DOI BIOPHOTONICS - MyBiozoommybiozoom.com/.../2016/04/Journal-of-Biophotonics_2013_Schulstudi… · BIOPHOTONICS The cutaneous ... abstain from smoking, ... bility of the measured

Galley P

roofs

FULL ARTICLE

Spectroscopic biofeedback on cutaneous carotenoidsas part of a prevention program could be effectiveto raise health awareness in adolescents

Ruo-Xi Yu*; 1, Wolfgang Kocher2, Maxim E. Darvin1, Monika Buttner3, Sora Jung1,Bich Na Lee1, Christoph Klotter4, Klaus Hurrelmann5, Martina C. Meinke1,and Jurgen Lademann1

1 Charite – Universitatsmedizin Berlin, Department of Dermatology, Venerology and Allergology, Center of Experimentaland Cutaneous Physiology (CCP), Chariteplatz 1, 10117 Berlin, Germany

2 Opsolution Spectroscopic Systems GmbH, Goethestr. 25–27, 34119 Kassel, Germany3 Elisabeth-Knipping-Schule, Mombachstraße 14, 34127 Kassel, Germany4 Fulda University of Applied Science, Fulda, Postfach 22 54, 36012 Fulda, Germany5 Hertie School of Governance, Friedrichstraße 180, 10117 Berlin, Germany

Received 19 August 2012, revised 5 October 2012, accepted 11 October 2013Published online zzz zzz 2013

Key words: antioxidants, biofeedback, carotenoids, free radicals, skin, dietetics, stress, adolescent, spectrum analysis

# 2013 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758

Journal of

BIOPHOTONICS

The cutaneous carotenoid concentration correlates withthe overall antioxidant status of a person and can beseen as biomarker for nutrition and lifestyle. 50 highschool students were spectroscopically measured fortheir cutaneous carotenoid concentrations initially in astatic phase, followed by an intervention phase with bio-feedback of their measured values, living a healthylifestyle and on healthy food this time. The volunteersshowed higher carotenoid concentrations than found inprevious studies. A significant correlation of healthy life-style habits and a high antioxidant status could be deter-mined. Subjects improved their nutritional habits andsignificantly increased their carotenoid concentrationduring intervention. Follow-up five months later showeda consolidation of the increase. The investigations showthat a healthy diet and a well-balanced lifestyle correlatewith a high cutaneous antioxidant concentration and thatspectroscopic biofeedback measurement of cutaneouscarotenoids as part of an integrated prevention programis a feasible and effective means to raise the healthawareness in adolescents.

Distribution of measured values in phase 1 and phase 2.Average value in phase 1: 5.73� 1.37; average value inphase 2: 6.45 � 1.12. a.u.: arbitrary units. The continuousline represents the approximated Gaussian bell curve.

* Corresponding author: e-mail: [email protected], Phone: +49 30 450 518 235, Fax: +49 30 450 518 918

213

30

95

Bio

Ph

oto

nic

sH

eft

11-1

2/2

01

3S

atz

pro

gra

mm

:3

B2

Art

.:D

OI0

01

34

/Yu

u:/p

_1

/jb

p/_

Ste

hsa

tz/0

01

34

/jb

p0

01

34

u.3

d

(Bu

nd

ste

g:2

2m

m,A

uß

en

ste

g:1

8.7

4m

m)

insg

esa

mt

12

Se

ite

n((

Be

arb

eit

un

gin

VM

wa

reU

niv

ers

alC

S3

))2

9.1

0.2

01

3B

ea

rb.:

Sch

.

Early View publication onwww.wileyonlinelibrary.com(issue and page numbers not yet assigned;citable using Digital Object Identifier – DOI)

J. Biophotonics 1–12 (2013) / DOI 10.1002/jbio.201200134

Galley P

roofs

1. Introduction

Reactive oxygen species (ROS) and other free radi-cals are well known to be a general metabolic by-product in all aerobic organisms. In human beings,they are mainly produced in the mitochondrial re-spiratory chain [1]. At low concentrations they playan important role in physiological signaling pathways[2] and disinfection [3]. Various internal and externalstress factors, e.g., inflammation [4], UV-irradiation[5, 6], extensive exercise [7–9] or alcohol consump-tion [10] can result in an excessive production ofhigh amounts of free radicals and consequentlycause oxidative stress which impairs the antioxidantdefense system and damages all cell components in-cluding lipids, proteins and DNA [11–13].

Antioxidants are enzymes and non-enzymaticsubstances including several vitamins, carotenoidsand flavonoids [14–16]. They form a defense systemagainst the development of oxidative stress by neu-tralizing free radicals before those can damage cellstructures [11, 17–19]. In the human skin, carote-noids are the biggest group of antioxidant substances[20]. Lycopene and beta-carotene represent the high-est percentage of cutaneous carotenoids and it hasbeen shown that lycopene has the highest antioxi-dant activity among all carotenoids [21].

