUKO, MD referred to a tertiary facility for management over a five-year H RAMNARAIN, MD period (January 2005 to December 2009). Ethical and hospital Women’s Health and HIV Research Group, Nelson R permission were obtained from the appropriate authorities. Mandela School of Medicine, University of KwaZulu-Natal, At every patient visit, relevant data were captured on a software Durban, South Africa package (Medicom, Medicom Solutions, India). Baseline J MOODLEY, MD, [email protected] data recorded included demographic obstetric information, AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 217

investigations done and all maternal and foetal complications. prosthethic valves, mean age 24 (15–45) years. Thirteen (21%) All data were captured onto a structured data form. presented in the first trimester; 37 (61%) in the second and 11 The policy states that pregnant women with MPHV receive (18%) in the third trimester. Fifty-three (85%) were aged ≤ 30 heparin in the first trimester, which is switched to warfarin in years and 34 (56%) were primigravidae. The demographic and the second trimester and then replaced by intravenous heparin at relevant clinical details are shown in Table 1.9 In addition to 37 weeks prior to delivery. Heparin is stopped six hours before prosthetic heart valve replacements, five patients had associated elective caesarean section (C/S) or induction of labour and medical conditions, namely systemic lupus erythematous (n = 1), re-started 12 hours post C/S or six hours post vaginal delivery tuberculosis (n = 1), insulin-dependent diabetes mellitus (n = 1), if no bleeding complication has occurred. Warfarin, usually 10 parathryoidism (n = 1) and epilepsy (n = 1). mg, is given on the first day, simultaneously with intravenous The dosage details of prophylactic anticoagulation therapy heparin and the doses adjusted until the INR is 2.5–3 on two are shown in Table 2 (n = 56). Two patients had bioprosthetic consecutive occasions; heparin is then stopped. All patients were valves and were not on any anticoagulation therapy, while three treated according to this policy. with MPHV had stopped anticoagulation of their own accord Descriptive statistics were used and all results are presented as prior to pregnancy. Forty-seven patients had isolated mitral frequencies, means and percentages. valve replacements, 11 had mitral and aortic valve replacements and one an aortic valve replacement for a mean duration of 10 Results (7–21) years. Five patients had thrombotic events. Four with isolated mitral Over the five-year study period, 1 021 hospital records of valve prostheses were found on admission to hospital to have patients with cardiac disease were identified. Sixty-one (6%) had thrombosis on echocardiography. The characteristics of four patients with mechanical heart valve thrombosis are shown in TABLE 1. BASELINE CHARACTERISTICS OF ALL PREGNANT Table 3. Three of these patients had repeat MPHV surgery prior WOMEN WITH PROSTHETIC HEART VALVES to delivery and one at the time of elective C/S. The details of all Characteristics Number (n = 61) valve replacement data are shown in Fig. 1. Maternal age (years) The fifth patient with a thrombotic event resulted in a Mean (range) 24 (16–45 ) maternal death. The brief details are as follows: a 24-year-old Age groups (years) P1G2 presented at 34 weeks gestational age, had a mitral valve 15–20 17 replacement and was on warfarin 5 mg daily. Following the 21–25 23 stabilisation of her INR at a warfarin dose of 2.5 mg for five 26–30 13 to seven days, she complained of severe headache and her INR > 30 8 was 6. The patient was given frozen plasma to stabilise her INR. Parity Shortly thereafter she complained of severe headache and had a P0 34 low Glasgow coma scale (GCS). CT scan revealed a large left P1 17 intracerebral bleed. A post mortem C/S was performed and a P2–3 4 2.6-kg live baby with good Apgar scores was delivered. P0+1 3 Echocardiography was performed in all patients. The mean P0+4 1 ejection fraction in 57 patients was 55% (range: 34–70) and in P1+1 2 four patients < 45%. Five patients showed dilated right ventricle Gestational age (weeks) on admission and right atrium. Another patient showed dilated right atrium; < 14 18 one had an ascending aortic aneurysm and another patient had 14–28 32 an aneurysm of the aortic root. Two patients had secondary 28–38 11 pulmonary hypertension; three had vegetations and were treated History of previous pregnancies for infective endocarditis. There were 41 live births, two of which ended in early Miscarriage 6 neonatal deaths. There were six stillbirths and 12 miscarriages. Intrauterine death 3 The mode of delivery and foetal outcomes are shown in Table 4. Stillbirth (MSB) 2 Neonatal death 2 TABLE 2. DOSAGE DETAILS OF ANTICOAGULATION HIV status USED BY PATIENTS ON PRESENTATION AT THE Negative 43 FIRST ANTENATAL VISIT (n = 56) Positive 16 Dose of anti- Trimester 1 Trimester 2 Trimester 3 coagulation drug (n = 12) n (%) (n = 35) n (%) (n = 9) n (%) CD4 > 200 cells/ml 14 Warfarin CD4 < 200 cells/ml 2 Declined 2 2.5 mg 1 (8) 0 (0) 0 (0) NYHA functional class 5 mg 6 (50) 23 (66) 5 (56) 1 49 7.5 mg 3 (25) 7 (20) 3 (33) 11 6 10 mg 0 (0) 1 (3) 1 (11) 111 4 40 mg 1 (8) 0 (0) 0 (0) 1V 2 7.5 mg alt 5 mg 1 (8) 2 (6) 0 (0) NYHA – New York Heart Association classification.9 10 mg alt 7.5 mg 0 (0) 2 (6) 0 (0) 218 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

TABLE 3. CHARACTERISTICS OF FOUR PATIENTS WITH MECHANICAL HEART VALVE THROMBOSIS Patient Age GA at 1st Age at valve Mode of Neonatal no (years) antenatal visit Parity surgery Anticoagulant at 1st antenatal visit delivery outcome 1 22 21 1 12 Defaulted on therapy prior to pregnancy NVD IUD 2 25 36 1 15 Warfarin 5 mg C/S Alive 3 21 12 0 11 Warfarin 5 mg NVD Alive 4 18 19 0 9 Defaulted on therapy prior to pregnancy NVD IUD IUD = intrauterine death; NVD = normal vaginal delivery; C/S – caesarean section.

= TABLE 4. DELIVERY DETAILS Prosthetic heart valves (n 61) Percent- Characteristics Number age Anaesthetic Mechanical heart Tissue heart = = Spinal 9 22 valves (n 59) valves (n 2) Epidural 28 68

General anaesthetic 4 10 Position of valve mitral aortic mitral and aortic mitral Live babies (n = 41) replacement (n = 44)* (n = 1) (n = 11) (n = 2) Spontaneous labour Prophylactic yes yes yes no Delivered vaginally 2 (epidural) 40 anticoagulation = Emergency caesarean 3 (1 epidural + 2 GA) 60 Valve thrombosis yes (n 4) no no no Maternal death yes (n = 1) no no no Induced labour *Three patients with MPHV stopped medication of their own accord prior Delivered vaginally 3 (epidural) 50 to pregnancy. Emergency caesarean 2 (2 epidural + GA) 50 Elective caesarean (n = 31) Fig. 1. Flow diagram showing position of prosthetic heart Epidural 21 67 valve replacement. Spinal 9 30 but not from those of affluent societies.11,12 A Canadian study Emergency (GA) 1 (failed spinal) 03 reported that the mean age at first antenatal booking was 32 Stillbirths (n = 6) years.13 This implies that the severity of rheumatic heart disease NVD (spontaneous) 2 33 may be greater in under-resourced countries, warranting valve Induced 4 67 replacement at an early age. Furthermore, rheumatic heart Miscarriages (n = 12) disease is uncommon in affluent countries and congenital Warfarin exposure in 1st trimester (NVD) 12 100 abnormalities form the majority of cardiac conditions seen in GA = general anaesthetic; NVD = normal vaginal delivery. pregnancy.13 The high number of pregnancies at an early age in our study There were four cases of warfarin embryopathy (three may also be due to cultural and socio-economic factors. Such presented for antenatal care in the second trimester and the other factors may have influenced late booking for maternity care, as in the third trimester). The maternal and foetal characteristics 37 (61%) patients attended antenatal care in the second trimester together with the dose of warfarin at admission and sonographic of pregnancy. Late booking for antenatal care and large patient findings are outlined in Table 5. The most common maternal numbers on warfarin throughout the first and second trimesters complications during the antenatal period and immediately post of pregnancy may indicate that women with cardiac disease do delivery are listed in Table 6. not receive adequate and/or consistent information on family planning, contraceptive services, sexually transmitted infections and the hazards of warfarin therapy in the first trimester. Discussion The challenge for health professionals in under-resourced In this retrospective audit on prosthetic heart valves in pregnancy, countries, irrespective of their medical discipline, is to ensure the mean age was 24 years and 56% were primigravidae. The that such information is provided, not only to the individual low mean age and high number of primigravidae are in keeping woman, but also to partners, families and the community at with studies originating from other under-resourced countries,4,10 large. Further, it begs the question whether a family planning

TABLE 5. CONGENITAL ABNORMALITIES DUE TO WARFARIN EMBRYOPATHY Patient Maternal GA (weeks) at 1st Foetal outcomes no age (years) antenatal visit Parity Anticoagulant and dose Sonography – congenital abnormalities of pregnancy 1 20 32 1 Warfarin 7.5 mg Choanal atresia/ microcephaly nasal hypoplasia ENND 2 22 26 1 Warfarin 5 mg/2. 5 mg Hydrocephalus, flattening of nasal bone polyhydraminos SB 3 36 10 2 Warfarin 7.5 mg Skeletal deformity of spine, nasal hypoplasia, hydrocephalus SB 4 30 29 0 Warfarin 5 mg Nasal hypoplasia, mid-facial hypoplasia diaphragmatic hernia ENND ENND = early neonatal death; SB = stillbirth; GA = gestational age. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 219

TABLE 6. MATERNAL COMPLICATIONS DURING THE replacement, an elective C/S was planned because she had ANTEPARTUM AND POSTPARTUM PERIOD severe cephalo-pelvic disproportion and a tight mitral stenosis, Complications Number Percentage requiring valve replacement. Both mother and baby did well. Antepartum These cases illustrate the high perinatal mortality associated with valve replacements during pregnancy. Epistaxis 2° anticoagulation* 3 5 Atrial fibrillation 8 14 The morbidity and mortality associated with anticoagulation with warfarin is also high and is probably due to high doses Infective endocarditis 6 10 of warfarin in the last two trimesters of pregnancy and the Valve thrombosis 4 7 immediate postpartum period. This is illustrated in our study by Warfarin embryopathy 4 7 five cases of thrombosis and the maternal death associated with Postpartum high doses of warfarin. C/S wound haemotomas 7 12 There is no current consensus as to the best approach to Primary postpartum haemorrhage 1 2 * anticoagulation in pregnant women with MPHV,19 as there are no Postpartum haemorrhage 2° anticoagulation* 3 1 large randomised studies to guide decision making.18 In women Data expressed as mean (range) or as number (percentage) with MPHVs, the types of anticoagulation that can be used Patients requiring blood transfusion. * during pregnancy include warfarin, UH and LMWH. Warfarin, a vitamin K antagonist, crosses the placental professional should be attached to cardiac clinics to provide and barrier, is teratogenic and has been associated with an increased reinforce appropriate information and prescribe a wide range of incidence of spontaneous abortion, prematurity, stillbirths and contraceptives where necessary. Similar recommendations have central nervous system developmental disorders.7 Furthermore, been made in the Saving Mothers Reports over the last decade it would appear that the pivotal period for the risk of foetal in South Africa.5 congenital abnormalities is between six and 12 weeks of It is difficult to relate the number of pregnancies to the time gestation, resulting in a 6–10% risk of embryopathy.21,22 There from valve replacement. The average age of patients was 24 are also reports suggesting that warfarin is associated with years, most were primigravidae and the average age of valve intracranial foetal bleeding and an increased incidence of replacement was 12 years. stillbirths.21,23,24 Nevertheless because of its ease of use and In our audit, the majority of patients were fitted with MPHV. monitoring, it is commonly prescribed in most under-resourced The probable reason for this was that MPHV are cheaper and countries. have greater longevity than bioprosthetic heart valves. However, Warfarin is also still used in affluent countries. North et al. because of the propensity of MPHV to undergo thrombosis, reported high foetal waste rates but low valve thrombotic rates in prophylactic anticoagulation is strongly recommended, if not their series of patients with MPHV using prophylactic warfarin mandatory.14-16 Most of our patients with MPHV had either therapy.25 If warfarin dosage does not exceed 5 mg daily, the risk St Jude or Orynx valves. These new-generation MPHV were of foetal warfarin embryopathy is extremely small.7 Vitale et designed to improve blood flow dynamics, prolong longevity and al. studied 58 pregnancies in 43 women with MPHV who took decrease thrombogenicity. However, our findings and those of warfarin ≤ 5 mg or > 5 mg (target INR 2.5–3.5) until delivery. others show that these new-generation MPHV are still associated There were significantly fewer foetal complications in women with thrombo-embolism.17-19 taking ≤ 5 mg warfarin. It was suggested that warfarin at doses There is therefore no doubt that prophylactic anticoagulant ≤ 5 mg to achieve a therapeutic INR may be safe during the first therapy is necessary, but which prophylactic anticoagulation trimester.7 regimen should be used in young women with MPHV, or should In our study, 29 (50%) patients were on warfarin ≤ 5 mg in women requiring heart valve replacement have bioprosthetic the first trimester and did not have congenital foetal anomalies, valves inserted prior to pregnancy? Two patients in our study had which was similar to reports from India, Oman and Lebanon.26,27 bioprosthetic valves. Although patients fitted with these valves However, the four (7%) patients who had embryopathies had do not require prophylactic anticoagulation, as they are less warfarin doses of > 5 mg daily. thrombogenic than MPHV, the main issues with bioprosthetic The risk of miscarriages and stillbirths are also reported to be valves are their cost, limited lifespan and the possibility of an high in women taking warfarin in the first trimester of pregnancy.8 increased risk of structural valve deterioration (SVD) during In our study, we had 41 live births, with two resulting in neonatal pregnancy.20 In addition, serious SVD can require re-surgery deaths, 12 miscarriages and six stillbirths. The miscarriages and during pregnancy to replace failing bioprosthetic valves, and stillbirths were probably related to high dosages of warfarin all such operations are associated with mortality and morbidity. and lack of close monitoring of anticoagulant therapy.21,24 This Elkayam and Bitar stated that about 50% of women of child- highlights the need for close and intensive monitoring of bearing age will require valve replacement owing to SVD seven warfarin during pregnancy, particularly at the time of switching to 10 years after the original operation.20 They also reported SVD from one type of anticoagulant to another in the first trimester in 47% of patients with a history of pregnancy compared with of pregnancy. only 14% in non-pregnant patients.20 In our study four patients Lack of intensive monitoring at the time of switching from required replacements; three prior to delivery. All three were one form of anticoagulant to another is also demonstrated by symptomatic, did not have antenatal care and did not respond the high wound haematoma rate in our study, namely seven to medical treatment. In two of the three cases, the pregnancies cases of wound haematomas, three of which required surgical ended in intrauterine deaths, while the baby in the third case intervention, and one case of broad ligament haematoma, which was born alive. In the fourth patient who had a repeat valve settled on conservative treatment. There were also four cases 220 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

30 TABLE 7. ANTICOAGULATION REGIMEN IN PREGNANT an anti-Xa of 1.0–1.2 U/ml. There were no thrombo-embolic WOMEN WITH MECHANICAL PROSTHETIC VALVES events in this study of 15 women with MHVP. These reports Pre-pregnancy follow a randomised study in South Africa, comparing UH with • Discuss anticoagulation regimen with patient enoxaparin, which was stopped prematurely because of two • Continue warfarin until pregnancy is achieved deaths from thrombo-embolism in the enoxaparin group. Anti- 31 • When menstruation is delayed, perform pregnancy test every few days Xa levels were measured but no dose adjustment was done. until positive or until menstruation (in order to detect pregnancy at an There is evidence that increasing doses of heparin are required early stage) with increasing gestational age because of the increased blood • Give patient and health professional responsible for anticoagulation volume and greater renal clearance as pregnancy progresses.30 written instructions about anticoagulation regimen during pregnancy There was one maternal death in our study. Earlier studies Sixth to 12th week of pregnancy originating from under-resourced countries have reported two • If warfarin daily dose is < 5 mg, consider continuation of warfarin maternal deaths in 312 patients studied,4 one in 229 patients,32 throughout pregnancy and 10 in 480 patients studied.33 • Alternatively, substitute warfarin with subcutaneous LMWH twice There was a high rate of maternal complications in our study. daily Four (7%) patients on warfarin ≥ 5 mg daily developed valve • Adjust LMWH dose to achieve peak anti-Xa levels of 0.7–1.2 U/ml four hours post dose thrombosis in the mitral position. The mitral position is prone to • If trough levels are sub-therapeutic with therapeutic peak levels, dose thrombosis and our audit confirms similar findings from other three times daily studies.31,34 • Check anti-Xa levels twice a month Throbo-embolic events and embryopathy are not the only 13th to 35th week of pregnancy risks that are associated with pregnancies in women with • Resume warfarin MHVP. Atrial fibrillation, infective endocarditis and C/S wound • Or use LMWH adjusted dose haematomas were relatively common complications observed in 36th week of pregnancy our study. • Substitute warfarin with subcutaneous LMWH twice daily • Adjust LMWH dose to achieve peak anti-Xa levels of 0.7–1.2 U/ml Conclusions four hours post dose This study confirms that the use of warfarin throughout pregnancy • If trough levels are sub-therapeutic with therapeutic peak levels, dose three times daily carries a significant risk of embryopathy. This risk may be • Check anti-Xa levels weekly greater with doses of > 5 mg but no definite conclusions can be drawn. In addition, the use of warfarin in the second trimester of pregnancy is associated with significant foetal losses, probably of postpartum haemorrhage (Table 6) following the switching due to poor monitoring and control of warfarin dosages. The of anticoagulants. These cases occurred at C/S, highlighting switching of warfarin to heparin at the time of delivery may be the need for intense monitoring of coagulation indices at associated with maternal complications. this time. McLintock et al. also reported high rates of ante- Recommendations for management of anticoagulation in and postpartum haemorrhagic complications using LMWH pregnant women with MHVP are found in guidelines produced throughout pregnancy.11 by the American College of Cardiology/American Heart A number of studies report that UH and LMWH therapy is Association.35 These guidelines are based on the opinions of safe for the foetus.21,23 Unfractionated heparin does not cross the experts. An anticoagulation regimen reported by Pieper et al. placenta and does not have the potential to cause foetal bleeding has been modified and shown in Table 7;8 and takes into account, or teratogenecity. Heparin is generally considered safer than the key points of the American guidelines. Large randomised warfarin during pregnancy in terms of embryopathy, however trials of dose-adjusted LMWH are necessary before firm the efficacy of heparin in the prevention of thrombo-embolic recommendations on an acceptable prophylaxis anticoagulation complications during pregnancy is contentious. Several reports regimen for prevention of thrombosis of MHVP can be made. indicate that its use is associated with high incidence of thrombo- embolic complications, including fatal valve thrombosis in high-risk pregnant women managed with subcutaneous UH and References LMWH therapy.18,19,22,23 1. Nassar AH, Hobeika EM, Abol Wssarmad HM, et al. Pregnancy Chan et al. reviewed pregnancy outcomes in women with outcome in women with prosthetic heart valves. Am J Obstet Gynecol MPHV and reported thrombo-embolic complications in 3.9% of 2004; 191: 1009–1013. pregnancies using warfarin only; 9.2% in women who received 2. Sbarouni E, Oakley CM. Outcome of pregnancy in women with valve prostheses. Br Heart J 1994; 71: 196–201. UH in the first trimester followed by warfarin, and 33% in 3. Nqayana T, Moodley J, Naidoo DP. Cardiac disease in pregnancy. 14 pregnancies treated with UH heparin throughout pregnancy. Cardiovasc J South Afr 2008; 19(3):145–151. Oran et al. reviewed pregnant women with MPHV managed 4. Malhotra M, Sharma JB, Tripathii R, Arora P, Arora R. Maternal and with LMWH and reported complications related to valve fetal outcome in vavular heart disease. Int J Gynaecol Obstet 2004: thrombosis in 10/81 pregnancies.28 Similarly, another review 84: 11–16. reported thrombotic events in 22% of pregnant women (n = 76) 5. Saving Mothers: A report of the National Committee on Confidential Enquiries into Maternal Deaths: 2005–2007. Pretoria: Department of managed with LMWH.29 Health, 2008. More recently, data are emerging that dose-adjusted LMWH 6. Shannon MS, Edwards MB, Long F, et al. Anticoagulant manage- (enoxaparin) may be administered safely in pregnancy when ment of pregnancy following heart valve replacement in the United there is a dosage adjustment throughout pregnancy to maintain Kingdom, 1986-2002. J Heart Valve Dis 2008; 17(5): 526–532. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 221

