Vascular damage and dysfunction in hypertensive emergencies

(1)

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

Amraoui, F.

Publication date

2017

Document Version

Final published version

License

Other

Link to publication

Citation for published version (APA):

Amraoui, F. (2017). Vascular damage and dysfunction in hypertensive emergencies.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

(2)

AMA

GE AND D

YSFUNC

TION IN HYPER

TENSIVE EMER

GENCIE

S

Fouad Amr

VASCULAR DAMAGE AND DYSFUNCTION

IN HYPERTENSIVE EMERGENCIES

UITNODIGING

Voor het bijwonen van de openbare verdediging van

het proefschrift

VASCULAR DAMAGE

AND DYSFUNCTION

IN HYPERTENSIVE

EMERGENCIES

Op donderdag 9 februari 2017 14:00 uur in de Agnietenkapel Oudezijds Voorburgwal 231 Amsterdam Receptie na afloop van de promotie Paranimfen: Sietse Boogmans Alex Wagemakers FOUAD AMRAOUI CLARISSENHOF 121, 1115 CD DUIVENDRECHT F.AMRAOUI@AMC.UVA.NL

(3)

VASCULAR DAMAGE AND DYSFUNCTION

IN HYPERTENSIVE EMERGENCIES

(4)

Author: Fouad Amraoui Cover: Mhamed Yahiaoui Layout: Gildeprint Printing: Gildeprint ISBN: 978-94-6233-530-1

Financial support by the Dutch Heart Foundation for the publication of this thesis is gratefully acknowdledged.

Further financial support for printing this thesis was kindly provided by: the Dutch Kidney Foundation, Astellas, Pfizer, Sanofi, Chipsoft, Stichting tot Steun Promovendi Vasculaire Geneeskunde and the Academic Medical Center

(5)

VASCULAR DAMAGE AND DYSFUNCTION

IN HYPERTENSIVE EMERGENCIES

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor

aan de Universiteit van Amsterdam op gezag van de Rector Magnificus

prof. dr. ir. K.I.J. Maex

ten overstaan van een door het College voor Promoties ingestelde commissie, in het openbaar te verdedigen in de Agnietenkapel

op donderdag 9 februari 2017, te 14:00 uur

door Fouad Amraoui geboren te Haarlem

(6)

Promotores: Prof. dr. S. Middeldorp Universiteit van Amsterdam Prof dr. J.M. van der Post Universiteit van Amsterdam

Co-promotores: Dr. B.J.H. van den Born Universiteit van Amsterdam Dr. L. Vogt Universiteit van Amsterdam

Overige leden: Prof. dr. E.T. van Bavel Universiteit van Amsterdam Prof. dr. F.J. Bemelman Universiteit van Amsterdam Prof. dr. H.R. Büller Universiteit van Amsterdam Dr. A.T. Lely Universiteit Utrecht Prof. dr. M. Nieuwdorp Universiteit van Amsterdam Prof. dr. Y.M. Smulders Vrije Universiteit Amsterdam

(7)

(8)

(9)

Chapter 1 Introduction and outline of this thesis 9

Part I Cardiovascular and renal complications of malignant hypertension

Chapter 2 Mortality and Cardiovascular Risk in Patients with a History of 25 Malignant Hypertension: a Case-Control Study

Chapter 3 Long-term renal outcome in patients with malignant hypertension: 37 a retrospective cohort study

Chapter 4 Value of retinal examination in hypertensive encephalopathy 51

Part II Role of anti-angiogenic factors in the pathophysiology of pre-eclampsia

Chapter 5 SFlt-1 elevates blood pressure by augmenting endothelin-1-mediated 65 vasoconstriction in mice

Chapter 6 Potential role for aspirin in the prevention of pre-eclampsia-like 83 syndrome in bevacuzimab-treated patients with colorectal carcinoma Chapter 7 Syndecan-1 and glycosaminoglycans in pre-eclamptic and 95

normotensive pregnancies

Chapter 8 Placental expression of heparan sulfate 3-O-sulfotransferase-3A1 113 in normotensive and pre-eclamptic pregnancies

Chapter 9 Microvascular glycocalyx dimension estimated by automated SDF 129 imaging is not related to cardiovascular disease

Chapter 10 Sphingolipid metabolism in pre-eclamptic and normotensive 143 pregnancies

Chapter 11 Summary and perspectives 159

Appendices Nederlandse samenvatting 173

Authors and affiliations 179

Dankwoord 183

Portfolio 186

(10)

(11)

1

Introduction and outline of this thesis

Fouad Amraoui, Liffert Vogt, Saskia Middeldorp and Bert-Jan H van den Born

(12)

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

(13)

1 INTRODUCTION

High blood pressure (BP) is one of the major risk factors for cardiovascular disease (stroke, myocardial infarction, heart failure), renal disease and mortality worldwide.1_{In the}

majority of patients, hypertension is caused by a combination of genetic, behavioral and environmental factors (primary or essential hypertension), while only a small proportion of hypertensive patients have a single identifiable cause of high BP (secondary hypertension). Most patients with hypertension are asymptomatic and BP lowering treatment in these patients is aimed at reducing the long-term risk of cardiovascular and renal disease. However, in some patients with hypertension, BP may rise to critical values, causing acute, sometimes irreversible, damage to target-organs such as brain, heart, and kidneys. Rapid diagnosis and treatment of this severe form of hypertension or “hypertensive crisis” is required to prevent or limit organ damage in the acute phase.2

Hypertensive crisis is frequently encountered at the emergency department and is defined as severe hypertension, usually >220 mmHg systolic and >120 mmHg diastolic BP with manifest (emergency) or imminent (urgency) target-organ damage.3_{There is no cut-off}

value for BP in the definition of hypertensive crisis, because development of acute organ damage depends more on the magnitude and velocity of BP elevation from baseline rather than on actual BP values at presentation. This means that acute organ damage is more likely to occur at lower absolute BP values in young and previously normotensive patients compared to older patients with chronic hypertension.

Hypertensive crisis is a heterogeneous disorder with a large variation in clinical presentation, depending on the type and extent of organ damage. Examples of acute organ damage include aortic dissection, ischemic and hemorrhagic stroke, acute heart failure with pulmonary edema, ischemic coronary disease, renal failure, retinopathy and thrombotic microangiopathy. Two relatively common forms of hypertensive crises comprise hypertensive crisis with retinopathy and pre-eclampsia. These two hypertensive emergencies are the focus of this thesis and will be introduced in more detail below.

Hypertensive crisis with retinopathy and pre-eclampsia are both considered to be hypertensive emergencies that require rapid treatment to prevent or limit acute organ damage. Hypertensive crisis with retinopathy is a renin-mediated form of hypertension that predominantly affects patients with untreated essential hypertension. Because of long-existing untreated hypertension, BP may reach extremely high values at presentation. Pre-eclampsia affects pregnant women in the second half of pregnancy. Anti-angiogenic factors produced by the placenta are responsible for endothelial dysfunction and hypertension in these women. Because these women are often previously normotensive, symptoms and hypertensive organ damage may already arise at relatively low BP values. Anti-angiogenic medication for the treatment of malignancies may also cause a pre-eclampsia-like

(14)

syndrome with hypertension and proteinuria. Despite these pathophysiological differences, hypertensive crisis with retinopathy and pre-eclampsia may be complicated by similar life-threatening conditions, including hypertensive encephalopathy (or eclampsia), renal failure and thrombotic microangiopathy (Figure 1).