It has been demonstrated that carotenoids canserve as marker substances for the general antioxi-dant status of human skin [22]. The various antioxi-dant substances form protection chains in the tissue,i.e., they are protecting each other against the de-structive action of free radicals [23, 24]. The cuta-neous carotenoid concentration also correlates withits concentration in the serum [25] and therefore canbe seen as a biomarker for the general antioxidantcapacity of the organism [22, 26]. It may be consid-ered as a looking glass for the health-related lifestyleand nutrition of a subject [27].

Since humans cannot produce carotenoids andother important antioxidants themselves, an antioxi-dant-rich and balanced nutrition plays an importantrole in their supply [28]. Positive effects on healthand beauty have been shown for various foodstuffsrich in antioxidants [29, 30].

At the same time, stress increases the ROS gen-eration and, subsequently, reduces the carotenoidconcentration in the skin. Thus, a person’s cutaneouscarotenoid concentration can be seen as marker forhis nutritional habits.

Recent studies suggest that a high concentrationof cutaneous carotenoids correlates with better skinappearance [31–33] and that decreased carotenoidconcentration may predispose to the development ofskin cancer [34, 35]. One study investigating volun-teers between 40 and 50 years found out that thosesubjects whose skin ageing values were highest ex-hibited low carotenoid concentrations [31]. If sub-

jects at the age of 50 decide to improve their life-style, this will influence their appearance over thenext decade, but even the healthiest diet will not re-juvenate their skin. Therefore, adolescents, who havethe potential to influence their future health by to-day’s behavior, are to be convinced by biofeedbackmeasurements to improve their nutritional behaviorand reduce their stressors.

Teenagers show a weaker orientation to the fu-ture than adults and are more focused on instant re-wards [36]. They often show a high-risk behavior to-wards risk factors for well-known diseases affectingtheir future health and economic welfare [37, 38].

Biofeedback in many forms has been approvedin studies and also in clinical routine as a useful in-strument to bring invisible physiological parametersto view and, consequently, influence a subject’s be-havior [39]. In a recent study on the antioxidantstatus of adults, a non-invasive biofeedback on cuta-neous carotenoid concentrations led to an enhance-ment of the volunteers’ health-related lifestyle andnutrition resulting in higher cutaneous carotenoidconcentrations in many of them [40]. These measure-ments became only possible by the use of non-inva-sive spectroscopic methods like resonance Ramanspectroscopy (RRS) and reflectance spectroscopy.

The question therefore arises, whether this bio-feedback effect on cutaneous carotenoids can befound in adolescents and whether the instant feed-back on their antioxidant status can serve as a meansfor behavioral health prevention in this age group.

2. Material and methods

2.1 Volunteers

The study was performed on 50 subjects (24 malesand 26 females) aged between 17 and 20 years. Theywere all students of the 11th and 12th grade of theElisabeth-Knipping School, a vocational grammarschool in Kassel, Germany. They were students fromfive different classes, of which four had dietetics andone bioengineering as a special field. All volunteershad a healthy skin type II according to Fitzpatrickclassification [41].

2.2 Study design

Prior to the measurements, the subjects were in-formed about the aims and procedure of the studyand they were asked to complete questionnairesabout their lifestyle and nutritional habits. Prior to

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758

R.-X. Yu et al.: Biofeedback on cutaneous carotenoids and lifestyle enhancement2

Journal of

BIOPHOTONICS

# 2013 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.biophotonics-journal.org

Galley P

roofs

the second phase, specific information about the im-portance of antioxidants and the influence of dietand stressful lifestyle factors on their level was giveto the students in the form of a lecture by Prof. La-demann (Dept. of Dermatology) and Prof. Klotter(Dept. of Nutritional Science).

The actual study took place in three phases. In afirst phase from February 14th to March 18th, 2011,the current state on the antioxidant status and thestudents’ lifestyle was ascertained. Prior to the sec-ond phase from March 21st to April 15th, 2011, thestudents were informed once more about the influ-ence of protective and stress factors on health ingeneral and especially on the antioxidant status. Thethird phase was performed in September 2011,where the students were instructed to work on ahealthy lifestyle and live on healthy food during theintervention phase. Moreover, they were asked toabstain from smoking, alcohol consumption and toavoid stress as far as possible. In addition, the stu-dents received a free lunch meal containing antioxi-dant-rich fruit and vegetables on every school dayduring the intervention phase.

The measurements were carried out during thestudents’ lunch breaks twice a week. Before eachmeasurement, the volunteers filled in a question-naire about their diets, lifestyle habits and possiblestress factors such as smoking, sunbathing, illness orpsychological stress over the last two days. Afterwashing their hands, the volunteers measured thecarotenoid concentration with the biozoom� spectro-metric device [42] on the thenar eminence of bothhands. The palm was chosen as measuring locationbecause of its accessibility, the high concentration ofcarotenoid antioxidants, the absence of strong differ-ences in the skin color of the ethnic groups [34, 43,44] and the pre-existence of reference concentrationsbased on former measurements on the palm both byour group [42, 45] and by other groups using theRRS [27]. The measuring area was restricted to thethenar eminence in order to improve the reproduci-

bility of the measured results. Measurements werecarried out on both hands in order to avoid a biascaused by concentration differences between leftand right, which had occurred in earlier measure-ments. The measured values of both hands were ar-ithmetically averaged, rounded in accordance withstandard commercial practice and then displayed ina scale from 1 to 12.

