Prof. Salvatore Novo,
Coronary microcirculation in patients with angina and risk factors yet with a coronary bed free from atherosclerotic stenosis can be puzzling; explore the causes of microvascular dysfunction, the means to detect and assess it, its relation to cardiac risk factors and determine clinical scenarios, prognosis and treatment.
Data show the clinical relevance of a condition that Harvey Kemp in 1973 labeled Cardiac Syndrome X (CSX) - angina, myocardial ischemic signs and normal or nearly normal coronaries on angiogram characterised this condition (1). Its pathological mechanisms are epicardial coronary spasm, coronary bridging, coronary slow-flow, Tako-tsubo and coronary microvascular dysfunction (2). Research of cardiac syndrome X has focused on microvascular dysfunction (CMVD), a common pathophysiological dysfunction of this condition together with an altered perception of pain, and especially on its role in the genesis of ischemia. New methods for assessing microcirculation are also being explored. Indeed, microvascular dysfunction causes microvascular angina; it requires the absence of demonstrable obstructive coronary artery disease (CAD) and an altered microvascular response to functional testing. It is primary, if microvascular angina is the unique cause of symptoms, and secondary, if it occurs in the setting of a specific disease (3). Stable microvascular angina (MVA) has recurrent effort angina, while instable or acute MVA presents as acute coronary syndrome, rest pain and in some cases an increase in mild myocardionecrosis markers (4).The 2013 ESC Guidelines on management of stable coronary arteries disease (SCAD) (5) dedicate specific consideration to angina with normal coronary arteries, and identify two clinical patterns as part of stable coronary disease: microvascular angina and vasospastic angina - which may involve in some cases only the microvascular district.
Coronary microcirculation is not usually the object of routine imaging however, being the major determinant of vascular resistance - 80% of total resistance is due to coronary microcirculation - its dysfunction may compromise myocardial perfusion. Indeed, the coronary pre-arterioles and arterioles - i.e. the coronary arteries <500 μm in diameter, physiologically modulate coronary blood flow (CBF) in response to neural, mechanical and metabolic factors (4-7); Possible causes for microvascular dysfunction are:
Various further observed alterations can combine in a variety of ways in individual patients and thus, determine the broad clinical spectrum observed in clinical practice of cardiac blood flow impairment:
On examination, myocardial ischemia is detected by stress ECG or Holter's dynamic monitoring, which show ST-segment depression and chest pain, but not by stress-echography, because the patchy distribution of perfusion abnormalities makes the abnormalities macroscopically undetectable. Ischemic areas may be pointed out by stress scintigraphy, which show stress induced perfusion defects, by magnetic resonance (MR) or positron emission tomography (PET).
Here are areas of study investigating the causes of CMVD:
Several studies investigate the association between cardiovascular risk factors and CMVD in women and their results are discordant:
Nevertheless, mounting data show that CVRF can affect microcirculation and are associated with CMVD, suggesting that they do so in at least a proportion of patients.
In hypertensive subjects especially:
In obese subjects with evidence of neither heart disease nor epicardial artery stenosis,:
The role of inflammation is also demonstrated in CSX patients without traditional cardiovascular risk factors: subjects with elevated CRP levels have significantly reduced CFR, compared to those with low PCR levels (28).
Microcirculation can't be investigated by angiogram, thus, several techniques for functional assessment, of coronary flow reserve,both invasive and non-invasive, have been proposed (29, 30) (Table 1). Microvascular function is evaluated by testing vascular flow responses to vasodilators stimuli using various techniques, and calculating Coronary Flow Reserve (CFR) as the ratio between hyperemic and basal coronary flow. The most widely used substances are adenosine and dipyridamole, to test endothelium-independent vasodilatation, acetylcholine (ACH) and cold pressure to the endothelium-dependent one. A CFR below 2.5 (although a threshold of 2.0 might be more specific and diagnostic) is usually considered as diagnostic of CMVD impairment. However, non-invasive techniques may lack sensitivity and specificity for diagnosis and are unable to differentiate between epicardial and microvascular abnormalities. A complete characterisation of coronary microvascular (CMV) function would also require, especially when vasodilator tests are normal or inconclusive, the assessment of response to vasoconstrictor stimuli (ACH, ergonovine), performable only during an invasive study, with the aim of exclusion of significant vasoconstriction of epicardial vessels and detection of microvessel spasm (4).