Figure 1: Differences and similarities among hypertensive crisis with retinopathy (orange), pre-eclampsia (green) and anti-angiogenic treatment-induced hypertension (grey).

Hypertensive crisis with retinopathy (malignant hypertension)

Traditionally, hypertensive crisis with retinopathy is referred to as malignant hypertension. In an observational study from 1939, when treatment of hypertension was not yet possible, Keith, Wagener and Barker showed that patients with severe hypertension and grade IV retinopathy had a remarkably poor prognosis, as more than 90% of patients died within 1 year.4_{Patients with severe hypertension and grade III retinopathy had a slightly better}

prognosis with 65% of patients deceased after 1 year. The term malignant hypertension was introduced for patients with grade IV retinopathy and accelerated hypertension for patients with grade III retinopathy. The difference in survival disappeared with the advent of antihypertensive medication leading to incorporation of both grade III and IV retinopathy in the definition of malignant hypertension since the 1980s.5,6

The annual incidence of malignant hypertension is currently 1-3 per 100.000 individuals and is more common in multi-ethnic populations.7,8_{Secondary causes of high blood pressure are}

more frequently observed in malignant hypertension compared to chronic non-malignant hypertension, but the majority of patients have unrecognized or uncontrolled primary hypertension. Limited access to healthcare and non-adherence to medication contribute to the development of malignant hypertension in these patients. These socio-economic

(15)

1

factors, together with the much higher prevalence of hypertension among patients of sub-Saharan African origin, result in a 4 times higher prevalence of malignant hypertension among sub-Saharan African immigrants.8

Pathophysiology

Clinical studies and animal experiments have shown that activation of the renin-angiotensin-aldosteron system (RAAS) plays a central role in the pathophysiology of malignant hypertension.9_{Virtually any cause of BP elevation, for example cessation of antihypertensive}

medication, can trigger RAAS activation if the BP elevation is high enough to induce endothelial damage. In the kidneys, damage to small arteries and arterioles leads to renal ischemia and consequently to renin secretion by juxtaglomerular cells and RAAS activation. At the same time, pressure natriuresis is initiated to limit BP elevation. Eventually, this leads to hypovolemia, thereby contributing to RAAS activation. The subsequently produced angiotensin II is a powerfull vasocontrictor that further elevates BP and gives rise to a vicious circle of BP elevation, endothelial damage, renal ischemia and RAAS activation (Figure 1).

Figure 2: Simplified representation of the vicious circle underlying hypertensive crisis with retinopathy

In addition to elevation of BP, angiotensin II has direct cytotoxic effects on the vessel wall via activation of pro-inflammatory cytokines such as IL-6 and NF-Kβ. Moreover, endothelial damage activates the coagulation cascade, leading to formation of thrombi that further obliterate the microcirculation and induce mechanical destruction of erythrocytes.

(16)

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39 Histopathological changes

Autopsy studies have indentified fibrinoid necrosis and myointimal proliferation of small arteries and arterioles as histopathological hallmark of malignant hypertension.10_Fibrinoid

necrosis reflects seepage of fibrin through necrotic vessel walls, whereas myointimal proliferation leads to narrowing of the vessel lumen. This vascular damage can be observed in virtually all organs. In addition, ischemic glomerular changes are observed in the kidneys (Figure 2), including collapsed glomerular tufts, widening of Bowman’s capsule with collagen depositions, hyperplasia of juxtaglomerular cells and tubular necrosis.

Figure 3: Histopathological changes in hypertensive crisis with acute renal insufficiency

Myointimal hyperplasia and fibrinoïd necrosis in an almost completely occluded arteriole (arraow). Collapsed glomerular tuft with widening of Bowman’s capsule (asterisk). Figure adapted from: B.J.H van den Born and G. van Montfrans, Malignant hypertension. In: Jörres et al. Management of Acute Kidney Problems pp 305-316, Spinger-Verlag Berin Heidelberg 2010

Clinical presentation and complications

Depending on the extent of organ damage, clinical presentation is highly variable. Patients may be asymptomatic, but can also present with seizures or cortical blindness. Most patients complain of headache and visual disturbances, followed by gastro-intestinal complaints such as nausea, vomiting and abdominal pain (Table 1).

(17)

1

Table 1: Presenting symptoms of patients with hypertensive crisis and retinopathy in Amsterdam, The Netherlands

Percentage

Headache 62

Visual disturbances 56

Gastro-intestinal complaints, including weight loss 36

Congestive heart failure 21

Neurological symptoms 10

Left ventricular hypertrophy 83

Severe renal insufficiency (creatinine >300 µmol/l) 31

Renal insufficiency (creatinine 115-300 µmol/l) 36

Thrombotic microangiopathy 32

Besides bilateral ischemic retinal lesions, patients with malignant hypertension frequently present with acute renal insufficiency, thrombotic microangiopathy and hypertensive encephalopathy. One third of patients with malignant hypertension have severe renal insufficiency at presentation and more than 60% of patients have proteinuria (>300mg/24 hrs). Although kidney replacement therapy may be required in the acute phase, recovery of renal function has been reported until two years after the hypertensive crisis with BP lowering treatment.11,12_{Nonetheless, a relatively large proportion of patients will progress}

to end-stage renal disease. Acute renal insufficiency and proteinuria is caused by ischemic glomerular damage resulting from fibrinoid necrosis and thrombotic occlusions of small arteries and arterioles. This thrombotic microagiopathy is clinically reflected by consumption of platelets and fragmentation of erythrocytes, which can be observed in approximately 30% of patients with malignant hypertension. As a result of this deleterious effect of thrombotic microangiopathy on the kidneys, presence of hemolysis and low platelet count is an important predictor of renal failure and possible recovery.12

Neurological symptoms consistent with hypertensive encephalopathy occur in approximately 10-15% of patients with malignant hypertension. Hypertensive encephalopathy is a clinical diagnosis based on 1) severe hypertension in combination with neurological symptoms such as lowered consciousness (delirium, agitation, lethargy, coma) or seizures 2) absence of an alternative explanation for the neurologic symptoms and 3) resolution of the symptoms after adequate BP control has been achieved. Histopathological changes in hypertensive encephalopathy include edema, microscopic haemorrhages and infartions.13_Cerebral

edema can be visualised on T2-weighted MRI showing a characteristic pattern of posterior reversible encephalopathy syndrome (PRES), Figure 3. PRES may also be observed in several other life-threatening diseases including thrombotic thrombocytopenic purpura, eclampsia, sepsis and can also be triggered by immunosuppressive drugs such as tacrolimus and cyclosporine. Diagnosing hypertensive encephalopathy may therefore be difficult. The

(18)

presence of severe hypertension and grade III of IV hypertensive retinopathy may help distinguish hypertensive encephalopathy from other causes of PRES.

