The effects of aging on the cutaneous microvasculature

The effects of aging on the cutaneous microvasculahxe Robert I. Kelly, MB, FACD: Rupert Pearse, BSc: Richard H. Bull, MB, MRCP, Jean-Luc Leveque, PhD, b, * Jean de RigaI, PhD,b and Peter S. Mortimer, MD, FRCF London, United Kingdom, and Aulnay-sous-Bois, France Background: Few studies have attempted to quantitatively assessin vivo changes in the microvasculature with age. Objective: The objective was to assessin vivo structural and functional changes in the cutaneous microvasculahue with aging and to analyze the contribution of the microvasculature to skin color. Methods: Video capillaroscopy, in conjunction with fluorescein angiography, and laserDoppler flowmetry were used to compare elderly and young normal volunteers. Skin color differences were assessedwith a handheld color reflectance meter. A photoexposed site, the forehead, and the relatively photoprotected ventral forearm were studied to differentiate photoaging from chronologic aging. Results: Dennal papillary loops were significantly reduced in old skin compared with young skin (forehead by 40%; forearm by 37%). Horizontal vessels showed increased volume fraction in elderly forehead and forearm skin. Laser-Doppler studies demonstrated no significant differences between young and old skin, indeed, hyperemic responsiveness appeared more rapid in the elderly. Color measurements showed elderly skin, particularly in men, to be significantly darker and redder. Conclusion: A marked loss in dermal nutritional vessel density and surface area for exchange is a feature of both chronologic aging and photoaging. (J AM ACAD DERMATOL 1995;33:749-56.)

With aging many skin components tend to diminish in size and number, and it has often been said that this change applies to the cutaneous microvascuIature.‘A However, few data support this belief. Previous assessments of microvasculature density have used two-dimensional frozen sections stained with alkaline phosphatase for histologic analysis. This method tends to overestimate the number of vessels. It has also been speculated that reduced cutaneous vascuIar responsiveness occurs with aging5 and that a reduced erythema response occurs after UVB exposure.6 The problem with such tests is the involvement of factors other than the vasculature in contributing to the response being measured. Other invesFrom the Depamnent a and Laboratories

of Dermatology, St. Georges Hospital, London, de Recherche de L’Oreal, Aulnay-sous-Bois.b

Supported

by L’Oreal,

Accepted

for publication

who also provided April

Reprint requests: P. S. Mortimer, Georges Hospital, Blackshaw Copyright 0190-9622/95

Chromameter.

MD, Department of Dermatology, St. Rd., London S.W.17 OQT, U.K.

0 1995 by the American $5.00 + 0

the Minolta

3, 1995.

16/l/65262

Academy

of Dermatology,

Inc.

tigators have assessed age-related differences in wheal resorption,7 disappearance of diffusable dye,8 clearance of radiolabeled material such as sodium 22,9 and blister development after topical application of 50% ammonium hydroxide1o and have generally found these to be delayed in the elderly. This delay may relate to reduced diffusion or transudation across vessel waIIs or reduced blood flow. However, changes in the extracellular matrix may also be relevant. It has been suggested that these changes can lead to diminished thermoregulation with vuhrerabihty to hypothermia, reduced capacity to clear antigens predisposing persons to airborne and photocontact dermatitis, and delayed development of inflammatory reactions potentiating the harmful effects of noxious agents.’ Others have suggested that changes in the cutaneous microvasculature may contribute to many of the physiologic deficits of aged skin2 Skin color is influenced greatly by intravascular hemoglobin, and it is often stated that increasing pallor with age is caused by a reduction in skin vasculature.’ Skin color can be measured accurately and 749

Journal of the American

750

Kelly et al.