Non invasive methods - Transthoracic Doppler echocardiography (TTDE) is the first screening test to identify significant impairment of CMV function. It allows the measurement of CBF velocity in the left anterior descending artery. CFR is defined as the ratio of hyperemic diastolic peak ﬂow velocity during maximal vasodilatation to basal ﬂow velocity. Results are comparable with those obtained by intracoronary Doppler ﬂow wire (ICDW) recording and PET. Myocardial perfusion and CFR are also assessable through cardiac MR, myocardial contrast, echocardiography and PET. These tests each have their own limitations in terms of ability in perfusion quantification, required expertise, availability, costs and potential risks; thus, they are second line methods in patients in whom TTDE is inconclusive or in whom assessment for specific aims is required (29).Invasive methods - during angiography it is possible to assess CMV function through several techniques: thermodilution or gas washout method and intracoronary Doppler (ICD) flow wire allow CBF and CFR quantification; intravascular Doppler ultrasonography (IVDUS) permits the direct visualisation of arterial walls and detection of atherosclerotic plaques that foregoes to the angiography. is the most used is the intracoronary recording of CBF by Doppler and pressure wires.Intracoronary Doppler is the most used techniques as it allows direct measurement of CBF velocity in single epicardial arteries. The product of CBF velocity and the cross-sectional arterial area gives a measure of CBF. Evaluations prior and post vasodilators infusion allows the CFR measurement. Through intracoronary Doppler and pressure sensor incorporated, it is possible to calculate the Index of Microvascular Resistance (IMR), defined as the distal coronary pressure multiplied by the hyperaemic mTT.
Angiogram allows microvascular function evaluation indirectly through some angiographic indexes.Myocardial blush is the myocardial opacification resulting by injection of dye into the coronary. Counting the number of heart cycles required for it to fade out, we achieve the Myocardial Blush Grade (MBG), which depends on the microcirculation resistance to the dye passage and the efficiency of venous drainage (32). Total Myocardial Blush Score is the sum of the MBG of each coronary territory and defines the overall microvascular functionality (33). The Timi Frame Count (TFC) is calculated on the basis of the number of frames required for dye to reach a standardized distal landmark of the considered coronary vessel and a correction factor depending on vessel length; it is related to the velocity wherewith dye fulfill epicardial vessels and index of microvascular district resistance (34). Similarly to the Total Myocardial Blush score, Total Timi Frame Count, is the sum of the three major coronary vessel scores, useful for a comprehensive view of the coronary microcirculation function (35).Table 1. Comparative summary of methods to investigate coronary microvascular function (29).
CBF, coronary blood flow; CMV, coronary microvascular; CMR, cardiac magnetic resonance; ICD, intracoronary Doppler; MCE, myocardial contrast echocardiography; TTDE, transthoracic Doppler echocardiography. - Poor for the item; + sufficient for the item; ++ good for the item; +++ very good for the item.
A finding of angina with a normal arteriogram may occur in the setting of a chronic (stable) or an acute (unstable) angina syndrome (4). In both cases, the first step in diagnosis is to differentiate from non-cardiac chest pain.
Chronic microvascular angina - About 50% of coronary angiograms in patients with stable angina show normal or non-obstructed (<50% stenotic) coronary arteries. This condition is much more common in women than in men, and may be due to vasospastic angina (Prinzmetal's variant angina and microvascular spasm) (36), or to chronic microvessels disease, either isolated or associated with systemic (amyloidosis, Fabry) or cardiac disease other than CAD. Chronic MVA is characterised by association with classical atherosclerotic risk factors, history of recurrent angina, often typical and mainly effort induced that require repeated diagnostic evaluation with non-invasive stress tests (frequently with abnormal results) and even coronary angiography. Differential diagnosis on clinical features is impossible, but some clues suggest MVA: chest pain persists for several minutes after effort interruption and/or shows poor or slow response to nitroglycerin, stress tests induce angina and ST-segment depression but no left ventricular contractile abnormalities at echocardiography, and sublingual nitrate administration prior exercise test execution determines earlier appearance of ECG abnormalities and/or angina. CMVD should be identified through non-invasive tests capable of highlighting absence of contractile abnormalities and exploring the vasodilator activity of coronary microcirculation: TTDE is the first choice technique. When vasodilators tests are normal or inconclusive, the response to vasoconstrictor stimuli should be assessed invasively; coronary angiogram is also necessary to obtain definite diagnosis of primary MVA, since rules out epicardial abnormalities (4). In Fig.1 ESC recommendations for diagnostic evaluation in angina without obstructive CAD are showed.