Figure 4: Cerebral edema in hypertensive encephalopathy, typically located in the parieto-occipital region

Cerebral edema in patients with hypertensive encephalopathy can be visualized using T2 weighted MRI. This characteristic radiological image is also referred to as posterior reversible encephalopathy syndrome (PRES) and may also be observed in other diseases including thrombotic thrombocytopenic purpura and eclampsia.

Treatment

Diagnostic procedures at presentation are aimed at identifying the presence and extent of organ damage and may provide information on the underlying cause of the hypertensive crisis. The primary goal of treatment is to limit or prevent further hypertensive organ damage rather than achieving normal BP values. A rapid but controlled lowering of BP with only 25% of the mean arterial pressure at presentation is required in the first 24 hours. Patients with malignant hypertension have been shown to have an impaired cerebral autoregulation, posing them at risk for cerebral hypoperfusion with too large drops in BP.3

(19)

1

Prognosis

The survival of patients with malignant hypertension has drastically improved with the availability of effective antihypertensive medication to 90% after 5 years.14,8_{Before the}

availability of adequate treatment, mortality was mainly caused by heart failure, renal failure and stroke. However, a current mortality rate of 10% after 5 years is still relatively high, considering the relatively young age of patients included in the studies. Whether patients with malignant hypertension still remain at increased risk of dying is unknown, as survival of these patients has not been compared with adequately matched controls without malignant hypertension. Although traditional cardiovascular risk factors such as excess smoking, dyslipidemia and uncontrolled hypertension have been associated with malignant hypertension,15,16_{it is presently unclear whether these traditional risk factors}

contribute to excess mortality.

In Part I of this thesis we aimed to quantify the excess mortality risk in patients with a history of malignant hypertension and to investigate whether traditional cardiovascular risk factors contribute to excess mortality in these patients by comparing them with age, sex and ethnicity matched controls from the same area of residence. Our second aim was to assess the long term renal outcome and its predictors in patients with malignant hypertension. Although renal function may recover after the acute phase of malignant hypertension, little data exist on the long term renal prognosis. Thirdly, presence of grade III or IV hypertensive retinopathy has been shown to coincide with disturbed cerebral autoregulation,17_{and is considered to distinguish hypertensive encephalopathy from other}

causes of encephalopathy. However, several case-reports suggested that grade III or IV hypertensive retinopathy may be absent in patients with hypertensive encephalopathy. We aimed to assess the frequency of grade III and IV hypertensive retinopathy in a relatively large series of patients with hypertensive encephalopathy.

Pre-eclampsia

Hypertension is a frequent complication of pregnancy, with 10% of all pregnant women having either chronic hypertension, gestational hypertension or pre-eclampsia, which may or may not be superimposed on chronic hypertension. Pre-eclampsia is characterized by hypertension and proteinuria in the second half of pregnancy. The definition of pre-eclampsia has recently been revised by the International Society for the Study of Hypertension in Pregnancy (ISSHP).18_{Pre-eclampsia is currently defined as new-onset hypertension after}

20 weeks gestation in combination with maternal organ damage, including proteinuria (>300mg/24hrs), renal insufficiency, neurological or haematological complications, uteroplacental dysfunction or fetal growth restriction. Pre-eclampsia is one of the leading causes of maternal and neonatal morbidity and mortality worldwide, especially in developing

(20)

countries.19_{Severe pre-eclampsia (BP >160/110mmHg) is a hypertensive emergency that}

requires immediate treatment to prevent potentially life-threatening complications such as eclampsia, stroke, renal failure and pulmonary edema.18

The pathophysiology of pre-eclampsia is likely multifactorial and still incompletely understood. The current hypothesis is that different and yet undetermined mechanisms may cause abnormal placentation, with incomplete remodelling of uterine spiral arteries by invading trophoblast cells. This leads to inadequate placental perfusion and hypoxia. In response to hypoxia, the placenta paradoxically produces anti-angiogenic factors, including the soluble Vascular Endothelial Growth Factor (VEGF) receptor FMS-like tyrosine kinase 1 (sFLT-1). Scavenging of VEGF from the maternal circulation by sFlt-1 leads to endothelial dysfunction and eventually maternal signs of pre-eclampsia.20

The mechanism by which scavenging of VEGF leads to BP elevation has been the subject of intensive investigation in recent years. Endothelin-1, a powerful vasoconstrictor, has been identified as a key mediator of hypertension induced by VEGF inhibition.21,22_{In addition,}

blockade of VEGF signaling has been shown to induce a shift in sphingolipid metabolism towards increased production of ceramide, a bioactive lipid with anti-angiogenic properties.23

Ceramide content is elevated in pre-eclampsia,24_{and has been linked to both hypertension}

and proteinuria in clinical and pre-clinical studies.25,26_{Whether ceramide is related to VEGF}

inhibition in women with pre-eclampsia and contributes to BP elevation and proteinuria, has not been studied previously.

VEGF signaling is regulated by the endothelial glycocalyx, which is composed of a complex network of membrane-bound proteoglycans and attached glycosaminoglycans.27,28,29,30

The glycocalyx lines the wall of both the micro-and macrovasculature and is involved in regulation of permeability, leucocyte adhesion, coagulation and vascular tone.31_{Changes in}

endothelial glycocalyx composition and function might also contribute to the pathogenesis of pre-eclampsia via disruption of VEGF signaling. Although several changes in glycocalyx composition have indeed been observed in placental tissue and plasma of pre-eclamptic women, the functional consequences of these abnormalities are presently unknown. A reason for the lack of studies addressing functional implications of altered glycocalyx composition in pre-eclampsia, might be the technical difficulty of glycocalyx volume estimation, which is invasive and time-consuming.32_{Recent technological advances have}

led to the development of a new method, allowing non-invasive real-time estimation of glycocalyx volume in the microcirculation of large study populations. This method has, however, not yet been validated in clinical studies.

The principal aim of Part II of this thesis was to investigate the mechanisms by which VEGF inhibition leads to BP elevation. To this end we aimed to assess whether alterations in glycocalyx composition contribute to sFLT-1-induced BP elevation and whether sFLT-1

(21)

1

induces changes in sphingolipid metabolism that subsequently contribute to hypertension and proteinuria in women with pre-eclampsia. A secondary aim was to investigate whether a recently developed technique for non-invasive real-time glycocalyx volume estimation could be used for assessment of the microcirculation in pregnant women with and without pre-eclampsia.

Outline of this thesis

Part I describes cardiovascular and renal complications of malignant hypertension. In Chapter 2, we compared cardiovascular risk factors, actual cardiovascular disease and

mortality of patients with a history of malignant hypertension with age, sex and ethnicity matched normotensive individuals and non-malignant hypertension patients. Chapter 3 addresses the long-term effect of malignant hypertension on renal function and assesses determinants of renal outcome. Patients presenting with malignant hypertension frequently have neurologic symptoms consistent with hypertensive encephalopathy. However, hypertensive encephalopathy may also exist without grade III or IV hypertensive retinopathy that define malignant hypertension. In Chapter 4, we evaluate what proportion of patients with hypertensive encephalopathy present without grade III or IV hypertensive retinopathy.

