Journal of Human Hypertension (2012) 26, 56–63 & 2012 Macmillan Publishers Limited All rights reserved 0950-9240/12 www.nature.com/jhh ORIGINAL ARTICLE Impairment of skin blood flow during post-occlusive reactive hyperhemy assessed by laser Doppler flowmetry correlates with renal resistive index

P Coulon1, J Constans2 and P Gosse1 1Service de Cardiologie et Hypertension Arte´rielle, University Hospital of Bordeaux, Hoˆpital Saint Andre´, Bordeaux, France and 2Service de Maladies Vasculaires, University Hospital of Bordeaux, Hoˆpital Saint Andre´, Bordeaux, France

We lack non-invasive tools for evaluating the coronary evaluated from an ambulatory measurement of the cor- and renal microcirculations. Since cutaneous Doppler rected QKD100–60 interval. We included 22 hypertensives laser exploration has evidenced impaired cutaneous micro- and 11 controls of mean age 60.6 vs 40.8 years. In this vascular responses in coronary artery disease and in population, there was a correlation between RI and basal impaired renal function, we wanted to find out if there zero to peak flow variation (BZ-PF) (r ¼À0.42; P ¼ 0.02) and was a link between the impairments in the cutaneous and a correlation between RI and rest flow to peak flow variation renal microcirculations. To specify the significance of the (RF-PF) (r ¼À0.44; P ¼ 0.01). There was also a significant rise in the renal resistive index (RI), which is still unclear, correlation between RI and the corrected QKD100–60 (r ¼ we also sought relations between RI and arterial stiffness. À0.47; P ¼ 0.01). The significant correlation between PORH We conducted a cross-sectional controlled study in a parameters and RI indicates that the functional modifica- heterogeneous population including hypertensive patients tions of the renal and cutaneous microcirculations tend to of various ages with or without a history of cardiovascular evolve in parallel during ageing or hypertension. The disease along with a healthy control group. The cutaneous relation between RI and arterial stiffness shows that RI is microcirculation was evaluated by laser Doppler flowmetry a compound index of both renal microvascular impairment of the post-occlusive reactive hyperhemy (PORH) and of and the deterioration of macrovascular mechanics. the hyperhemy to heat. The renal microcirculation was Journal of Human Hypertension (2012) 26, 56–63; evaluated by measurement of the RI. Arterial stiffness was doi:10.1038/jhh.2010.117; published online 20 January 2011

Keywords: microcirculation; skin blood flow; laser Doppler flowmetry; renal resistive index