Ideally, there were eight measurements per per-son and phase. In 10 subjects, an additional ninthmeasurement took place at the end of phase 2 (Day2.Z), because their first measurements of the secondphase accidently started even before the first lunchmeal was offered. Unexpected for the students, therewas a third measurement phase, the follow-up, inSeptember 2011 in order to review the sustainabilityof the effects achieved in phase 2. Measurementshad been carried out by the students themselves asthe device was easy to handle.

The subjects were informed about their own re-sults not before the beginning of the interventionphase in order to prevent a premature biofeedback-effect. In phases 2 and 3, the volunteers were imme-diately informed about their values after each meas-urement.

Written consent was obtained from the studentsand their parents prior to the study. All procedureswere in accordance with the Helsinki Declaration of1975 as revised in 2008. A detailed description ofthe study progress can be seen in Figure 1.

2.3 Measuring method

Major problems with previous measuring methodsfor the determination of the antioxidant concentra-tion in humans are the invasiveness of these meth-ods, e.g., the high performance liquid chromatogra-phy for which blood or skin samples need to betaken [46, 47], the immobility of spectroscopic de-

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758

Figure 1 Study design

J. Biophotonics (2013) 3

FULLFULLARTICLEARTICLE

# 2013 by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.biophotonics-journal.org

Galley P

roofs

vices like the RRS and the low sensitivity of othermeasurement methods [48]. Moreover, the classicalanalytical methods are cost-intensive and time con-suming. For a study under in vivo conditions in aschool environment with adolescents, a portable,non-invasive and fast device for determining the ca-rotenoid concentration in human skin was needed.In the present study, we used the newly developedbiozoom� spectroscopic system (Opsolution GmbH,Kassel, Germany) [42]. The working principle of thebiozoom� device is based on reflectance spectro-scopy using a light emitting diode (LED) as a sourceof irradiation. The LED emits a blue light spectrumranging between 440 and 490 nm, which correspondsto the absorption maximum of carotenoids. The va-lue of the small dip in the diffusely reflected spec-trum measured in the range between 458 and472 nm is associated with the concentration of caro-tenoids in the skin.

The measured values were presented in an ordi-nal scale ranging from 1 to 12, with 12 correspondingto 1.2 nmol g�1 skin, which is the highest concentra-tion measured with the RRS device, which had beendeveloped by the Center of Experimental & AppliedCutaneous Physiology of the Charite – Universitats-medizin Berlin, Germany for the quantitative deter-mination of the carotenoid concentration in humanskin [49]. With that system, more than 10,000 volun-teers and patients, respectively, had been investi-gated over a period of ten years. The biozoom� systemwas calibrated using the RRS, and the comparison ofthe results measured with the biozoom� device andthe RRS on the same volunteers showed a good linearcorrelation (Pearson’s R ¼ 0:88) in earlier measure-ments as previously described in detail by our group[42].

The measurement with the prototype used in thestudy takes about 1.5 minutes for both hands. The fi-nal product is planned to require only a few seconds.In order to display the measured result, an addi-tional output device with Bluetooth� technology(smartphone, tablet or PC) is necessary. In the pre-sent study, a laptop with a prototype output programwas used. The final retail price of the measuringdevice is not yet decided, but it is designed to beaffordable for private consumers. For schools, thedeveloper plans special rates or sponsorships. InFigure 2, the use of the measuring system is demon-strated.

2.4 Statistical analysis

IBM SPSS� Statistics 19.0.0 (SPSS Inc., Chicago, IL)and Microsoft Excel 2008 for Macintosh were usedfor statistical analysis and generation of diagrams.The Mann-Whitney test was used for statistical ana-lysis of independent samples and the Wilcoxon testfor paired samples. For correlation analysis, Pear-son’s R and R-squared were calculated. A score rang-ing from 0 to 16 was conceived in order to evaluatethe influence of multiple lifestyle habits such as nu-tritional or smoking habits and illness on the meas-ured values. Higher scores signify better lifestylehabits and health (Table 1).

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758

Figure 2 Measuring procedure with the biozoom� device.The measurement is push-button activated and the wire-less device sends the results to a computer via Bluetooth�.

Table 1 Health related lifestyle score1

Score: 0 1 2

Breakfast before school2 �1 2–3 �4Cooking at home2 �1 2–4 �5Fruit consumption3 �1 2–4 �5Vegetables consumption3 �1 2–4 �5Soft drink consumption3 �5 2–4 �1Sweets consumption3 �5 2–4 �1Smoking habits daily occasional neverDays of illness4 >50% 15–50% <15%

1 Calculated through answers given to questions on life-style and nutritional habits. The score ranges from 0 to 16.2 In days per week. Students were asked how many daysper week they eat breakfast before school or cook athome.3 In portions per day. Students were asked how many reg-ular portions of different foods they consume on a normalday.4 The percentage of attended measuring days on whichthe volunteers have indicated illness on the measuring dayor up to 3 days before.


