A. Gallo

Effects of age, blood pressure and antihypertensive treatments on retinal arterioles remodeling assessed by adaptive optics.

Background: In humans, adaptive optics camera enables precise large-scale noninvasive retinal microcirculation evaluation to assess ageing, blood pressure and antihypertensive treatments respective roles on retinal arterioles anatomy. Method: We used adaptive optics camera rtx1 (Imagine-Eyes, Orsay, France) to measure wall thickness, internal diameter and to calculate wall-to-lumen ratio (WLR) and wall cross-sectional area of retinal arterioles. This assessment was repeated within a short period in two subgroups of hypertensive individuals without or with a drug-induced blood pressure drop. Results: In 1000 individuals, mean wall thickness, lumen diameter and WLR were 23.2 ± 3.9, 78.0 ± 10.9 and 0.300 ± 0.054 &mgr;m, respectively. Blood pressure and age both independently increased WLR by thickening arterial wall. In opposite, hypertension narrowed lumen in younger as compared to older individuals (73.2 ± 9.0 vs. 81.7 ± 10.2 &mgr;m; P < 0.001), whereas age exerted no influence on lumen diameter. Short-term blood pressure drop (−29.3 ± 17.3/−14.4 ± 10.0 mmHg) induced a WLR decrease (−6.0 ± 8.0%) because of lumen dilatation (+4.4 ± 5.9%) without wall thickness changes. By contrast, no modifications were observed in individuals with stable blood pressure. In treated and controlled hypertensives under monotherapy WLR normalization was observed because of combined wall decrease and lumen dilatation independently of antihypertensive pharmacological classes. In multivariate analysis, hypertension drug regimen was not an independent predictor of any retinal anatomical indices. Retinal arteriolar remodeling comprised blood pressure and age-driven wall thickening as well as blood pressure-triggered lumen narrowing in younger individuals. Conclusion: Remodeling reversal observed in controlled hypertensives seems to include short-term functional and long-term structural changes.

Is lomitapide a life-saving drug in homozygous familial hypercholesterolemia

In the paper published in this review, Robert Leipold et al. conducted a modelling analysis of the potential effect of lomitapide, on major adverse cardiovascular events and survival in homozygous familial hypercholesterolemia. This rare and life-threatening disease has been defined phenotypically on the basis of an untreated (low-density lipoprotein (LDL)-cholesterol plasma concentration of more than 13mmol/L (>500mg/dL), or a treated LDL-cholesterol concentration of 8mmol/L or greater ( 300mg/dL), and the presence of cutaneous or tendon xanthomas before the age of 10 years, or the presence of elevated LDL-cholesterol levels consistent with heterozygote familial hypercholesterolemia in both parents. Patients with homozygous familial hypercholesterolemia are at extremely high risk of cardiovascular disease, with severe events occurring during childhood or young adulthood if untreated. From the pre-statin era, the management of homozygous familial hypercholesterolemia has evolved following the widespread availability of statins, ezetimibe, as well as lipoprotein apheresis. However, despite these treatments most patients do not reach their LDL-cholesterol targets. These patients may benefit from new therapeutic options such as proprotein convertase subtilisin-kexin type 9 (PCSK9) antibodies, lomitapide and mipomersen (microsomal transfer protein inhibitors) and cholesterol-ester transfer protein inhibitors. The latter group has not yet obtained market approval. Subject to consideration of benefit/ versus risk and cost, which may differ from country to country, such pharmacotherapies may ultimately translate to improved clinical outcome for patients with this disease. Randomised studies assessing the impact of these new therapeutic options on clinical events are not feasible and might even be unethical. However, the cost of the above-mentioned new therapeutic options is also very high and it is most useful to approach by modelling their potential benefit. This is obviously important for the homozygous but also the heterozygote form in which no clinical outcome trials have yet been conducted. However, we should keep in mind that homozygous hypercholesterolemia is heterogeneous in terms of LDL-cholesterol reduction upon treatment. In this respect it is important to look specifically at the benefit in the most interesting subgroup with null mutations which does not respond to PCSK9 inhibitors but has a significant decrease with lomitapide. The modelling analysis considered cardiovascular events and effect on mortality. However, the disease represents a considerable burden for patients, not only due to cardiovascular disease but also due to physical signs and limitations together with a number of psychosocial factors, treatment-related issues and impact on the education and employment situation. In addition, the cost of treatment including long-term weekly or bi-weekly apheresis is extremely high. The model is based on LDL-cholesterol reduction. The risk of cardiovascular events is strongly related to cholesterol burden, i.e. the average exposure to cholesterol throughout life which is a function of plasma cholesterol levels and duration of exposure. As a causal relationship between LDL-cholesterol and cardiovascular events is well established, the benefit of hypolipidemic drugs can indeed be evaluated by the average effect on plasma LDLcholesterol levels. The modelling was based on a 38% plasma LDL-cholesterol reduction, which is, according to the authors, a conservative approach because the maximal efficacy might be associated with up to a 50% decrease of plasma LDL-cholesterol levels. When analysing effect on events and survival it is important to have long-term confirmation of the average LDL-cholesterol decrease in a real-life setting. The experience of the