Introduction renal resistive index (RI) calculated from pulsed Doppler flow at the interlobular arteries have been In daily clinical practice, the possibilities of evalu- proposed for evaluating the renal microcircula- ating the quality of the microcirculation of certain tion.1–3 However, we still lack simple tools for territories of ‘vascular’ patients are relatively limited evaluating the coronary and cerebral microcircu- compared with the many tools for evaluating ‘large lations. In this context, study of the cutaneous vessels’. However, even in the absence of significant microcirculation has been the subject of numerous impairment of the large vessels, a deterioration in investigations.4,5 Laser Doppler flowmetry (LDF) can the microcirculation might cause certain compli- determine alterations in the blood flow of the cations such as impaired renal function, blindness, capillaries over a small area of skin after applying neuropathy, myocardiopathy and mental dysfunc- various stimuli. This technique has evidenced tion. Examination of the ocular fundus, measure- deteriorations during ageing, diabetes and hyperten- ment of the transcutaneous oxygen pressure and sion.6–12 However, this technique will only be of capillaroscopy have been the main clinical tools for value if the cutaneous microcirculation can be used evaluating the microcirculation. More recently, the as a mirror of the vital territories (cerebral, renal and coronary), which are more difficult to access. Correspondence: Dr P Coulon, Service de Cardiologie et Hyper- Indeed, Holowatz et al.13 recently urged for study tension Arte´rielle, University Hospital of Bordeaux, Hoˆpital Saint of the relations between the pathological modifica- Andre´, 1 Rue Jean Burguet, Bordeaux 33075, France. tions of these different microcirculations. E-mail: [email protected] Received 2 September 2010; revised 23 November 2010; accepted To study the relations between the renal and 26 November 2010; published online 20 January 2011 cutaneous microcirculations, we carried out an PORH assessed by LDF correlates with RI P Coulon et al 57 exploration of the cutaneous microcirculation by (small angle thermostatic probe 457, PERIMED LDF followed by measurement of the RI of healthy FRANCE, Craponne, France) fixed on the skin subjects and hypertensive patients of different previously cleansed with acetone by a transparent ages. In parallel, the same subjects benefited from sticky strip, connected to a laser Doppler module measurement of arterial rigidity by analysis of (Periflux System 500, PERIMED FRANCE) con- corrected QKD100–60 interval, obtained from ambula- nected to a computer. This probe includes a heat tory measurement of blood pressure (ABPM). channel, which can heat an area of skin from a Indeed, it has recently been suggested that RI might thermal module (Peritemp 400S Heater, PERIMED be a marker of the cardiovascular risk in the hyper- FRANCE). The results are expressed in arbitrary tensive.14 We suggest that this could be accounted laser Doppler units (U) or in unit seconds (Us) for by its relations with both the micro- and the for the area under the curve of occlusion (AO) macrocirculations. and of post-occlusive hyperhemy (AH). The varia- tions of cutaneous blood flow during applications of the various vasoactive stimuli were recorded and Materials and methods then processed semi-automatically using Perisoft This was a pilot study conducted on a small sample software (PERIMED FRANCE). of patients and healthy controls. Fitting of device. The patients who had fasted for at least 4 h, lay on their backs for 15 min in an air- Studied population conditioned quiet room whose temperature ranged Taking into account the exploratory character of this from 21 to 24 1C. An arm-band was fixed on the right study, the population of hypertensives had different arm. The laser Doppler probe was placed on the ages in order to survey the range of normal to highly ventral side of the right forearm 10 cm from the wrist impaired microcirculation. We studied two quite crease, on the centre line away from hairy areas and distinct groups: hypertensive patients and healthy surface veins of the patient. controls. Recording. We performed a continuous recording Inclusion criteria for patients of laser Doppler flow during 12 min, starting with  Treated or untreated hypertensives with elevated the rest flow (RF) for 3 min then throughout the ABPM (24 h mean BP4130/80). application of the vasoactive stimuli.  Those with good quality ABPM-QKD recordings, namely 480% of the 96 measurements programmed Vasoactive stimuli. Post-occlusive reactive hyperh- over 24 h validated after manual elimination of emy (PORH): Inflation of the arm-band to a supra- aberrant measurements. Verification of good detec- systolic pressure, around 200 mm Hg for 3 min tion of the last Korotkoff sound corresponding to the provided a biological zero of the laser flow (BZ) as diastolic pressure by the microphone by analyzing well as the area under the curve of the occlusion the slope of the QKD/diastolic blood pressure (DBP) phase (AO). Then abrupt deflation enabling analysis relationship, which must be negative. of the various criteria of PORH: peak flow (PF), rise  Renal echo-Doppler of good quality defined by the time to the peak flow (TM), the time of half-rise to ability to formally exclude renal arterial stenosis peak (TH1), the time of half-return to the resting and a Doppler flow recording providing a reliable flow (TH2) and the area under the curve of the post- measurement of systolic and diastolic velocities occlusive hyperhemy zone (AH). at six different sites (three measurements per Maximum hyperhemy by heating the cutaneous kidney: upper, middle and lower segments). zone of interest to 44 1C: After the return to the resting flow following the PORH (individually Inclusion criteria for controls assessed but in general in 3 min following deflation of the arm-band) powering up the heating module  Good quality ABPM-QKD recording (same criteria as for the patients) and normal blood pressure connected to the probe bringing the area of skin under study to 44 1C, enabled recording of a plateau (24 h mean blood pressure o130/80). of hyperhemy, with a peak flow (PF 44 1C) and the  Absence of self-reported cardiovascular, renal or average over the third minute of heating (mean other chronic disease. Renal echo-Doppler of good 44 1C). quality (same criteria as for the patients). Exclusion criteria Data processing. The data were analyzed semi- automatically with the Perisoft software. The zones  Renal arterial stenosis. were positioned manually corresponding to the  Single kidney or other renal abnormality. resting flow, flow on occlusion of the humeral artery, the PORH and the third minute of heating Evaluation of the cutaneous microcirculation to 44 1C. The software calculated laser Doppler flows Materials. Recording of the cutaneous blood flow in arbitrary laser Doppler unit at rest (RF), at the of a small surface of skin with a laser Doppler probe time of occlusion (BZ), at the peak after deflating the