Journal of

BIOPHOTONICS


Galley P

roofs

3 Results

3.1 Compliance and attendance

In general, the volunteers were highly motivatedand all of the participating students took part on allmeasurements when they were at school. If studentswere absent from school on a measuring day due toillness or other reasons, values were not ascertainedfor that day. In phase 1, an average attendance of88% was reached. Due to illness of many students,attendance was lower in the second half of phase 2and phase 3. The questionnaires accompanying themeasurements were mostly duly completed. Ques-tions about alcohol consumption, however, were of-ten not answered completely.

In the follow up five months after the end of theintervention, two students had already left schooland did not take part in these measurements.

During the first phase, the students were in-structed to continue with their regular lifestyle ha-bits. However, 14 of them reported that they livedhealthier than normal during phase 1, as the studysetting had raised their health consciousness. 36 ofthe volunteers claimed to have actively changedtheir lifestyle habits in phase 2 beyond the lunchmeal at school. All volunteers took part at theschool lunch offered in phase 2, nine of them daily,the rest missed it at least once.

3.2 Measured results in the steady phase

In the first phase, the median carotenoid concentra-tion of each subject ranged between 3 and 9. The

average of the median values of the volunteers was5.73� 1.37, which is significantly higher than theaverage value of 5.36� 1.81 determined by Meinkeet al. with the RRS in 151 healthy volunteers agedbetween 18 and 75 years in Berlin, Germany [45].

The measured values were steady over the wholephase and for every measuring day the median valueof all subjects together was 6. There were no signifi-cant differences based on age or gender, nor be-tween students with dietetics and those with bioengi-neering as special field.

23 of the volunteers smoked at least occasionallyduring phase 1, ten of them on a daily basis. 35 ofthem drank alcohol on one or more days duringphase 1. The distribution of the measured values isshown in Figure 3a, b.

3.3 Influencing factors on the carotenoidconcentration

The measured carotenoid concentrations correlatedwith the volunteers’ lifestyles. Several well-knownhealth-related lifestyle factors were assessed throughquestionnaires; and it became apparent that high caro-tenoid concentrations are associated with better nutri-tion, non-smoking and good health. However, the in-dividual influence factors on their own, e.g., theincidence of illness (Pearson’s R ¼ �0:178; p ¼ 0:23)or consumption of fruits (Pearson’s R ¼ 0:273;p ¼ 0:06) do not correlate significantly with the mea-sured values, even though a clear tendency was found.

However, we assumed that different factors havea synergetic effect on the antioxidant status as theyhave on health in general and therefore created ascore that takes multiple influencing factors into ac-

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758 (a) (b)

Figure 3 (a), (b) Distribution ofmeasured values in phase 1 andphase 2. Average value in phase 1:5.73� 1.37; average value in phase2: 6.45 � 1.12. a.u.: arbitrary units.The continuous line represents theapproximated Gaussian bell curve.




Galley P

roofscount (Table 1). The median of the measured valuescorrelates significantly with this health-related life-style score (Pearson’s R ¼ 0:303; p ¼ 0:03) showingthat the combination of several health-promotinglifestyle habits and circumstances, namely non-smok-ing, healthy nutritional habits and the absence of ill-ness has a positive effect on cutaneous antioxidantconcentrations (Figure 4), which confirms earlierfindings [40, 50].

Moreover, an association of high carotenoid con-centration and good psychosocial status, which wasascertained through questionnaires, could be found.Students with high carotenoid concentrationsshowed higher satisfaction and better performanceat school, claiming more often that they had enoughtime for themselves and to spend with hobbies andfriends. The subjective view on the current overallquality of life correlates significantly with the mea-sured antioxidant concentration. On a scale from 1to 10, 1 being the worst imaginable and 10 the bestimaginable life, subjects indicated where they seethemselves at the moment (Figure 5). A higher score

is significantly associated with a higher carotenoidconcentration (Pearson’s R ¼ 0:328; p ¼ 0:02).

There was no correlation between BMI andmeasured concentrations. According to findings ofMeinke et al. [45], a BMI of over 30 is correlatedwith a significantly lower carotenoid concentration.However, only one subject in the study at hand hada BMI of over 30.