Part II of this thesis describes translational research into the roles of VEGF inhibition,

glycocalyx composition and sphingolipid metabolism in the pathophysiology of pre-eclampsia. The production of anti-angiogenic factors by the placenta, including sFLT-1, contributes to endothelial dysfunction and BP elevation in pre-eclampsia. In Chapter 5, we explore the mechanisms of sFlt-1 induced BP elevation in mice. Interestingly, patients receiving anti-VEGF treatment for malignancies may develop a pre-eclampsia-like syndrome with hypertension and proteinuria. The results of animal experiments described in chapter 5 point towards an important role of vasocontractile prostanoids in anti-VEGF-induced hypertension. In Chapter 6, we translate this observation, by a post-hoc analysis of the effect of aspirin on BP in patients receiving anti-VEGF treatment with bevacizumab in a randomized clinical trial. In Chapter 7 we compare the composition of the endothelial glycocalyx among pre-eclamptic and normotensive pregnant women by measuring circulating amounts of syndecan-1 and glycosaminoglycans including heparan sulphate, dermatan sulphate and keratan sulphate. We assess whether changes in glycocalyx composition have clinical implications in pre-eclampsia and investigate whether syndecan-1 deficiency contributes to sFlt-1-induced BP elevation in mice. Abnormal placentation, leading to inadequate perfusion and placental hypoxia likely precedes maternal endothelial dysfunction in pre-eclampsia. Chapter 8 focuses on differences in placental glycocalyx composition in placental tissue obtained from normotensive pregnant women and pre-eclamptic women. Potential clinical implications of differences in placental glycocalyx composition are described in this chapter. The glycocalyx volume, as estimated by non-invasive visualisation of the sublingual

(22)

microcirculation, has been shown to be related to endothelial function in selected groups of patients. However, the method for this analysis was cumbersome and difficult to use in large populations. In Chapter 9 we evaluate whether a newly developed device with software for automated estimation of glycocalyx volume can be used for clinical studies in pre-eclampsia. For this purpose we aimed to validate this technique in a large population, including healthy individuals as well as patients with overt cardiovascular disease. The novel system was designed for real-time estimation of glycocalyx volume by including over 3000 visible measurement sites in the sublingual microcirculation. VEGF inhibition has been shown to stimulate the production of ceramide, which is an anti-angiogenic sphingolipid. Ceramide has been shown to be elevated in plasma of women with pre-eclampsia and has previously been linked to hypertension and proteinuria. In Chapter 10 we compare sphingolipid metabolism among pre-eclamptic and normotensive pregnancies and investigate whether excess ceramide relates to hypertension and proteinuria in women with pre-eclampsia.

(23)

1 REFERENCES

1. Lim, S. S. et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet (London, England) 380, 2224–60 (2012). 2. Vaughan, C. J. & Delanty, N. Hypertensive emergencies. Lancet 356, 411–417 (2000). 3. van den Born, B. J. H. et al. Dutch guideline for the management of hypertensive crisis - 2010

revision. Neth. J. Med. 69, 248–255 (2011).

4. Keith, N. M., Wagener, H. P. & Barker, N. W. Some different types of essential hypertension: their course and prognosis. Am. J. Med. Sci. 268, 336–45 (1974).

5. Ahmed, M. E., Walker, J. M., Beevers, D. G. & Beevers, M. Lack of difference between malignant and accelerated hypertension. Br. Med. J. (Clin. Res. Ed). 292, 235–7 (1986). 6. McGregor, E., Isles, C. G., Jay, J. L., Lever, A. F. & Murray, G. D. Retinal changes in malignant

hypertension. BMJ 292, 233–234 (1986).

7. Lip, G. Y., Beevers, M. & Beevers, G. The failure of malignant hypertension to decline: a survey of 24 years’ experience in a multiracial population in England. J. Hypertens. 12, 1297–305 (1994).

8. van den Born, B.-J. H., Koopmans, R. P., Groeneveld, J. O. & van Montfrans, G. A. Ethnic disparities in the incidence, presentation and complications of malignant hypertension. J.

Hypertens. 24, 2299–304 (2006).

9. Kincaid-Smith, P. Malignant hypertension. J. Hypertens. 9, 893–9 (1991).

10. Sanerkin, N. G. Vascular lesions of malignant essential hypertension. J. Pathol. 103, 177–84 (1971).

11. Mourad, G., Mimran, A. & Mion, C. M. Recovery of renal function in patients with accelerated malignant nephrosclerosis on maintenance dialysis with management of blood pressure by captopril. Nephron 41, 166–169 (1985).

12. Van Den Born, B. J. H., Honnebier, U. P. F., Koopmans, R. P. & Van Montfrans, G. A. Microangiopathic hemolysis and renal failure in malignant hypertension. Hypertension 45, 246–251 (2005).

13. Chester, E. M., Agamanolis, D. P., Banker, B. Q. & Victor, M. Hypertensive encephalopathy: a clinicopathologic study of 20 cases. Neurology 28, 928–39 (1978).

14. Lane, D. a, Lip, G. Y. H. & Beevers, D. G. Improving survival of malignant hypertension patients over 40 years. Am. J. Hypertens. 22, 1199–1204 (2009).

15. Isles, C. et al. Excess smoking in malignant-phase hypertension. Br. Med. J. 1, 579–81 (1979). 16. E., E. et al. Dyslipidaemia in patients with malignant-phase hypertension. QJM - Mon. J. Assoc.

Physicians 94, 327–332 (2001).

17. Immink, R. V. et al. Impaired cerebral autoregulation in patients with malignant hypertension.

Circulation 110, 2241–5 (2004).

18. Tranquilli, A. L. et al. The classification, diagnosis and management of the hypertensive disorders of pregnancy: A revised statement from the ISSHP. Pregnancy Hypertension: An

International Journal of Women’s Cardiovascular Health 4, 97–104 (2014).

19. Souza, J. P. et al. Moving beyond essential interventions for reduction of maternal mortality (the WHO Multicountry Survey on Maternal and Newborn Health): A cross-sectional study.

The Lancet 381, 1747–1755 (2013).

20. Powe, C. E., Levine, R. J. & Karumanchi, S. A. Preeclampsia, a disease of the maternal endothelium: The role of antiangiogenic factors and implications for later cardiovascular disease. Circulation 123, 2856–2869 (2011).

(24)

21. George, E. M. & Granger, J. P. Endothelin: key mediator of hypertension in preeclampsia. Am.

J. Hypertens. 24, 964–969 (2011).

22. Kappers, M. H. W. et al. The vascular endothelial growth factor receptor inhibitor sunitinib causes a preeclampsia-like syndrome with activation of the endothelin system. Hypertension

58, 295–302 (2011).

23. Petrache, I. et al. Ceramide upregulation causes pulmonary cell apoptosis and emphysema-like disease in mice. Nat. Med. 11, 491–8 (2005).

24. Melland-Smith, M. et al. Disruption of sphingolipid metabolism augments ceramide-induced autophagy in preeclampsia. Autophagy 11, 653–669 (2015).

25. Spijkers, L. J. A. et al. Hypertension is associated with marked alterations in sphingolipid biology: A potential role for ceramide. PLoS One 6, (2011).