Journal of Human Hypertension PORH assessed by LDF correlates with RI P Coulon et al 58 arm-band (PF) and during the plateau of hyperhemy small sample size and a non-normal distribution, a to 44 1C (maximum flow: PF 44 1C and average of the non-parametric test (Spearman’s r) was employed plateau over the third minute of heating: mean for the correlation analysis. 44 1C). It also calculated the variation (%) between RF and PF and between BZ and PF, the rise time to the peak (TM) and times of half-rise (TH1) and Results half-fall (TH2) in seconds as well as the area under the curve of the zone of occlusion (AO) and of the Description and comparison of the two groups zone of post-occlusive reactive hyperhemy (AH) in The study included 22 patients and 11 controls, Units  second and AH/AO ratio. whose main characteristics are presented in Table 1. A cutaneous laser Doppler recording is illustrated in Measurement of RI. This was conducted immedi- Figures 1 and 2. The patients were older, had a ately after the cutaneous LDF with an iE33 echo- greater body weight, a higher SBP and a shorter graph (Philips, Amsterdam, The Netherlands) and a corrected QKD100–60. In all, 3 patients were diabetic S5–1 probe (Sector Array transducer 5–1 MHz, and 20 patients were receiving treatment with one Philips). Echo-Doppler of the kidneys and the renal or more antihypertensive drugs (ACEI: 3 patients, arteries excluded arterial stenosis or a single kidney. ARA2: 6 patients, calcium inhibitor: 10 patients, Pulse Doppler flows were recorded in the inter- diuretic: 8 patients and b-blocker: 3 patients) and 7 lobular arteries in the upper, middle and lower third had a history of cardiovascular disease (stroke, of the left and right kidneys determining maximum arteritis of lower limbs and coronaropathy). With systolic velocity (MSV) and end-diastolic velocity respect to the LDF data, there was no significant (EDV). RI was calculated from the formula ((MSV- difference between the two groups, that is, in RF or EDV)/MSV)) and the mean of six readings was used in AH and AO. On the other hand, BZ–PF variation for the statistical analysis. and RF–PF variation, as well as the AH/AO ratio were significantly lower in the patient groups (531 ABPM coupled with the measurement of QKD vs 867; P ¼ 0.001), (200 vs 385; P ¼ 0.004), (0.69 vs interval. This was carried out on the same day as 1.23; P ¼ 0.02), indicative of an impairment of their the other measurements or in the previous month. A PORH compared with the controls. TH1 was also Dyasis Integra (Novacor, Rueil-Malmaison, France) significantly longer in the patients (1.9 vs 1; instrument was used. The arm-band was placed on P ¼ 0.003). TM, TH2 and PF 44 1C did not differ the left arm. Blood pressure was measured every significantly between the two groups. 15 min over 24 h by an auscultatory method. Auto- matic coupling with ECG recording enabled for each Table 1 Principle characteristics of the two groups measurement calculation of the QKD interval, inter- val between the onset of the QRS to the last sound Patients Control P detected in diastole by the microphone positioned (n ¼ 22) (n ¼ 11) on the brachial artery. From the set of measurements the following values were calculated: mean systolic Age (years) 60.6±18 40.8±14 o0.05 blood pressure over 24 h (mean SBP), mean diastolic Gender (F) 45.5% 45.5% NS Weight (kg) 77.9±15 72.5±4 o0.05 blood pressure over 24 h (mean DBP), mean pulse Height (cm) 168.2±8 169.8±11 NS pressure over 24 h (mean PP), mean blood pressure Mean SBP (mm Hg) 133.8±20 111±8 o0.0001 over 24 h (mean MBP) and the mean QKD interval Mean DBP (mm Hg) 78.4±15 72.5±4 o0.05 over 24 h. The software also calculated the corrected Mean PP (mm Hg) 55.5±11 38.8±7 o0.0001 QKD100–60corr (ms) 182.8±12 204.7±15 o0.05 QKD100–60 automatically, the theoretical QKD interval RI 0.69±0.1 0.60±0.05 NS for a SBP of 100 mm Hg at a heart rate of 60 beats per RF (U) 17.2±7 12.6±5NS minutes, corrected by the duration of QRS complex BZ (U) 9.2±6 6.0±1.6 o0.05 according to formula QKD corr ¼ QKD -(QRS AO (Us) 1525±1099 1172±669 NS 100–60 100–60 ± ± duration (ms)-80). This index is a validated indicator PF (U) 53.0 20 56.9 33 NS RF–PF variation (%) 200±137 385±160 o0.05 of arterial stiffness, correlated with cardiovascular BZ–PF variation (%) 531±372 867±298 o0.05 risk in hypertensive patients. TH1 (s) 1.87±1.1 1.05±0.3 o0.05 TM (s) 9.04±3.5 8.77±2.5 NS TH2 (s) 18.97±5.9 18.75±4.0 NS Statistical analysis AH (Us) 736±843 1242±578 NS The statistical analyses were performed using SPSS AH/AO 0.69±0.8 1.23±0.6 o0.01 software version 17 (SPSS Inc, Chicago, IL, USA). Mean 44 1C (U) 128±82 124±50 NS 1 ± ± We carried out the descriptive analysis of the groups PF 44 C (U) 175 108 160 73 NS of patients and controls and then compared Abbreviations: AO, occlusion area under the curve; AH, hyperemia their characteristics by Mann–Whitney U-test. We area under the curve; BZ, basal zero; mean 44 1C, mean flow during also carried out univariate regression analyses to the third minute of heating; NS, non-significant; PF, peak flow; PF 1 identify relations between RI and the various laser 44 C, peak flow during heating; QKD100–60, QKD100–60 interval corrected by QRS duration; RI, renal resistive index; RF, rest flow; ; Doppler parameters and then between RI and the TH1, time to half hyperemia; TM, time to peak; TH2, time of half ABPM-QKD parameters. Taking into account the return to rest.