7. Vitale N, De Feo M, De Santo LS, Pollice A, Tedesco N, Cotrufo M. Circulation 2003; 107: 1240–1246. Dose-dependent fetal complications of warfarin in pregnant women 24. Cotrufo M, De Feo F, De Santos LS, et al. Risk of warfarin during with mechanical heart valves. J Am Coll Cardiol 1999; 33: 1637–1641. pregnancy with mechanical heart valves. Obstet Gynecol 2002; 99(1): 8. Pieper PG, Balci AP, Van Dijk AP. Pregnancy in women with prosthetic 35–40. valves. Neth Heart J 2008; 16: 406–411. 25. North RA, Sadler I, Stewart AW, McCowan LM, Kerr AR. White HD. 9. Criteria Committee of the New York Association. Nomenclature and Long term survival and valve-related complications in young women Criteria for Diagnosis of Disease of the Heart and Great Vessels, 6th with cardiac valve replacement. Circulation 1999; 99: 2669–2676. edn. Boston: Brown, 1964. 26. Geelani MA, Singh S, Verma A, Nagesh A, Betigeri V, Nigam M. 10. Hameed A, Karaalp IS Tummala PP, et al. The effect of valvular heart Anticoagulation in patients with mechanical valves during pregnancy. disease on maternal and fetal outcome of pregnancy. J Am Coll Cardiol Asian Cardiovasc Thorac Ann 2005; 13: 30–33. 2001; 37: 893–899. 27. Al-Lawati AA, Venkitraman M, Al-Delaime T, Valliathu J. Pregnancy 11. McLintock C, McCowan LME, North RA. Maternal complications and mechanical heart valves replacement: dilemma of anticoagulation. and pregnancy outcomes in women with mechanical prosthetic valves Euro J Cardiothorac Surg 2002; 22: 223–227. treated with enoxaprin. Br J Obstet Gynecol 2009; 16: 1585–1592. 28. Oran B, Lee-Parritz A, Ansell J. Low molecular weight heparin for the 12. Quinn J, Von Kiemperer K, Brooks R, Peebles D, Walker F, Cohen H. prophylaxis of thromboembolism in women with prosthetic mechanical Use of high intensity adjusted dose low molecular weight heparin in heart valves during pregnancy. Thromb Haemost 2004; 92: 747–751. women with mechanical heart valves during pregnancy: a single centre 29. James AH, Brancazio LR, Gehring TR, Wang A, Ortel TL. Low experience. Haematologica 2009; 94(11): 1608–1612. molecular weight heparin for thrombo-prophylaxis in pregnant women 13. Silversides CK, Colman JM, Sermer M, Siu SC. Cardiac risk in preg- with mechanical heart valves. J Matern Fetal Neonatal Med 2006; 19: nant women with rheumatic mitral stenosis. Am J Cardiol 2003; 91: 543–549. 1382–1385. 30. Chitsike RS, Jacobsen BF, Manga P, Rhemtula HA, Moodley S, Toweel 14. Chan WS, Anand S, Ginsberg JS. Anticoagulation of pregnant women GD. A prospective trial showing the safety of adjusted dose enoxaparin with mechanical heart valves: a systemic review of the literature. Arch for thromboprophylaxsis of pregnant women with mechanical pros- Intern Med 2000; 160: 191–196. thetic heart valved. Clin Appl Thromb Hemost 2010, June13. (Epub 15. Caceres-Loriga FM. Prosthetic valve thrombosis: is it time for a new ahead of print). consensus conference? Euro J Echocardiogr 2008; 9: 413–414. 31. Elkayam U, Singh H, Irani A, Akhter MW. Anticoagulation in pregnant 16. Danik S, Fuster V. Anticoagulation in pregnant women with prosthetic women with prosthetic heart valves. J Cardiovasc Pharmacol Ther heart valves (Review). Mt Sinai J Med 2004; 71: 322–329. 2004; 9: 107–115. 17. Kulik A, Rubens FD, Wells PS, Kearon C, Mesana TG, van Berkom 32. Rahaman J, Rahman MS, Al-Suleiman SA, Al-Jama FE. Pregnancy J, Lam BK. Early postoperative anticoagulation after mechanical complicated by maternal cardiac disease: a review of 274 patients. valve replacement: a systemic review. Ann Thorac Surg 2006; 81(2): Obstet Gynecol 2000; 20(3): 243–245. 770–781. 33. Sahwney H, Aggarwal N, Vasishta K, Sharma Y, Grover A. Maternal 18. Warnes CA. Prosthetic heart valves. In: Steer PJ, Gatzoulis MA, Baker and perinatal outcome in rheumatic heart disease. Int J Gynaecol Obstet P, eds. Heart Disease in Pregnancy. London: RCOG, 2006: 157–168. 2003: 81(2): 153–156. 19. Maxwell C, Sermer M. Mechanical heart valves and pregnancy. Fetal 34. Sahnoun-Trabelsi I, Jimenez M, Choussat A, Roudaut R. Prosthetic Matern Med Rev 2007; 18(4): 311–331. valve thrombosis in pregnancy. A single-center study of 12 cases. Arch 20. Elkayam U, Bitar F. Valvular heart disease and pregnancy part 11; pros- Mal Coeur Vaiss 2004; 97: 305–331. thetic valves (Review). J Am Coll Cardiol 2005; 46: 403–410. 35. Bonow RO, Carabello BA, Chatterjee K, de Leon A Jr, Faxon DP, 21. Khamooshi AJ, Kashfi F, Hoseini S, Tababaei MB, Javadpour H, Noohi Freed MD, et al. ACC/AHA 2006 guidelines for the management F. Anticoagulation for prosthetic heart valves in pregnancy. Is there an of patients with valvular heart disease: a report of the American answer? Asian Cardiovasc Thorac Ann 2007; 15(6): 493–496. College of Cardiology/American Heart Association Task Force on 22. Maxwell C, Poppas A, Dunn E, Sermer M. Pregnancy, mechanical heart Practice Guidelines (writing Committee to revise the 1998 guidelines valves and anticoagulation navigating the complexities of management for the management of patients with valvular heart disease) devel- during gestation. In: Rosene-Montella K, Keeley EJ, Barbour LA, Lee oped in collaboration with the Society of Cardiovascular anesthesi- RV, eds. Medical Care of Pregnant Patient. 2nd edn. Philadelphia: ologists endorsed by the Society for Cardiovascular Angiography and American College of Physicians, 2007: 344–355. Interventions and the Society of Thoracic Surgeons. J Am Col Cardiol 23. Hung L, Rahimtoola SH. Prosthetic heart valves and pregnancy. 2006; 48: c1–148.

HEALTHCARE 222 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Review Article

Endothelial dysfunction: the early predictor of atherosclerosis MASHUDU MUDAU, AMANDA GENIS, AMANDA LOCHNER, HANS STRIJDOM

Abstract Atherosclerosis is a chronic progressive vascular disease, Since the discovery in the 1980s that nitric oxide (NO) is in characterised by plaque formation and subsequent fissure, fact the elusive endothelium-derived relaxing factor, it has erosion or rupture of the plaque with thrombosis of the plaque 3 become evident that NO is not only a major cardiovascular surface. A complication of coronary atherosclerosis can be the signalling molecule, but that changes in its bioavailability are development of myocardial ischaemia and ultimately myocardial 4 crucial in determining whether atherosclerosis will develop infarction. Hypertension, tobacco use, dyslipidaemia, diabetes or not. Sustained high levels of harmful circulating stimuli mellitus, physical inactivity and obesity, all of which are associated with cardiovascular risk factors such as diabe- associated with the development of atherosclerosis and IHD, are tes mellitus elicit responses in endothelial cells that appear considered to be the top risk factors for cardiovascular mortality 2 sequentially, namely endothelial cell activation and endothe- worldwide. lial dysfunction (ED). Chronic exposure to cardiovascular risk factors and the ED, characterised by reduced NO bioavailability, is now harmful circulating stimuli associated with these conditions recognised by many as an early, reversible precursor of overwhelms the defense mechanisms of the vascular endothelium, atherosclerosis. The pathogenesis of ED is multifactorial; hence compromising its integrity and ultimately initiating 5 however, oxidative stress appears to be the common under- endothelial dysfunction (ED). Mounting evidence is pointing lying cellular mechanism in the ensuing loss of vaso-active, to ED as one of the major pathophysiological links between inflammatory, haemostatic and redox homeostasis in the exposure to cardiovascular risk factors and the development of 6 body’s vascular system. The role of ED as a pathophysi- atherosclerotic disease (Fig. 1). ological link between early endothelial cell changes associ- ED is commonly associated with reduced nitric oxide (NO) ated with cardiovascular risk factors and the development bioavailability, and hence an inability of the endothelium to of ischaemic heart disease is of importance to basic scientists and clinicians alike. Keywords: endothelium, endothelial dysfunction, nitric oxide EXPOSURE TO CARDIOVASCULAR RISK FACTORS: bioavailability, eNOS uncoupling, oxidative stress, atheroscle- • Hyperglycaemia/insulin resistance (diabetes mellitus) • Smoking rosis • Hypertension Submitted 7/6/11, accepted 11/11/11 • Obesity • Hyperlipidaemia Cardiovasc J Afr 2012; 23: 222–231 www.cvja.co.za DOI: 10.5830/CVJA-2011-068 If sustained and not reversed

Between 1995 and 2004, cardiovascular diseases accounted for DEVELOPMENT OF ENDOTHELIAL CELL CHANGES: about 195 deaths per day in South Africa. Particularly disturbing • Endothelial activation is that cardiovascular mortality is expected to escalate by a • Endothelial dysfunction staggering 41% in the working age group (35–64 years) in the Associated with reduced NO bioavailability; 1 pro-; pro-oxidative stress; South African population by the year 2030. In 2004, the World pro-coagulation; pro-inflammation Health Organisation (WHO) reported cardiovascular diseases/ ischaemic heart disease (IHD) to be the leading cause of death If sustained and not reversed worldwide and that cardiovascular deaths are envisaged to 2 escalate to 23.4 million by the year 2030. PROGRESSION TO ATHEROSCLEROSIS: • Plaque formation and sequelae Department of Biomedical Sciences, Division of Medical • Myocardial ischaemia Physiology, Faculty of Health Sciences, Stellenbosch • Myocarial infarction University, Stellenbosch, South Africa MASHUDU MUDAU, MSc Fig. 1. Exposure of endothelial cells to cardiovascu- AMANDA GENIS, MSc lar risk factors and the resultant pathophysiological AMANDA LOCHNER, PhD changes, i.e. endothelial activation and dysfunction, with HANS STRIJDOM, BMedSc, MB ChB, PhD, [email protected] progression to atherosclerosis if risk-factor exposure is sustained. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 223

initiate in response to vasodilatory stimuli such as effects.12 The vasodilatory state is mediated by factors such as acetylcholine or shear stress. It represents an initial reversible nitric oxide (NO), endothelium-derived hyperpolarising factor step in the development of atherogenesis, and for this reason, (EDHF) and prostacyclins, while a vasoconstrictory state is early clinical identification of ED may become an important tool mediated by factors such as endothelin-1 (ET-1), angiotensin in the prevention or reversal of progression to atherosclerosis II and thromboxane A2.7,12 Of these endothelial-derived factors, and IHD.7 NO, which was originally identified as the endothelial-derived ED comprises a loss of balance between endothelial- relaxing factor (EDRF), has since evoked much interest as it derived vasodilatory and vasoconstrictory factors, where is considered to be the most potent endogenously synthesised the pro-vasoconstrictory state becomes dominant, leading to vasodilator in the body, and a key marker of endothelial function progressive pathophysiological changes. These changes appear and dysfunction. as sequentially occurring responses in endothelial cells, also The vasoactive factors released by endothelial cells and their referred to by some as the endothelial activation–dysfunction– effects are summarised in Table 1. In view of the physiological, injury triad.8,9 Collectively, these endothelial changes exhibit pathophysiological and clinical importance of NO in vascular pro-inflammatory, pro-oxidant, proliferative, pro-coagulation physiology, we have included a brief discussion on the physiology and pro-vascular adhesion features.7,10 of NO below.

The endothelium: a functional organ Nitric oxide The vascular endothelium consists of approximately 1–6 × 1013 The realisation in the 1980s that the identity of EDRF was in fact endothelial cells and accounts for about 1 kg of total body weight. NO was rather astonishing, as NO had until then been perceived For many years after its discovery, the endothelium was believed as nothing more than a toxic environmental pollutant found to be an inert, semi-permeable barrier between circulating in cigarette smoke, exhaust fumes of motor cars and harmful blood and the underlying sub-endothelial tissues.11 However, gases generated by industrial processes.13,14 The ground-breaking extensive research has since revealed a far more complex role discovery that NO is also synthesised in the body and functions for the endothelium. We now know that the endothelium is in as a chemical messenger with important physiological effects fact a metabolically active organ, playing a crucial role in the introduced a novel paradigm in cardiovascular physiology and maintenance of vascular homeostasis by releasing a variety of pathophysiology. In addition to its vasodilatory properties, NO vasoactive factors that can either dilate or constrict the blood was also found to exert anti-inflammatory and cardioprotective vessels, depending on the type of the stimulus.7 effects.15 Vascular homeostasis entails keeping a tightly controlled Owing to its gaseous and free-radical nature, NO is able balance between a vasodilatory state, which is often associated to diffuse easily between cells and tissues and react with a with anti-oxidant, anti-inflammatory and anti-thrombotic effects variety of molecules in the body.13 NO is synthesised from on one hand, and a vasoconstrictory state on the other, which is the amino acid L-arginine by a family of enzymes known as associated with pro-oxidant, pro-inflammatory and pro-thrombotic nitric oxide synthase (NOS).14 The NOS enzyme occurs as

TABLE 1. OVERVIEW OF ENDOTHELIUM-DERIVED VASO-ACTIVE FACTORS Endothelium-derived factors Physiological effects Enzymatic source and mechanism of action Nitric oxide (NO) • Potent vasodilator • Synthesised by the enzymes: eNOS, nNOS and iNOS, • Inhibits inflammation, VSMC proliferation and migra- with eNOS the major endothelial source of NO during tion, platelet aggregation and adhesion, and leukocyte physiological conditions adhesion • Diffuses from endothelial cells to underlying VSMCs • Regulates where it binds to soluble guanylyl cyclase, leading to • Regulates cardiac metabolism a cascade of events that ultimately result in vascular • Cardioprotective during ischaemia–reperfusion injury relaxation

Prostacyclin (PGI2) • Vasodilatory agent • Derived from arachidonic acid by cyclooxygenase-2 • Inhibits platelet aggregation (COX-2) Endothelium-derived • Exerts vasodilatory effects, particularly in small arter- • Its identity is still under suspicion with proposed hyperpolarising factor ies of diameter ≤ 300 μm candidates such as potassium ions and hydrogen (EDHF) peroxide • Causes relation of VSMCs by means of membrane hyperpolarisation Endothelin-1 (ET-1) • A potent vasoconstrictor • Synthesised by endothelin-converting enzyme

• Exerts its effects via two receptors: ETA expressed

on endothelial cells and ETB on VSMCs. ETA recep-

tors promote vasoconstriction, whereas ETB receptors promote NO production and ultimately reduction in ET-1 production

Thromboxane A (TXA2) • A potent vasoconstrictor • Derived from arachidonic acid by COX-1 Angiotensin ll • A potent vasoconstrictor • Synthesised by angiotensin converting enzyme

• Elicits its effects via two receptors: AT1 which promotes vasoconstriction and cell proliferation, and

AT 2 which antagonises the effects of AT1 224 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Anti-inflammation: mechanism in the underlying vascular smooth muscle cells leukocyte adhesion (VSMCs) and ultimately, relaxation19 (Fig. 2). and migration Vascular lumen Failure of the eNOS protein to dimerise, or the absence of Anti-oxidation Anti-platelet aggregation some of the cofactors mentioned above will lead to the enzyme NO – catalysing the formation of O2 instead of NO, a mechanism 20 Endothelial cells referred to as eNOS uncoupling (Fig. 3). As will be discussed later, eNOS uncoupling is an important mediator of ED during a eNOS pathophysiological state.