Retinal arteriolar remodeling evaluated with adaptive optics camera: Relationship with blood pressure levels

AIM To research a retinal arterioles wall-to-lumen ratio or lumen diameter cut-off that would discriminate hypertensive from normal subjects using adaptive optics camera. PATIENTS AND METHODS One thousand and five hundred subjects were consecutively recruited and Adaptive Optics Camera rtx1™ (Imagine-Eyes, Orsay, France) was used to measure wall thickness, internal diameter, to calculate wall-to-lumen ratio (WLR) and wall cross-sectional area of retinal arterioles. Sitting office blood pressure was measured once, just before retinal measurements and office blood pressure was defined as systolic blood pressure>=140mmHg and diastolic blood pressure>=90mmHg. ROC curves were constructed to determine cut-off values for retinal parameters to diagnose office hypertension. In another population of 276 subjects office BP, retinal arterioles evaluation and home blood pressure monitoring were obtained. The applicability of retinal WLR or diameter cut-off values were compared in patients with controlled, masked, white-coat and sustained hypertension. RESULTS In 1500 patients, a WLR>0.31 discriminated office hypertensive subjects with a 0.57 sensitivity and 0.71 specificity. Lumen diameter<78.2μm discriminated office hypertension with a 0.73 sensitivity and a 0.52 specificity. In the other 276 patients, WLR was higher in sustained hypertension vs normotensive patients (0.330±0.06 vs 0.292±0.05; P<0.001) and diameter was narrower in masked hypertensive vs normotensive subjects (73.0±11.2 vs 78.5±11.6μm; P<0.005). CONCLUSION A WLR higher than 0.31 is in favour of office arterial hypertension; a diameter under<78μm may indicate a masked hypertension. Retinal arterioles analysis through adaptive optics camera may help the diagnosis of arterial hypertension, in particular in case of masked hypertension.