26. Merscher, S. & Fornoni, A. Podocyte pathology and nephropathy - sphingolipids in glomerular diseases. Frontiers in Endocrinology 5, (2014).

27. Lamorte, S. et al. Syndecan-1 promotes the angiogenic phenotype of multiple myeloma endothelial cells. Leukemia 26, 1081–1090 (2012).

28. Cohen, T. et al. VEGF121, a vascular endothelial growth factor (VEGF) isoform lacking heparin binding ability, requires cell-surface heparan sulfates for efficient binding to the VEGF receptors of human f melanoma cells. J. Biol. Chem. 270, 11322–11326 (1995).

29. Gitay-Goren, H., Soker, S., Vlodavsky, I. & Neufeld, G. The binding of vascular endothelial growth factor to its receptors is dependent on cell surface-associated heparin-like molecules.

J. Biol. Chem. 267, 6093–6098 (1992).

30. Sela, S. et al. Local retention versus systemic release of soluble VEGF receptor-1 are mediated by heparin-binding and regulated by heparanase. Circ. Res. 108, 1063–1070 (2011).

31. Nieuwdorp, M. et al. The endothelial glycocalyx: a potential barrier between health and

vascular disease. Current opinion in lipidology 16, (2005).

32. Nieuwdorp, M. et al. Measuring endothelial glycocalyx dimensions in humans: a potential novel tool to monitor vascular vulnerability. J Appl Physiol 104, 845–852 (2008).

(25)

Part I

Cardiovascular and renal complications

of malignant hypertension

(26)

(27)

2

Mortality and Cardiovascular Risk in Patients

with a History of Malignant Hypertension:

a Case-Control Study

Fouad Amraoui, Niels V. Van Der Hoeven, Irene G. M. Van Valkengoed, Liffert Vogt, and Bert-Jan H. Van Den Born

(28)

ABSTRACT

The survival of patients with malignant hypertension (MHT) has considerably improved over the past decades. Data regarding the excess risk of mortality and the contribution of conventional cardiovascular risk factors are lacking. We retrospectively assessed cardiovascular risk factors and all-cause mortality in 120 patients with a history of MHT and compared them with 120 normotensive (NT) and 120 hypertensive (HT) age, sex and ethnicity-matched controls. Total cholesterol, LDL cholesterol and BMI were lower in MHT patients compared to HT controls, whereas blood pressure, HDL cholesterol and smoking habit were similar. Median eGFR was lower in MHT patients compared NT and HT controls (both p<0.01). Annual incidence of all-cause mortality per 100 patient-years was higher in MHT patients (2.6) compared to NT (0.2) and HT (0.5) controls (both p<0.01). Mortality of patients with a history of MHT remains high compared to NT and HT controls. Patients with MHT had a more favourable cardiovascular risk profile compared to hypertensive controls, but a higher prevalence of renal insufficiency.

(29)

2 INTRODUCTION

Malignant hypertension (MHT) is a hypertensive emergency characterized by severe hypertension and acute microvascular complications including grade III or IV hypertensive retinopathy. If left untreated, 5-year survival is less than 5% mainly because of stroke, myocardial infarction, congestive heart failure and end-stage renal disease.[1-3] With the availability of antihypertensive drugs and improved patient care, mortality has been markedly reduced to ~10% after 5 years.[2;4] This is however still considerable, given the relatively young study populations, with age varying between 40 and 50 years at presentation.[2;5] Previous cohort studies, including our own, have shown that renal dysfunction is an important predictor of mortality in patients with MHT [2;4], while other studies suggested a role of traditional cardiovascular risk factors such as excess smoking, decreased levels of high-density lipoprotein (HDL) and poor blood pressure (BP) control.[6-9] However, most of these studies lacked a control population thereby limiting the internal validity. Nonetheless, insight into the excess risk of CVD and mortality in patients with a history of MHT is required to identify which preventive measures may further improve outcome of this extreme phenotype of hypertension related organ damage.

Therefore, the principle aim of this study was to quantify the excess mortality risk in patients with a history of MHT. The second aim was to investigate whether traditional cardiovascular risk factors contribute to the increased risk. To this end, we compared cardiovascular risk factors and all-cause mortality of patients with a history of MHT with age-, gender- and ethnicity-matched normotensive (NT) and hypertensive (HT) controls.

METHODS

Participants

We used a case-control design to compare patients with a history of MHT with NT and HT controls. The selection of patients with a history of MHT has been previously described. [4] Briefly, we searched the database of a large teaching hospital in Amsterdam, The Netherlands. The diagnosis at discharge is recorded in this database according to the International Classification of Diseases codes (ICD). All charts of patients admitted between August 1992 and January 2010 with MHT or related diagnoses were reviewed for clinical criteria of MHT including 1) diastolic BP ≥120 mmHg and 2) presence of grade III or IV hypertensive retinopathy.[10] Excluded were patients <18 years, pregnant women and patients who were already on dialysis before admission. Patients referred from elsewhere were excluded to prevent referral bias.

(30)

Patients with a history of MHT were individually matched for age, sex and ethnicity with NT and HT control subjects from the SUNSET study (Surinamese in The Netherlands: Study on Ethnicity and Health), a large population-based study among non-institutional adults.[11] The SUNSET study was carried out between 2001-2003 to assess the cardiovascular risk profile of 35-60 year old people of European, African and South-Asian origin in Amsterdam, in the catchment area of our hospital. In total, 1383 persons of South-Asian, African and European origin participated in an interview and physical examination and were followed through the national medical registration database until December 23rd 2007. Self-reported ethnicity was used for classification in ethnic groups. Black subjects were from sub-Saharan African descent, mainly from Ghana and Nigeria. Asian subjects were mainly from the Sub-Indian continent, whereas white subjects were of West-European ancestry. Patients with a history of MHT who were either younger or older than NT and HT controls from SUNSET were matched with controls closest to their own age.

Data collection and definitions

Vital status was assessed by inquiry of the municipal administration registries. For patients with a history of MHT, the cause of death was retrieved from the medical file or from general physicians. In addition, we recorded follow-up data on cardiovascular events of these patients. Data derived within three months after admission of patients with a history of MHT were censored, because death or cardiovascular events occurring during this period could be attributable to the acute episode of MHT. For NT and HT controls data on the cause of death or the number of cardiovascular events were not available.

All conventional cardiovascular risk factors including age, gender, ethnicity, systolic and diastolic BP, body mass index (BMI), smoking, lipid profile, statin use, fasting glucose, presence of diabetes mellitus, plasma creatinine and proteinuria were assessed at the entry of the SUNSET study for NT and HT controls. For patients with a history of MHT, age, gender, ethnicity, smoking status and presence of left ventricular hypertrophy were assessed at initial admission. Left ventricular hypertrophy was considered present when detected by cardiac ultrasound or by ECG according to the Sokolow-Lyon criteria. Systolic and diastolic BP, lipid profile, statin use, fasting glucose, prevalence of diabetes, BMI, plasma creatinine and proteinuria were documented during a follow-up visit at the outpatient department using a standardized risk assessment identical to the SUNSET study. Values assessed more than 2 years after admission were excluded. The median time between admission and cardiovascular risk assessment was 5 months, with an interquartile range (IQR) of 2-10 months after presentation.