NO Vasodilation Cardiovascular risk factors associated with VSMCs the development of ED cGMP Type 1 diabetes mellitus and insulin resistance/type 2 diabetes mellitus Fig. 2. Synthesis of NO, downstream mechanisms and Both type 1 and type 2 diabetes are independent risk factors physiological effects. NO is synthesised by eNOS in the for the development of accelerated atherosclerosis, IHD and endothelial cells and diffuses into the underlying vascu- cardiovascular disease in general.11 Similarly, type 1 diabetes lar smooth muscle cells (VSMCs), where it activates the second messenger, cyclic guanosine monophosphate mellitus, insulin resistance and type 2 diabetes mellitus have (cGMP). Further downstream, signalling eventually leads been shown to be strongly associated with the development of to VSMC relaxation and vasodilation. In addition, NO ED.21 In fact, the temporal progression from insulin resistance to regulates vascular homeostasis by anti-oxidation, anti- inflammatory and anti-platelet aggregation effects. type 2 diabetes mellitus has been postulated to be mirrored by the progression of ED to atherosclerosis.22 three isoforms, namely neuronal NOS (nNOS), inducible NOS ED observed in diabetes mellitus is primarily attributable to 14,16,17 – (iNOS) and endothelial NOS (eNOS). Physiologically, (1) oxidative stress (increased O2 generation due to upregulated eNOS and nNOS are constitutive, calcium-dependent enzymes expression of NADPH oxidase), and (2) increased formation and continuously produce low levels of NO. On the other hand, of advanced glycation end-products (AGEs).23,24 Aside from iNOS is calcium independent, its expression is provoked by scavenging NO, causing decreased NO bioavailability and – inflammatory cytokines, and it produces large amounts of NO, producing peroxynitrite, O2 also modifies the activity and about 1 000-fold more than eNOS or nNOS.13 This can have regulation of eNOS, and promotes vascular smooth muscle cell potentially harmful consequences as excess NO can react with (VSMC) proliferation and inflammation.24 – the free radical superoxide anion (O2 ), yielding a harmful and Hyperglycaemia, as occurs in diabetes mellitus, results highly reactive species, peroxynitrite.13 in non-enzymatic glycation of intracellular and extracellular All NOS isoforms require cofactors such as (6R)-5,6,7,8- proteins and lipids, which leads to the generation of AGEs. The tetrahydrobiopterin (BH4), flavin adenine dinucleotide (FAD), latter subsequently accumulate in the vascular wall and reduce flavin mononucleotide (FMN), and iron protoporphyrin IX NO activity by quenching NO.24,25 AGEs also bind to specific (haem).18 Of the three isoforms, it has been proposed that surface receptors, called receptors for AGEs (RAGE), which are eNOS is the major isoform responsible for NO production expressed on cells such as monocytes, macrophages and VSMCs, under physiological conditions in the cardiovascular system and resulting in the amplification of an inflammatory response,24,25 endothelial cells in particular, leading to the classical signalling increased vascular permeability and oxidative stress.25

A B e– e– e– e– HOOC– NADP+ FAD FMN HOOC– NADP+ FAD FMN e– Monomer e– Monomer

N2H– H2N–

NO + 2+ O2 + – 2+ O2 + ↓ BH4 Haem/Fe O2 Haem/Fe citrulline l-arginine l-arginine Hommodimer: eNOS = Zn eNOS = Coupled ONOO– BH3 Zn Uncoupled

O2 + 2+ NO + O2 + ↓ 2+ – Haem/Fe BH4 Haem/Fe O2 l-arginine citrulline l-arginine

–NH2 –NH2 Monomer e– Monomer e– FMN FAD NADP+ -COOH FMN FAD NADP+ -COOH e– e– e– e–

Fig. 3. Coupled and uncoupled eNOS. (A) In the presence of sufficient levels of substrates and co-factors, and the absence of harmful reactive species, eNOS monomers will form a dimerised, coupled enzyme and produce physi- ological amounts of NO. (B) Decreased levels of the substrate, L-arginine and/or harmful effects exerted by increased levels of ONOO–, cause failure of the enzyme to dimerise, leading to the uncoupling of eNOS and the production of – O2 instead of NO. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 225

Furthermore, hyperglycaemia is also known to activate reduced or absent in hypertensive patients.37 In addition to this, protein kinase C (PKC), which decreases eNOS activity, leading Iaccarino et al.38 observed decreased protein kinase B (PKB)/ to reduced NO and increased ET-1 production.24 In the setting Akt-dependent activation of eNOS in a model of spontaneously of ED, ETB receptor-mediated vasodilatory effects of ET-1 are hypertensive rats (SHR). blunted (refer to Table 1) and therefore the vasoconstrictory state In a recent study, the role of oxidative stress and ED in the predominates.26 PKC also enhances the expression of adhesion development of hypertension in spontaneously hypertensive rates molecules such as ICAM, VCAM and E-selectin,24 which is was investigated.35 The results showed that early treatment with associated with endothelial cell activation. the antioxidant reservatrol was associated with reduced oxidative ED has been reported to occur early in insulin resistance.22 stress markers, improved endothelium-dependent vasodilatation Often insulin resistance is associated with central adiposity and an attenuation in the development of hypertension in these and hence the metabolic syndrome, i.e. hypertriglyceridaemia, animals. low high-density lipoprotein (HDL) levels, high low-density lipoprotein (LDL) levels and hypertension, all of which could Smoking potentially favour the development of ED and eventually atherogenesis.22 Tobacco smokers exhibit decreased NO bioavailability, increased levels of ox-LDL, and impaired flow-mediated vasodilation, phenomena which are all highly suggestive of ED.39 Passive Hyperlipidaemia smoking has recently also been implicated in impairment of Hyperlipidaemia constitutes increased circulating lipids including endothelial function.39,40 It appears that the harmful effects of cholesterol and triglycerides, a state which can predispose to ED. smoking on endothelial cells are dose dependent and reversible Possible mechanisms underlying hyperlipidaemia-induced ED upon smoking cessation.39 As with other cardiovascular disease – include: (1) upregulation of NADPH oxidase, increased O2 risk factors, oxidative stress appears to be the major mechanistic production and oxidative stress, (2) increased plasma levels of link between smoking and ED.39,41 asymmetric dimethylarginine (ADMA),25 and (3) oxidation of Cigarette smoke is rich in free radicals and directly delivers LDL.27 ADMA is an endogenous inhibitor of eNOS and competes free radicals to the body. Besides being the supplier of free with L-arginine for the same binding site on eNOS, thus resulting radicals, cigarette smoke facilitates endogenous release of ROS – 41,42 in eNOS uncoupling, increased O2 production and hence via activation of inflammatory cells. Furthermore, smoking decreased NO production. Plasma concentrations of ADMA has been reported to decrease the levels of HDL cholesterol, have been reported to be increased in hypercholesterolaemia,28,29 which is known to have anti-endothelial dysfunction and anti- and this compound is considered to be both a marker and risk atherosclerotic properties.43 factor of ED.28 In addition to scavenging NO, excess O – modifies LDL 2 Aging cholesterol to form oxidised LDL (ox-LDL), which plays a major role in the development of endothelial activation and Increasing age has been recognised as one of the factors that 43,44 atherogenesis.30 Ox-LDL has been reported to promote ET-1 predisposes to ED. With aging, the ability of the endothelium 45 production,31 expression of adhesion molecules and chemo- to produce NO is reduced. Furthermore, some studies have attractants, as well as VSMC migration and proliferation.27 reported reduced expression and activity of eNOS as well as Furthermore, ox-LDL can be engulfed by macrophages forming decreased expression of a major downstream target molecule of foam cells which adhere to the vessel wall and contribute to the NO, soluble guanylyl cyclase (sGC) in VSMCs, and its activity in 45 initiation of an atherosclerotic plaque.27 older animals. In addition to the decreased NO production, other Both LDL and ox-LDL have been shown to increase the endothelial-derived relaxing factors (EDRFs) (prostacyclin and activity of S-adenosylmethionine-dependent methyltransferases, EDHF) are also reduced, while endothelial-derived contracting which lead to increased ADMA synthesis. Therefore, LDL and factors (EDCFs) such as ET-1 and COX-derived prostanoids, 44,45 ox-LDL may be accountable for the increased plasma levels of and ROS production are increased. Plasma levels of ADMA ADMA in hypercholesterolaemia.32 LDL or ox-LDL can also 33,34 upregulate caveolin-1 synthesis and thus inhibit eNOS activity – ↑ cholesterol + O2 (Fig. 4).

Hypertension ox-LDL ED is a prominent underlying feature of hypertension,35 and patients with hypertension have been shown to demonstrate Endothelial cells: ↑ ADMA, ↑ caveolin–1: + Macrophages blunted forearm blood flow in response to vasodilatory stimuli ↑ endothelin–1 eNOS coupling Foam cells such as acetylcholine and bradykinin,36 which is indicative ↑ adhesion molecules ↓ eNOS activation ↓ of ED. Increased production of ROS and endothelial-derived NO bioavailability contracting factors (EDCFs) such as ET-1, angiotensin II, PGH2 and TXA2, and decreased NO bioavailability are all observed in Endothelial Endothelial Atherosderotic patients with hypertension.26,36 activation dysfunction plaque Shear stress is known to be one of the most important Fig. 4. Pathophysiological effects and the interplay mechanisms of inducing NO-mediated vasodilation in both – between increased plasma cholesterol and O2 levels, and the micro- and macrovasculature. However, this response is endothelial cell responses. 226 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

are also known to rise with increased age.45 essential hypertension, as well as in chronic smokers.20 One of the mechanisms contributing to reduced NO levels in In addition to peroxynitrite-induced eNOS uncoupling, other aging may be the increased activity of arginase I.43,44 Arginase I is oxidants such as hydrogen peroxide have also been shown to an enzyme that catalyses conversion of L-arginine to L-ornithine uncouple the enzyme. Therefore, during conditions of oxidative and urea, and it thus competes with eNOS for L-arginine.43 stress, eNOS deviates from its role of being an essential regulator – Hence, the increased activity of this enzyme as observed with of the functioning of the cardiovascular system to being an O2 advancing age may result in uncoupling of eNOS, reduced releasing enzyme. A vicious circle therefore develops, whereby 43,44 – NO production and hence ED. Clearly the balance between uncoupled eNOS synthesises O2 at the expense of NO, further EDRFs and EDCFs is lost with advancing age, establishing aggravating oxidative stress. aging as a risk factor for the development of ED. Moreover, Inflammation is another common underlying mechanism of aging is often associated with co-morbid conditions such ED.53 Under physiological conditions, the endothelium regulates as diabetes, hypertension and hypercholesterolaemia, further vascular inflammation (including expression of adhesion exacerbating the risk of developing ED, atherosclerosis and molecules and leukocyte adhesion) via the release of NO.54 It is ultimately cardiovascular diseases.44 therefore more likely that ED will promote sustained vascular inflammation, which is detrimental to the vascular system. However, several studies have reported that inflammation also Proposed mechanisms of ED promotes ED and it is therefore recognised as a novel risk factor Oxidative stress appears to be the common underlying cellular for cardiovascular diseases.53,55 mechanism for the development of ED in all the risk factors There seems to be a causal relationship between oxidative discussed above. According to the literature, cardiovascular stress and inflammation. Oxidative stress may amplify vascular risk factors are associated with upregulation of ROS sources, inflammation signalling pathways,56 and conversely inflammatory – 7,20 especially NADPH oxidase. However, other sources of cells increasingly release O2 . Inflammation is often associated ROS such as xanthine oxidase, cyclooxygenase (COX) and with the overexpression of inflammatory cytokines such as mitochondria also play a role.23 In fact, eNOS per se becomes a tumour necrosis factor-alpha (TNF-α) and interleukin-1 (IL-1). potential ROS generator when in the uncoupled state.20 Harmful These inflammatory cytokines in turn prompt endothelial cells effects of oxidative stress include increasing VSMC proliferation or macrophages to express adhesion molecules such as VCAM- (resulting in thickening of the vascular wall), endothelial cell 1 and ICAM-1, MCP-1, interleukin-6 (IL-6) resulting in a state apoptosis, and increased expression and activity of matrix of endothelial activation, which is a precursor of ED57 (Fig. 1). metalloproteinases, which are involved in the establishment of The role of TNF-α in ED has received considerable attention an atherosclerotic plaque.39 in recent years, and is now well appreciated. High levels of Oxidative stress comprises increased rates of oxidant TNF-α have been associated with cardiovascular diseases such as production and decreased levels of antioxidant activity [e.g. acute myocardial infarction, chronic heart failure, atherosclerosis 58 superoxide dismutase (SOD), vitamin C and E, etc.].46 Under and myocarditis. Increased TNF-α levels are also significantly physiological conditions, the enzyme SOD regulates the levels correlated with obesity, which is an independent risk factor for 59 – 47 – ED. This inflammatory cytokine has been reported to promote of O2 . However, increased generation of O2 overwhelms the – defensive mechanisms of SOD, leaving O2 free to react with other molecules, particularly NO, for which it has a greater affinity.47 NADPH oxidase – Mitochondria O2 is implicated in the direct induction of ED by the Xanthine oxidase scavenging of NO, leading to the production of the highly reactive and harmful reactive nitrogen species (RNS), peroxynitrite.48 In – fact, the reaction between O2 and NO has been reported to SOD 9 Catalase occur much faster (rate constant = 6.7 × 109 m/s) than that of K = 2.0 × 10 – 9 49 – dismutation of O by SOD (rate constant = 2.0 × 10 m/s). O2 = H2O2 H2O + O2 2 K = High levels of peroxynitrite are injurious to the cells, oxidatively 6.7 × damaging DNA, lipids and proteins. In addition to being 10 cytotoxic, peroxynitrite damages the intricate eNOS structure, 9

– – – leading to eNOS uncoupling, which further perpetuates the ED NO = ONOO NO2 + OH vicious circle50 (Fig. 5). Peroxynitrite has been reported to oxidise the essential eNOS Protein nitration uncoupling cofactor of eNOS, BH4 to its inactive form, trihydrobiopterin Apoptosis – 20,50,51 Necrosis radical (BH3 ), which in turn leads to uncoupling of eNOS. Furthermore, peroxynitrite may oxidise the zinc thiolate cluster in the centre of the eNOS enzyme, resulting in the loss of the Fig. 5. Oxidative and nitro-oxidative stress. Superoxide – zinc ion and the formation of disulfide bonds between the anion (O2 ) released from sources such as NADPH oxidase, mitochondria and xanthine oxidase is dismutat- enzyme monomers, and thus disruption of the binding site for ed to hydrogen peroxide (H2O2) by superoxide dismutase 20,52 BH4 and L-arginine (Fig. 3). Vitamin C is able to recycle (SOD), which is then converted to water and oxygen by – 50,51 – 2 BH3 to BH4, and supplementation with BH4 has been catalase. However, O has a higher affinity for NO than reported to restore endothelial function in conditions such as SOD, and when in excess, it preferentially combines with 51 NO to produce peroxynitrite with various pathophysio- insulin resistance, hypercholesterolaemia, diabetes mellitus and logical consequences. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 227

ROS formation via NADPH oxidase and xanthine oxidase.60 the independent prognostic value of vascular/endothelial function For example, Gao et al. reported that TNF-α induces ED via measurements, the authors conceded that more and larger human increased NADPH oxidase activity in coronary arterioles of mice studies should be undertaken to confirm this finding. with type 2 diabetes.61 In addition, TNF-α has been implicated in the downregulation of eNOS expression (and therefore Biomarkers decreased NO production) by accelerating eNOS mRNA degradation.34,60,62 According to Zhang et al., ED observed in Although many biomarkers of endothelial function have been myocardial ischaemia–reperfusion injury may be attributable to identified, only some may have potential clinical use. Therefore, increased TNF-α expression via the enhancement of xanthine the measurement of endothelial biomarkers remains a tool oxidase activity.63 mainly utilised in the experimental animal and in vitro laboratory Other harmful stimuli-induced mechanisms associated with setting. the development of increased oxidative stress and decreased eNOS activation/eNOS uncoupling include activation of Reduction of NO bioavailability cholesterylester transport protein (CETP), downregulation of Reduction in endothelial-derived NO production or bioavailability lipoprotein lipase, downregulation of peroxisome-proliferator represents a measurable parameter that is suggestive of the activated receptor (PPAR), downregulation of protein kinase development of ED. Laboratory-based studies most often make A (PKA), activation of caveolin, activation of rho-kinase and use of indirect NO measurements (measurements of metabolic 64 downregulation of sphingosine-1-phosphate. products of NO) to confirm ED, such as nitrogen oxide levels67 or levels of stable degradation products, nitrite and nitrate.68 Many Assessment of endothelial function researchers also measure expression and/or activation of eNOS, Direct mechanical endothelial function the main enzymatic source of NO in endothelial cells, to further 69,70 measurements validate their findings. In humans, changes in NO production can also be detected Direct endothelial function measurement in humans has the in plasma by measurement of NO metabolites. Kiettisanpipop potential to become an important clinical tool in diagnosing et al. reported a decrease in plasma NO metabolite levels in or predicting the development of cardiovascular disease in the patients with severe pulmonary hypertension compared to presence or absence of cardiovascular risk factors. Furthermore, those with moderate hypertension.71 Oestrogen replacement in recognition of evidence pointing towards a pathophysiological therapy has been shown to increase plasma NO levels while link between ED and IHD, a number of human experiments have decreasing ET-1 levels and thus improving endothelial function been conducted in which clinical assessment of ED is explored in postmenopausal women.72 Heiss et al. demonstrated that as a possible predictor or prognostic marker of cardiovascular plasma levels of NO derivatives (nitrosyl/nitroso species) were 65 events. decreased in patients with endothelial dysfunction.73 It remains An ideal method for the direct measurement of endothelial to be seen, however, whether the measurement of plasma NO or function should be safe, cost-effective, non-invasive, repeatable, NO-derived metabolites will become a widely used clinical tool 5,65 reproducible and standardised between laboratories. Current with predictive properties. methods of assessing endothelial function include flow-mediated dilation (FMD), forearm plethysmography, finger-pulse plethysmography, pulse curve analysis and quantitative coronary ADMA angiography65 (Table 2). ADMA has emerged as a mediator, independent risk factor and, In a recent review article, FMD was recognised as a commonly from a clinical perspective, potentially promising marker of undertaken, non-invasive technique to assess endothelial ED.29 As explained earlier, ADMA is endogenously synthesised function.66 In this article, a meta-analysis of 14 studies (more via methylation of arginine residues in the nuclear proteins74 than 8 300 subjects) showed that FMD was strongly predictive and competitively inhibits eNOS, resulting in decreased NO of future cardiovascular events. Despite the evidence in favour of production, which may induce eNOS uncoupling. Synthesis