ADAPTIVE OPTICS CAMERA ENABLES TO DESCRIBE DIFFERENT PATTERNS OF RETINAL VASCULATURE IN HYPERTENSION AND TYPE 2 DIABETES

Objective: Retinal vasculature is a well-known target of early organ damage in hypertension and diabetes. Adaptive optics, a totally noninvasive, accurate method, allows the precise evaluation of retinal arteriolar and venular networks. The aim of this study was to describe arteriolar and venular vasculature in hypertensive and diabetic subjects by comparison to controls using adaptive optics camera. Design and method: Hypertensive patients (group H), diabetic subjects without overt diabetic retinopathy (group D) and non-hypertensive non-diabetic control subjects (group C) were recruited. Adaptive Optics RTX1® Camera was used to capture three consecutive images along the supero-temporal arteriole and vein (avoiding any arterio-venous nicking) and measure arteriolar and venular internal diameter (ID) in order to calculate AVR. For arteries, Wall Thickness (WT) was also measured to calculate Wall-to-Lumen Ratio (WLR) and Wall Cross-Sectional Area (WCSA). Hypertension was defined according to the presence of any antihypertensive treatment, diabetes was defined by any antidiabetic treatment or HbA1c > 6.5% on two samples. Results: 129 patients were included (53 group H, 38 group D and group C). Despite the same BP levels in groups H and D, Hypertensive patients had a significant arteriolar narrowing and diabetics patients an arteriolar dilation (group H 85.4 ± 13 &mgr;m vs group D 97.7 ± 12 &mgr;m and group C 93.3 ± 12 &mgr;m, p = 0.015) but a similar venular ID was observed in all groups (group H 126.2 ± 17 &mgr;m vs group D 124.5 ± 18 &mgr;m vs group C 124.4 ± 18 &mgr;m).Hypertensive subjects had a lower AVR (0.68 vs group D 0.77 vs group C 0.76, p = 0.004) and a higher WLR (0.283 ± 0.04 group H, 0.258 ± 0.03 group D and 0.257 ± 0.04 group C).In diabetics, an increased WCSA was also observed (4327.2 &mgr;m2 group D vs 3823.7 &mgr;m2 group H vs 4008.0 group C), no differences were found in WT (23.6 ± 3 &mgr;m group H vs 24.9 ± 3 &mgr;m group D vs 24.2 ± 4 &mgr;m group C). Conclusions: Two different microvascular patterns were observed in this study: arteriolar inward remodeling characterizes hypertension and arteriolar eccentric hypertrophy is observed in diabetes. Adaptive Optics represents a powerful source of new microvascular markers and may contribute to a revival of the fundoscopy in hypertension and diabetes.

RETINAL ARTERIOLAR MICRO-CONSTRICTIONS EVALUATED WITH ADAPTIVE OPTICS: A NOVEL MARKER IN HYPERTENSION

Objective: Retinal arteriolar remodeling is an early marker of subclinical target organ damage in arterial hypertension. Through adaptive optics which is totally noninvasive and highly accurate, it is possible to measure changes in arteriolar diameter within 1 &mgr;m accuracy. The aim of this study was to evaluate a new marker describing internal diameter variability of the supero-temporal arteriole in hypertensive patients before and after blood pressure control. Design and method: Adaptive Optics RTX1® Camera (ImagineEye, Orsay, France) was used to capture three consecutive images along the supero-temporal arteriole. Wall Thickness (WT) and internal diameter (ID) were measured to calculate Wall-to-Lumen Ratio (WLR) and Wall Cross-Sectional Area (WCSA). A coefficient of variation (CV) for ID was calculated for each group by the following formula: (standard deviation ID/mean ID)*100 over three consecutive measurements. Subjects with a CV ID > 75% were classified as irregular. Uncontrolled hypertensive subjects in the irregular group were given an antihypertensive pharmacological treatment and were reevaluated 1 month after. Results: 44 patients were analyzed (mean age 47.7 ± 11). Median CV ID in the irregular group was 11% [IQR 9.0–15.0] as compared to 2.0% (regular group) [IQR 1.0–4.0], p < 0.001. Patients in the arteriolar irregular group had an increase in home blood pressure (148.3/96.3 vs 130.7/ 82.6 mmHg, p < 0.01). They had significantly decreased ID (82.24 ± 13.5 vs 89.9 ± 14.8, p = 0.01) and increased WLR (0.311 ± 0.07 vs 0.262 ± 0.04, p = 0.025) whereas no differences were observed in WT and WCSA. At one-month follow-up, along with a significant blood pressure reduction, which was associated with arteriolar enlargement and WLR reduction, a decrease in median ID CV was observed (11%[IQR 9.0–15.0] to 4.2% [IQR 1.8–6.05], p = 0.014). Figure. No caption available. Conclusions: Arteriolar micro-constrictions are observed in a subset of hypertensive patients with the use of adaptive optics camera. A decrease in blood pressure is accompanied with their disappearance. Beyond classical retinal microvascular remodeling indexes, Adaptive Optics may allow the definition of novel markers of microvascular remodeling that are associated with hypertension.