Journal of Human Hypertension PORH assessed by LDF correlates with RI P Coulon et al 59

140 TM 120 PF 100 80 TH1 TH2 PU 60 40 TR 20 RF TL B 0 00:02:00 00:04:00 00:06:00 00:08:00

Figure 1 Laser Doppler flow of a control at rest and during PORH. (a) Resting flow (RF). (b) Biological zero (BZ) during the occlusion. (c) PORH. A full colour version of this figure is available at the Journal of Human Hypertension journal online.

1 PU ΣΣ 200 Occlusion 175 150 125 100 75 50 25

00:01:00 00:02:00 00:03:00 00:04:00 00:05:00 00:06:00 00:07:00 00:08:00 00:09:00 00:10:00 00:11:00

Figure 2 Laser Doppler flow of a control at rest, occlusion, PORH, and then in heating to 44 1C (d). A full colour version of this figure is available at the Journal of Human Hypertension journal online.

0.80 0.80

0.70 0.70 RI RI

0.60 0.60

R2 Linéaire = 0.158 R2 Linéaire = 0.138 0.50 0.50

0 200 400 600 0 200 400 600 800 1000 1200 1400 RF-PF variation % BZ-PF variation % Figure 4 Relationship between RI and RF–PF variation Figure 3 Relationship between RI and BZ–PF variation (r ¼À0.44; P ¼ 0.01). (r ¼À0.42; P ¼ 0.02).

Relations between the renal and cutaneous RF–PF variation (r ¼À0.44; P ¼ 0.01) (Figure 4) and microcirculations between RI and TH1 (r ¼ 0.35; P ¼ 0.048). There was The univariate regression analyses carried out no correlation between RI and the AH/AO ratio between RI and LDF parameters evidenced a (r ¼À0.29; P ¼ 0.1), or with any of the other para- significant correlation between RI and BZ–PF varia- meters of the cutaneous laser Doppler examination. tion (r ¼À0.42; P ¼ 0.02) (Figure 3), between RI and In fact the parameters that were impaired in the

Journal of Human Hypertension PORH assessed by LDF correlates with RI P Coulon et al 60 vasodilation obtained by heating to 44 1C were R2 Linéaire = 0.198 comparable between the two groups. By contrast, the functional response to a 3-min ischaemia 0.80 (named as PORH) was significantly impaired in the patients, with a slower and less intense response (longer TH1, lower RF–PF variation, lower BZ–PF 0.70 variation and lower AH/AO ratio). This was in- dicative of a reduced functional response of the RI cutaneous microcirculation after ischaemia in the hypertensive patients. 0.60 In contrary to Carberry et al.,7 our results did not show a reduction in maximum laser Doppler flow during cutaneous heating in the hypertensives 0.50 compared with the controls (160 vs 175, P ¼ NS). This discordance can be accounted for by methodo-