Renal function was estimated according to the Modification of Diet in Renal Disease (MDRD) formula.[12] Macroalbuminuria was defined as urinary protein excretion >300 mg/day on 24-hour urine or >200 mg/L on a morning spot sample. All laboratory tests in patients

(31)

2

with MTH and the NT and HT controls were performed in the hospital’s central laboratory according to local protocols.

Statistical analysis

Continuous data were expressed as mean and SD or median and interquartile range (IQR) for variables with a skewed distribution. Categorical data were expressed as number and percentages. Differences between groups for continuous variables were assessed by a one-way ANOVA with post-hoc LSD correction for parametric or Dunnets post-hoc correction for non-parametric distributions. Chi-square tests were used for categorical variables. Annual incidence rates were calculated for mortality to account for differences in follow-up duration. The annual incidence rates were expressed as the number of events per 100 person-years of follow-up. To assess the mortality over time, Kaplan-Meier plots were generated to express five-year survival. The log rank test was used to assess differences in all-cause mortality between groups. SPSS software was used for all analyses (Statistical Package for the Social Sciences, version 19.0, Inc. Chicago, Illinois, USA). A p-value <0.05 was considered significant.

RESULTS

Characteristics of patients with malignant hypertension at presentation

A total of 120 patients admitted with MHT were included with a mean age of 44 years (range 19-79), 83 (69%) were male, and 60 (50%) were from West-European ancestry. Mean BP at admission was 230±23/145±17 mmHg. Neurologic symptoms consistent with hypertensive encephalopathy were present in 11 (9%) patients and 66 (55%) patients had grade IV hypertensive retinopathy. Left ventricular hypertrophy was present in 95 (79%) patients. Hypertension was diagnosed prior to admission in 65 (54%) patients and 39 (33%) patients were treated with antihypertensive medication. Median plasma creatinine at admission was 2.0 mg/dL with an interquartile range of 1.2–4.5 mg/dL. A primary renal disease could be identified in 9 patients (8%), and renovascular disease was diagnosed in 7 patients (6%).

Comparison of cardiovascular risk profiles at baseline

Patients with a history of MHT were well-matched for age, sex and ethnicity with HT and NT controls (Table 1). Systolic and diastolic BP levels during follow-up were higher in patients with a history of MHT compared to NT controls (both p<0.01), but were not different from HT controls (p=1.0 for systolic and p=0.3 for diastolic BP). BMI of MHT patients was similar compared to NT controls and lower compared to HT controls (p<0.01). Smoking habits did not differ between the groups. Patients with a history of MHT had lower total cholesterol

(32)

and LDL cholesterol levels compared to HT and NT controls (P<0.01), while statins were more frequently prescribed to patients with a history of MHT (9%) compared to HT (3%) and NT (2%) controls (p=0.02). After excluding MHT patients who used statins, mean plasma total cholesterol and LDL cholesterol remained significantly lower in patients with a history of MHT compared to HT controls (p<0.01 for both total cholesterol and LDL cholesterol). HDL cholesterol was similar in all groups. Triglycerides were comparable in patients with a history of MHT and HT, but lower in NT controls (p<0.01). There were no differences in fasting glucose levels among groups (p=0.19), however, diabetes mellitus was more prevalent in HT controls compared to both NT controls and MHT patients (p<0.01).

Table 1. Baseline Characteristics

NT HT MHT p-value

Patients, No 120 120 120

-Follow-up, months (IQR) 66 (62-70) 67 (62-70) 62 (24-103) 0.54

Age 44 (8) 44 (6) 44 (12) -Male, No. (%) 83 (69%) 83 (69%) 83 (69%) -Black, No. (%) White, No (%) Asian, No (%) 57 (48%) 60 (50%) 3 (3%) 57 (48%) 60 (50%) 3 (3%) 57 (48%) 60 (50%) 3 (3%) -Systolic BP, mmHg 116 (12) 144 (15) ¶ _{144 (23)} ¶ _<0.01 Diastolic BP, mmHg 75 (7) 93 (8) ¶ _{91 (15)} ¶ _<0.01 Total cholesterol, mg/dL 204 (40) 214 (43) 196 (44)† _<0.01 LDL cholesterol, mg/dL 130 (37) 133 (40) 114 (38) ¶ † _<0.01 HDL cholesterol, mg/dL 56 (14) 55 (16) 54 (18) 0.50 Triglycerides, mg/dL 90 (51) 131 (99) ¶ _{132 (96)}¶ _<0.01

Statin perscribed, No. (%) 2 (2) 4 (3) 10 (8) <0.05 Antihypertensive drugs, No (%) 0 24 (20%) 81 (68%)† _<0.01

Fasting plasma glucose, mg/dL 97 (23) 103 (22) 97 (23) 0.19 Diabetes mellitus, No (%) 5 (4%) 12 (10%) 5 (4%) <0.01 Body mass index, kg/m2 _{25.9 (4.8)} _{28.2 (5.4)} ¶ _{26.1 (5.1)}† _<0.01

Current Smoker, No (%) 67 (56%) 60 (50%) 52 (43%) 0.19 Plasma creatinine, mg/dL (IQR) 0.9 (0.8-1.0) 0.9 (0.8-1.0) 2.0 (1.2-3.4) ¶ † _<0.01

eGFR mL/min/1.73m2_(IQR) _{82 (71-112)} _{83 (71-109)} _{33 (14-68)} ¶ † _<0.01

Macroalbuminuria, No (%) 0 4 (3%) 66 (55%) <0.01 Cardiovascular risk factors at first follow-up visit in patients with a history of malignant hypertension (MHT) as compared to baseline values of age-, gender- and ethnicity-matched normotensive (NT) and hypertensive (HT) controls from the same area of residence. Values are mean with SD, median with IQR or numbers and percentage. SD, standard deviation; IQR, interquartile range; eGFR, estimated glomular filtration rate. ¶_{p<0.05 versus NT,}†_{p<0.05 versus HT.}

(33)

2

Median estimated glomerular filtration rate (eGFR) of patients with a history of MHT (33, IQR 14-68 mL/min/1.73m2) was lower compared to NT (82, IQR 71-112 mL/min/1.73m2) and HT (83, IQR 71-109 mL/min/1.73m2) controls (p<0.01). Macroalbuminuria was more often present in patients with a history of MHT compared to NT and HT controls (p<0.01). Sixteen out of 120 patients (13%) with MHT needed kidney replacement therapy at the follow-up visit compared to none in the groups with NT and HT controls.

Cardiovascular events in patients with a history of MHT

During a median follow-up of 62 months, cerebrovascular accidents were the most frequently observed cardiovascular event in patients with a history of MHT (n=5). Other events that occurred in MHT patients were myocardial infarction (n=4), angina pectoris (n=2), and peripheral artery disease (n=2). In addition, two cardiovascular events (1 stroke and 1 case of angina pectoris) occurred within three months after admission.