TABLE 2. CLINICAL DETECTION TECHNIQUES OF ENDOTHELIAL FUNCTION Method Brief description Forearm plethysmography Involves intrabrachial infusion of endothelial-dependent vasodilators such as acetylcholine, meta- choline, substance P and bradykinin, with subsequent measurement of changes in endothelial function of forearm arterioles Flow-dependent dilation of the brachial artery This method employs a high-resolution ultrasound to quantify flow-mediated dilation of the brachial artery Finger-pulse plethysmography (ENDO-PAT) A novel non-invasive technique that measures changes of the pulse-wave amplitude during reac- tive hyperaemia. Low pulse-wave amplitudes are associated with compromised endothelial func- tion and are therefore good predictors of cardiovascular disease Pulse curve analysis A non-invasive technique that relies on the measure of arterial stiffness to quantify endothelial function Quantitative coronary angiography following An invasive approach of quantifying endothelial function, which involves intracoronary infusion intracoronary infusion of acetylcholine of the endothelium-dependent vasodilator, acetylcholine, and subsequent measurment of the vaso- motor response 228 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

of ADMA is catalysed by protein arginine methyltransferases both of which yield tyrosine-nitrating compounds.80,81 Via (PRMTs) and its degradation is catalysed by dimethylarginine formation of these compounds, peroxynitrite leads to nitration 75 dimethylaminohydrolases (DDAH). (addition of a NO2 group) of tyrosine residues of proteins, DDAH levels are often decreased in a variety of cardiovascular leading to formation of nitrotyrosine.82 diseases, which leads to the upregulation of ADMA. For Under normal conditions, low levels of free or protein-bound example, treatment of endothelial cells with TNF-α, ox-LDL and nitrotyrosine are detectable, which may indicate low levels of glucose has been reported to diminish the activity of DDAH.74,75 oxidants and nitrating species produced during physiological Furthermore, PRMTs and DDAH are ROS-sensitive; the activity processes. However, significant nitrotyrosine upregulation is of the DDAH is impaired, whereas the activity of the PRMTs is observed in conditions that are associated with nitroxidative increased in conditions of oxidative stress.75 stress such as inflammation, cardiovascular disease (including Indeed, increased plasma levels of ADMA have been ED and atherosclerosis) and neurodegenerative disorders.80 documented in patients with conditions such as hyperlipidaemia, Tyrosine nitration may modify the structure and function of hypertension, coronary artery disease, stroke and end-stage renal proteins, leading to alterations in catalytic activity of enzymes, disease.75 Furthermore, ADMA was found to be significantly production of antigenic epitopes, and impaired cell signal elevated in patients with unstable angina, and reduced plasma transduction.82 levels of ADMA at six weeks post-percutaneous coronary It has recently been proposed that nitrotyrosine levels intervention was found to be possibly indicative of a reduced risk can be clinically detected in urine samples using a surface of recurrent cardiovascular events.76 plasmon resonance (SPR) sensor83 or high-performance liquid A recent study investigated the prognostic value of ADMA chromatography.84 However, nitrotyrosine measurements in the with regard to cardiovascular disease and death (fatal or non-fatal context of ED research remain confined to the experimental myocardial infarction, coronary insufficiency, angina pectoris, laboratory setting. stroke or TIA, intermittent claudication or heart failure) in Framingham Offspring study participants.77 Although ADMA Other biomarkers of ED/vascular injury was significantly associated with all-cause mortality in this Recently, the European Society of Cardiology Working Group on population, the study could not find an association between Peripheral Circulation published a position statement on methods ADMA and cardiovascular disease incidence. for evaluating endothelial function.85 In this comprehensive review, several biochemical markers and assays that are used to Circulating endothelial cells and endothelial examine different aspects of endothelial function were discussed. microparticles The working group mooted plasma ADMA levels as a potential Circulating endothelial cells (CECs), which are mature cells biomarker of endothelial function; however the authors cautioned that have detached from the endothelium, represent a novel that currently, direct endothelial function measurements remain a biomarker of endothelial injury.78,79 In a healthy person, the superior indicator and should not be replaced by plasma ADMA endothelium is constantly refurbished at a replication rate of level measurements due to inconsistent prognostic data obtained < 1% and levels of CECs are very low. Studies using a flow- with the latter. activated cell sorter (FACS) isolation technique reported CECs Another biomarker with potential clinical application is ranging from 50–7 900 cells/ml in healthy individuals and up to oxidised LDL levels; however this biomarker also presents with 39 100 cells/ml in individuals with vascular diseases.79 some limitations. It is difficult to determine ox-LDL levels in Potential mechanisms underlying endothelial cell detachment vivo and the ability of elevated plasma ox-LDL to independently may be mechanical injury, action of proteases and/or cytokines, predict the development of coronary heart disease is still defective endothelial cell adhesion to the extracellular matrix, equivocal. cellular apoptosis, and injurious actions of cardiovascular risk In a recent study on a model of rat carotid injury, proteomic factors, such as occur during the induction of ED. Increased analysis of blood proteins showed significantly differential levels of CECs are associated with ED, cardiovascular diseases expression of vitamin D binding protein (VDBP), aldolase A and a variety of other diseases.78,79 Apoptotic CECs expressing (aldo A) and apolipoprotein E (ApoE) two weeks after injury.86 surface marker (CD146) have been reported to be increased in Reduced circulating levels of all three of these plasma markers patients with cardiovascular disease.65 were associated with the presence of vascular injury and may Other circulating cellular markers of endothelial injury represent novel markers of ED; however, further research is include endothelial microparticles (EMPs), which are small cell necessary. membrane vesicles released into the circulation by activated or 5,65 apoptotic cells. Patients with hypertension and coronary artery Summary of assessment of endothelial function disease have been reported to demonstrate high levels of EMPs,65 With the development of an ever-increasing number of and according to Tramontano et al., statins can diminish the measurement techniques of endothelial function (both direct release of EMPs in cultured coronary endothelial cells. mechanical endothelial function assessment and measurement of biomarkers of endothelial function), most authors agree Nitrotyrosine upregulation that more and larger human-based studies are necessary to In addition to its ability to directly uncouple the eNOS enzyme, validate their clinical usefulness. The overall objective of such which can lead to ED, peroxynitrite undergoes protonation to form studies should ultimately be to establish standardised protocols peroxynitrous acid (ONOOH), or it can combine with carbon allowing for the clinical diagnosis of ED, and quantification of – dioxide (CO2) to form nitroso-peroxocarboxylate (ONOOCO2 ), cardiovascular risk, followed by the reversal of ED by means of AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 229

anti-ED therapies. We are not there yet. The various endothelial demonstrated in investigations where addition of an eNOS function assessment tools at our disposal should be compared inhibitor (L-NMMA) to pravastatin-treated hypercholesterol- in studies that account for, among others, pathophysiological aemic patients hampered the endothelium-dependent relevance and reproducibility, predictability in diverse patient vasodilation.93 Moreover, statins have been reported to stimulate populations, ease of use, cost-effectiveness, and risk assessment activity of PKB/Akt, a major upstream activating signalling abilities that are superior to the tools currently in use.85 molecule of eNOS, and increase stability of eNOS mRNA, thus Another shortcoming in our understanding of the clinical enhancing eNOS expression.33 significance of ED is the lack of studies where the effects In addition to statins and NO donors (discussed above), other of therapies that specifically target endothelial biology are drugs that have been shown to improve endothelial function investigated. In this regard, promising observations were made include angiotensin converting enzyme (ACE) inhibitors, in a study which showed that dietary supplementation with angiotensin receptor (AT1 receptor) blockers, peroxisome the NO-donor, L-arginine significantly improved endothelium- proliferator-activated receptor-γ (PPAR-γ) agonists, antioxidants dependent dilatation in young adults with hypercholesterolaemia.87 and oestrogen replacement.22,92 In a recent review article, Similarly, in another study on rabbits, dietary supplementation Balakumar et al. identified potentially novel target sites for the with L-arginine prevented hypercholesterolaemia-induced ED by pharmacological improvement of vascular function.64 These target augmentation of NO-production.88 sites include rho-kinase, poly (ADP ribose) polymerase, protein tyrosine phosphatase (PTPase), Akt, protein kinase A (PKA), caveolin, cholesterylester transfer protein (CETP), lipoprotein Progression of ED to atherosclerosis lipase, sphingosine-1-phosphate (S1P), advanced glycation ED has emerged as a potentially valuable prognostic tool in end-product (AGE) and transketolase, geranylgeranyltransferase predicting the development of atherosclerosis and ultimately (GGT), epoxide hydrolase and Janus kinase (JAK). IHD.89 The progression from the early changes observed in compromised vascular endothelium (endothelial activation and dysfunction) to atherosclerosis is complex and multifactorial.89 Conclusion The healthy, intact endothelium is a highly selectively In view of the ever-increasing prevalence of ischaemic heart permeable barrier and does not promote leukocyte adhesion disease in the developed and developing world, it has become and invasion, or platelet aggregation and adhesion.53 However, imperative to identify and investigate mechanisms of early, as the endothelium progresses to a dysfunctional state, vascular potentially reversible pre-atherosclerotic changes in the homeostasis becomes impaired, leading to reduced anti-oxidant, endothelium. To date, the most clearly defined and well- anti-inflammatory and anti-thrombotic properties (due to reduced understood early precursor of atherosclerosis is ED. In fact NO bioavailability), enhanced endothelial permeability (barrier ED can be regarded as the primum movens of atherosclerotic dysfunction), upregulated pro-inflammatory cytokine levels, and disease. Several cellular mechanisms and markers of ED that expression of adhesion molecules such as VCAM-1 and ICAM- could potentially lead to the development of early detection 1, which facilitate leukocyte adhesion to the endothelium.53 and therapeutic interventions have been determined. However, Leukocyte adhesion represents one of the first steps in the more research aiming at improving our understanding of ED initiation of atherosclerosis. After adhering to the endothelium, is necessary in order to establish its detection and reversal as leukocytes (monocytes and lymphocytes) cross the endothelium essential and routinely utilised future tools in the prevention of and migrate into the intima.54,90 Migration to the intima is IHD. mediated by chemo-attractants such as monocyte chemotactic protein-1 (MCP-1).91 Upon reaching the intima, monocytes This study was sponsored in part by the Medical Research Council (MRC) of transform into macrophages and express receptors that facilitate South Africa, National Research Fund (NRF) of South Africa and the Harry uptake of lipids. Uptake and accumulation of lipids lead to Crossley Foundation, University of Stellenbosch. the transformation of macrophages into foam cells, which initiate an atherosclerotic lesion and further enhance release of References inflammatory cytokines.54,90 Through these complex mechanisms, 1. Steyn K. Heart Disease in South Africa. In: Fourie JM (ed). Chronic a cascade of events, which begins with the formation of an diseases of lifestyle unit. Medical Research Council (MRC) media data early atherosclerotic lesion, leading to an advanced lesion document. MRC, 2007. 2. World Health Organisation (WHO). Global health risks: Mortality and characterised by a plaque formation ensues.90 burden of disease attributable to selected major risks. Geneva: WHO, 2009. Anti-endothelial dysfunction therapies 3. Lahoz C, Mostaza JM. Atherosclerosis as a systemic disease. Rev Esp Cardiol 2007; 60(2): 184–195. It has been shown that interventions such as lifestyle modification 4. Choi D, Hwang K-C, Lee K-Y, Kim Y-H. Ischemic heart disease: (exercise and diet) and various classes of pharmacological drugs current treatments and future. J Contr Release 2009; 140: 194–202. can improve endothelial function, and in some instances reduce 5. Deanfield JE, Halcox JP, Rabelink TJ. Endothelial function and the risk of cardiovascular diseases.22,92 The anti-ED properties of dysfunction: Testing and clinical relevance. Circulation 2007; 115: statins have been extensively studied. Statins such as pravastatin, 1285–1295. atorvastatin, simvastatin and fluvastin have all been shown 6. Yang G, Lucas R, Caldwell R, Yao L, Romero MJ, Caldwell RW. Novel mechanisms of endothelial dysfunction in diabetes. J Cardiovasc Dis to play an important role in correction of ED by improving Res 2010; 1(2): 59–63. endothelium-dependent vasodilation, in addition to their plasma 7. Chhabra N. Endothelial dysfunction – A predictor of atherosclerosis. cholesterol-lowering effects.33 Internet J Med Update 2009; 4(1): 33–41. Indeed, statin-induced eNOS activating effects have been 8. Bijl M. Endothelial activation, endothelial dysfunction and premature 230 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