Academy

of Dermatology November 1995

Fig. 2. Native capillaroscopy. A, Forehead of young Fig. 1. Fluorescein angiography. A, Forehead field of young subject shows prominent regularly arranged dermal papillary loops (white arrow) and horizontal vessels (clear arrow). B, Forehead field of elderly subject shows disorganized vascular pattern with marked tortuosity of horizontal vessels (dear arrow) and fewer dermal papillary loops (white arrow). (Bar, 0.1 mm.)

reproducibly with a handheld tristimulus colorime ter.11-16 Previous studies objectively assessing skin color cl-ranges with age have yielded conflicting results. Warren et al.17 found no difference in forehead skin color with age, whereas other studies’*, l9 have noted elderly exposed skin to be darker and redder. Intravital capillaroscopy is a noninvasive method of obtaining images of the microvasculature in vivo. The addition of fluorescein angiography improves contsast and permits better visualization of the vasculature.20-22 It has significant advantages over histologic methods because of better visualization of the three-dimensional vascular network in vivo. In this study in vivo capillaroscopy and laser Doppler flowmetry were used to assess struchtral and functional changes in the cutaneous microvasculature with aging. We have also used a handheld

subject shows many orderly arranged dermal papillary loops (arrow) and some horizontal vessels (arrowhead), which contrasts with B, forehead of elderly subject who demonstrates prominent, tortuous, horizontal vessels. Capillary loops are sparse. (Bar, 0.1 mm.) color-reflectance meter to assess skin color differences between young and old subjects and have analyzed the role that changes in the cutaneous microvasculature play in contributing to these color differences. A photoexposed site, the forehead, and the relatively photoprotected ventral aspect of the forearm were studied with each technique. METHODS Subjects

Two age groups were compared; each consisted of 13 subjects. The elderly group consisted of five men and eight women ‘whose ages ranged from 65 to 88 years (mean, 74.9 years), and the young group consisted of four men and nine women whose ages ranged from 18 to 26 years (mean, 23.2 years). Exclusion criteria included hypertension, diabetes, psoriasis, peripheral vascular disease, epilepsy, patients with an active dermatosis (in particular acne rosacea) in the areas being examined, and subjects taking vasoactive medications.

of the American Academy of Dermatology Volume 33, Number 5, Part 1

Journal

Table

Kelly et al.

751

I. Vascular density* Capillary

Forehead fluorescein Dots Lines Forearm fluorescein Dots Lines Forehead native Dots Lines Forearm native Dots Lines

density

(vessels

[No.]/mm2)

Young (mean 2 SEM)

Elderly (mean 2 SEM)

N=13 42.85 rt 3.24 42.23 2 2.39 N=5 27.20 ? 1.11 22.80 ? 2.99 N=7 25.14 k 3.10 25.57 2 3.10 N=6 24.67 2 0.99 25.00 2 2.34

N=13 25.46 k 2.09 40.46 i 2.18 N=5 17.2 -c 1.28 29.8 2 2.48 N=8 13.25 t 1.53 31.5 k 2.71 N=5 8.4 k 2.15 33.6 + 2.15

Dots, DermaJ papillary loops; lines, horizontal vessels; SEM, standard *Densities were determined on video prints of fluorescein angiographic

error of the mean. and native capillaroscopic

fields. p Values

p Value 0.0001 0.5892 0.0005 0.1230 0.0033 0.1291 0.0002 0.0260 were calculated

with the Student

t test. All subjects were white. The young group consisted of healthy medical students. The old group was randomly selected with respect to occupation and sun exposure. None of these subjects, however, had had any outdoor occupation. All were of English nationality, had lived most of their lives in Britain, and had only occasional holidays in sunny climates. Subjects in both age groups varied in skin type with skin types I, II, and III being equally represented. No subjects had skin type IV or above.

Subject

assessment

Each subject was assessedclinically regarding degree of skin pigmentation, erythema, and numbers of visible telangiectases. Elderly subjects were scored clinically regarding the degree of photoaging on the basis of the severity of solar elastosis, the number of solar lentigines, and the extent of file and coarse wrinkling. The data were recorded on linear analog scales,and subjects were scored along a 7 cm line with values ranging from 0 to 7 as a measure of the extent or severity of each feature. The distance in centimeters from the left end of the line to the scoring mark was the score. A score of 0 meant that the feature was not visible, whereas 7 was equivalent to the most severe change. Only one observer was used tbroughout the study for consistency.