3.3 Measured results in the interventionphase

In the second phase, seven subjects had a lowermedian measured value than in phase 1 and twelvesubjects kept their median value. 30 volunteers in-creased their median value in phase 2, eight of themby two or more points. The average of the medianmeasured value of phase 2 was 6.45 � 1.12 thus ex-ceeding the median measured in phase 1 by 0.71.The measured values increased during the secondphase (days 2.1–2.Z in Figure 8 below). The averageof the median values for the second half of phase 2,i.e. day 2.5 to 2.Z amounts to 6.67 � 1.03, which iseven 0.94 points higher than in phase 1 (Days 1.1–1.8). The development of the measured values of allsubjects together throughout phases 1 and 2 can be

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758

Figure 4 Correlation of the health-related lifestyle scorewith the measured values in phase 1. The score takes intoaccount smoking, nutritional habits and the incidence ofillness. Pearson’s R ¼ 0:303; P ¼ 0:03; a.u.: arbitrary units.* This subject claimed that she already improved her nutri-tion and lifestyle habits during the first phase of the studyas she ate less meat than usual and took food supplementsin the period of the first two measurements. Those changeswere not taken into account in the health-related lifestylescore since the score is calculated on the basis of the infor-mation given before the measurements took place.

Figure 5 Correlation of the overall current view on thequality of life and the measured carotenoid concentrationin phase 1. The volunteers were asked to answer the ques-tion how content they are on a scale from 0 to 10 with0 points being the worst and 10 points the best imaginablelife possible. Pearson’s R ¼ 0:328; p ¼ 0:02; a.u.: arbitraryunits.


Journal of

BIOPHOTONICS


Galley P

roofsseen in Figure 6. The approximately bell-shaped dis-tribution of measured values in phases 1 and 2 isshown in Figures 3a, b.

In general, subjects with a low baseline in phase 1achieved the most significant increase in their valuesin phase 2 (Figure 7). However, the ranking in rela-tion to the median value was not significantly chan-ged. The extent to which the median measured con-centration changed was not dependent on age orgender.

3.4 Changes in behavior and circumstancesin the intervention phase

Apart from higher carotenoid concentrations, thereare more changes in the second phase towards thebaseline. More than half of the subjects reported infree text questionnaires that they had felt healthierin the second phase. One female volunteer reportedthat with the healthier lifestyle in phase 2, her skinappearance improved.

According to their answers in the questionnaires,36 of the volunteers tried actively to enhance theirlifestyle in the intervention phase beyond the free

lunch meal at school. In general, the subjects con-sumed more fruit and vegetables in the second phasecompared to the first.

5 of the 23 smokers from phase 1 did notsmoke in phase 2, which means that 22% of thesmokers from phase 1 quit smoking at least for theperiod of the intervention phase. Also the numberof cigarettes smoked was lower in phase 2. Priorto the intervention, 10 of the students smoked ona daily basis. Two of them reduced their cigaretteconsumption radically in phase 2, their medianmeasured values increased by 1.5 and 2 points, re-spectively, which is significantly higher than theaverage increase of 0.71. The volunteers also re-duced their alcohol consumption. 27 subjects drankalcohol on one or more days in phase 2 comparedto 35 subjects in phase 1, also the amount of alco-hol was lower in phase 2.

In phase 2, the volunteers wrote more schoolexams. 24 subjects stated that they had a signifi-cantly higher level of stress due to exam prepara-tions in phase 2. On average, the subjects wrotetwo exams per week in the first and five examsper week in the second phase. Another uncontrol-lable change in phase 2 was the higher incidenceof illness in the students. Teachers reported thatmany subjects suffered from a flu-like infection inphase 2.

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758

Figure 6 Box plots of the measured values of all subjectsover the course of phase 1 (1.1–1.8) and phase 2 (2.1–2.Z). Phase 1 is displayed in red, phase 2 in green. 2.Z:additional measuring day for ten subjects, whose firstmeasurements of the second phase started even before thefirst lunch meal was offered. Average value phase 1:5.73� 1.37; average value phase 2: 6.45 � 1.12; 1 a.u.: arbi-trary units.

Figure 7 Correlation of the concentration increase fromphase 1 to phase 2 with the measured values in phase 1.Blank circles: one volunteer with that value; light blue cir-cles: two volunteers with the same value; dark blue circles:three or more volunteers with the same value. Pearson’sR ¼ 0:576; p < 0:01; a.u.: arbitrary units.




Galley P

roofs

3.5 Follow-up

In a third measuring phase five months after the in-tervention, the sustainability of the achieved effectson the carotenoid concentration and the lifestyle en-hancements was reviewed. 47 volunteers took partin the two-week follow-up measurements; two hadalready dropped out of school.

Only 13 students smoked during the follow-up,which is not only fewer than at the beginning of thestudy but also than in the intervention phase. Oneformer smoker and one non-smoker had droppedout of school. Significant changes in the occurrenceof illness and diet were not observed in comparisonto the second phase.

A consolidation of the measured values can befound. The average of the median of all subjects par-ticipating in phase 3 is higher by 0.42 � 1.10 pointsthan in phase 2 (p ¼ 0:002). Excluding the twodropped out subjects, it is higher by 0.40 points.Compared to phase 1, it is 1.14 points higher. Thechange of the measured values from phase 2 tophase 3 is not as significant as from the first to thesecond phase (change in phase 2: þ0:71 points,p < 0:001).