atherosclerosis in systemic autoimmune disease. Neth J Med 2003; 60: mechanisms of impaired nitric oxide bioactivity. In: Loscalzo J (ed). 273–277. Molecular Mechanisms of Atherosclerosis. London: Taylor and Francis 9. Rabelink TJ, De Boer CH, van Zonneveld AJ. Endothelial activation 2005: 95–110. and circulating markers of endothelial activation in kidney disease. Nat 35. Bhatt SR, Lokhandwala MF, Banday AA. Resveratrol prevents endothe- Rev Nephrol 2010; 6: 404–414. lial nitric oxide synthase uncoupling and attenuates development of 10. Bonetti PO, Lerman LO, Lerman A. Endothelial dysfunction: a hypertension in spontaneously hypertensive rats. Eur J Pharmacol marker of atherosclerotic risk. Arterioscler Thromb Vasc Biol 2003; 2011; 667(1-3): 258–264. 23: 168–175. 36. Tang EH, Vanhoutte PM. Endothelial dysfunction: a strategic target in 11. Esper RJ, Nordaby RA, Vilarino JO, Paragano A, Cacharon JL, the treatment of hypertension? Pflugers Arch 2010; 459(6): 995–1004. Machado RA. Endothelial dysfunction: a comprehensive appraisal. 37. Paniagua OA, Bryant MB, Panza JA. Role of endothelial nitric oxide Cardiovasc Diabetol 2006; 5: 4. in shear stress-induced vasodilation of human microvasculature: dimin- 12. Strijdom H, Lochner A. Cardiac endothelium: More than just a barrier! ished activity in hypertensive and hypercholesterolemic patients. SA Heart 2009; 6(3): 174–185. Circulation 2001; 103: 1752–1758. 13. Strijdom H, Chamane N, Lochner A. Nitric oxide in the cardiovascular 38. Iaccarino G, Ciccarelli M, Sorriento D, et al. AKT participates system: a simple molecule with complex actions. Cardiovasc J Afr in endothelial dysfunction in hypertension. Circulation 2004; 109: 2009; 20: 303–310. 2587–2593. 14. Bruckdorfer R. The basics about nitric oxide. Mol Aspect Med 2005; 39. Puranik R, Celermajer DS. Smoking and endothelial function. Progr 26: 3–31. Cardiovasc Dis 2003; 45: 443–458. 15. Moncada S, Palmer RM, Higgs EA. The discovery of nitric oxide as the 40. Barnoya J, Glantz SA. Cardiovascular effects of secondhand smoke: endogenous . Hypertension 1988; 12: 365–372. nearly as large as smoking. Circulation 2005; 111: 2684–2698. 16. Balligand J-L, Cannon PJ. Nitric oxide synthases and cardiac muscle: 41. Burke A, FitzGerald GA. Oxidative stress and smoking-induced vascu- autocrine and paracrine influences. Arterioscler Thromb Vasc Biol lar injury. Progr Cardiovasc Dis 2003; 46: 79–90. 1997; 17: 1846–1858. 42. Antoniades C, Tousoulis D, Stefanadis C. Smoking in Asians: it doesn’t 17. Bryan NS, Bian K, Murad F. Discovery of the nitric oxide signaling stop at vascular endothelium. Int J Cardiol 2008; 128: 151–153. pathway and targets for drug development. Front Biosci 2009; 14: 1–18. 43. Katusic ZS. Mechanisms of endothelial dysfunction induced by aging, 18. Alderton WK, Cooper CE, Knowles RG. Nitric oxide synthases: struc- role of arginase I. Circ Res 2007; 101: 640–641. ture, function and inhibition. Biochem J 2001; 357: 593–615. 44. Herrera MD, Mingorance C, Rodrίguez- Rodrίguez R, et al. Endothelial 19. Dudzinski D, Michel T. Life history of eNOS: partners and pathways. dysfunction and aging: an update. Ageing Res Rev 2010; 9: 142–152. Cardiovasc Res 2007; 75(2): 247–260. 45. Vanhoutte PM. Ageing and endothelial dysfunction. Eur Heart J Suppl 20. Förstermann U, Munzel T. Endothelial nitric oxide synthase in vascular 2002; 4: A8–A17. disease: from marvel to menace. Circulation 2006; 113: 1708–1714. 46. Pennathur S, Heinecke JW. Oxidative stress and endothelial dysfunction 21. Potenza MA, Gagliardi S, Nacci C, et al. Endothelial dysfunction in in vascular disease. Curr Diabetes Rep 2007; 7: 257–264. diabetes: From mechanisms to therapeutic targets. Curr Med Chem 47. Landmesser U, Harrison DG, Drexler H. Oxidant stress – a major 2009; 16: 94–112. cause of reduced endothelial nitric oxide availability in cardiovascular 22. Hsueh WA, Lyon CJ, Quinones MJ. Insulin resistance and the endothe- disease. Eur J Clin Pharmacol 2006; 63: 13–19. lium. Am J Med 2004; 117: 109–117. 48. Yokoyama M. Oxidant stress and atherosclerosis. Curr Opin Pharmacol 23. Guzik TJ, Mussa S, Gastaldi D, et al. Mechanisms of increased vascular 2004; 4: 110–115. superoxide production in human diabetes mellitus, role of NAD(P)H 49. Huang PL. Endothelial nitric oxide synthase and endothelial dysfunc- oxidase and endothelial nitric oxide synthase. Circulation 2002; 105: tion. Curr Hypertens Rep 2003; 5: 473–480. 1656–1662. 50. Kuzkaya N, Weissmann N, Harrison DG, et al. Interaction of peroxyni- 24. Avogaro A, de Kreutzenberg S, Fadini G. Endothelial dysfunction: trite, tetrahydrobiopterin, ascorbic acid, and thiols. J Biol Chem 2003; causes and consequences in patients with diabetes mellitus. Diabetes 278: 22546–22554. Res Clin Pract 2008; 82: S94–S101. 51. Katusic ZS. Vascular endothelial dysfunction: does tetrahydrobiopterin 25. Soldatos G, Cooper ME, Jandeleit-Dahm KAM. Advanced-glycation play a role? Am J Physiol Heart Circ Physiol 2001; 281: 981–986. end products in insulin-resistant states. Curr Hypertens Rep 2005; 7: 52. Zou M-H, Shi C, Cohen RA. Oxidation of the zinc-thiolate complex 96–102. and uncoupling of endothelial nitric oxide synthase by peroxynitrite. J 26. Versari D, Daghini E, Virdis A, et al. Endothelium-dependent contrac- Clin Invest 2002; 109: 817–826. tions and endothelial dysfunction in human hypertension. Br J 53. Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Pharmacol 2009; 157: 527–536. Circulation 2002; 150: 1135–1143. 27. Sawamura T. LOX-1, a lectin-like oxidized LDL receptor identified 54. Osto E, Cosentino F. The role of oxidative stress in endothelial dysfunc- form endothelial cells, in endothelial dysfunction. Int Congr 2004; tion and vascular inflammation. In: Ignarro LJ (ed). Nitric Oxide: 1262: 531–534. Biology and Pathobiology. 2nd edn. London: Academic Press 2010: 28. Böger RH, Bode-Böger S, Szuba A, et al. Asymmetric dimethlyargi- 705–754. nine (ADMA): a novel risk factor for endothelial dysfunction, its role 55. Szmitko PE, Wang C-H, Weisel RD, et al. New markers of inflam- in hypercholesterolemia. Circulation 1998; 98: 1842–1847. mation and endothelial cell activation, part 1. Circulation 2003; 108: 29. Böger RH, Vallance P, Cooke JP. Asymmetric dimethlyarginine 1917–1923. (ADMA): a key regulator of nitric oxide synthase. Atherosclerosis 56. Madamanchi NR, Vendrov A, Runge MS. Oxidative stress and vascular Suppl 2003; 4: 1–3. disease. Arterioscler Thromb Vasc Biol 2005; 25: 29–38. 30. Zeibig S, Li Z, Wagner S, et al. Effect of the ox-LDL binding protein 57. Blake GJ, Ridker PM. Novel clinical markers of vascular wall inflam- Fc-CD68 on plaque extension and vulnerability in atherosclerosis. Circ mation. Circ Res 2001; 89: 763–771. Res 2011; 108(6): 695–703. 58. Meldrum DR. Tumor necrosis factor in the heart. Am J Physiol 31. Boulanger CM, Tanner FC, Bea ML, et al. Oxidized low density Regulatory Intergrative Comp Physiol 1998; 274: 577–595. lipoproteins induce mRNA expression and release of endothelin from 59. Fain JN, Bahouth SW, Madan AK. TNF-α release by the nonfat cells of human and porcine endothelium Circ Res 1992; 70: 1191–1197. human adipose tissue. Int J Obes 2004; 28: 616–622. 32. Warnholtz A, Mollnau H, Oelze M, et al. Antioxidants and endothelial 60. Zhang H, Park Y, Wu J, et al. Role of TNF-α in vascular dysfunction. dysfunction in hyperlipidemia. Curr Hypertens Rep 2001; 3: 53–60. Clin Sci 2009; 116: 219–230. 33. Davignon J, Ganz P. Role of endothelial dysfunction in atherosclerosis. 61. Gao X, Belmadani S, Picchi A, et al. Tumor necrosis factor-α induces Circulation 2004; 109: 27–32. endothelial dysfunction in Leprdb mice. Circulation 2007; 115: 34. Hamburg NM, Vita JA. Endothelial dysfunction in atherosclerosis: 245–254. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 231

62. Stenvinkel P. Endothelial dysfunction and inflammation – is there a 77. Böger RH, Sullivan LM, Schwedhelm E, Wang TJ, Maas R, Benjamin link? Nephrol Dial Transplant 2001; 16: 1968–1971. EJ, et al. Plasma asymmetric dimethylarginine and incidence of 63. Zhang C, Xu X, Potter BJ, Wang W, et al. TNF-α contributes to cardiovascular disease and death in the community. Circulation 2009; endothelial dysfunction in ischemia/reperfusion injury. Arterioscler 119(12): 1592–1600. Thromb Vasc Biol 2006; 26: 475–480. 78. Boos CJ, Lip GYH, Blann AD. Circulating endothelial cells in cardio- 64. Balakumar P, Kaur T, Singh M. Potential target sites to modulate vascu- vascular disease. J Am Coll Cardiol 2006; 48: 1538–1547. lar endothelial dysfunction: current perspectives and future directions. 79. Erdbruegger U, Haubitz M, Woywodt A. Circulating endothelial cells: a Toxicology 2008; 245(1–2): 49–64. novel marker of endothelial damage. Clin Chim Acta 2006; 373: 17–26. 65. Münzel T, Sinning C, Post F, et al. Pathophysiology, diagnosis and 80. Souza JM, Peluffo G, Radi R. Protein tyrosine nitration-functional prognostic implication of endothelial dysfunction. Ann Med 2008; 40: alteration or just a biomarker? Free Radic Biol Med 2008; 45: 357–366. 180–196. 81. Hurst JK. Whence nitrotyrosine? J Clin Invest 2002; 109: 1287–1289. 66. Green DJ, Jones H, Thijssen D, Cable NT, Atkinson G. Flow-mediated 82. Pacher P, Beckman JS, Liaudet L. Nitric oxide and peroxynitrite in dilation and cardiovascular event prediction: does nitric oxide matter? health and disease. Physiol Rev 2007; 87: 315–424. Hypertension 2011; 57(3): 363–369. 83. Jin J, Wang C, Tao Y, et al. Determination of 3-nitrotyrosine in human 67. Kim F, Gallis B, Corson MA. TNF-alpha inhibits flow and insulin sign- urine samples by surface plasmon resonance immunoassay. Sensors aling leading to NO production in aortic endothelial cells. Am J Physiol Actuators B 2011; 153: 164–169. Cell Physiol 2001; 280(5): C1057–1065. 84. Molavi B, Mehta JL. Oxidative stress in cardiovascular disease: 68. Bove K, Neumann P, Gertzberg N, Johnson A. Role of ecNOS-derived molecular basis of its deleterious effects, its detection and therapeutic NO in mediating TNF-induced endothelial barrier dysfunction. Am J consideration. Curr Opin Cardiol 2004; 19: 488–493. Physiol Lung Cell Mol Physiol 2001; 280(5): L914–922 85. Lekakis J, Abraham P, Balbarini A, Blann A, Boulanger CM, Cockcroft 69. Xia Z, Luo T, Liu HM, Wang F, Xia ZY, Irwin MG, Vanhoutte PM. J, et al. Methods for evaluating endothelial function: a position state- L-arginine enhances nitrative stress and exacerbates tumor necrosis ment from the European Society of Cardiology Working Group on factor-alpha toxicity to human endothelial cells in culture: prevention Peripheral Circulation. Eur J Cardiovasc Prev Rehabil 2011 Mar 11. by propofol. J Cardiovasc Pharmacol 2010; 55(4): 358–367. [Epub ahead of print]. 70. Goodwin BL, Pendleton LC, Levy MM, Solomonson LP, Eichler DC. 86. Huang NF, Kurpinski K, Fang Q, Lee RJ, Li S. Proteomic identification Tumor necrosis factor-alpha reduces argininosuccinate synthase expres- of biomarkers of vascular injury. Am J Transl Res 2011; 3(2): 139–148. sion and nitric oxide production in aortic endothelial cells. Am J Physiol 87. Clarkson P, Adams MR, Powe AJ, Donald AE, McCredie R, Robinson Heart Circ Physiol 2007; 293(2): H1115–1121. J, et al. Oral L-arginine improves endothelium-dependent dilation 71. Kiettisanpipop P, Lertsapcharom P, Chotivittayatakorn P, et al. Plasma in hypercholesterolemic young adults. J Clin Invest 1996; 97(8): levels of nitric oxide in children with congenital heart disease and 1989–1994. increased pulmonary blood flow. J Med Assoc Thai 2007; 90: 2053– 88. Javanmard SH, Nematbakhsh M, Sanei MH. Early prevention by 2057. L-arginine attenuates coronary atherosclerosis in a model of hyper- 72. Best PJM, Berger PB, Miller VM, et al. The effect of estrogen replace- cholesterolemic animals; no positive results for treatment. Nutr Metab ment therapy on plasma nitric oxide and endothelin-1 levels in post- (Lond) 2009; 6: 13. menopausal women. Ann Intern Med 1998; 128: 285–288. 89. Yang Z, Ming X-Z. Recent advances in understanding endothelial 73. Heiss C, Lauer T, Dejam A, et al. Plasma nitroso compounds are dysfunction in atherosclerosis. Clin Med Res 2006; 4: 53–65. decreased in patients with endothelial dysfunction. J Am Coll Cardiol 90. Ruberg FL, Loscalzo J. Inflammation and atherothrombosis. In: 2006; 47: 573–579. Loscalzo J (ed). Molecular Mechanisms of Atherosclerosis. London: 74. Landim MBP, Filho AC, Chagas ACP. Asymmetric dimethylarginine Taylor and Francis 2005: 45–60. (ADMA) and endothelial dysfunction: implications for atherogenesis. 91. Libby P. Changing concepts of atherogenesis. J Intern Med 2000; 247: Clinics 2009; 64: 471–478. 349–358. 75. Sydow K, Münzel T. ADMA and oxidative stress. Atherosclerosis Suppl 92. Widlansky ME, Gokce N, Keaney JF, et al. The clinical implications 2003; 4: 41–51. of endothelial dysfunction. J Am Coll Cardiol 2003; 42: 1149–1160. 76. Krempl TK, Maas R, Sydow K, Meinertz T, Böger RH, Kähler J. 93. John S, Schlaich M, Langefelm M, et al. Increased bioavailabil- Elevation of asymmetric dimethylarginine in patients with unstable ity of nitric oxide after lipid lowering therapy in hypercholester- angina and recurrent cardiovascular events. Eur Heart J 2005; 26(18): olemic patients: a randomized, placebo-controlled, double blind study. 1846–1851. Circulation 1998; 98: 211–216.

HEALTHCARE 232 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Conference Report

Expert report on the 22nd European meeting on hypertension and cardiovascular protection, London, 26–29 April 2012 Following the release of the updated of resistant hypertension’, intends to scoring system for cardiovascular risk NICE (British Hypertension Society) facilitate a better understanding of the assessment, it was fascinating to hear guidelines last year and the South African effectiveness, safety, limitations and how the scoring system was developed. guidelines recently, this meeting presented issues still to be addressed with RDN. Prof RB D’Agostino, mathematician an opportunity to hear opinion leaders The meeting covered issues of and statistician, discussed the validation analyse, debate and discuss the best way cardiovascular protection. The cardio– of this score in the non-Framingham to detect and manage hypertension in the ankle vascular stiffness index (CAVI) population; the transportability and 21st century. is a non-invasive technique used in external validation, which allows one Regarding treatment, the issue Japan to assess arterial stiffness. It is to use it for other populations such of diuretics was highlighted. The expressed as a ratio between the internal as non-Caucasians. It has also been NICE guidelines advocate the use of pressure in blood vessels (blood pressure) validated for non-US populations such as thiazide-like diuretics (indapamide) and changes in vascular diameter, and the Chinese. and chlorthalidone. The South African measures pulse-wave velocity between guidelines also advocate these agents; the heart and femoral artery. The clinical Acute and chronic cardiovascular however the choice of diuretic still implications are whether it will be a responses to endurance and includes low-dose hydrochlorothiazide. useful tool to assess arterial disease Prof Franz Messerli, an eminent American independent of blood pressure levels. ultra-endurance exercise opinion leader, was asked what is used in There is also an association between During this session the Morganroth his country and the answer was thiazides, CAVI and atherosclerosis. All the work hypothesis was revisited. The question often in fixed combination. has previously been done in Asians and was asked whether the heart adapts The lesson is to use the thiazides with Japanese, and data were presented on differently to endurance and resistance caution, be aware of the metabolic side the reliability of this index in a cohort of exercise. effects and understand that the agents now 4 000 Swiss subjects. Of note is that the In the 1970s, using only non-two- advocated have been shown to have better index increases with age and women have dimensional guided echocardiography, outcomes data. Prof G Bakris stressed a lower score than men. it was concluded that left ventricular the need to ensure that patients are not A workshop dedicated to Bill Kannel, mass index is increased in athletes who hypokalaemic, as this contributes to father of the Framingham study, was undertake either isotonic or isometric vasoconstriction of vessels and resistant delivered by co-workers. Prof S Franklin, exercise. The mechanism underlying the hypertension. It is important to correct nephrologist, discussed the five seminal increased left ventricular mass related potassium levels before adding in more articles published regarding blood to chamber dilatation in the endurance therapy. pressure, ageing and cardiovascular athlete versus an increase in left A focus of the meeting was renal disease, and highlighted findings that ventricular wall thickness in the athlete nerve denervation (RDN) in resistant changed our understanding of the disease. undertaking isometric exercise. hypertension. Prof M Esler gave the Prof D Levy, involved with cardiovascular Use of more modern technology Bjorn Folkow award lecture titled ‘The gene therapy, highlighted the genotype such as cardiac magnetic resonance clinical physiology of the sympathetic and phenotype genome-wide association imaging (MRI) confirmed that cyclists, nervous system: no longer a promissory data from this study. swimmers and runners, at peak training, note in hypertension’. He highlighted Although it was predicted in June 2000 did demonstrate left ventricular chamber the pathophysiology of the sympathetic that within 10 years one would be able to dilatation (42–62 mm) and that the nervous system and why surgical find out what particular genetic conditions septal wall thickness ranged from 6–14 splanchnicectomy, done from 1934 to patients have, the individual single- mm. In addition, right ventricular mass 1960, did not work. nucleotide polymorphisms identified do was increased, mainly as a result of an Renal denervation using radiofrequency not really contribute to the magnitude increase in end-diastolic volume. There laser is a minimally invasive percutaneous of hypertension. Those identified for was balanced eccentric hypertrophy of procedure characterised by short recovery systolic hypertension only contribute both the right and left ventricle. However, times and the absence of systemic side to a 1-mmHg increase, and diastolic to in athletes undertaking strength training, effects. It appears that this technique 0.5 mmHg. There are associations with there was no evidence of concentric left will be considered an adjunctive therapy stroke risk and left ventricular mass but ventricular hypertrophy. In fact less than in the future. The European Society of no association with renal markers of 2% had a left ventricular wall thickness Hypertension position paper, ‘Renal hypertension. in excess of 13 mm. denervation – an interventional therapy Regarding the Framingham In another session, Rob Shave from AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 233