125.00 150.00 175.00 200.00 225.00 250.00 logical differences. Indeed, cutaneous heating stems QKD100-60corr from the vasodilation from the inhibition of smooth arteriolar muscle tone via the action of nitric oxide, Figure 5 Relationship between RI and QKD100–60corr (r ¼À0.44; which generally appears in two successive peaks, P ¼ 0.01). the first during the first 3 min of heating, the second more intense but delayed by almost 30 min.15,16 In our study, we only recorded laser Doppler flow patients compared with controls were those whose during the first 3 min of heating to 44 1C whereas distribution was correlated with that of RI. There Carberry recorded this flow over 40 min at a was also a negative correlation between BZ–PF temperature of 42 1C. Serne´17–19 using videocapil- variation and age (r ¼À0.37; P ¼ 0.03) and between laroscopy showed that hypertensives have both RF–PF variation and age (r ¼À0.41; P ¼ 0.02). functional and structural capillary loss in the skin. The laser Doppler flow corresponding to the first peak of vasodilatation that we recorded may not Relations between RI and the QKD parameters correspond to the maximum vasodilation and may The univariate analysis detected a significant not be a good witness of capillary density. It is also negative correlation between RI and the corrected possible that our sample was too small to evidence a QKD100–60 (r ¼À0.44; P ¼ 0.01) (Figure 5) and be- difference between these two groups. tween RI and mean PP (r ¼ 0.51; P ¼ 0.003). There As in our study, Stewart20 failed to find any was no correlation between RI and mean MBP difference between a group of patients with chronic (r ¼À0.26; P ¼ 0.15). renal insufficiency and a control group, either for maximum flow and the area under the PORH curve, or for the flow at the first peak of vasodilation to Discussion heat. By contrast, he noted a significant decrease in Relation between the states of the renal and cutaneous maximum flow of the second peak in the renal microcirculations patients. Our recording was perhaps too short to We evidence here for the first time a significant observe the flow corresponding to the maximum correlation between the renal microcirculation eval- vasodilation. uated by the calculation of the RI and the cutaneous It should also be borne in mind that 20 out of our microcirculation evaluated by the post-occlusive 22 patients were receiving antihypertensive treat- hyperhemy reaction observed by LDF (correlation ment. It is possible that these treatments modify the between RI and BZ–PF (r ¼À0.42; P ¼ 0.02) and PORH and the hyperaemic reaction to heating, and it correlation between RI and RF–PF (r ¼À0.44; is known that antihypertensive agents such as ACEI, 21 P ¼ 0.01)). Although this result needs to be confirmed but not dihydropyridines can modify the RI. We in larger populations, it indicates that the microcircu- deliberately examined the patients without altering lations in these two territories evolve in steps. their treatment in order to carry out our micro- circulatory evaluation under normal conditions. To our knowledge there are no reports on the impact of Impairment of PORH but not hyperhemy at 44 1C in the the various classes of antihypertensive agents on the hypertensive cutaneous microcirculation. This study included a population of young healthy volunteers with no history of cardiovascular dis- ease, and a group of older hypertensive patients, Cutaneous microcirculation as a model of the total including three diabetics and seven with a history microcirculation? of cardiovascular disease. The resting laser Doppler Several authors have pointed out the interest of flow and those at the time of the maximum a simple and non-invasive evaluation of the