9 in 47 students kept the median measured valuefrom phase 2; another 9 of them had a lower and 29a higher median value than in the interventionphase. Compared to baseline, 5 of the 47 remainingsubjects had lower levels, 7 maintained their pre-vious value, and 35 subjects increased their value, 16of them by two or more points.

3.6 Individual changes in the volunteers

While in the previous paragraphs the results of thegroups were explained, the individual kinetics of thecarotenoid concentrations in selected volunteers arediscussed in this paragraph.

Some volunteers experienced significant lifestylechanges or very unusual situations in the course ofthe study. The question therefore arises whether sig-nificant changes can also be seen in the course oftheir carotenoid concentration. In this section, thosevolunteers are looked at more closely.

Subject A is the subject with the lowest measuredvalues in both phases (Figure 8). She does notsmoke, rarely drinks alcohol and usually eats a lot offruit and vegetables. In the first phase, where her va-lues varied greatly, she was sick on five of eight meas-urement days. Also her fruit and vegetable con-sumption was very irregular in phase 1, she ate moreof both in phase 2. She had a lot of private stress inboth phases and said that especially the days aroundmeasuring day 2.4 involved “a lot of stress, anxietyand worry” due to conflicts with friends and family.

Subject B had a relatively low level in phase 1and experienced the highest increase of all the sub-jects in the course of the second phase (Figure 8).She ate more fruit, vegetables, drank more fruitjuices, ate fewer sweets and took food supplementson three days in phase 2. In the second phase, sheexperienced a little more stress at school due toexam but fewer conflicts at home, which had beenvery burdening in phase 1. Overall, she was in a bet-ter mood in phase 2 and experienced it as less stress-ful than phase 1.

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758

Figure 8 Measured carotenoid con-centration over the course of phases1 and 2. Volunteer A: Phase 1 med-ian: 3; phase 2 median: 4. VolunteerB: Phase 1 median: 4.5; phase2 median: 6. Volunteer C: Phase1 median: 9; phase 2 median: 9. *Vo-lunteer C stated that she ate lessfruits and vegetables than normallyaround these days.


Journal of

BIOPHOTONICS


Galley P

roofs

Subject C had the highest values in both phase 1and phase 2 (Figure 8). Especially in phase 1, sheate significantly more vegetables and fruit than theaverage subject, did not smoke or drink in bothphases. She never experienced any private stress inboth phases and describes herself as a contentedperson. In both phases, she was preoccupied withexam preparations for three days, which is less thanmost of the subjects. The volunteer claimed that shealready had changed her diet in the first phase inso far as she ate less meat than usual and took foodsupplements in the period of the first two measure-ments. Her consumption of fruit and vegetables de-creased in phase 2 but was still above average.Especially on the days prior to measuring day 2.3in phase 2, when the measured value was lowerthan on the other measuring days, she ate little fruitand vegetables.

After the second phase, the students were askedto write down their thoughts on the study anon-ymously. The following quotes are taken from thoseprotocols:

“You are really surprised when you see: ‘OhGod, my value is now at two or three!’ But then youhave the motivation to eat better from now andthereby achieve a better value.”

“Especially, I have reduced stress, now I start ear-lier to study when we are going to write an examand try to be more relaxed in general.”

“By the better diet I got a better skin complex-ion.”

“It‘s not something, where you need years to seewrinkles or diseases. You have immediate results,the better physical condition is also there immedi-ately.”

“I started eating fruits and vegetables every dayand tried to reduced fat in my diet.”

“We have developed a real ambition to improveour [carotenoid] values.”

4. Discussion

The pilot study at issue was carried out in order toinvestigate the influence of lifestyle habits on carote-noid concentrations in adolescents. In addition it ex-plored to what extent biofeedback measurements in-fluence those lifestyle habits in terms of healthynutrition and stress reduction in adolescents.

It provides the findings that carotenoid concen-trations in this cohort are variable but in averagestable over the period of four weeks and higher thanin the average adult population.

The results also indicate that lifestyle habits andstressful events are reflected in cutaneous carotenoidconcentrations. Most of the results on factors influ-encing the cutaneous carotenoid concentration are

consistent with former findings [40, 45], in particularthat higher carotenoid concentrations can be foundin non-smokers and subjects with healthy nutritionand little stress. Contrary to Meinke et al. [45], nodifferences between male and female subjects werefound. A possible explanation is that adolescents ofboth genders have more similar lifestyle habits thanadults since they are co-educated and their parentsand school mainly determine their diet. Moreover, inthis age group, the distribution of smokers is morebalanced between the sexes whereas there are signif-icantly more male than female smokers within olderadults.