Cardiff in the United Kingdom asked the The ACE inhibitor–CCB combination eating. Patients who have less sleep have question whether there is potential for was better even in those patients with a lower leptin but higher ghrelin levels. The harm from too much exercise. Troponin relative tachycardia. latter is produced by the stomach and is released after prolonged exercise. After causes the sensation of hunger. Sleep- the London Marathon, 56% of athletes deprived people had no difference in Cardiovascular risk in have troponin levels above the cut-off for energy expenditure versus those who myocardial infarction. He concluded that obstructive sleep apnea were not sleep deprived. Obese patients troponin release was ‘physiological and Obstructive sleep apnea (OSA) is found report less sleep than lean patients and not related to any functional abnormality in 25% of heart failure patients, 60% of this has been linked to the pathogenesis of the heart’. patients who have had a stroke, 85% of of obesity. In the Nurses Health study, After endurance exercise, such as a patients with resistant hypertension, 50% those who slept less than five versus more marathon or a 100-miler, the pressure in of hypertensives and 30% of patients with than seven hours per night had a 1.5-fold the right ventricle was increased, while acute coronary syndrome. About 30% of increased risk of developing obesity. there was a reduction in strain, strain obese patients have unrecognised OSA. Another study conducted by Somers rate and diastolic function in the right Patients with hypertension, OSA and evaluated subjects sleeping less or more ventricle. Speckle tracking confirmed an obesity are at increased risk of atrial than seven hours per night. In those who abnormality of radial strain in the septum fibrillation, sudden cardiac death, reduced were sleep deprived, 50% snored, 45% with septal flattening. The intense duration cardiac function, coronary disease and fell asleep in the day for no reason and caused acute dysfunction of the right atrio-ventricular block. There is a nine- 5% fell asleep while driving in the last 30 ventricle and although the right ventricular fold increase of developing the metabolic days of observation. function had usually normalised by the syndrome and more silent cerebrovascular If subjects sleep less and have a end of the first week, chronic structural lesions. disrupted sleep cycle, less insulin is changes, as demonstrated on MRI, with There is much more blood pressure secreted by the pancreas, even though reduced right ventricular function were and heart rate variability in patients blood glucose levels are higher after found in some of the most practiced with OSA. Alternating bradycardia and meals than in those who are not sleep athletes. tachycardia at night on 24-hour ECG deprived. The metabolic rate is reduced Beta-adrenergic receptor desensitisa- monitoring is another clue to the presence and the subjects have higher nighttime tion has been found after exercise and of OSA. and daytime blood pressure. Chronic this attenuates cardiac function because Clues to the diagnosis of OSA include sleep deprivation (< five hours per night) of reduced chronotropic and inotropic excessive daytime sleepiness, choking at was associated with a two-fold increase in response. Furthermore, changes in night, night sweats, nocturnal diuresis, incidence of acute coronary syndrome in loading conditions did not explain the motor activity during sleep, headaches, the Nurses Health study. reduced function of the right ventricle. dry mouth, impotence and cognitive Highly trained athletes differ impairment. Clinical signs relate from amateur athletes in that cardiac to the waist and neck circumference, Hypertension in the elderly remodelling is more profound and may hypertrophy of the tonsils, adenoids, The VALISH trial enrolled patients with persist despite detraining. Cardiac MRI the soft palate and even the uvula. isolated systolic hypertension (age range has shown myocardial fibrosis in athletes Contributing lifestyle parameters include 70–84 years). These patients were divided with an extensive history of competition excessive alcohol consumption, smoking, into two groups. In one group, target in endurance sport. The long-term clinical sedatives, as well as sleeping in a supine systolic blood pressure was < 140 mmHg significance of the fibrosis is unknown position. whereas in the second group the target and warrants further study. Avoiding these factors is key to systolic blood pressure was between 140 Prof Neil Poulter discussed the management. Continuous positive airway and 149 mmHg. They found no difference ASCOT trial, which has influenced the pressure via a nasal mask should be in outcome with strict versus less strict newest set of NICE/BHS guidelines. In administered for five to six hours per blood pressure control (< 140 vs 140–149 these guidelines, patients under the age night to improve outcome. mmHg). of 55 years should be given an ACE Prof Somers then discussed sleep, There was some discussion about the inhibitor as first-line therapy, whereas obesity and cardiovascular disease. In J-shaped curve regarding diastolic blood older or black patients should start initial 1910, the average sleep duration was pressure in elderly patients. The data therapy with a calcium channel blocker. If nine hours per night and in 2005 it was suggest that if diastolic blood pressure the blood pressure remains uncontrolled, 6.5 hours per night. A third of people are is lowered to < 80 mgHg in patients who the next step would be the combination sleep deprived on a daily or regular basis. have coronary heart disease at baseline, of an ACE inhibitor and calcium channel In an elegant experiment conducted at the there is an increase in cardiovascular blocker. This combination was associated Mayo Clinic, it was shown that less sleep mortality. with a more favorable outcome than was associated with higher caloric intake Prior to blood pressure measurement, the beta-blocker and thiazide–diuretic (500 calories a day) and this translated patients must have rested for at least five combination. Blood pressure variability into 3 kg of extra fat accumulation in one minutes and the readings are repeated. In was much less using the former month. the elderly, particular emphasis should be combination and was associated with Normally, leptin levels increase at night placed on standing blood pressure. an improved cardiovascular outcome. during sleep and this tells the brain to stop The HYVET trial recruited patients 234 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

over the age of 80 years with a systolic (140–160/< 90 mmHg) as trials have only in systolic blood pressure is a strong blood pressure of > 160 mmHg. Patients recruited patients with a systolic blood predictor of stroke, independent of mean were eligible for the trial if the systolic pressure > 160 mmHg. systolic blood pressure. Increased residual blood pressure was > 160 mmHg but not variability in systolic blood pressure if the standing systolic blood pressure in patients with treated hypertension is was < 140 mmHg. In this study, target Blood pressure variability associated with a high risk of vascular blood pressure was < 150/80 mmHg. This refers to visit–visit variability, within- events. Active treatment with indapamide-based visit variability, as well as variability over To most effectively prevent stroke, therapy was associated with a significant a longer period of time such as a week blood pressure-lowering drugs should reduction in total mortality (25%), or a month. The greater the variability in reduce mean blood pressure without cardiovascular mortality (35%), stroke blood pressure, the higher the risk. No increasing variability. Ideally they should mortality (60%) and heart failure (50%). specific number was given for the cut-off do both. The JATOS trial, published in measure which predicted higher versus Hypertension Research in 2008, lower variability. Dr Naomi Rapeport, confirmed that a systolic blood pressure A good technique to assess variability [email protected] of < 140 mmHg was associated with entailed home blood pressure monitoring Specialist physician, Milpark Hospital, more deaths in those over the age of 75 with three successive measurements a Johannesburg years. It was finally concluded that the day, over a period of seven days. The target blood pressure in the very elderly greater the variability in readings, the Dr Shirley Middlemost, is 150/80 mmHg. Currently, there are higher the risk. [email protected] no data on the optimal blood pressure Recent data published in the Lancet in Cardiologist, Hermanus MediClinic, goal in patients with mild hypertension 2010 confirmed that visit–visit variability Western Cape

SUBSCRIBE to CVJAfrica & SAJDVD Fully searchable data set of CVJAfrica (2001-2012) & SAJDVD (2004-2012).

You can subscribe to both these journals online.

The electronic version of the journal is available at www.diabetesjournal.co.za SEPTEMBER/OCTOBER 2011

www.cvja.co.za JANUARY/FEBRUARY 2012 E www.cvja.co.za VOL 22 NO 5 SEAS R DI VOL 23 NO 1 A CardioVascular Journal of Africa (official journal for PASCAR) Volume 9 Number 1 & VASCUL TES CardioVascular Journal of Africa (official journal for PASCAR) ABE AFRICA SAJDVD Featured in this issue AFRICA New antithrombotic drugs ‡ Under-treatment of hypercholesterolaemia The South African Journal of DI ‡ Anticoagulation in diabetic Mutation associated with hypertension patients March 2012 ‡ Echocardiographic findings in marfan syndrome

Gliptins in type 2 diabetes ‡ Effect of handgrip training on blood pressure Cardiovascular Journal of Africa . Vol 22, No 5, Septemb 5, No 22, Vol . Africa of Journal Cardiovascular ‡ Cardiovascular risk profiles of untreated hypertensives Risk of cardiovascular disease in the Bellville pilot ‡ Anomalous origin of the left pulmonary artery study ‡ Double-chambered right ventricle Serum levels of RAGE and TNF-α in type 2 diabetes

Serum tumour necrosis factor-α in type 2 diabetes

Improving warfarin control

Liraglutide: an incretin 2011 er/October mimetic

Printed by Durbanville Commercial Printers Tel: 021 946 4074

D monthly e-alerts

News

WIN AN IPAD!

Diabetes Educator’s Focus

To enter, subscribe to SAJDV

Achieving Best Practice

Ethics Focus

Reviews

www.cvja.co.za – subscription is 200 ZAR for 24/7 access to the CVJAfrica since 2001. www.diabetesjournal.co.za – subscription is 200 ZAR for 24/7 access to the SAJDVD. Major Advances in Cardiology

2010 2005 PREAMI SH I T Perindopril and Remodelling in Elderly f with Acute Myocardial Infarction

2008 2005 BEAUTI f UL

2008 2003

HYPERTENSION IN THE VERY ELDERLY TRIAL

2007 2001

ACTION IN DIABETES VASCULAR DISEASE: PRETERAX AND DIAMICRON MR CONTROLLED EVALUATION

ONCE DAILY ONCE DAILY PREXUM® PREXUM® PLUS 4 - 10 mg COVERSYL® COVERSYL® Perindopril 4 mg/Indapamide 1.25 mg Perindopril

S3 PREXUM® PLUS Tablets. Perindopril 4 mg and Indapamide 1,25 mg. Reg. No. 38/7.1.3/0028. S3 COVERSYL® PLUS Tablets. Perindopril 4 mg and Indapamide 1.25 mg. Reg. No.: 33/7.1.3/0363. S3 PREXUM® 4 mg Tablets. Perindopril 4 mg. Reg. No. 36/7.1.3/0020. S3 PREXUM® 10 mg Tablets. Perindopril 10 mg. Reg. No. A39/7.1.3/0233. S3 COVERSYL® 4 mg Tablets. Perindopril 4 mg. Reg. No.: X/7.1.3/314. S3 COVERSYL® 10 mg Tablets. Perindopril 10 mg. Reg. No. A39/7.1.3/0236. S3 CORALAN® 7,5 mg Tablets. Ivabradine 7,5 mg. Reg. No. A39/7.1.4/4011. For full prescribing information, refer to package insert approved by medicines regulatory authority. NAME AND BUSINESS ADDRESS OF THE HOLDER OF THE CERTIFICATE: SERVIER LABORATORIES SOUTH AFRICA (Pty) Ltd. Reg. No. 72/14307/07. Building Number 4, Country Club Estate, 21 Woodlands Drive, Woodmead 2191. PO Box 930, Rivonia 2128, Republic of South Africa. Tel: +27 (0) 861 700 900. Fax: +27(0)11 525 3401. 144592 Why you should publish with CVJA

• Increased international exposure (indexed in Pubmed, Medline, Scopus, Embase and Thompson Reuters/ISI)

• Quick return on submissions

• Changing patterns of heart disease in Africa get more exposure than in other journals

• Acceptance of diabetes studies as vascular studies in CVJA

• African studies are given preference

• Well-illustrated interventional studies are needed for CME in Africa (full website support capability)

• Instructions for authors on www.cvja.co.za

• A PowerPoint presentation for new authors: 'How to write a scientific paper'

• Submit your manuscript online at www.cvja.co.za

Contact us on [email protected] AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 e1

Case Report

Non-obstructive membranes of the left atrial appendage TANIA BORDONALI, ALBERTO SAPORETTI, ENRICO VIZZARDI, ANTONIO D’ALOIA, ERMANNA CHIARI, LIVIO DEI CAS

Abstract Measurement of the two-dimensional (2D) LAA area is The left atrial appendage (LAA) is a blind-ending, complex not reproducible or helpful, in view of this complex structure. structure distinct from the body of the left atrium and is Improved imaging techniques and the use of biplane and sometimes regarded as a minor extension of the atrium. multi-plane TEE have allowed visualisation of the LAA, which However, it should routinely be analysed as part of a trans- previously was difficult to do using other imaging methods. oesophageal echocardiographic examination. In this study The presence of membranes on the left LAA cavity is rare, and 9 we describe the presence of a non-obstructive membrane their origin is not clear. Embryologically, the trabecular LAA is traversing the cavity of the LAA, found incidentally on trans- a remnant derived from the left wall of the primary atrium, oesophageal echocardiography. which forms during the third and fourth weeks of embryonic development.10 The main smooth-walled left atrial cavity develops Keywords: left atrial appendage, transoesophageal echocardiog- later and is formed from an outgrowth of the pulmonary veins.10 raphy, left atrial thrombosis Reviewing variations in size and morphological characteristics 1,3,7,8 Submitted 6/6/10, accepted 31/5/11 of the LAA, no mention was made of these membranes. When performing a TEE in patients with atrial fibrillation, Cardiovasc J Afr 2012; 23: e1–e2 www.cvja.co.za the pectinate muscles should not be confused with a thrombus. DOI: 10.5830/CVJA-2011-020 Because the accuracy of LAA thrombus detection with TEE is important in the pre-cardioversion evaluation of these patients, it A 70-year-old man without traditional risk factors presented is vital to know what variations, especially in location and size to our emergency room with palpitations and dyspnoea. A of the pectinate muscle, exist in the normal anatomy of the LAA. 12-lead electrocardiogram showed atrial fibrillation with a high Larger pectinate muscles (≥ 1 mm) occur in 97% of LAAs and ventricular rate. Two-dimensional echocardiogram revealed a constitute another potential pitfall in TEE imaging of the LAA. hypertrophic left ventricle with reduced left ventricular function Small (< 1 mm) pectinate muscles are seen only in the first and (ejection fraction 45%). Pre-cardioversion transoesophageal last decades of life.11 echocardiography (TEE) showed an enlarged left atrial appendage The differential diagnosis of long, thin structures in the LAA (LAA) with moderate spontaneous echo contrast and a long, thin may therefore include prominent pectinate muscles, side lobe structure traversing the body of the LAA without obstructing it. artefacts and partial resolution of thrombi. The images in our This image was first ascribed to a thrombus, so oral case, however, seemed different. anticoagulation was started. After four weeks on good The clinical origin and implications of these membranes are anticoagulation, a second TEE was performed and the same not certain, however, they may represent an anatomical variant image with the same dimensions was noted. After electrical that the echocardiographer should be aware of. cardioversion the patient recovered his ventricular function and was discharged on standard therapy.

Discussion When investigating the LAA by TEE, it is important to keep in mind that the LAA is a three-dimensional (3D), multi-lobed structure.1-8 Therefore, evaluation should include imaging in multiple planes, including orthogonal views, in order to image the entire complex 3D structure.

Istitute of Cardiology, University of Brescia, Spedali Civili Brescia, Italy TANIA BORDONALI, MD, [email protected] ALBERTO SAPORETTI, MD ENRICO VIZZARDI, MD ANTONIO D’ALOIA, MD ERMANNA CHIARI, MD LIVIO DEI CAS, MD Fig. 1. Left atrial appendage in transoesophageal echo- cardiography. e2 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

References geal echocardiography. Am J Cardiol 1996; 78: 774–778. 6. Al-Saady NM, Obel OA, Camm AJ. Left atrial appendage: structure, 1. Agmon Y, Khandheria BK, Gentile F, et al. Echocardiographic assess- function, and role in thromboembolism. Heart 1999; 82: 547–555. ment of the left atrial appendage. J Am Coll Cardiol 1999; 34: 7. Veinot JP, Harrity PJ, Gentile F, et al. Anatomy of the normal left atrial 1867–1877. appendage: A quantitative study of age-related changes in 500 autopsy 2. Pollick C, Taylor D. Assessment of left atrial appendage function by : Implications for echocardiographic examination. Circulation transesophageal echocardiography: Implications for the development 1997; 96: 3112–3115. of thrombus. Circulation 1991; 84: 223–231. 8. Ernst G, Stollberger C, Abzieher F, et al. Morphology of the left atrial 3. Yao SS, Meisner JS, Factor SM, et al. Assessment of left atrial append- appendage. Anat Rec 1995; 242: 553–561. age structure and function by transesophageal echocardiography: A 9. Coughlan B, Lang RM, Spencer KT. Left atrial appendage stenosis. J review. Echocardiography 1998; 15: 243–256. Am Soc Echocardiogr 1999; 12: 882–883. 4. Mugge A, Kuhn H, Nikutta P, et al. Assessment of left atrial appendage 10. Al-Saady NM, Obel OA, Camm AJ. Left atrial appendage: structure, function by biplane transesophageal echocardiography in patients with function, and role in thromboembolism. Heart 1999; 82: 547–555. nonrheumatic atrial fibrillation: Identification of a subgroup of patients 11. Meissner I, Whisnant JP, Khandheria BK, et al. Prevalence of potential at increased embolic risk. J Am Coll Cardiol 1994; 23: 599–607. risk factors for stroke assessed by transesophageal echocardiography 5. Rubin DN, Katz SE, Riley MF, et al. Evaluation of left atrial appendage and carotid ultrasonography: The SPARC study. Mayo Clin Proc 1999; anatomy and function in recentonset atrial fibrillation by transesopha- 74: 862–869. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 e3

Case Report

Mobile atheromatous plaque of the aortic arch diagnosed by transthoracic echocardiography prior to coronary artery bypass surgery Which one would you choose: scepticism or wishful thinking?