It should be remembered that a normal or near-normal angiogram does not necessarily rule out the presence of a large ‘hidden’ atherosclerotic burden, or future development of atherosclerosis stenosis. The IVDUS shows that non obstructed coronary arteries have often diffuse CAD with wall positive remodeling (wall thickens outwards without notching the lumen) (42). In patients with diffuse epicardial disease but no relevant proximal stenosis, the Fractional Flow Reserve (FFR, the ratio between distal coronary pressure and aortic pressure during maximal coronary vasodilation; a value ≤0.8 is indicative of downstream perfusion as limited to become inadequate if oxygen demand increases) (31) may be helpful to identify hemodynamic relevant coronary plaques, although no yet obstructive, and to avoid ascribing patient’s symptoms to microvascular disease (37).
Instable microvascular angina - CMVD may presents also as acute coronary syndrome, with de novo or worsening angina pattern or recurrent rest induced or mild effort induced chest pain. Microvascular angina is responsible of 10% of male and 25% of female of acute coronary syndrome and ‘normal’ coronary angiograms. Differential diagnosis should rule out transient thrombosis, coronary embolism, epicardial spasm, and two specific clinical syndromes: microvascular spasm angina and stress-related cardiomyopathy. The diagnosis requires the evidence of a cardiac ischemic origin of symptoms, suggested by new abnormalities on standard ECG which gradually reverse to normal, and sometimes by mild elevation of serum myocardionecrosis markers, the evidence of CMVD, obtained by the assessment of response to vasoconstrictor and vasodilator stimuli, and the exclusion of epicardial coronary spasm or transient coronary thrombosis. The former is achievable with provocative tests; the latter is only assumable by exclusion (4). Coronary embolism, due to atrial fibrillation or flutter, is rare, but as atrial fibrillation is often clinically unrecognized, the frequency of this mechanism of may be underestimated (38). Microvascular spasm angina (vasospastic angina) is diagnosed by intracoronary acetylcholine infusion, that reproduces angina and ST-segment changes but not epicardial spasm, and Takotsubo disease is recognizable for peculiar clinical and morphologic features. However, both these conditions are related to microvascular dysfunction, the former to diffuse coronary microvascular spasm (11), the latter to acute severe coronary microvascular constriction (39).
Specific timelines regarding prognosis are controversial because of the heterogeneity of population included in different studies, in which presence and extent of CAD, left ventricular function impairment, and/or pathogenetic mechanism underlying angina are not taken into consideration. Small studies carried out in patients with stable microvascular angina found rates of major cardiovascular events comparable to those of the general population (4), with the exception of re-admissions for angina (40); A more recent study records higher rate of adverse cardiovascular events in patients with SCAD and normal coronary arteries or diffuse non-obstructive stenosis compared with that in subject without CAD (41). Approximately 20% to 30% of patients with stable MVA experience progressive worsening of symptoms (4), leading to meaningful impairment of life quality. Among subjects with angina and a normal angiogram, subgroups at high risk seem to be those with documented ischemia, mild CAD, and microvascular dysfunction. The risk of epicardial events appears during the long term (>2 years) follow-up (36). Regarding unstable MVA, prognosis is not well known. A prospective study comparing patients with unstable and stable MVA found no major cardiac events in either group at a mean follow-up of 36 months, a similar proportion of rehospitalisation for chest pain, although 71% of patients with stable MVA complain of persistence or recurrence of angina, versus only 32% of patients with unstable MVA (4).
All patients with microvascular angina should achieve optimal coronary risk factor control. Symptomatic treatment is empirical because of the limited knowledge of its causes and the lack of conclusive therapeutic trials. Moreover, the susceptibility of symptoms to medical treatment is variable and it is necessary to experiment with different drug combinations before achieving control.
Traditional anti-ischemic drugs are the first step in medical treatment (4).