Intravital

capillaroscopy

Equipment.” We used an Epi-illuminated microscope giving final magnification of 102 times on a television monitor linked to a video camera and a super-VHS video recorder, a video timer, and a video printer. For fluorescein angiography an excitation filter (450 to 500

nm) and a blue barrier filter (5 15 run) were inserted across the light beam. Microscopy. Examinations were performed in a temperature-controlled room (22.0” to 24.0” C). Two sites were examined: the forehead, 2 cm above the supraorbital ridge and 2 cm lateral to the midline (photoexposed), and the ventral forearm, approximately 5 cm below the elbow flexure (photoprotected). The site was marked before rnicroscopic visualization of the vasculature was done. The subjects lay on a couch for 30 minutes before the study to acclimatize to controlled conditions. The area examined was immobilized within a vacuum pillow and covered with liquid paraffin to reduce surface glare. Four fields around the mark were examined under native capillaroscopy, and after an intravenous bolus of sodiumfluorescein 20% solution was administered (0.3 and 0.2 ml/L blood volume for forearm and forehead, respectively), one field was studied with fluorescein angiography. Analysis. One fluorescein and four native fields were studied in each patient. The results of the native fields were averaged. Vascular density was measured by counting vessels on captured printout images of the recordings of examined fields. Vessels were counted as dermal papillary loops (“dots”) or as horizontal vessels (“lines”). These were counted within an area equivalent to 1 mm2 of skin, magnified by a factor of 102, and expressed as density of vessels per square millimeter. The “dots” therefore represent nutritional exchange vessels, whereas L‘hnes” represent postcapillary venules, ascending arterioles, and horizontally oriented capilhuies and are part of the subpapillary plexus. Morphometry was done as previously described. 23To assessthe heterogeneity of tissue perfusion, video recordings were analyzed to measure (1) the time for the fluorescein to appear in the skin vessels

Journal

752

Table

II. Volume fraction* Young (mean +. SEM)

Old (mean 2 SEM)

Forehead N=12 N=13 Dots 0.0791 lr 0.0067 0.0494 + 0.0037 Lines Forearm Dots

0.0571 t 0.0123 N=S 0.0481 +: 0.0036

Lines

0.1121 k 0.0124 N=S 0.0266 k 0.0029

p Value 0.0006 0.0021 0.0016

0.0223 -C0.0043 0.1013 -t 0.0069
Dermal papillary loops; lines, horizontal vessels;SEM, standard error of the mean. *Volume fraction was assessedby morphomeny. An acetatesheet with 941 equally spacedpoints equivalent to 1 mm’ of skin was placed over fluorescein angiography prints. Points falling on “dots” and “lines” were counted and then divided by 941 to give a fractional measurement. Dots,

after injection (Ti) and (2) the time taken from initial appearance of fluorescein to filling of 90% of the vessels (T90). Prints were examined visually with respect to tortuosity, dilatation, and regularity of the vessel pattern. These differences were assessedqualitatively.

Laser Blood

of the American

Academy

Kelly et al.

Doppler

flowmetry

flow (red blood cell flux) was measured

with a

laser Doppler flowmeter (Moor Instruments MBF 3D, Axminster, Devon, U.K.) with built-in printer and connected to a computer. One probe was attached to the forehead and another to the ventral forearm. Basal flow levels were recorded, after which time blood flow was occluded for 3 minutes. On the forehead this procedure was done by exerting pressure on a ring around the probe until a new steady reading was recorded and on the forearm by inflating a sphygrnomanometer cuff above the systolic pressure. After 3 minutes the pressure was removed, resulting in postischemic reactive hyperemia. This procedure was performed in 10 young (five male and five female) and 10 old (three male, seven female) subjects and repeated in seven others to assessthe reproducibility of the technique. Analyses. The following features were assessed:basal flow level as a measure of resting blood flow, the reactive hyperemic peak flow to assessthe degree of vascular responsiveness, the time from pressure release to reach the reactive hyperemic peak, and the time taken to return from the peak to a new baseline. These times were used mainly as measures of the rate of response. The area under the reactive hyperemic curve (from pressure release until return to baseline) was also calculated. Flow (red blood cell flux) was measured in arbitary units and time in seconds.