Journal of Human Hypertension PORH assessed by LDF correlates with RI P Coulon et al 61 cutaneous microcirculation which might mirror the observed impairments will need confirmation in renal microcirculation and even the cardiac and exploratory studies. cerebral ones.5,13 Nevertheless, the structural mod- ifications (arteriolar remodelling and capillary loss) may not well evolve similarly in territories subjected to quite different haemodynamic constraints. Our Place of RI in the evaluation of the vascular patient results cannot answer this question. On the other Our study also showed for the first time a significant hand, by evidencing a significant correlation be- correlation between RI and the arterial rigidity tween PORH and RI our results indicate that the evaluated by the corrected QKD100–60 (r ¼À0.44; functional modifications of the renal and cutaneous P ¼ 0.01). The corrected QKD100–60 is an indication microcirculations during ageing and hypertension of arterial stiffness, correlated with the cardiovas- may occur in steps. cular risk of hypertension, independently of the The cutaneous microcirculation is subjected to conventional cardiovascular risk factors.28 One of major physiological variations, essentially due to its its major advantages is that it is independent of role in thermoregulation. The study of the PORH the instantaneous value of the blood pressure in under conditions at rest several hours after ingestion contrast to the measurement of the pulse wave of food and at a strictly controlled temperature, and velocity.29 In a rabbit kidney perfused ex vivo, by referencing the maximum flow to the resting flow Tublin et al.30 noted a closer link between RI and or the zero biological one nevertheless seems a the index of pulse pressure (PP/SBP) than with renal simple and non-invasive way of evaluating impair- vascular resistances, indicating that central hemo- ment in the microcirculation.4,22 By showing the dynamic factors have an important role in the prognostic value of the measurement of the media/ alteration in RI. By contrast, showing a correlation lumen ratio of the small arteries of gluteal fat, between RI and histological study of the renal Rizzoni et al.23 demonstrated that study of the tissues31 and correlation between RI and renal structure of peripheral small arteries would be of vascular resistances in patients with renal insuffi- value for the assessment of cardiovascular risk in ciency3 several authors have showed the importance such patients. This study demonstrated that evalua- of impairments in the renal microvascular bed in tion of the microcirculation of peripheral territories modifications of RI. In a recent leading article, is of real interest for vascular evaluation of hyper- Hausberg et al.32 summarized the debate opposing tensive patients. Biopsy of gluteal fat appears rather those for which RI reflects renal vascular resistances impractical for large-scale studies designed to and the impairments of the renal vascular bed and stratify risk. By contrast, cutaneous laser Doppler those for which it is an indication of impairment of is a non-invasive tool, which could be used the arterial system in general. Our results showed a clinically. It would of interest to find out whether link between RI and arterial stiffness, as well as a the functional impairments we describe at the relationship between RI and the microcirculation. cutaneous level by laser Doppler have a similar These results put RI in its anatomical place, between value to that of the media/lumen ratio. That would the large vessels and the microcirculation. RI is the seem likely if the impairment in the cutaneous ratio of the difference between MSV and the EDV of microcirculation evolves in parallel with that in the blood in the interlobular arteries to MSV of the brain, heart and kidneys. same flow ((MSV-EDV)/MSV). The rate of travel of a Various teams have demonstrated concomitant fluid is related to the gradient of pressure between impairment of the cutaneous microcirculation and two sectors. This pressure gradient depends on myocardial perfusion. Sax et al.24 and Pedrinelli the hemodynamic power, which has a stable (MBP) et al.25 have both shown that patients presenting and a pulsatile component (PP), but also on the chest pains despite angiographically normal large peripheral resistance, which is affected by the coronary trunks (Syndrome X) have elevated vascu- functional and structural state of the renal micro- lar resistances compared with controls following circulation. RI is thus a compound index indicative ischaemia of the forearm. Jung et al.26 also showed a of both microvascular (arteriolar remodelling, en- deterioration of PORH in patients receiving heart dothelial dysfunction and capillary loss) and transplants. Shamim-Uzzaman et al.27 noted a macrovascular impairments (arterial stiffness and delayed and reduced PORH in patients with elevation in pulse pressure) occurring with age, coronary diseases compared with controls. Taken hypertension and diabetes. together, these findings are in favour of the link The existence, in hypertensive patients, of a between abnormalities of the cutaneous microcircu- significant correlation between RI and impairments lation and those (less easy to observe) of the of target organs (urinary albumin/creatinine ratio, coronary microcirculation. the index of left ventricular mass and intima media While noting the close links between the renal thickness14,33) points to RI being a marker of and cutaneous microcirculations, our study brings cardiovascular risk.33 Our results are consistent with an additional argument in favour of exploration this view by showing that RI is an integral of the cutaneous microcirculation in ‘vascular’ compound index of both micro- and macrovascular patients. However, the prognostic value of the impairments.