Coronary microvascular dysfunction is a pathological mechanism of cardiac syndrome X, which causes various types of cardiac flow impairments which account for the various presentations. Microvascular dysfunction is a heterogeneous group of disorders related to several mechanisms – still largely speculative, operating alone or in combination, in the different cases. Of note, it's unknown why exposition to CVRF determines in some subject epicardial coronary arteries disease (CAD) and in others microvascular disease, or whether CMVD is the early step of CAD that progresses invariably or whether it is a distinct pathology. How to distinguish CMVD from SCAD to avoid unnecessary angiograms has also yet to be determined. Prospective studies are needed to characterise very specifically the different patients and identify subgroups at increased risk and their own prognosis. Fig 1. Investigations recommended by ESC in patients with suspected coronary microcircular disease (A) and vasospastic angina (B) (5).
1. Left ventricular function in patients with the anginal syndrome and normal coronary arteriograms. Kemp HG Jr. Am J Cardiol. 1973; 32: 375-6. 2. "Microvascular angina" as a cause of chest pain with angiographically normal coronary arteries. Cannon RO 3rd, Epstein SE. Am J Cardiol. 1988; 61:1338-43. Review. 3. Coronary microvascular dysfunction. Camici PG, Crea F. N Engl J Med. 2007; 356: 830-40. Review.4. Primary coronary microvascular dysfunction: clinical presentation, pathophysiology, and management. Lanza GA, Crea F. Circulation. 2010; 121:2317-25.5. 2013 ESC guidelines on the management of stable coronary artery disease: the Task Force on the management of stable coronary artery disease of the European Society of Cardiology. Task Force Members, Montalescot G, Sechtem U, Achenbach S, Andreotti F, Arden C, Budaj A, et al. Eur Heart J. 2013;34(38):2949-3003. 6. Coronary arteriolar myogenic response is independent of endothelium. Kuo L, Chilian WM, Davis MJ. Circ Res. 1990; 66:860-6.7. Therapeutic advances in myocardial microvascular resistance: unravelling the enigma. Patel B, Fisher M. Pharmacol Ther. 2010; 127:131-47.8. Impaired coronary vasodilator responsiveness as a cause of lactate production during pacing-induced ischemia in patients with angina pectoris and normal coronary arteries. Greenberg MA, Grose RM, Neuburger N, Silverman R, Strain JE, Cohen MV. J Am Coll Cardiol. 1987; 9:743-51.9. Both endothelium-dependent and endothelium-independent function is impaired in patients with angina pectoris and normal coronary angiograms. Chauhan A, Mullins PA, Taylor G, Petch MC, Schofield PM. Eur Heart J. 1997; 18: 60-8.10. Evidence of endothelial dysfunction in coronary resistance vessels in patients with angina pectoris and normal coronary angiograms. Motz W, Vogt M, Rabenau O, Scheler S, Lückhoff A, Strauer BE. Am J Cardiol. 1991; 68: 996-1003.11. Angina pectoris caused by coronary microvascular spasm. Mohri M, Koyanagi M, Egashira K, Tagawa H, Ichiki T, Shimokawa H, Takeshita A. Lancet 1998; 351:1165-9.12. Abnormal coronary vasomotion during exercise in patients with normal coronary arteries and reduced coronary flow reserve. Bortone AS, Hess OM, Eberli FR, Nonogi H, Marolf AP, Grimm J et al. Circulation. 1989; 79: 516-27.13. Coronary slow-flow causing transient myocardial hypoperfusion in patients with cardiac syndrome X: long-term clinical and functional prognosis. Fragasso G, Chierchia SL, Arioli F, Carandente O, Gerosa S, Carlino M, et al. Int J Cardiol. 2009;137:137-44.14. Relation between cardiovascular risk factors and coronary microvascular dysfunction in cardiac syndrome X. Sestito A, Lanza GA, Di Monaco A, Lamendola P, Careri G, Tarzia P, et al. J Cardiovasc Med (Hagerstown). 