Calorimetric

assessment

Skin color was measured with a handheld reflected light calorimeter (Minolta Chromameter model CR-200,

of Dermatology November 1995

Osaka, Japan) in the L*a*b* system.“-‘6 The color isexpressed in three-dimensional space and is defined by its reflectance (L*) and its chromaticity (a* and b*). The L* value gives the level of brightness ranging from total black (L* = 0) to total white (L* = 100). The a* value represents the balance between red (positive values up to +lOO) and green (negative values up to -100). The b* value represents the balance between yellow (positive values up to + 100) and blue (negative values up to -100). Before use the calibration’ channel of the calorimeter was brought to standard white-plate level. Six readings were taken from the forehead and forearm on each subject. The results were averaged to give the final result. Ten young subjects (seven female, three male) and 11 elderly subjects (six female, five male) were studied.

Statistical

methods

All results comparing young skin with old skin were analyzed with the Student t test. Signikance level in each case was taken to be 0.05. Correlation between findings was assessedwith the Pearson product moment correlation coefficient.24

RESULTS Vascular morphologic

appearance

In the elderly a striking loss of regularly arranged capillary loops was seen, whereas horizontal vessels appeared tortuous, elongated, disorganized, and often grossly dilated (Figs. 1 and 2). Capillary

density

Fluorescein angiography. In the elderly forehead a 40% reduction was seen in the density of dermal papillary loops (“dots”) with no significant reduction in the horizontal vessels (“lines”). In the forearm a 37% reduction was seen in the dermal papillary loops in the elderly and no significant reduction in horizontal vessels. Native microscopy. In the forehead it was more difficult with native microscopy to obtain video prints of sufficient clarity that allowed accurate assessment of capillary density. For this reason, examinations of only seven young subjects and eight elderly subjects were analyzed. Dermal papillary loops were reduced by 47% in the elderly. No significant difference was seen in the density of the horizontal vessels visualized in the two groups. In the forearm a significant (66%) reduction was seen in

dermal papillary loops in the elderly. The density of visualized horizontal vessels was significantly increased in the elderly (Table I).

Journal of the American Academy of Dermatology Volume 33, Number 5, Part 1

Table III. Perfusion heterogeneity versus elderly subjects*

Kelly et al.

in young

Table IV. Laser Doppler flowmetry versus elderly subjects* Young (mean + SEM)

N=13

Time to initial filling (Ti) Forehead Forearm Time to 90% filling (T90) Forehead Forearm

N=13

19.19 k 0.66 31.53 2 2.30 N=5

22.02 L 0.95 29.77 2 2.68 N=5

0.0210 0.6330

2.64 ? 0.43 9.63 2 1.83

3.64 +- 0.50 12.73 +- 1.14

0.1400 0.1890

SEM, Standard error of the mean. *Ti is the time taken for fluorescein to first appear in the skin vessels after intravenous injection. T90 is the time taken from initial appexante of fluorescein to filling of 90% of the vessels. p Values were calculated with the Student t test.

Volume

fraction

The volume fraction of the dermal papillary loops was significantly reduced in both the forehead and forearm in elderly skin. Horizontal lines, in contrast, showed increased volume fraction in both the elderly forehead (p = 0.0021) and forearm 07 < 0.0001) (Table II).

Forehead Basal flow Peak flow Time to peak Time to normal Area under curve Forearm Basal flow Peak flow Time to peak Time to normal Area under curve

N=lO 66.8 2 361.00 + 49.11 ? 370.33 t

7.251 36.25t 8.88$ 37.20

753

in young

Elderly (mean -+ SEM)

N=lO 89.75 t 339.00 2 28.88 f 297.88 2

P

Value

18.97t 50.637 2.64$ 30.71$

0.25 0.71 0.056 0.16

198735 t 155248 83695 2 12543

0.41

N=lO 21.56 74.05 12.25 152.44

2 t !I 2

9.89-t 9.301 1.76$ 85.80$

4189 ? 434

N=lO 13.09 69.90 13.20 60.27

t-c -c ?