[PP.19.16] SHORT-TERM RETINAL ARTERIOLAR LUMEN CHANGES ARE UNRELATED TO PERIPHERAL VASCULAR RESISTANCE CHANGES IN TREATED HYPERTENSIVE PATIENTS

Objective: The microcirculation is the major site of the total peripheral resistance (TPR); the arteriolar vasomotricity is its determinant. While some correlations between TPR and retinal arteriolar remodeling have been shown in a chronic setting, no prospective data exists regarding short-term changes in retinal arteriolar lumen and total resistances. Our aim was to evaluate the relations between changes in TPR and retinal microvasculature anatomical indices in uncontrolled hypertensive patients. Design and method: 29 hypertensive patients (16 males, 13 females, mean age 48 ± 14) underwent Adaptive Optics imaging using RTX1 camera® (ImagineEye, Orsay, France) to measure Wall Thickness (WT), Internal Diameter (ID), Wall Cross Sectional Area (WCSA) and Wall-to-Lumen Ratio (WLR) of retinal arterioles. The Integrated Hemodynamic System (Hotman System®) was used to calculate Stroke Systemic Vascular Resistance Index (SSVRI), an index of TPR. Among them, 10 uncontrolled hypertensive subjects (6 males, 4 females, mean age 50 ± 12) had a follow-up assessment 1 month after introduction or increase of an antihypertensive therapy. Results: At baseline, mean Systolic BP (SBP) and Diastolic BP were 147 ± 21 and 84.9 ± 13 mmHg respectively. Retinal WT, ID, WLR and WCSA were 21.8 ± 3.6 &mgr;m, 74.5 ± 12 &mgr;m, 0.299 ± 0.060 and 2954 ± 765 &mgr;m2, respectively. Mean SSVRI was 142.4 ± 40 dyn*sec/cm5. In the whole population, no correlation was found between ID and SSVRI (Spearmans &rgr; = 0.292, p = 0.124). Baseline and follow up BP levels, retinal arteriolar anatomical indices and resistance characteristics are shown in Table 1. Figure. No caption available. At follow-up, a significant decrease in SBP was observed (−13 ± 7.5% p = 0.008) in association with a significant lumen dilatation of retinal arterioles (+11.6 ± 7.6%, p = 0.018). No significant changes were observed in other retinal parameters or in SSVRI although vasodilator drugs were used. Conclusions: Retinal arteriolar lumen dilatation was observed after short-term brachial BP reduction, and was not related to TPR decrease. This suggests that 1) the changes in retinal arteriolar network may be more easily detected than TPR changes, and/or that 2) lumen dilatation or retinal arterioles changes occur to a greater extent than at the peripheral level.