Comparison of all-cause mortality

Eighteen patients with a history of MHT died during follow-up. One patient died within three months after admission of a malignancy and was excluded from survival-analysis. Causes of death of the remaining 17 patients with a history of MHT included cardiovascular events (n=6), malignancy (n=2), infectious disease (n=2), renal failure (n=2), and was uncertain for five patients. Of the control subjects, one NT and three HT subjects died during follow-up. Annual incidence rate of all-cause mortality per 100 years of follow up was significantly higher in patients with a history of MHT (2.6) compared to both NT (0.2) and HT (0.5) controls (both p<0.05, Table 2). Log rank test of 5-year all-cause mortality showed a higher mortality in patients with a history of MHT compared to both NT and HT controls (both p<0.05, Figure 1).

Table 2. Annual Incidence Rates of All-cause Mortality

Total observation years NT677 HT665 MHT648

Deaths 1 3 17

Annual incidence rate 0.2 0.5 2.6

RR (95%CI) compared to NT 1 3.1 (0.3-29.4) 17.8 (2.4-133.6)* RR (95%CI) compared to HT 0.3 (0.0-3.1) 1 5.8 (1.7-19.8)** Annual incident rate of all-cause mortality per 100 person-years of follow-up in patients with a history of malignant hypertension (MHT) as compared to age-, gender- and ethnicity-matched normotensive and hypertensive controls from the same residence area; RR, relative risk; CI, confidence interval; NT, normotensive controls; HT, hypertensive controls; MHT, patients with malignant hypertension, *p< 0.05 compared to NT. **p< 0.05 compared to HT.

(34)

There was no difference in the average age of deceased patients with a history of MHT compared to NT and HT controls. Comparison of deceased and surviving patients with a history of MHT showed that patients who died during follow-up tended to be older (51±19 vs. 43±11 years), were less often male (41% versus 75%, p=0.01) and tended to smoke more often (71% versus 46%). There were no significant differences in BP (152±24 /90±17 vs. 143±21/89±13 mmHg), and other cardiovascular risk factors between deceased and surviving patients with MHT. Estimated GFR (42 [IQR 16-105] vs. 37 [IQR 25-58] mL/ min/1.73m2) was similar in deceased and surviving patients with a history of MHT, whereas macroalbuminuria tended to be present more often in deceased patients (65% vs. 54%).

Figure 1: 5-year all-cause mortality for each group

Legend: NT, normotensive controls; HT, hypertensive controls; MHT, patients with a history of malignant hypertension, *p<0.05 on log-rank test for MHT compared to NT and HT. The inner panel shows an enlargement of the outer figure.

DISCUSSION

We show that despite considerable improvement in survival over the past decades, patients with a history of MHT remain at increased risk of dying compared to age, sex, and ethnicity matched NT and HT controls. Cardiovascular risk factors seem of little influence on the excess mortality, as total and LDL cholesterol, obesity and prevalence of diabetes mellitus were higher in HT controls compared to patients with MHT, while smoking habit was comparable. BP levels were also similar in MHT patients and HT controls, suggesting that adherence to antihypertensive medication in patients with a history of MHT may have improved after admission to the hospital. However, renal function was significantly impaired in patients

(35)

2

with MHT compared to NT and HT controls, suggesting that renal dysfunction may be an important contributor to the higher mortality rate observed in patients with a history of MHT.

There is ample evidence that renal dysfunction increases the risk of cardiovascular and all-cause mortality, with both decreased eGFR and proteinuria contributing individually to this increased risk.[13-16] In fact, all-cause mortality is similar in patients with chronic kidney disease (CKD) when compared to diabetic patients without CKD.[17] In the present study, we observed that patients with a history of MHT who died during follow-up, had more often macroalbuminuria compared to surviving patients with a history of MHT. In addition, 16 (13%) patients with a history of MHT were on kidney replacement therapy.

Hypertension is associated with clustering of cardiometabolic risk factors, including obesity, diabetes and dyslipidemia.[18] HT control subjects indeed had an increased cardiometabolic risk as demonstrated by the higher BMI, higher LDL and triglyceride levels and higher prevalence of diabetes mellitus compared to NT controls. However, there was no evidence of clustering of cardiometabolic risk factors in patients with a history of MHT except for higher triglyceride levels. The apparent lack of cardiovascular risk factor clustering in patients with a history of MHT contradicts previous reports, which showed that these patients had higher plasma triglycerides, lower plasma HDL cholesterol, and smoked more often compared to either non-malignant HT or NT controls.[6-9] Differences in the proportion of smoking patients could be explained by temporal changes in smoking behaviour as studies on associations between MHT and cardiovascular risk date back over 30 years ago. In addition, previous studies did not use matched control subjects to account for differences in socioeconomic status or cultural background, potentially influencing cardiometabolic risk factors and smoking behaviour. In the present study, control groups were derived from the same residence area and were individually matched for age, sex and ethnicity with MHT patients to limit differences in socioeconomic status and cultural background. With regard to the aforementioned difference in HDL cholesterol, timing of the blood collections may have been relevant as HDL levels were previously assessed in the acute phase of MHT. Because HDL cholesterol is an acute phase reactant, the lower HDL cholesterol levels in that study may have been influenced by the inflammatory response associated with MHT. To avoid influence of these acute effects, cardiovascular risk profile was completed with a fasting venous blood sample after patients were discharged from the hospital and BP lowering treatment was instituted.

Despite the lack of an unfavourable cardiovascular risk profile compared to HT and NT controls, 13 (11%) patients with a history of MHT suffered from cardiovascular events during follow-up. The implication of this discrepancy between estimated cardiovascular risk and the observed number of cardiovascular events is that risk predictors based on traditional risk factors such as the Framingham or SCORE underestimate the risk in patients with a

(36)

history MHT. Our data show that patients with a history of MHT should be considered as high risk patients and suggest, in line with ESH recommendations, that prediction models for cardiovascular risk should be avoided in these patients.

This study has both strengths and limitations. Our study is the first to compare the survival and cardiovascular risk factors of a large group of consecutive patients with a history of MHT with that of NT and HT controls. Limitations include firstly its retrospective nature. Due to coding errors some patients with a history of MHT could have been missed. To overcome this, we performed a sensitivity analysis showing that no patients with a history of MHT who visited the emergency room between 1992–2008 were missed. [4] Secondly, the age range of SUNSET participants was limited to 35-60 year, whereas the patients with a history of MHT were aged 19-79. Nonetheless, most patients with a history of MHT that fell outside this age range were younger than 35, implying a lower mortality risk, and the mean age of deceased patients was similar among all groups. Thridly, the recruitment window of NT and HT controls from SUNSET was considerably smaller leading to a much smaller variation in follow-up time compared to the patients with a history of MHT. To account for this, the annual incidence rate of all-cause mortality was calculated. Because the median follow-up time was similar, we estimate that the influence on our results is limited. Finally, average follow-up BP was similar in patients admitted from 1992 to 2000 compared to those admitted from 2001 to 2010 (data not shown), indicating that introduction of new antihypertensive medication during the recruitment did not change BP control rate.

In conclusion, mortality is increased in patients with a history of MHT compared to matched normotensive and hypertensive controls. Patients with MHT had a favourable cardiovascular risk profile compared to HT controls, but had severe renal dysfunction. Since uncontrolled hypertension is the only modifiable predictor of long-term renal outcome and mortality in patients with MHT [4], tight BP control should be the primary goal in the management of patients with a history of MHT.