A possible explanation for the high baseline con-centrations in our volunteers might be that the sub-jects were within a relatively homogeneous group ofyoung and in general healthy students with a com-paratively high level of education. Moreover, the vo-lunteers were students from a school, where nutri-tional science is one of the main focuses, which mayhave influenced their nutritional habits. Also, 14 sub-jects claimed they had lived healthier than usual dur-ing the first phase and it is not unlikely that morevolunteers unwillingly did so. Moreover, only 10 in49 (20.4%) volunteers smoked on a daily basis atbaseline. The smoking prevalence in the generalGerman population is 32% for men and 22% forwomen [51]. In the study of Meinke et al. [45], 28 in108 (25.9%) volunteers, who had given statementsabout their smoking habits, were smokers and thosehad a significantly lower carotenoid concentration(4.52� 1.77) than the non-smokers (5.69� 1.78). Itis therefore likely that the low smoking prevalencealso contributed to higher values in the present in-vestigations.

The study at hand indicates that a biofeedbacksystem on cutaneous carotenoids as part of an inte-grated health prevention program is able to help in-creasing the motivation to adopt a healthy lifestylein high school students. Moreover, its use is corre-lated with an increase in the cutaneous carotenoidconcentration and therefore the general antioxidantstatus. Furthermore, a consolidation of these twomajor results can still be seen after five months.Those findings correspond to former findings onadults [40].

To our knowledge, the study at hand is the firstone investigating the antioxidant status of highschool students and also the first study, in which thesubjects carried out the measurements themselves.There has only been one study investigating antioxi-dant concentrations in US-American preschool chil-dren, which was the first of its kind being performedin a school environment [52]. This pilot project hasproven that the spectroscopic device is suitable foruse at school and by students themselves and haspaved the way for the development and implementa-tion of further field studies.

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758




Galley P

roofs

5. Limitations

The results cannot be fully generalized since theinvestigations involved a rather small number ofvolunteers, all within a very homogeneous groupwith regard to age, education and habitat. Anotherlimitation is the subjectivity of the answers statedin the questionnaires, especially when it comes tothe evaluation of the health state or the influenceof stressing events. Also, the students filled in thequestionnaires about the previous two or threedays by memory, which limits the accuracy of theanswers.

Another limitation of the impact of the studymight be the limited practicability of a multi-modalprevention program including changes in the cater-ing, especially in countries, were school lunches areprovided by a well-established industry.

Although a significant increase in the carotenoidconcentration can be seen in the volunteers, there isonly limited evidence for each individual influencingfactor on this result. Also, the influence of the bio-feedback on specific aspects of lifestyle was not ac-quired in particular. However, the results indicateseveral possible influencing factors on the increaseof the carotenoid concentration in phases 2 and 3. Inphase 1, non-smokers, volunteers with a high con-sumption of fruit or vegetables, and volunteers whooften cook at home had higher measured values.Since students smoked less and ate more vegetablesand fruit in the second phase, there is reason to pre-sume that those facts contributed to the increase intheir values. Further investigations with larger popu-lations in a controlled environment and with a moresensitive acquisition of individual changes areneeded in order to make clear statements on thoseand other specific influencing factors.

Even though volunteers had significantly highermeasured values after intervention, the effect ofbiofeedback and consequent lifestyle modificationmight be underestimated. On the one hand, thevolunteers were exposed to a higher prevalence ofillness and a higher stress level due to significantlymore exams in phase 2. On the other hand, manyof the volunteers had changed their lifestyle al-ready in the first phase as mentioned above, know-ing that they would be measured. This might havecontributed both to the high baseline values and asmaller increase than possible in the interventionphase.

As mentioned above, an increase of the carote-noid concentration was particularly noticeable inthe second half of phase 2. On the one hand, thiscan be explained by the simultaneous start of theintervention and the measurements of the secondphase, so that the intervention-related effects werenot reflected in the early measurements of phase

2. Earlier investigations on the kinetics of carote-noids have shown that it takes up to three daysuntil the positive effects are reflected in the skin[40]. On the other hand, the increase in the courseof phase 2 can be interpreted as the effect of thegradual change in the subjects’ lifestyles. Some vo-lunteers stated that they felt like being in a compe-tition and compared their results with fellow stu-dents, one volunteer said in an interview that theytherefore “developed a genuine ambition to im-prove [their] values??” in the course of the inter-vention phase [53].

The consolidation of both the healthy lifestyleand the increased carotenoid levels is particularlyastonishing since no measurements were taken with-in the five months before the follow-up measure-ments. A continuous biofeedback might have pre-sumably contributed to an even greater increasewithin that time. However, the increase towards thefollow up might not be only due to the sustainabilityof the biofeedback effect since former findings haveshown significant seasonal variations of cutaneouscarotenoid concentrations with higher concentra-tions in the summer and autumn months presum-ably due to the better availability of fresh fruit andvegetables in temperate latitudes [27, 40]. Neverthe-less, the “seasonal increase” was not pronouncedduring the spring months [40], when measurementphases 1 and 2 were performed. However, in Sep-tember, the “seasonal increase” is maximal, whichcould have influenced the results of phase 3. Thus,it cannot be said whether seasonal influences or bio-feedback and previous health education events hadmore influence on the carotenoid increase in theskin for phase 3.