AC HATEMI, O OMAY, M BASKURT, S KÜCÜKOGLU, B ÖZ, K SÜZER

Abstract than 4 mm) and the complexity of the plaque (ulcerative and/ A routine pre-operative chest X-ray of a patient admitted to or mobile) are important risk factors for unexplained arterial our institution for an elective coronary artery bypass opera- embolic events such as stroke, transient ischaemic attack and tion revealed a mildly dilated mediastinal silhouette, which peripheral emboli, together with multi-organ failure and death. led the cardiovascular surgery resident to schedule emer- Furthermore, mobile atheromas of the aortic arch are associated gency transthoracic echocardiography (TTE), with a special with increased peri-operative strokes in patients undergoing 2 note asking for detailed evaluation of the ascending aorta cardiac surgery. Stroke incidence was found to be around 25% and aortic arch. TTE revealed a mobile atheroma at the in patients with mobile plaques of the aortic arch, while it was 3 aortic arch, which obliged the cardiac surgery team to modi- only 2% in patients with quiescent non-mobile plaques. fy their strategy to combined hemi-arcus aortae replacement Potential aetiological risks independently associated with and coronary artery bypass grafting (CABG). Although with complex plaque formation are advanced age, history of transoesophageal echocardiography (TEE) a small portion hypertension, hypercholesterolaemia, increased body mass index, of the ascending aorta may be obscured by the trachea, TEE diabetes, and past or present tobacco use. Similarly, established provides higher resolution images than TTE. Therefore one risk for stroke occurrence are advanced age, male gender, previous can conclude that TEE is the imaging modality of choice for stroke history, heredity, hypertension, smoking, diabetes mellitus, 4,5 detecting aortic atheromatous plaques but in patients with carotid artery disease, coronary artery disease and polycythaemia. low risk for stroke and aortic atheromas, a detailed TTE We can clearly conclude that risk factors for atherosclerosis may be sufficient for the pre-operative assessment. and stroke overlap. In fact, in cardiac patients without clinical evidence of severe atherosclerotic disease, a high prevalence of Keywords: aortic arch, atherosclerosis, aorta, echocardiogra- combined aortic and carotid plaques was reported.4 Surgeons phy, circulatory arrest, coronary artery bypass grafts, CABG should consider these patients as strong candidates for Submitted 26/1/10, accepted 3/6/11 pre-operative and postoperative athero-embolic complications. Transoesophageal echocardiography (TEE), which is a safe Cardiovasc J Afr 2012; 23: e3–e5 www.cvja.co.za and relatively less invasive procedure with a very low risk of DOI: 10.5830/CVJA-2011-027 complication is the modality of choice for the diagnosis of aortic atheromas, although CT, MRI and intra-operative epi-aortic The aorta is an important source of athero-emboli, as recent ultrasonography are known to be complementary examination studies have confirmed the strong correlation between severe techniques.6 The progress in TEE technology has enabled surgeons aortic atheromatous plaques and stroke/death in the elderly.1 to obtain a detailed view of the aorta pre- and peri-operatively, to The dimensions of the aortic arch atheroma (larger and thicker quantify atheromatous plaques according to their thickness and the presence of mobile components, therefore classifying them Department of Cardiovascular Surgery, Institute of as simple or complex. In one study, TEE was able to find aortic Cardiology, Istanbul University, Istanbul, Turkey arch atheromatous disease in 55% of patients with a normal chest AC HATEMI, MD, PhD, [email protected] X-ray, and 91% of those had heavily calcified aortic knobs.7 O OMAY, MD We assumed that TEE evaluation of the aorta is useful in older K SÜZER, MD patients with risk factors for stroke and those with radiographic Department of Cardiology, Institute of Cardiology, Istanbul evidence of aortic calcification, to determine the presence of University, Istanbul, Turkey severe atheromatous disease of the aortic arch pre-operatively. M BASKURT, MD However, TEE is a semi-invasive procedure, which mostly S KÜCÜKOGLU, MD requires sedation, is not always readily available, and may not be Department of Pathology, Cerrahpasa Faculty of Medicine, suitable for haemodynamically unstable patients. In this report, Istanbul University, Istanbul, Turkey we highlight that in some patients such as ours but not in all, B ÖZ, MD TTE may be used instead of TEE in this manner. e4 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Fig. 1. Pre-operative chest X-ray showing a dilated and mildly calcified aortic arch (arrows). Case report A 65-year-old man with a history of stable angina was admitted to our institution for an elective CABG operation. He had hypertension, hypercholesterolaemia, diabetes mellitus controlled with oral medication and a smoking history of 50 years. The patient had no neurological complaints. The central nervous system and cardiovascular system were normal. On his chest X-ray, the aortic arch seemed dilated and Fig. 3. Photograph taken at the end of the opera- mildly calcified (Fig. 1). Coronary angiography revealed tion. Hemi-arcus replacement (no: 28 gel-coated dacron triple-vessel disease with a left main coronary artery lesion vascular graft) with proximal coronary anastomoses of of 60%. In our protocol, all patients scheduled for CABG the saphenous grafts constructed directly to the aortic graft. SVG: saphenous vein graft, Ao: dacron vascular operations are simultaneously scheduled for a pre-operative graft. TTE examination for evaluation of their valvular and ventricular functions. In this patient with a mildly enlarged mediastinal The surgical strategy was modified due to these findings silhouette on chest X-ray, the referring cardiovascular surgeon and the arterial cannulation site was moved to the innominate involved the echocardiograhy laboratory for a detailed artery with a regular two-staged venous cannulation, followed evaluation of the ascending aorta and aortic arch. TTE by a hemi-arcus aorta replacement with a quadruple CABG performed at our institution showed minimal aortic regurgitation (left internal thoracic artery–left anterior descending artery with an ejection fraction of 60% and a mobile atheroma bypass graft, aorta–diagonal artery–obtuse marginal branch at the aortic arch with minimal aortic dilatation (Fig. 2).

A B C D Fig. 4. Macroscopic and histopathological view of the aortic arch specimen. A. Excisional aortic specimen with the mobile atheroma in the aortic arch (arrow) showing rupture of the tunica intima, atheromas with ulceration, Fig. 2. Transthoracic echocardiographic examination and a pedunculated thrombus formation attached to the revealing a mobile atheroma at the aortic arch with mini- arterial wall. B, C and D. Histopathological examination of mal aortic dilatation. Asc Ao: ascending aorta, LCCA: the specimen showing atherosclerotic intimal changes, left common carotid artery, LSCA: left subclavian artery, chronic fibrosis and full-thickness degeneration of the Desc Ao: descending aorta, M: mobile atheroma. artery. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 e5

of the circumflex artery bypass with saphenous vein graft, to the aorta). The small portion of the ascending aorta that is aorta–right coronary artery bypass with sapheneous vein graft) masked by the trachea near the origin of the innominate artery (cardiopulmonary bypass time: 123 min, total circulatory arrest may not be seen on TEE and only 2% of the plaques may be time at 18°C: 25 min, cross-clamp time: 81 min). missed with this modality.11 After distal coronary anastomoses, a segment from the As a significant proportion of aortic plaques of stroke supracoronary aorta to the left subclavian artery ostium was patients can be demonstrated with TTE, the necessity for a excised and the hemi-arcus was replaced with a number 28 TEE evaluation, which is a relatively invasive procedure, will gel-coated dacron vascular graft. Lastly, proximal coronary automatically diminish. Unfortunately, combined TEE and TTE anastomoses of the saphenous grafts were performed directly to examinations of the aortic arch may still be mandatory in a the aortic graft (Fig. 3). No operative/postoperative embolic or subgroup of patients, to rule out the presence of atherosclerotic other complications were experienced following the successful plaque. TTE imaging of the aortic arch is also useful for operation. sequential evaluation of the plaques already identified with Pathological examination of the specimen revealed TEE, and therefore helps physicians to omit repetitive TEE macroscopically: rupture of the tunica intima, an atheromatous examinations. plaque with ulceration, and a pedunculated thrombus formation attached to the arterial wall, and microscopically: atherosclerotic Conclusion intimal changes, chronic fibrosis, and full-thickness degeneration Routine TTE evaluation is a valuable modality, particularly for of the artery (Fig. 4). The patient was discharged from the ICU elderly candidates, for aortic cannulation in open-heart surgery. on the third postoperative day and from hospital on the 15th In our case, the pre-operative TTE examination of the patient postoperative day. The patient presented for routine cardiology enabled the surgeon to make the correct and custom-designed and cardiovascular surgery follow up without any complaints. operative decision, which assured a safe procedure and better surgical outcome. Discussion References Although several variables were identified as risk factors for 1. Sharifkazemi MB, Aslani A, Zamirian M, Moaref AR. Significance of peri-operative stroke, the majority of strokes occur in patients aortic atheroma in elderly patients with ischemic stroke. A hospital- where no definitive aetiological factors can be identified. All based study and literature review. Clin Neurol Neurosurg 2007; 109(4): patients undergoing cardiac surgery may have atherosclerotic 311–316. aortic plaques with no clinical evidence, and these are a potential 2. Tunick PA, Kronzon I. Atheromas of the thoracic aorta: clinical and source of serious peri-operative or postoperative athero-embolic therapeutic update. J Am Coll Cardiol 2000; 35(3): 545–554. 3. Barbut D, Lo YW, Hartman GS, Yao FS, Trifiletti RR, Hager DN, et al. complications. Thorough pre-operative echocardiographic Aortic atheroma is related to outcome but not numbers of emboli during evaluation of the patient, and particularly of the elderly, is crucial coronary bypass. Ann Thorac Surg 1997; 64(2): 454–459. for an uneventful surgical outcome. 4. Tribouilloy C, Peltier M, Colas L, Senni M, Ganry O, Rey JL, Lesbre TTE compares favourably with TEE in the identification of JP. Fibrinogen is an independent marker for thoracic aortic atheroscle- atheromatous plaques of the aortic arch and distal ascending rosis. Am J Cardiol 1998; 81(3): 321–326. aorta, although it is less effective in identifying simple plaques 5. Katz ES, Tunick PA, Rusinek H, Ribakove G, Spencer FC, Kronzon I. Protruding aortic atheromas predict stroke in elderly patients of the proximal ascending aorta. The demonstration of aortic undergoing cardiopulmonary bypass: experience with intraoperative plaque with TEE has a sensitivity of 91%, specificity of 82%, transesophageal echocardiography. J Am Coll Cardiol 1992; 20(1): and positive and negative predictive values of 72 and 95%, 70–77. respectively.8 However, TEE, which is a sensitive technique to 6. Daniel WG, Erbel R, Kasper W, Visser CA, Engberding R, Sutherland determine protruding aortic atheromas with or without a mobile GR, et al. Safety of transesophageal echocardiography. A multicenter component, cannot always visualise plaques located in the distal survey of 10,419 examinations. Circulation 1991; 83(3): 817–821. ascending aorta and proximal aortic arch.9 7. Marschall K, Kanchuger M, Kessler K, Grossi E, Yarmush L, Roggen S, et al. Superiority of transesophageal echocardiography in detecting Weinberger et al. reported that TTE could be used to visualise aortic arch atheromatous disease: identification of patients at increased plaques in the distal ascending aorta and aortic arch, and risk of stroke during cardiac surgery. J Cardiothorac Vasc Anesth 1994; particularly plaques at the junction of the ascending aorta and 8: 5–13. aortic arch that could be obscured by the bronchi and may be 8. Tribouilloy C, Peltier M, Colas L, Rida Z, Rey JL, Lesbre JP. missed by TEE.9 TTE was able to detect simple plaques undetected Multiplane transoesophageal echocardiographic absence of thoracic with TEE, particularly in the proximal ascending aorta. Konstadt aortic plaque is a powerful predictor for absence of significant coronary et al. reported that up to 42% of the ascending aorta cannot artery disease in valvular patients, even in the elderly. A large prospec- tive study. Eur Heart J 1997; 18(9): 1478–1483. be visualised by TEE, so potential embolic plaques can be 9. Weinberger J, Azhar S, Danisi F, Hayes R, Goldman M. A new nonin- 10 missed by that modality. All complex plaques, morphologically vasive technique for imaging atherosclerotic plaque in the aortic arch: similar and visualised with TEE were also demonstrated with an initial report of stroke patients by transcutaneous real-time B-mode TTE. Both techniques are able to identify the plaques as ultrasonography. Stroke 1998; 29; 673–676. pedunculated or proliferative and to picture their mobility. 10. Konstadt SN, Reich DL, Quintana C, Levy M. The ascending aorta: To our knowledge, there is no study comparing TTE how much does transesophageal echocardiography see? Anesth Analg 1994; 78: 240–244. with TEE in detecting mobile atheromatous plaques. Most 11. Krinsky GA, Freedberg R, Lee VS, Rockman C, Tunick PA. Innominate echocardiographers feel that TEE is more accurate than TTE artery atheroma: a lesion seen with gadolinium-enhanced MR angiog- for the critical measurement of plaque thickness and for the raphy and often missed by transesophageal echocardiography. Clin diagnosis of mobile thrombi (high resolution and proximity Imaging 2001; 25: 251–257. e6 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Case Report

Successful stenting of catheter-induced unprotected left main coronary artery dissection G ERTAŞ, E URAL, WJ VAN DER GİESSEN

Abstract common femoral artery for haemodynamic support during the Catheter-induced left main coronary artery (LMCA) dissec- procedure. A floppy coronary guide wire could not be advanced tion is a dramatic, although uncommon complication of diag- to the distal left anterior descending (LAD) artery, so with the nostic coronary angiography and requires prompt treatment. wire in the first septal branch, a bare-metal stent (4 × 24-mm We describe a case of iatrogenic occlusive dissection of the Liberté) was implanted in the LMCA towards the LAD (Fig. LMCA during coronary angiography, treated by subsequent 2A). Circumflex coronary arterial flow was restored but because percutaneous recanalisation. of a spiral dissection, there was no flow in the distal LAD (Fig. 2B). Keywords: LMCA, coronary artery dissection, stent Subsequent attempts to advance the wire were unsuccessful. Submitted 30/12/09, accepted 26/6/11 Leaving the first wire in the mid-LAD, a second floppy wire could easily be advanced to the distal vessel. A second bare- Cardiovasc J Afr 2012; 23: e6–e7 www.cvja.co.za metal stent (3 × 32-mm Liberté) was implanted, overlapping with DOI: 10.5830/CVJA-2011-033 the first stent. TIMI 3 flow was obtained (Fig. 2C). Post-procedure troponin levels remained negative and the echocardiogram revealed no change in wall-motion abnormality Case report compared to baseline. After an uneventful period, the patient A 58-year-old female with a 10-year history of hypertension was was discharged on post-procedural day nine. She remained admitted with stable angina pectoris (Canadian Cardiovascular asymptomatic during one year of follow up. Society class II). On admission, the patient’s arterial pressure was 130/70 mmHg and her heart rate was 84 beats/min. The ECG showed negative T waves in leads V4–6. An exercise treadmill A B test was inconclusive. Transthoracic echocardiography revealed global left ventricular hypokinesia with a left ventricular ejection fraction of 40%. She underwent diagnostic coronary angiography. Left coronary angiography revealed a normal arterial tree (Fig. 1A). After the first two contrast injections, the patient experienced severe chest pain and became hypotensive. The electrocardiogram showed ST-segment elevation. Coronary injection revealed there was no contrast medium passing beyond the distal left main coronary artery (LMCA) (Fig. 1B). Catheter-induced LMCA dissection Fig. 1. (A) Left coronary angiography revealed a normal arterial tree. (B) Coronary injection revealed there was no was concluded and because of the haemodynamic deterioration, contrast medium passing beyond the distal LMCA. urgent percutaneous coronary intervention was performed. After a loading dose of 600 mg clopidogrel and 300 mg of A B C aspirin, an intra-aortic balloon pump was inserted from the left

Department of Cardiology, Faculty of Medicine, Bezmialem Vakif University, Istanbul, Turkey G ERTAŞ, MD, [email protected] Department of Cardiology, Medical Faculty, and Interventional Cardiology Research and Application Unit, Kocaeli University, Kocaeli, Turkey E URAL, MD Fig. 2. (A) A bare-metal stent from the LMCA to the LAD Department of Cardiology, Thoraxcentre, Erasmus MC, was implanted. (B) After stent placement, flow to the Rotterdam, and Interuniversity Cardiology Institute of the circumflex artery was restored, but there was no flow in Netherlands, ICIN-KNAW, Utrecht, the Netherlands the distal LAD because of the dissection. (C) A second WJ VAN DER GİESSEN, MD, PhD bare-metal stent was implanted slightly overlapping with the first stent. TIMI 3 flow was obtained. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 e7

Discussion term survival has been reported for percutaneous coronary Catheter-induced LMCA dissection is a rare complication of intervention in patients with LMCA dissection. No cardiac diagnostic coronary angiography, but has a very high mortality deaths have occurred during a mean follow up of 30 months, rate if not treated. In previous reports, authors have defined with 84.6% angiographic success. Therefore, particularly in PCI 2 this situation as the ‘nemesis of interventional cardiologists’.1 centres, it should be considered as the first treatment option. The incidence of catheter-induced LMCA dissection has been reported as less than 0.1% in several articles,2,3 but more than 10 References times higher in patients with left main coronary artery stenosis 1. Jain D, Kurowski V, Katus HA, Richardt G. Catheter-induced dissection 4 (1.05%). of the left main coronary artery, the nemesis of an invasive cardiologist Vigorous contrast injections while the catheter is not co-axial, a case report and review of the literature. Z Kardiol 2002; 91: 840–845. not using a soft-tipped catheter, deep intubation, and careless 2. Cheng CI, Wu CJ, Hsieh YK, Chen YH, Chen CJ, Chen SM, et al. catheter manipulation can all cause LMCA dissection. After Percutaneous coronary intervention for iatrogenic left main coronary LMCA stenosis, patients with hypertension, Marfan syndrome, artery dissection. Int J Cardiol 2008; 126: 177–182. congenitally unicuspid and bicuspid aortic valves, and cystic 3. Dunning DW, Kahn JK, Hawkins ET, O’ Neill WW. Iatrogenic coronary artery dissections extending into and involving the aortic root. Cathet medial necrosis have been reported to have a higher risk factor Cardiovasc Intervent 2000; 51: 387–393. 5 for dissection. 4. Kovac JD, de Bono DP. Cardiac catheter complications related to left main stem disease. Heart 1996; 76: 76–78. 5. Awadalla H, Sabet S, El SA, Rosales O, Smalling R. Catheter-induced Conclusion left main dissection incidence, predisposition and therapeutic strategies Traditionally, CABG has been the treatment of choice, but experience from two sides of the hemisphere. J Invasive Cardiol 2005; it delays the time until normal flow is restored. Good long- 17: 233–236. e8 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

Case Report

New images in carcinoid heart disease M KLOBUČIĆ, MH PAAR, RŠ PADOVAN, J VINCELJ, B FILA

Abstract deposition of fibrous tissue on the endocardial surfaces of the Echocardiography was the main imaging technique in the heart. They are mainly found on the right side of the heart and are diagnosis and follow up of carcinoid heart disease but in the truly pathognomonic. Echocardiography was the main imaging last decade magnetic resonance imaging (MRI) has evolved technique in the diagnosis and follow up of CHD, but in the last into a new diagnostic modality. Most of the reported MRI decade, magnetic resonance imaging (MRI) has evolved into a features were similar to those observed by echocardiography new diagnostic modality. – tricuspid and/or pulmonary valve thickening and immobil- ity with consequent valvular dysfunction and right heart Case report enlargement. To our knowledge, this is the first report describ- A 62-year-old woman was admitted to surgery in 1996 with ing endocardial enhancement of the right cardiac chambers, symptoms of small-bowel obstruction. Exploration of the tricuspid valve and subvalvular apparatus, which corre- sponds with histologically seen fibrous carcinoid plaques. Keywords: carcinoid heart disease, magnetic resonance imag- ing, echocardiography Submitted 30/1/11, accepted 18/7/11 Cardiovasc J Afr 2012; 23: e8–e10 www.cvja.co.za DOI: 10.5830/CVJA-2011-038