1.43t 6.69? 1.523 6.75$

4973 t- 504

0.71 0.36 0.67 0.25 0.26

“Red blood cell flux was measured at rest (basal flow) and after an ischemic period of 3 minutes the reactive hyperemic peak flow was measured. Time to peak represents the time taken from the end of the ischemic period to reach the postischemic peak. Time to normal is the time taken from the peak flow to the new baseline. Area under curve represents the total magnitude of the hyperemic flux response. p Values were calculated with the Student t test. tArbitrary units. $Seconds.

Tissue perfusion Time to appearance of fluorescein was significantly faster in the forehead skin of young subjects. No difference was noted in the time taken from the initial appearance of fluorescein in the vessels to filling of 90% of vessels in both forehead and forearm. Thus no difference was seen in the heterogeneity of tissue perfusion between the two age groups (Table III). Laser

Doppler

flowmetry

No significant difference was seen in any of the following features in old and young subjects in either forehead or forearm: basal flow levels, posthyperemic peak flow, time taken to return to a new baseline, and area under the posthyperemic curve. However, the time from pressure release to reach the posthyperemic peak was less in the elderly forehead (p = 0.056), w hi ch is consistent with more rapid vasodilatation. This difference was not found in the forearm (Table IV). Calorimetric

results

The elderly showed a significant decrease in the L* value (luminescence) on the forehead and fore-

arm and an increase in the a* value (redness) in the forehead and forearm that did not reach significance. No significant difference was seen in the b* value between the two groups in both forehead and forearm (Table V). Within the old group a significant reduction was seen in the L* value on the forehead of men compared with that of women and an increase in the a* value in the men compared with the women. No significant sex-related difference was seen in the young forehead with respect to L* value (female, 65.968 t 0.625; male, 65.347 + 1.037, p = 0.599) and a* value (female, 10.32 + 0.378; male, 10.757 -+ 0.184, p = 0.611) (Table VI). Clinical

evaluation

Clinical measurements were scored from 0 to 7. The young group had a mean erythema score of 4.33 (SEM k 0.47) and the elderly 4.59 (t 0.48). The mean telangiectasia score in the young was 0.92 (+ 0.28) and in the elderly 3.03 (2 0.55). For global photoaging in the elderly the mean score was 3.66 (+ 0.5).

Journal of the American

754

Academy

Kelly et al.

Table VI. Sex-related differences in skin color in elderly forehead skin

Table V. Calorimetric differences between young and elderly subjects* (mean

Forehead L” a* b* Forearm L”

a* b”

Young 2 SEM)P

N=lO 65.79 -t 10.45 + 13.64 t N=lO 68.44 L 5.71 k 11.63 ”

Elderly (mean t SEM)t

Female (mean t SEM)*

p Value

N=ll 0.51 0.36 0.63 0.69 0.32 0.66

61.82 5 12.99 i 13.42 5 N=Il 65.83 2 6.98 ?12.26 +

of Dermatology November 1995

N=6 0.93 1.15 0.67

0.0018 0.058 1 0.814

0.80 0.54 0.56

0.0246 0.0653 0.4728

*The color is defined by its reflectance (L*) and its chromaticity (a* and b*). The L* value gives the level of brightness. The a* value represents the balance between red (positive) and green (negative). The b* value represents the balance between yellow (positive) and blue (negative). p Values were calculated with the Student t test. tArbitmy units.

Correlations Significant statistical correlations were found between the forehead a* (erythema) value and clinical assessment of erythema (r = 0.73; p = 0.002), number of telangiectases as assessed clinically (Y = 0.76; p = O.OOl), the forehead laser Doppler basal flow (Y = 0.67; p = 0.0013), and the forehead laser Doppler posthyperemic peak (Y= 0.47; p = 0.0327). No significant correlation was found between vascular density and calorimetric values L* and a* or chronologic age or severity of photoaging within the elderly group. DISCUSSION This study is the first to quantify changes in cutaneous microvascular density with age and the first to document forehead skin capillary density in vivo. Previous studies have concluded, on the basis of qualitative rather than quantitative data, that aging is associated with a marked reduction in the density of the cutaneous microvasculature.1-3 It has also been suggested that impairment of skin vascular dilatory responsiveness to various stimuli occurs with aging.5, 6 Authors have speculated that these changes lead to a marked increase in skin pallor-l and to many of the physiologic deficits of elderly skin2 In this study vessels were counted as either dermal papillary loops (CLs) or horizontal vessels (HVs). Dermal papillary loops represent nutritional exchange vessels, whereas HVs consist mainly of postcapillary venules but also of horizontal capillaries and small arterioles. It is not possible to distin-