Journal of Human Hypertension PORH assessed by LDF correlates with RI P Coulon et al 62 10 Serne EH, de Jongh RT, Eringa EC, RG IJ, Stehouwer What is known about this topic CD. Microvascular dysfunction: a potential patho- K Resting and maximal forearm skin blood flows are reduced in hypertension.7 physiological role in the metabolic syndrome. Hyper- K Post-occlusive reaction hyperhemy (PORH) is impaired in tension 2007; 50: 204–211. CAD and chronic renal failure but the relation between the 11 Rossi M, Carpi A, Galetta F, Franzoni F, Santoro G. renal and cutaneous microcirculation states are The investigation of skin blood flowmotion: a new unknown.20,27 approach to study the microcirculatory impairment in K RI elevation significance is still under debate.32 vascular diseases? Biomed Pharmacother 2006; 60: 437–442. What this study adds 12 Larkin SW, Williams TJ. Evidence for sensory nerve K There is a relation between the renal microcirculation involvement in cutaneous reactive hyperemia in hu- evaluated by the RI and the cutaneous microcirculation function evaluated by the PORH indicating that the mans. Circ Res 1993; 73: 147–154. microcirculations in these territories may evolve in parallel. 13 Holowatz LA, Thompson-Torgerson CS, Kenney WL. K There is also a significant correlation between RI and The human cutaneous circulation as a model of

arterial stiffness evaluated by the corrected QKD100–60. generalized microvascular function. J Appl Physiol K These results suggest that 1/RI is a compound index 2008; 105: 370–372. indicative of both microvascular and macrovascular 14 Tedesco MA, Natale F, Mocerino R, Tassinario G, impairments; 2/cutaneous microcirculation could be a Calabro R. Renal resistive index and cardiovascular mirror of the ‘general microcirculation’, but PORH organ damage in a large population of hypertensive alteration prognostic significance in hypertensives must be patients. J Hum Hypertens 2007; 21: 291–296. studied. 15 Kellogg Jr DL, Liu Y, Kosiba IF, O’Donnell D. Role of nitric oxide in the vascular effects of local warming of the skin in humans. J Appl Physiol 1999; 86: 1185–1190. Conflict of interest 16 Minson CT, Berry LT, Joyner MJ. Nitric oxide and neurally mediated regulation of skin blood flow during The authors declare no conflict of interest. local heating. J Appl Physiol 2001; 91: 1619–1626. 17 Serne EH, Gans RO, ter Maaten JC, Tangelder GJ, Donker AJ, Stehouwer CD. Impaired skin capillary References recruitment in essential hypertension is caused by both functional and structural capillary rarefaction. 1 Norris CS, Barnes RW. Renal artery flow velocity Hypertension 2001; 38: 238–242. analysis: a sensitive measure of experimental and 18 Antonios TF, Singer DR, Markandu ND, Mortimer PS, clinical renovascular resistance. J Surg Res 1984; 36: MacGregor GA. Rarefaction of skin capillaries in 230–236. borderline essential hypertension suggests an early 2 Veglio F, Provera E, Pinna G, Frascisco M, Rabbia F, structural abnormality. Hypertension 1999; 34: 655–658. Melchio R et al. Renal resistive index after captopril 19 Antonios TF. Generalized microvascular disease test by echo-Doppler in essential hypertension. Am J in essential hypertension: evidence from studies Hypertens 1992; 5: 431–436. of cutaneous microcirculation. In: Levy, Struijker- 3 Petersen LJ, Petersen JR, Ladefoged SD, Mehlsen J, Boudier (eds). Role of Micro and Macrocirculation in Jensen HA. The pulsatility index and the resistive Target Organ Damage in Diabetes and Hypertension, index in renal arteries in patients with hypertension Chapter 5. Wiley-Blackwell: Chichester, UK, 2009, pp and chronic renal failure. Nephrol Dial Transplant 57–69. 1995; 10: 2060–2064. 20 Stewart J, Kohen A, Brouder D, Rahim F, Adler S, 4 Cracowski JL, Minson CT, Salvat-Melis M, Halliwill JR. Garrick R et al. Noninvasive interrogation of micro- Methodological issues in the assessment of skin vasculature for signs of endothelial dysfunction in microvascular endothelial function in humans. Trends patients with chronic renal failure. Am J Physiol Heart Pharmacol Sci 2006; 27: 503–508. Circ Physiol 2004; 287: H2687–H2696. 5 Levy BI. Commentary on viewpoint: the human 21 Leoncini G, Martinoli C, Viazzi F, Ravera M, Parodi D, cutaneous circulation as a model of generalized micro- Ratto E et al. Changes in renal resistive index and vascular function. J Appl Physiol 2008; 105: 380; urinary albumin excretion in hypertensive patients author reply 389. under long-term treatment with lisinopril or nifedi- 6 Rossi M, Carpi A, Di Maria C, Galetta F, Santoro G. pine GITS. Nephron 2002; 90: 169–173. Spectral analysis of laser Doppler skin blood flow 22 Kubli S, Waeber B, Dalle-Ave A, Feihl F. Reproduci- oscillations in human essential arterial hypertension. bility of laser Doppler imaging of skin blood flow as a Microvasc Res 2006; 72: 34–41. tool to assess endothelial function. J Cardiovasc 7 Carberry PA, Shepherd AM, Johnson JM. Resting and Pharmacol 2000; 36: 640–648. maximal forearm skin blood flows are reduced in 23 Rizzoni D, Porteri E, Boari GE, De Ciuceis C, Sleiman I, hypertension. Hypertension 1992; 20: 349–355. Muiesan ML et al. Prognostic significance of small- 8 Farkas K, Kolossvary E, Jarai Z, Nemcsik J, Farsang C. artery structure in hypertension. Circulation 2003; 108: Non-invasive assessment of microvascular endothelial 2230–2235. function by laser Doppler flowmetry in patients 24 Sax FL, Cannon III RO, Hanson C, Epstein SE. with essential hypertension. Atherosclerosis 2004; Impaired forearm vasodilator reserve in patients with 173: 97–102. microvascular angina. Evidence of a generalized dis- 9 Holowatz LA, Kenney WL. Local ascorbate adminis- order of vascular function? N Engl J Med 1987; 317: tration augments NO- and non-NO-dependent reflex 1366–1370. cutaneous vasodilation in hypertensive humans. Am 25 Pedrinelli R, Spessot M, Lorenzoni R, Marraccini P, J Physiol Heart Circ Physiol 2007; 293: H1090–H1096. L’Abbate A, Salvetti A et al. Forearm vasodilatory