2011;12:322-7.15. Abnormal cardiac adrenergic nerves function in patients with syndrome X detected by [123I] metaiodobenzylguanidine myocardial scintigraphy. Lanza GA, Giordano A, Pristipino C, Calcagni ML, Meduri G, Trani C, et al. Circulation. 1997;96:821-6.16. Evidence of parasympathetic impairment in some patients with cardiac syndrome X. Gulli G, Cemin R, Pancera P, Menegatti G, Vassanelli C, Cevese A. Cardiovasc Res. 2001;52:208-16. 17. Role of abnormal pain sensitivity and behavioral factors in determining chest pain in syndrome X. Pasceri V, Lanza GA, Buffon A, Montenero AS, Crea F, Maseri A. J Am Coll Cardiol. 1998;31:62-6. 18. Comparison of cardiovascular risk factors and biochemical profile in patients with cardiac syndrome X and obstructive coronary artery disease: A propensity score-matched study. Vasheghani-Farahani A, Nouri N, Seifirad S, Sheikh Fathollahi M, Hakki E, Alidoosti M, et al. ARYA Atheroscler. 2013; 9: 269-73. 19. Coronary microvascular reactivity is only partially predicted by atherosclerosis risk factors or coronary artery disease in women evaluated for suspected ischemia: results from the NHLBI Women's Ischemia Syndrome Evaluation (WISE). Wessel TR, Arant CB, McGorray SP, Sharaf BL, Reis SE, Kerensky RA, et al.; NHLBI Women's Ischemia Syndrome Evaluation (WISE). Clin Cardiol. 2007; 30:69-74.20. Relation between epicardial fat thickness and coronary flow reserve in women with chest pain and angiographically normal coronary arteries. Sade LE, Eroglu S, Bozbaş H, Ozbiçer S, Hayran M, Haberal A, et al. Atherosclerosis. 2009; 204: 580-5. 21. Determinants of reduction of coronary flow reserve in patients with type 2 diabetes mellitus or arterial hypertension without angiographically determined epicardial coronary stenosis. Galderisi M, Capaldo B, Sidiropulos M, D'Errico A, Ferrara L, Turco A, et al. Am J Hypertens. 2007;2012:1283-90. 22. Effects of Essential Hypertension on coronary Microcirculation: Focus on a Population of Hypertensives Affected by Microvascular Angina. Quagliana A, Evol S, Triolo OF, Piraino D, Pace G, Tona GR et al. J Clinic Experiment Cardiol 2012;3:176.23. Effects of type 2 diabetes mellitus on coronary microvascular function and myocardial perfusion in patients without obstructive coronary artery disease. Marciano C, Galderisi M, Gargiulo P, Acampa W, D'Amore C, Esposito R, et al. Eur J Nucl Med Mol Imaging. 2012;39:1199-206.24. Comparison of coronary artery flow impairment in diabetic and hypertensive patients with stable microvascular angina. V. Sucato, S. Evola, A. Quagliana, G. Novo, G. Andolina, P. Assennato, et al. Eur Rev Med Pharmacol Sci 201425. Mechanism of angina pectoris in patients with systemic hypertension and normal epicardial coronary arteries by arteriogram. Scheler S, Motz W, Strauer BE. Am J Cardiol. 1994; 73: 478-82. 26. Systemic inflammation is related to coronary microvascular dysfunction in obese patients without obstructive coronary disease. Tona F, Serra R, Di Ascenzo L, Osto E, Scarda A, Fabris R, et al. Nutr Metab Cardiovasc Dis. 2014;24:447-53.27. Obesity, inflammation and brachial artery flow-mediated dilatation: therapeutic targets in patients with microvascular angina (cardiac syndrome X). Ong P, Sivanathan R, Borgulya G, Bizrah M, Iqbal Y, Andoh J, et al. Cardiovasc Drugs Ther. 2012; 26:239-44. 28. Inflammation and microvascular dysfunction in cardiac syndrome X patients without conventional risk factors for coronary artery disease. Recio-Mayoral A, Rimoldi OE, Camici PG, Kaski JC. JACC Cardiovasc Imaging. 2013; 6:660-7. 29. Methods to investigate coronary microvascular function in clinical practice. Lanza GA, Camici PG, Galiuto L, Niccoli G, Pizzi C, Di Monaco A, et al. On behalf of Gruppo di Studio di Fisiopatologia Coronarica e Microcircolazione, Società Italiana di cardiologia. J Cardiovasc Med (Hagerstown). 2013;14:1-18. 