(37)

2 REFERENCES

1. Keith NM, Wagener HP, Barker NW. Some different types of essential hypertension: their course and prognosis. AM J Med Sci. 1939;196:332-339.

2. Lane DA, Lip GY, Beevers DG. Improving survival of malignant hypertension patients over 40 years. Am J Hypertens. 2009;22:1199-1204.

3. Lip GY, Beevers M, Beevers DG. Complications and survival of 315 patients with malignant-phase hypertension. J Hypertens. 1995;13:915-924.

4. Amraoui F, Bos S, Vogt L, van den Born BJ. Long-term renal outcome in patients with malignant hypertension: a retrospective cohort study. BMC Nephrol. 2012;13:71.

5. van den Born BJ, Koopmans RP, Groeneveld JO, van Montfrans GA. Ethnic disparities in the incidence, presentation and complications of malignant hypertension. J Hypertens 2006;24:2299-2304.

6. Bloxham CA, Beevers DG, Walker JM. Malignant hypertension and cigarette smoking. BMJ. 1979;1:581-583.

7. Edmunds E, Landray MJ, Li-Saw-Hee FL, Hughes BA, Beevers DG, Lip GY. Dyslipidaemia in patients with malignant-phase hypertension. QJM. 2001;94:327-332.

8. Isles C, Brown JJ, Cumming AM, Lever AF, McAreavey D, Robertson JI et al. Excess smoking in malignant-phase hypertension. BMJ. 1979;1:579-581.

9. Tuomilehto J, Elo J, Nissinen A. Smoking among patients with malignant hypertension. BMJ. (Clin Res Ed) 1982;284:1086.

10. World Health Organization. Arterial hypertension. WHO Tech Rep Ser. 1978;628:57.

11. Bindraban NR, van Valkengoed IG, Mairuhu G, Holleman F, Hoekstra JB, Michels BP et al. Prevalence of diabetes mellitus and the performance of a risk score among Hindustani Surinamese, African Surinamese and ethnic Dutch: a cross-sectional population-based study. BMC Public Health. 2008;8:271.

12. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med. 1999;130:461-470.

13. Chronic Kidney Disease Prognosis Consortium. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet. 2010;375:2073-2081.

14. Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med. 2004;351:1296-1305.

15. Tonelli M, Muntner P, Lloyd A, Manns BJ, James MT, Klarenbach S et al. Using proteinuria and estimated glomerular filtration rate to classify risk in patients with chronic kidney disease: a cohort study. Ann Intern Med. 2011;154:12-21.

16. K/DOQI clinical practice guidelines on hypertension and antihypertensive agents in chronic kidney disease. Am J Kidney Dis. 2004;43:S1-290.

17. Tonelli M, Muntner P, Lloyd A, Manns BJ, Klarenbach S, Pannu N et al. Risk of coronary events in people with chronic kidney disease compared with those with diabetes: a population-level cohort study. Lancet. 2012;380:807-814.

18. Weycker D, Nichols GA, O’Keeffe-Rosetti M, Edelsberg J, Khan ZM, Kaura S et al. Risk-factor clustering and cardiovascular disease risk in hypertensive patients. Am J Hypertens. 2007;20:599-607.

(38)

(39)

3

Long-term renal outcome in patients with malignant

hypertension: a retrospective cohort study

Fouad Amraoui, Sarah Bos, Liffert Vogt and Bert-Jan van den Born

(40)

ABSTRACT

Background: Malignant hypertension is frequently complicated by renal insufficiency.

Although the survival of this hypertensive emergency has improved, recent data on renal outcome and its predictors are lacking. We assessed renal outcome and its predictors in patients with malignant hypertension.

Methods: Retrospective analysis of patients with malignant hypertension in Amsterdam,

the Netherlands between August 1992–January 2010. Follow-up data on vital status, renal function and blood pressure (BP) were obtained from the outpatient department and from general practitioners. The primary composite endpoint was end-stage renal disease (ESRD) defined as the start of kidney replacement therapy (KRT) or ≥ 50% decline of estimated glomerular filtration rate (eGFR). The secondary endpoint was all cause mortality.

Results: A total of 120 patients admitted with malignant hypertension were included.

After a median follow-up period of 67 months (IQR 28 to 108 months) the primary endpoint was reached by 37 (31%) patients, whereas 18 patients (15%) reached the secondary endpoint. Twenty-nine (24%) patients started KRT and 8 patients (7%) had an eGFR decline ≥ 50%. After the acute phase (> 3 months after admission), initial serum creatinine and follow-up BP were the main predictors of future ESRD with hazard ratios of 6.1 (95% CI, 2.2–17) for patients with initial serum creatinine ≥ 175 µmol /L and 4.3 (95% CI, 1.4–14) for patients with uncontrolled hypertension.

Conclusions: Progressive renal function decline leading to ESRD remains a major threat

to patients with malignant hypertension. BP control during follow-up was an important modifiable predictor of renal outcome.

(41)

3 BACKGROUND

Malignant hypertension is a hypertensive emergency usually defined by the presence of severe hypertension in combination with ischemic retinal changes consistent with grade III or IV hypertensive retinopathy according to the Keith, Wagener and Barker classification [1]. Malignant hypertension is frequently complicated by renal dysfunction [2,3]. In most cases this is secondary to malignant hypertension with renal biopsy specimens typically showing acute ischemic changes secondary to medial hypertrophy, intimal hyperplasia and fibrinoid necrosis of small arteries and arterioles [4]. Although the survival of malignant hypertension has considerably improved with the advent of antihypertensive therapy, end-stage renal disease (ESRD) remains a significant cause of morbidity and mortality [5,6]. Depending on the setting and patient characteristics between 18% and 41% of patients may require kidney replacement therapy (KRT) during the acute phase [6,7].

One would expect that because of advances in the awareness, treatment and control of hypertension in the population at large the incidence of malignant hypertension and its renal complications would have declined. However, we and others have shown that malignant hypertension remains relatively common in large multi-ethnic communities and in urban sub-Saharan African populations. In the Netherlands, the relative contribution of malignant hypertension to the total number of patients starting KRT has increased by 40% in the past two decades [8]. Annual reports of the European Dialysis and Transplant Association and the US Renal Data System show that hypertension is an important cause of ESRD, but the role of hypertensive crises are not specified [9,10].

Previous studies have demonstrated recovery from acute renal dysfunction in some patients with malignant hypertension and ESRD [11,12]. However, data concerning long-term renal outcome are limited. Moreover, the influences of modifiable predictors such as blood pressure (BP) control have not been established. In this study, we aimed to assess long-term renal outcome and predictors of ESRD and renal function decline in patients with malignant hypertension.

METHODS

Participants

Included were patients presenting with malignant hypertension at a large teaching hospital, serving a multi-ethnic community in Amsterdam, the Netherlands. Patients were recruited between August 1992 and January 2010. The methods for the selection strategy have been previously described [6]. Briefly, the hospital database, in which the diagnosis at discharge is recorded according to the International Classification of Diseases codes (ICD) was searched.