[PP.19.15] RETINAL ARTERIOLES HYPERTROPHIC REMODELING IN UNCONTROLLED ACROMEGALY

Objective: Acromegaly is characterized by an increased cardiovascular risk. Hypertrophic effects of Growth Hormone (GH)/Insulin-like Growth Factor 1 (IGF1) system on the heart and large arteries are known but little data on microcirculation exist. Our aim was to assess retinal arteriolar remodeling in acromegaly patients using adaptive optics camera. Design and method: Adaptive Optics RTX1® Camera (ImagineEye, Orsay, France) was used to measure Wall Thickness (WT), Internal Diameter (ID) and to calculate Wall Cross Sectional Area (WCSA) and Wall-to-Lumen Ratio (WLR) on retinal arterioles of patients with acromegaly. As IGF1 is gender and age-dependent, an IGF1/normal value ratio (IGF1r) was generated for each patient. Acromegaly patients were then stratified according to their IGF1r: patients with IGF1r> = 1 were defined as uncontrolled, patients with IGF1r<1 were defined as having controlled acromegaly. Moreover, non-acromegaly control subjects matched for age/gender/diabetes/blood pressure levels and antihypertensive treatments were also recruited. Results: 80 patients and controls were recruited. Mean age was 51 ± 12 years and 54% were men. Subjects with uncontrolled acromegaly exhibited hypertrophic remodeling with increased WLR, WT and WCSA compared to both controlled patients and control subjects (Table). Figure. No caption available. No differences in ID were found between controls and patients with controlled acromegaly. Mean IGF1r was higher in uncontrolled subjects compared to subjects with controlled acromegaly (1.3 ± 0.46 vs 0.72 ± 0.18, p < 0.001). Moreover, IGF1r value was positively associated to WLR (r2 = 0.3, p < 000.1) and negatively to lumen (r2 = 0.1, p = 0.02) while there was a trend towards a positive association with WCSA and WT. Conclusions: Subjects with uncontrolled acromegaly exhibit hypertrophic arteriolar retinal remodeling associated with IGF1 levels increase. Normal retinal arteriolar anatomy has been found in patients with a controlled disease suggesting a potential reverse remodeling under treatment.

[OP.5A.02] SHORT AND LONG-TERM EFFECTS OF ANTIHYPERTENSIVE TREATMENT ON HUMAN RETINAL ARTERIOLE REMODELING EVALUATED WITH ADAPTIVE OPTICS CAMERA.

Objective: To evaluate the effects of blood pressure (BP) drop on retinal arteriole remodeling in hypertensive patients in a short-term vs long-term follow-up. Design and method: 57 hypertensive subjects without diabetes were consecutively enrolled and evaluated at baseline, 6 weeks (short-term) and 40 weeks (long-term). Among them, 28 had stable BP and 29 had BP drop after prescription of antihypertensive therapy after baseline visit. Adaptive Optics Camera using RTX1® (ImagineEye, Orsay, France) was used to measure Wall Thickness (WT), Internal Diameter (ID), Wall Cross Sectional Area (WCSA) and Wall-to-Lumen Ratio (WLR). Results: In the group of patients with “BP drop” at baseline, mean SBP/DBP, WT, ID, WLR and WCSA were 157/86 mmHg, 24.7 ± 4.7 &mgr;m, 73.3 ± 8.8 &mgr;m, 0.342 ± 0.07 and 3352 ± 943 &mgr;m2 respectively. At short and long-term follow up SBP decrease was – 13% and −21% and DBP −7% and −12%, respectively. In this group, while WCSA remained stable at 6 weeks a statistically significant reduction in WLR (−6%) was observed due to lumen dilatation (+2.4%) and wall thinning (−4%). After 40 weeks also, while WCSA still remained stable, a further WLR decrease was observed compared to baseline (−11%) also related to both further lumen dilatation (+5.6%) and arterial wall thickness decrease (−6.5%). No changes in BP or in retinal arteriole anatomical indices were observed in patients with a stable BP throughout. Conclusions: After antihypertensive treatment induced BP drop, the observed WLR decrease is due to lumen dilatation and wall thinning without any significant WCSA change. This suggests that this retinal arteriole reverse eutrophic remodeling depends on changes in BP and myogenic tone, wall components rearrangements more than on wall mass changes.

CO-01: Retinal arterioles remodeling evaluated by adaptive optics camera in humans and its relationships with age, blood pressure and cardio-vascular risk factors