30. A review of methods for assessment of coronary microvascular disease in both clinical and experimental settings. Pries AR, Habazettl H, Ambrosio G, Hansen PR, Kaski JC, Schächinger V, et al. Cardiovasc Res. 2008; 80:165-74. Review. 31. Guidelines on myocardial revascularization. Task Force on Myocardial Revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS); European Association for Percutaneous Cardiovascular Interventions (EAPCI), Wijns W, Kolh P, Danchin N, Di Mario C, Falk V, Folliguet T, et al. Eur Heart J. 2010; 31:2501-55.32. Angiographic evaluation of myocardial perfusion in patients with syndrome X. Atmaca Y, Ozdemir AO, Ozdol C, Oguz D, Gulec S, Kumbasar D, et al. Am J Cardiol. 2005; 96:803-5.33. Total blush score: a new index for the assessment of microvascular perfusion in idiopathic dilated cardiomyopathy. Atmaca Y, Duzen V, Ozdol C, Altin T, Tulunay C, Ertas F, et al. Coron Artery Dis. 2008; 19:181-5.34. TIMI frame count: a quantitative method of assessing coronary artery flow. Gibson CM, Cannon CP, Daley WL, Dodge JT Jr, Alexander B Jr, Marble SJ, et al. Circulation. 1996; 93:879-88. 35. Stable microvascular angina: instrumental evaluation of coronary microvascular dysfunction with coronary angiography and myocardial scintigraphy. Sucato V, Evola S, Novo G, Bronte E, Novo S. Int J Cardiol. 2014;171: e127-8.36. Coronary microvascular dysfunction in the clinical setting: from mystery to reality. Herrmann J, Kaski JC, Lerman A. Eur Heart J. 2012; 33:2771-2782b.37. Importance of diffuse atherosclerosis in the functional evaluation of coronary stenosis in the proximal-mid segment of a coronary artery by myocardial fractional flow reserve measurements. Rodés-Cabau J, Gutiérrez M, Courtis J, Larose E, Déry JP, Côté M, et al. Am J Cardiol. 2011; 108:483-90.38. ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: The Task Force for the management of acute coronary syndromes (ACS) in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC). Hamm CW, Bassand JP, Agewall S, Bax J, Boersma E, Bueno H, et al.; ESC Committee for Practice Guidelines. Eur Heart J. 2011; 32:2999-3054.39. Reversible coronary microvascular dysfunction: a common pathogenetic mechanism in Apical Ballooning or Tako-Tsubo Syndrome. Galiuto L, De Caterina AR, Porfidia A, Paraggio L, Barchetta S, Locorotondo G, et al. Eur Heart J. 2010; 31: 1319-27. 40. Angina pectoris and normal coronary arteriograms: clinical presentation and hemodynamic characteristics. Kaski JC, Elliott PM. Am J Cardiol. 1995; 76:35D-42D. Review.41. Stable angina pectoris with no obstructive coronary artery disease is associated with increased risks of major adverse cardiovascular events. Jespersen L, Hvelplund A, Abildstrøm SZ, Pedersen F, Galatius S, Madsen JK, et al. Eur Heart J. 2012;33(6):734-44.42. An intravascular ultrasound analysis in women experiencing chest pain in the absence of obstructive coronary artery disease: a substudy from the National Heart, Lung and Blood Institute-Sponsored Women's Ischemia Syndrome Evaluation (WISE). Khuddus MA, Pepine CJ, Handberg EM, Bairey Merz CN, Sopko G, Bavry AA, et al. J Interv Cardiol. 2010; 23:511-9.
Salvatore Novo, Annalisa Graceffa, Vincenzo Sucato, Angelo Quagliana, Salvatore Evola, Enrico Bronte, Giuseppina NovoChair of Cardiovascular Disease, University of Palermo, Division of Cardiology, Center for the early Diagnosis of preclinical and multifocal Atherosclerosis, Regional reference Center for the Diagnosis and Care of Heart Failure, University Hospital “P. Giaccone” of Palermo, Italy
Authors' disclosures: None declared.Other reportsE-journal article (2014) stable primary angina on stable-primary-anginaMechanisms and clinical implications of coronary microvascular remodeling (2010)
Our mission: To reduce the burden of cardiovascular disease
© 2017 European Society of Cardiology. All rights reserved