L” a* *Arbitrary

64.16 t- 0.55 10.30 + 0.75

Male (mean 2 SEM)*

p Value

N=5 59.01 t 0.84 16.15 ?I 1.37

0.001 0.004

units.

guish these different vessel types by their capillaroscopic appearance. Fluorescein angiography revealed a significant decline in CLs in the elderly. The reason for this finding is not known, but it occurs in association with flattening of the dermoepidermal junction. Although it is possible that true loss of capillaries occurs, some may atrophy, whereas others may reorient from vertical loops into horizontal capillaries. It appears that this process is to some extent reversible as indicated by the reappearance of CLs in diseases such as psoriasis and after treatment with topical retinoic acid, which restores the rete pegs and dermal papillae.25 Nevertheless it would seem that in the .elderly a significant loss of capillary surface area occurs for nutritional exchange, and this loss could have significant physiologic consequences. The forehead CLs density as determined by fluorescein angiography was less than that reported in a previous histologic study by Moretti. 26 This finding can be explained by the counting of only perfused vessels in capillaroscopy because only these vessels are visualized. However, there is also a tendency for overcounting of vessels on histologic evaluation because of twisting and turning of vessels that results in a single vascular lumen appearing more than once in a two-dimensional section. Shrinkage of histologic sections may also artefactually raise capillary density counts. The ventral aspect of the forearm was studied to compare a relatively sun-protected site with the photoexposed forehead. Fluorescein angiography also demonstrated a dramatic decline in CLs at this site, suggesting that this depletion occurs with both chronologic aging and photoaging. No significant decline in HVs was seen in either forearm or forehead in elderly skin. Native capillaroscopy proved less reliable than fluorescein angiography in evaluating differences in vessel density. Clear representative vascular fields were often difficult to obtain on the forehead under native capillaroscopy, particularly in young skin that

Joumal of the American Academy of Dermatology Volume 33, Number 5, Part 1

was less transparent. Fluorescein angiography, in comparison, provided clearer delineation of vessels in both young and old skin, thus providing more comparable results. Irregularities in skin pigmentation tend to obscure vessels under native microscopy. In addition, fluorescein angiography permits visualization of plasma-perfused vessels, whereas in native microscopy only vessels containing red blood cell columns can be seen. These factors contribute to the higher counts generally obtained with fluorescein angiography compared with those obtained with native microscopy. The morphologic appearance of the vascular bed was dramatically different in the old skin. Not only did a striking loss of regularly arranged papillary loops occur, but the HVs were often grossly dilated and appeared elongated, tortuous, and disorganized. Similar findings have been noted in the nail bed and bulbar conjunctiva with aging27, 28 and in the skinr” The apparent increase in horizontal vessels in elderly forearm skin by native capillaroscopy (but not fluorescein angiography) probably reflects the greater transparency of older skin because of atroPhY. We are not aware of any previous investigations that have assessed change in vascular volume with age. A significant reduction with age would be expected, and this reduction was found with respect to the volume of the dermal papillary loops, which was significantly reduced in both forehead and forearm in the elderly consistent with a marked loss of nutritive exchange vasculature. In contrast, a significant increase occurred in volume fraction occupied by the HV in both sites. In the forehead this increase occurred despite no difference in the IIV density between young and old skin as measured by fluorescein angiography and is thus consistent with dilatation of these vessels, which was a striking morphologic feature. Because of the difticultY in visualizing the horizontal vessels in the young under native microscopy, we considered that fluorescein angiography offered a valid comparison of the horizontal vessels in the two age groups. The heterogeneity of tissue perfusion was assessed by the time taken from the first appearance of dye within a vessel to the filling of 90% of the vessels. In certain conditions such as peripheral vascular disease this time can be prolonged.22 No difference was found, however, between young and old skin, suggesting that changes in the cutaneous vasculature did not alter the heterogeneity of tissue perfusion.