BACKGROUND Microvascular vessels and microvascular remodeling play a major role in blood pressure (BP) regulation and peripheral tissue oxygen delivery. The adaptive optics camera (AOC), a novel technique of fundus image analysis enables the noninvasive characterization of microvascular remodeling in the retina on a large scale. Our objective was to assess the role of Blood Pressure (BP), ageing, diabetes and other risk factors on retinal arteriolar remodeling using AOC. METHODS In 1.000 consecutive subjects in primary prevention with risk factors, we used the new RTX1(®) AOC (Imagine-Eyes, Orsay, France) and a semi-automated segmentation software to measure Wall Thickness (WT), Internal Diameter (ID) and to calculate Wall-to-Lumen Ratio (WLR) and Wall Cross Sectional Area (WCSA) on retinal arteriolar microvasculature. BP was assessed on standard conditions using oscillometric device during the AOC assessment. Hypertension was defined as the presence of Systolic BP>140mmHg and/or Diastolic BP>90mmHg. Standardized clinical and biological examinations were performed to assess for the presence of diabetes, dyslipidemia, obesity and current smoking which were defined according to guidelines. RESULTS Overall, ageing was associated with an increase in WLR due to sole WT increase. Subjects with office hypertension (n=313) had inward eutrophic remodeling with increased WLR (0.325±0.049 vs 0.292±0.056) due to higher WT, lower ID and overall stable WCSA. Diabetics (n=180) presented with hypertrophic remodeling indicated by higher WLR due to a significant increase in WT and WCSA concomitant to a normal BP. There was no significant remodeling associated with other cardiovascular risk factors. In multivariate analysis, BP and age remained independent positive correlates of WLR and WT whereas age had no influence on ID and only BP negatively correlated to ID CONCLUSIONS: Hypertension, age and diabetes are associated with increase in WLR and WT. Assessment OF WT and ID are mandatory to perform a correct interpretation of WLR increase. OAC could represent a promising tool for interventional studies on treatment-induced remodeling regression.

P4-33: Variations of retinal arteriolar wall-lumen ratio depend on mechanism of blood pressure changes

BACKGROUND A recently developed non-invasive opto-electronic technology (Adaptive Optics Camera, AOC) provides a morphologic analysis of the retinal arteriolar tree. Coupled with a validated semi-automated segmentation software it allows reproducible and reliable measurements of wall thickness (WT), internal diameter (ID), wall cross sectional area (WCSA) and calculation of wall-to-lumen ratio (WLR). Our aim was to assess the determinants of WLR along the temporal retinal artery and in case of acute Blood Pressure (BP) rise or drug-induced short term BP drop. METHODS We studied 3 groups of patients : 1/ We consecutively enrolled 48 patients and calculated WLR before the first bifurcation (parental vessel) and on its large and small post-bifurcation branches on the right eye temporal superior artery. 2/ In a subgroup of 10 patients WLR was measured at rest and right after exercise dynamic test. 3/ In 55 hypertensive subjects AOC assessment was repeated after a median of 49 days in 27 subjects with stable BP and 38 days in 28 subjects with a drop in BP secondary to antihypertensive drug adaptation. WT and ID were directly measured using non-invasive rtx1(®) AOC (ImagineEyes-Orsay, France) and a dedicated software analysis system was used to calculate WLR and WCSA. BP was measured at the time of AOC examination in standard conditions. Hypertension was defined as the presence of Systolic Blood Pressure (SBP)>140mmHg and/or Diastolic blood pressure (DBP)>90mmHg. RESULTS 1/ In the first group, in both normo- and hypertensive patients, a significant decrease in ID, WT and WCSA was observed on post-bifurcation branches whereas WLR remained constant (table 1, next page). 2/ In the second group, while SBP acutely increased of 47±8mmHg, ANOVA analysis did not show any significant changes in WLR. 3/ At 6 weeks, in the follow-up group with a short-term SBP drop (-19.3±7.7%), a decrease of WLR (-7.0±8.0mmHg) due to ID increase (+5.6±5.9%) without significant changes in WT and WSCA were observed. No changes were observed in the group with a stable BP (see table next page). CONCLUSIONS WLR, the index of remodeling in retinal small arteries, is constant along the arterial tree despite changes in WT and ID. While WLR did not change during acute exercise-induced BP surge, it decreased due to a ID increase after a short-term drug-induced BP drop. Those results confirm the homogeneity of remodeling along the retinal arteriolar tree and that mechanisms of BP changes have different effects on remodeling.