Carcinoid tumours are rare (1.2 to 2.1 per 100 000 general population per year), slow-growing, malignant tumours arising from the neural crest amine precursor uptake and decarboxylation cells (APUD cells), in particular from those situated in the gastrointestinal tract. When these tumours metastasise to the liver, their vasoactive products (5-hydroxytryptamine or serotonin, histamine, tachykinins, prostaglandins etc.), which are regularly metabolised in the endothelium of the liver and lung vasculature, can reach the circulation and thereafter produce symptoms of carcinoid syndrome (CS), characterised by dermal flushing, diarrhoea, bronchospasm and valvular carcinoid heart disease (CHD). The latter is the most serious manifestation of CS, leading to heart failure and death in 40% of all affected. Cardiac lesions are a consequence of serotonin-induced

Department of Internal Medicine, Bjelovar General Hospital, Croatia M KLOBUČIĆ, MD, [email protected] Department of Diagnostic and Interventional Radiology, University Hospital Center Zagreb and School of Medicine, University of Zagreb, Zagreb, Croatia MH PAAR, MD RŠ PADOVAN, MD Fig.1. Echocardiography demonstrating severe thicken- Institute of Cardiovascular Diseases, Dubrava University ing, retraction, shortening and almost complete immobi- Hospital, Zagreb, Croatia lisation of the tricuspid valve (TV) leaflets (a), leading to severe tricuspid regurgitation (b). Severe dilatation and J VINCELJ, MD trabeculation of the right ventricle (RV) with flattening Department of Surgery, Bjelovar General Hospital, Croatia of the interventricular septum (c). Characteristic dagger- shaped signal of tricuspid regurgitation; mild tricuspid B FILA, MD stenosis (d). RA – right atrium, LV – left ventricle. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 e9

Fig. 2. Magnetic resonance imaging. FSE T1-weighted black-blood four-chamber-view image presenting thickened and retracted tricuspid valve leaflets (white arrows, a). Systolic (b), and diastolic (c) phase of ciné SSFP four-chamber- view images demonstrating the signal void of tricuspid regurgitation (black arrowhead) and tricuspid stenosis (white arrowhead). A small pericardial effusion (asterisk) and enlargement of the right atrium (RA) and right ventricle (RV) were also observed. Endocardial delayed enhancement of the right atrium, tricuspid valve leaflets and right ventricle (black arrows, d). abdominal cavity revealed a large, inoperable tumour of the of the pericardium was noted and MR imaging of the heart was terminal ileum. A carcinoid tumour was diagnosed from the performed to exclude pericardial involvement. Cardiac MRI histological specimens. The first liver metastases were detected was performed using the 3.0-Tesla MRI system (Magnetom in 1997. Trio, Siemens Medical Solutions, Erlangen, Germany) with After the first symptoms of CS (flushing, diarrhoea, oedema a 12-element cardiac array coil, breath-hold technique and of the ankles) emerged in 2002, somatostatine analogue therapy electrocardiographic gating. After obtaining axial half-Fourier was introduced. This led to good control of the CS symptoms, acquisition single-shot turbo spin-echo (HASTE) images and and the urinary levels of 5-hydroxyindoleacetic acid (5-HIAA), localiser images in three planes, ciné loops of long- and short- a byproduct of serotonin degradation used for the detection axis views were acquired using the steady-state free precession and follow up of patients with carcinoid tumours, significantly technique (SSFP). In addition, anatomical ‘black-blood’ fast- decreased. spin-echo (FSE) T1-weighted and short-tau inversion recovery In 2007 the patient started to complain of fatigue and (STIR) T2-weigted four-chamber view images were obtained. dyspnoea on exertion. An echocardiographic examination (Aloka Delayed enhancement was evaluated 10 to 15 minutes after SD 4000 and Philips i22) revealed pathognomonic signs of CHD intravenous injection of a ‘double dose’ (0.2 mmol/kg) of – severe thickening, retraction, shortening and almost complete gadopentetate dimeglumine (Magnevist, Bayer Schering Pharma immobilisation of the tricuspid valve leaflets (Fig. 1a) with AG, Berlin, Germany) using the inversion prepared gradient- severe tricuspid regurgitation (Fig. 1b), mild tricuspid stenosis, echo technique with inversion time set to nullify the myocardial pulmonary valve stenosis and severe dilatation and trabeculation signal. Systolic function of both ventricles was measured using of the right ventricle (Fig. 1c). semi-automatic contour-detection software (ARGUS, Siemens The continuous-wave Doppler signal analysis showed a Medical Solutions, Forchheim, Germany). characteristic dagger-shaped signal of tricuspid regurgitation with The right atrium was severely enlarged, while the leaflets of a maximal velocity of 240 cm/s, demonstrating rapid decline in the tricuspid valve were thickened and barely mobile (Fig. 2a). pressure difference between the right ventricle and right atrium The right ventricle was enlarged with diastolic flattening of the (Fig. 1d). As a result of volume overload of the right ventricle, interventricular septum due to right ventricle volume overload. diastolic flattening of the interventricular septum was noticed. Signal voids caused by the turbulent flow through the tricuspid No pericardial effusions were detected but hyperechogenicity valve during and diastole were observed, pointing to e10 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

significant tricuspid regurgitation and stenosis, respectively (Fig. a few reports of the MRI features of this condition have been 2b, c). published.4-7 Most of these features are similar to those observed After intravenous administration of gadolinium-based contrast echocardiographically, as shown in our patient – tricuspid and/ agent, delayed enhancement of the tricuspid valve and endocardial or pulmonary valve thickening and immobility with consequent layer of the right cardiac chambers was evidenced, indicating a valvular dysfunction and right heart enlargement. fibrous carcinoid plaque deposited on the endocardium (Fig. 2d). A single article reporting contrast-enhancement characteristics The global systolic function of both ventricles was maintained. of CHD was found where delayed enhancement of the tricuspid A small circular pericardial effusion was present, without septal leaflet, mostly marked at the annulus, was demonstrated.4 pericardial thickening or pathological enhancement. No signs of In our patient, endocardial enhancement of the right atrium, right left-sided carcinoid heart disease were found. ventricle including the endocardial surface of the subvalvular apparatus, and both sides of the tricuspid valve was observed. Discussion Carcinoid tumours are slow growing and indolent, but Conclusion development of CHD symptoms dramatically reduces life We believe that this type of late enhancement, not previously expectancy. Standard palliative therapy in CS, therapy with described in the literature, perfectly corresponds with the well- somatostatine analogues, although reducing levels of serotonin known histological features of CHD. and achieving good control of other CS symptoms in almost 80% of patients, does not induce regression of the cardiac manifestations and most of the patients with carcinoid tumours References die from progressive right heart failure as a consequence of 1. Miller D, Farah MG, Liner A, Fox K, Schluchter M, Hoit BD. The rela- regurgitation and stenosis of the tricuspid and pulmonary valves. tion between quantitative right ventricular ejection fraction and indices of tricuspid annular motion and myocardial performance. J Am Soc As surgical intervention with annuloplasty or valve Echocardiogr 2004; 17: 443–447. replacement is often the only therapeutic option, it is important 2. Maceira AM, Prasad SK, Khan M, Pennell DJ. Reference right ventric- to follow up the patients regularly to detect the proper timing ular systolic and diastolic function normalized to age, gender and body for surgery. In recent reports, operative risk has declined to surface area from steady state free precession cardiovascular magnetic below 10% when surgical intervention was performed before resonance. Eur Heart J 2006; 27: 2879–2888. the development of severe right heart failure, in even mildly 3. Mirowitz SA, Gutierrez FR. MR and CT diagnosis of carcinoid heart symptomatic patients where the referral to surgery was based on disease. Chest 1993; 103: 630–631. 4. Martos R, Ridge C, Quinn M, Dodd J. Cardiac carcinoid: tricuspid severe valvular dysfunction and signs of right ventricular volume delayed hyperenhancement on cardiac 64-slice multidetector CT overload. and magnetic resonance imaging. Ir J Med Sci, published online: 22 Therefore, the most precise imaging technique should be January 2009. doi: 10.1007/s11845-008-0273-5. used for evaluation of right ventricular volumes and function. 5. Franzen D, Boldt A, Raute-Kreinsen U, Koerfer R, Erdmann E. While interobserver variability of quantitative assessment Magnetic resonance imaging of carcinoid heart disease. Clin Cardiol of the right ventricular ejection fraction by two-dimensional 2009; 32: E92–E93. 6. Mollet NR, Dymarkowski S, Bogaert J. MRI and CT revealing carci- echocardiography is up to 15 ± 13%,1 MRI shows much lower noid heart disease. Eur Radiol 2003; 13(Suppl 6): L14–L18. 2 values, around 4%, and should be in our opinion, the reference 7. Bastarrika G, Cao MG, Cano D, Barba J, de Buruaga JDS. Magnetic standard for follow up of patients with CHD. resonance imaging diagnosis of carcinoid heart disease. J Comput Since the first description of the MRI findings of CHD,3 only Assist Tomogr 2005; 29: 756–759. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 e11

Case Report

Aortic dissection, a complication during successful angioplasty of chronic total occlusion of the right coronary artery, was treated conservatively S CHUNLAI, PR STELLA, A BELKACEMI, P AGOSTONI

Abstract was 80% stenosis in the first obtuse marginal branch and Aortic dissection is an uncommon but potentially lethal 30% stenosis in the distal left circumflex coronary artery. The complication that can occur during the engagement of a mid-portion of the RCA was occluded, with poor collateral flow guiding catheter. We report a case of a 59-year-old woman to the distal RCA (Fig. 1). with acute aortic dissection due to percutaneous coronary After discussion with the cardiac surgeons, we decided on a intervention. This resulted in a retrograde extension of the staged PCI; first a PCI of the RCA and then a PCI of the obtuse dissection into the proximal part of the ascending aorta. With marginal branch was planned. A 7 french Judkins JR4 guiding haemodynamic stability, we decided to treat the aortic dissec- catheter (Bosotn Scientific) was used in combination with a tion medically. As a result, a complete resolution of the aortic 0.014-inch Pilot 50 and Persuader 3 guide wire to cross the dissection was documented by coronary angiography and the occlusion of the RCA. After successful recanalisation with the follow-up computed tomography scan. CTO wire, we changed to a normal floppy guide wire (Cougar, Medtronic) and performed gradual pre-dilations with a 2.0 × Keywords: aortic dissection, percutaneous coronary intervention 10-mm Sapphire, 3.0 × 15-mm Sapphire, 3.0 × 30-mm Pantera Submitted 15/6/11, accepted 6/9/11 and 3.50 × 15-mm XTRM Force balloon. Then we deployed three Endeavor REsolute stents (2.5 × 24 mm/2.75 × 24 mm/3.0 Cardiovasc J Afr 2012; 23: e11–e13 www.cvja.co.za × 30 mm) from the distal to mid-RCA. During placement of the DOI: 10.5830/CVJA-2011-050 second stent, the guiding position was lost and re-inserted. At the end of the procedure, we noticed an aortic dissection Dissection of the ascending aorta is a very rare iatrogenic in the proximal part of the ascending aorta (DeBakey type II, complication of percutaneous coronary intervention (PCI).1 We Stanford type A) with good antegrade flow and no residual report here on a case of accidental aortic dissection that occurred stenosis in the RCA (Fig. 2). The patient reported slight transient during PCI of a chronic total occlusion (CTO) lesion. chest pain not associated with nausea, vomiting, shortness of breath or palpitations. Her blood pressure was 120/75 mmHg (equal on both arms) and the pulse 78 beats per minute. Case report A 59-year-old woman with stable angina on limited effort was admitted to our hospital for PCI of a CTO lesion, which was located at the level of the mid-right coronary artery (RCA) after a previous non-Q-wave acute myocardial infarction. Echocardiography showed normal left ventricular systolic function without regional wall-motion abnormalities. Her coronary risk factors were hypertension, diabetes, hypercholesterolaemia, family history of coronary artery disease and smoking. On coronary angiography, the left main coronary artery and left anterior descending coronary artery were normal. There

Department of Cardiology, Second Affiliated Hospital of Soochow University, SuZhou, China S CHUNLAI, MD Department of Cardiology, University Medical Center Utrecht, Utrecht, the Netherlands S CHUNLAI, MD, [email protected] PR STELLA, MD, PhD A BELKACEMI, MD Fig. 1. Coronary angiography showed the mid-portion of P AGOSTONI, MD, PhD the RCA occluded, with poor collateral flow to the distal RCA. e12 CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 AFRICA

A B

Fig. 2. During injection of the right coronary artery, angiography showed Stanford type A (DeBakey type II) dissection in the proximal part of the aorta (A), and good anterograde flow in the RCA (B). Since both haemodynamically and angiographically there pericardial effusion was present with no evidence of tamponade were no changes, even after 25 minutes, we decided to stop (Fig. 4). the procedure and not bring the patient in for surgery. Cardiac The patient was discharged uneventfully and seen in the computed tomography (CT) revealed a type A dissection with cardiology outpatient clinic after 10 days. She had no complaints, a predominantly intramural haematoma. The dissection did with stable systolic blood pressure between 100 and 110 mmHg. not extend into the coronary arteries and cerebral vessels, but Four weeks later, during the scheduled second procedure for stopped at the origin of the arteria brachiocefalica (Fig. 3). PCI of the right circumflex coronary artery, aortography showed Because of her haemodynamic stability, we decided to treat no sign of the dissection in the proximal part of the aorta, and a the aortic dissection medically, so the patient was transferred well-maintained result of the RCA-CTO (Fig. 5). to the coronary care unit. Metoprolol tartrate 300 mg/day, telmisartan 80 mg/day, amlodipine 5 mg/day, enteric-coated Discussion aspirin 100 mg/day and clopidogrel 75 mg/day were given to the Aortic dissection is a rare complication after percutaneous patient. After five days, a follow-up CT showed a stable situation coronary intervention,1 with a reported incidence of 0.03 to with a considerable amount of sub-intimal haemorrhage at the proximal part of aorta, with unchanged dimensions. A small

Fig. 4. Five days after the event, a computed tomographic scan showed decreased echogenity with a consider- able amount of sub-intimal haemorrhage at the proximal Fig. 3. Computed tomographic scan revealed a type A part of the aorta. The thrombosed false lumen was not dissection with a predominantly intramural haematoma. expanding to arch. AFRICA CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012 e13

A B

Fig. 5. After four weeks, aortic angiography showed that the dissection was absent in the proximal part of the aorta (A), and the RCA was patent (B).

0.1%.2,3 The exact mechanism of aortic dissection caused by a stable haemodynamic status, a conservative, medical treatment PCI is unclear. Aortic dissections generally relate to retrograde was chosen with strict blood pressure control in order to prevent extension of a proximal dissection of a coronary artery. further extension. Forceful manipulation of guide wires, vigorous injection of the contrast medium, and accidental manipulation of the guiding catheter have also been suggested to trigger coronary and aortic Conclusion dissections during PCI.1 In this case, the aortic dissection may Aortic dissection is an uncommon but potentially lethal have been caused by mechanical trauma caused by the tip of the complication that can occur during PCI. Provided there is good guiding catheter while re-inserting into the ostium of the RCA. haemodynamic stability and no interference with other vessels Acute dissections involving the ascending aorta (DeBakey type coming out of the ascending aorta, vigilant medical therapy, as I and II, Stanford type A) are considered surgical emergencies described in our case report, may be preferable to surgery. requiring swift repair of the aortic root or reconstruction of the ascending aorta and the arch to improve prognosis. By contrast, References dissections confined to the descending aorta (DeBakey type II, 1. Ruda-Vega M. Aortic dissection – exceedingly rare complication of Stanford type B) are treated medically unless progression of the coronary angioplasty. Catheter Cardiovasc Diagn 1997; 42(4): 416. dissection, intractable pain, organ hypo-perfusion or extra-aortic 2. Yip HK, Wu CJ, Yeh KH, et al. Unusual complication of retrograde blood are demonstrated.4 dissection to the coronary sinus of Valsalva during percutaneous revas- If a surgical strategy is selected, there is a high risk cularization: a single center experience and literature review. Chest of potentially life-threatening bleeding complications due 2001; 119: 493–501. to aggressive anti-platelet treatment for these stent-treated 3. Alfonso F, Almeria C, Fernandez-Ortiz A, et al. Aortic dissection occurring during coronary angioplasty: angiographic and transesopha- patients. With β-blockers, in combination with other anti- geal echocardiographic findings. Cathet Cardiovasc Diagn 1997; 42: hypertensives, blood pressure levels between 100 and 120 412–415. mmHg are achievable.4 As this dissection remained localised to 4. Ince H, Nienaber CA. Diagnosis and management of patients with the anterolateral portion of the ascending aorta, combined with aortic dissection. Heart 2007; 93(2): 266–270. “THE CURRENT AHA DIETARY GUIDELINES RECOMMEND COMBINED EPA & DHA IN A DOSE OF APPROXIMATELY 1000 mg/DAY IN PATIENTS WITH CHD “ Lavie et al, Omega - 3 Polyunsaturated Fatty Acids and Vol. 54,No. 7 2009, August 11, 2009 585– 594 TRIMEGA™

* HIGH STRENGTH EPA & DHA 300 MG EPA AND 200 MG DHA IN A SINGLE CAPSULE

* ENSURED PATIENT COMPLIANCEWITH CONVENIENTBD DOSING

* SABS TESTED FOR DECLARED CONTENT OF ACTIVE MOEITIES.

* FORMULATION ALLOWS QUICK AND EASY DISSOLUTION.

* NO ERUCTATION

For further information contact

pH PHARRM AF ICA HEALTHCARE (PTY) LTD Reg. No. 1993/003911/07 PRIVATE BAG X8, ROSETTENVILLE 2130

TOLL FREE 0800 601 098 MAY 2012 VOL 23 NO 4

AFRICA

www.cvja.co.za CardioVascular Journal of Africa (official journal for PASCAR)

• Endothelial dysfunction

• Cardiac surgery and haematological malignancies

• Cardiovascular risk factors in coronary artery disease

• Changes in NT-proBNP

• Effusive–constrictive pericarditis in Ibadan

• Pneumonia in post cardiac surgery

• Cardiovascular disease in the Seychelles, Tanzania and Mauritius

PUBLISHED ONLINE: • Mobile atheromatous plaque of the aortic arch Cardiovascular Journal of Africa . Vol 23, No 4, May 2012 • Images in carcinoid heart disease Printed by Durbanville Commercial Printers Tel: 021 946 4074 Tel: Printed by Durbanville Commercial Printers