Journal of Human Hypertension PORH assessed by LDF correlates with RI P Coulon et al 63 capacity in patients with syndrome X: a comparison time of Korotkoff sounds. Blood Press Monit 1996; 1: with normal and hypertensive subjects. J Hypertens 193–195. Suppl 1989; 7: S92–S93. 30 Tublin ME, Tessler FN, Murphy ME. Correlation 26 Jung F, Mrowietz C, Labarrere C, Schuler S, Park JW. between renal vascular resistance, pulse pressure, Primary cutaneous microangiopathy in heart recipi- and the resistive index in isolated perfused rabbit ents. Microvasc Res 2001; 62: 154–163. kidneys. Radiology 1999; 213: 258–264. 27 Shamim-Uzzaman QA, Pfenninger D, Kehrer C, 31 Platt JF, Ellis JH, Rubin JM, DiPietro MA, Sedman AB. Chakrabarti A, Kacirotti N, Rubenfire M et al. Altered Intrarenal arterial Doppler sonography in patients with cutaneous microvascular responses to reactive hyper- nonobstructive renal disease: correlation of resistive aemia in coronary artery disease: a comparative study index with biopsy findings. AJR Am J Roentgenol with conduit vessel responses. Clin Sci (Lond) 2002; 1990; 154: 1223–1227. 103: 267–273. 32 Hausberg M, Lang D, Barenbrock M, Kosch M. What do 28 Gosse P, Lasserre R, Minifie C, Lemetayer P. Prognostic Doppler indices of renal perfusion tell us for the evalua- value of QKD interval corrected by QRS duration in tion of renal disease? JHypertens2005; 23: 1795–1797. hypertensive patients. Arch Mal Coeur Vaiss 2005; 98: 33 Pontremoli R, Viazzi F, Martinoli C, Ravera M, Nicolella C, 747–750. Berruti V et al. Increased renal resistive index in patients 29 Gosse P, Braunstein C, Clementy J. Beyond blood with essential hypertension: a marker of target organ pressure measurements: monitoring of the appearance damage. Nephrol Dial Transplant 1999; 14: 360–365.

Journal of Human Hypertension