Kelly et al. 755 Laser Doppler flowmetric results were not significantly different in young and old skin, including basal flow levels and the posthyperemic peak flow levels. The laser Doppler signal penetrates to an estimated depth of more than 1 mm, thus giving information about deeper vessels and arteriovenous anastomoses,2g-31 which are not assessed by capillaroscopy. CLs probably contribute little or nothing to the laser Doppler flux signal, and it is thus not surprising that the CL depletion had no impact on red blood cell flux as assessed by laser Doppler flowmetry. The more rapid posthyperemic response in the elderly suggests that elderly vasculature is capable of a more rapid dilatory response compared with the young. This finding contrasts with earlier work that concluded that elderly vasculature shows delayed and diminished vasodilatory responses to various stimuli5, 6 However, these studies assessed differences in inflammation-induced erythema and thus added nonvascular variables. Calorimetric results showed elderly skin to be significantly darker (L* value) than young skin in both forehead and forearm and would thus appear to occur with both chronologic aging and photoaging. Epidermal atrophy and increased transparency may contribute to this, leading to increased visibility of the dermal components. An increase in redness (a* value) also occurred in the elderly subjects (although not reaching statistical significance), which can be explained by increased cutaneous transparency combined with HV dilatation despite CL loss that contributes little to cutaneous erythema.32 The elderly men had significantly redder skin than the women. These results suggest that rather than skin pallor as an inevitable feature of aging from vascular depletion as has been suggested previously,’ many persons, particularly men, become more plethoric. Colorimenic results correlated well with clinical assessments of facial erythema and with the number of visible telangiectases as judged clinically. The extent to which reduced CL density influences the clinical and physiologic changes of aged skin remains speculative. It nevertheless seems possible that such a dramatic loss of superficial nutritional exchange vessels may contribute to degenerative changes of aging skin including epidermal and appendageal atrophy. This loss may also contribute to delay in wheal resorption,7 disappearance of diffusable dye,8 clearance of radiolabeled material? and blister development after topical application of diffusable dyelo in the elderly. It could also lead to

Journal of the American

756

Kelly et al.

reduced capacity to clear antigens, predisposing persons to airborne and photoinduced contact dermatitis that is more prevalent in the elderly and to delayed development of inflammatory reactions. ’ This study, however, found little evidence on the basis of assessment of blood flow vasodilatory responsiveness to implicate cutaneous vascular aging changes as a major factor in such physiologic deficits as diminished thermoregulation, as has been suggested previously. ’

15.

16. 17.

We thank Mr. S. McCuigan for his expert advice on statistical methods and Professor J.R. Levick for his instructive comments.

18.

REFERENCES

19.

1. Kligman AM. Perspectives and problems in cutaneous gerontology. J Invest Dermatol 1979;73:39-46. 2. Gil&rest BA. Aging of skin. In: Fitzpatrick TB, Eisen AZ, Wolff K, et al, eds. Dermatology in medicine. 3rd ed. New York: McGraw-Hill, 1987:146-53. 3. Montagna W, Carlisle MS. Structural changes in aging human skin. J Invest Dermatol 1979;73:47-53. 4. Ellis RA. Aging of the human scalp. In: Montagna W, Ellis RA, eds. The biology of hair growth. New York: Academic Press, 1958469. 5. Grove GL, Lavker R, Hoezle E, et al. Use of non-intrusive tests to monitor age associated changes in human skin. J Sot Cosmet Chem 1981;32:15-20. 6. Gil&rest BA, Stoff JS, Soter NA. Chronologic aging alters the response to W-induced inflammation in human skin. J Invest Dermatol 1982;79: 1 l-5. 7. Aschner DM. IirtradermaJ salt solution test in elderly patients. Exp Med Surg 1960;18:17-20. 8. Heite HJ. Der Functionzusand der kleinen Gefaesse im Lichte gliechzeitig gemessener Quaddel- und fluorescein Resorptions geschwindigkeit. Arch Klin Exp Dermatol 1957;206:216-26. 9. Cbristophers E, Khgman AM. Percutaneous absorption in aged skin. In Montagna W, ed. Advances in biology of skin VI. New York Pergamon Press, 1964:163-75. 10. Grove GL, Duncan S, Kligman AM. Effect of ageing on the blistering of human skin with ammonium hydroxide. Br J Dermatol 1982;107:393-400. 11. Seitz JC, Whitmore CG. Measurement of erythema and tanning responses in human skin using a tri-stimulus colorimeter. Dermatologica 1988;177:70-5. 12. Westerhof W, Van Hasselt BAAM, Kammeijer A. Quantification of UV-induced erythema with a portable computer controlled chromameter. Photodermatology 1986;3: 310-4. 13. Deleixhe-Mauhin F, Krezinski JM, Rorive G, et al. Quantification of skin color in patients undergoing maintenance hemodialysis. J AM ACAD DERMATOL 1992;27:950-3. 14. Sernp J, Agner T. Calorimetric quantitication of erythema:

20. 21. 22. 23.

24. 25. 26. 27. 28. 29. 30.

31. 32.

Academy

of Dermatology November 1995

a comparison of two calorimeters (Lange MicroColor and Minolta Chroma Meter CR-ZOO) with a clinical scoring scheme and laser-Doppler flowmetry. Clin Exp Dermatol 1990;15:267-72. Wilhelm KP, Surber C, Maibach HI. Quantification of sodium lauryl sulfate irritant dermatitis in mar-a comparison of four techniques: skin reflectance, transepidermal loss, laser Doppler flow measurement, and visual scores. Arch Dermatol Res 1989;281:293-5. Weatberall IL, Coombs BD. Skin color measurements in terms of CIELAB color space values. J Invest Dermatol 1992;99:468-73. Warren R, Gartstein V, K&man AM, et al. Age, sunlight, and facial skin a histologic and quantitative study. J AM ACAD DERMATOL 1991;25:75 I-60. Adhoute H, de Rigal J, Marchand JP, et al. Influence of age and sun exposure on the biophysical properties of the human skin: an in vivo study. Photodermatol Photoimmunol Photomed 1992;9:99-103. Pauly G, Gillon V, Jollif JC, et al. Fundamental study of skin colour in women, variation with age, and cutaneous area. In Dunhill MS, ed. Cosmetic science in the 1990’s and beyond; vol 1 SCC. 1990~336-68. Luetolf 0, Bull RH, Bates DO, et al. Capillary underperfusion in chronic venous insufficiency: A cause for leg ulceration? Br J Dermatol 1993;128:249-54. Bull RH, Bates DO, Mortimer PS. Intravital video-capillaroscopy for the study of the microcirculation in psoriasis. Br J Dermatol 1992;126:436-45. Bollinger A, Fagrell B. Clinical capiharoscopy. Toronto: Hogrefe and Huber, 1990: l-62. Aheme WA, Dunill MS. Point counting and estimation of volume fraction. In: Dunhill MS, ed. Morphometry. London: Edward Arnold, Hodder, and StoughtonLtd 1988:3345. Bland JM, Ahman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10. Khgman AM, Dogadkina D, Lavker RM. Effects of topical tretinoin on non-sun-exposed protected skin of the elderly. J AM ACAD DERMATOL 1993;29:25-33. Morretti G, Ellis RA, Mascon H. Vascular patterns in the skin of the face. J Invest Dermatol 1959;31:103-12. Tassi G, Maggi G, DeNicola P. Microcirculation in the elderly. Int Angiol 1985;4:275-83. Merlen JF. Capillaroscopy at the nail bed in functioning people aged 70 and over. Int Angiol 1985;4:285-8. Engelhart M, Kristensen JK. Evaluation of cutaneous blood flow responses by 133xenon washout and laser-doppler flowmeter. J Invest Dermatol 1983;80:12-5. Hales JRS, Roberts RGD, Westerman RA, et al. Evidence for skin microvascular compartmentalization by laser doppler and photoplethysmographic techniques. Int J Microcirc Clin Exp 1992;12:99-103. Braverman IM, Keh A, Goldminz D. Correlation of laser Doppler wave patterns with underlying microvascular anatomy. J Invest Dermatol 1990;95:283-6. Ryan TJ. The microcirculation of the skin in old age. Geronto1 Clin 1966;8:327-37.