The examples of volunteers as shown in Section3.6 are individual cases only. In some subjects, noreasonable explanation could be found for thecourse of antioxidant concentrations. However, meas-urements took place only every three to four days;and the course of antioxidants on the days betweenthe measurements is unknown. At the same time,the questionnaires cover questions about habits andevents on the days before each measurement. A re-cent study investigating the kinetics of cutaneouscarotenoids showed that the decrease of the concen-tration can be as fast as within two hours after astressful event, but the increase after the supplemen-tation of carotenoid-rich products takes one to threedays [40]. Therefore, investigations with more fre-quent measurements are needed in order to getfurther into the question of the influence of certainevents.

The study at hand only covered a small numberof volunteers and only shows a few significant re-sults. Nonetheless, tendencies can clearly be seen.Further investigations with more cases are needed inthis field.

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758


Journal of

BIOPHOTONICS


Galley P

roofs

6. Conclusion

In conclusion, the present results provide evidencethat health-promoting lifestyle habits and a goodstate of physical and mental health are associatedwith a high concentration of cutaneous carotenoids.Positive factors such as good psychological statusand consumption of antioxidant-rich food, and nega-tive factors such as sleep deprivation, cigarettesmoking and alcohol consumption, multifactoriallyinfluence the carotenoid concentration. Therefore,the concentration of cutaneous carotenoids can beseen as a useful indicator for general health.

At the end of the study, the students were askedto summarize their experience gained from the ap-plication of the biozoom� device. Nearly all studentsrevealed that they are reluctant to follow good ad-vice from their parents or teachers, whereas they arevery serious about the information gained from theirown body. They reported also that group measure-ments result in a competition for the best values.

The biozoom� system was appreciated by the vo-lunteers and asserted itself to be a quick and non-in-vasive option to measure carotenoid concentrations.It provides the potential to serve as a readily accessi-ble “looking glass” for general health and could beused within prevention and training programs.

Acknowledgements We would like to extend our gratitudeto Michael Hamm, professor at the Department for Nutri-tional Sciences at the University of Applied Sciences,Hamburg, Germany for his scientific contribution and sup-port and to OptecNet Deutschland e.V., KompetenznetzeOptische Technologien, Lubeck office for supporting ourstudy and for their helpful discussion.

Author biographies Please see Supporting Information on-line.

References

[1] L. Mandelker, Oxidative stress: the role of mitochon-dria, free radicals, and antioxidants. Preface, Vet ClinNorth Am Small Anim Pract 38, ix–xi (2008).

[2] V. J. Thannickal, B. L. Fanburg, Am J Physiol LungCell Mol Physiol 279, L1005–1028 (2000).

[3] T. Akaike, Rev Med Virol 11, 87–101 (2001).[4] O. . Aruoma, M. Grootveld, T. Bahorun, Biofactors

27, 1–3 (2006).[5] R. Ogura, M. Sugiyama, J. Nishi, N. Haramaki, J In-

vest Dermatol 97, 1044–1047 (1991).[6] N. S. Ranadive, I.A. Menon, S. Shirwadkar, S. D. Per-

sad, Inflammation 13, 483–494 (1989).[7] K. J. Davies, A. T. Quintanilha, G. A. Brooks, L. Pack-

er, Biochem Biophys Res Commun 107, 1198–1205(1982).

[8] T. Ashton, C. C. Rowlands, E. Jones, I. S. Young,S. K. Jackson, B. Davies, J.R. Peters, Eur J Appl Phy-siol Occup Physiol 77, 498–502 (1998).

[9] H. B. Vierck, M. E. Darvin, J. Lademann, A. Reis-shauer, A. Baack, W. Sterry, A. Patzelt, Eur J ApplPhysiol (2012).

[10] M. E. Darvin, W. Sterry, J. Lademann, A. Patzelt,Skin Pharmacol Physiol 26, 45–51 (2013).

[11] T. P. Devasagayam, J. C. Tilak, K. K. Boloor, K. S.Sane, S. S. Ghaskadbi, R. D. Lele, J Assoc PhysiciansIndia 52, 794–804 (2004).

[12] M. Valko, M. Izakovic, M. Mazur, C. J. Rhodes, J. Tel-ser, Mol Cell Biochem 266, 37–56 (2004).

[13] M. E. Darvin, W. Sterry, J. Lademann, T. Vergou, Mo-lecules 16, 10491–10506 (2011).

[14] J. J. Thiele, C. Schroeter, S. N. Hsieh, M. Podda,L. Packer, Curr Probl Dermatol 29, 26–42 (2001).

[15] M. Darvin, L. Zastrow, W. Sterry, J. Lademann, SkinPharmacol Physiol 19, 238–247 (2006).

[16] Y. Yamashita, Y. Okano, T. Ngo, P. Buche, A. Sirvent,F. Girard, H. Masaki, Skin Pharmacol Physiol 25, 78–85 (2012).

[17] J. Lademann, S. Schanzer, M. Meinke, W. Sterry,M. E. Darvin, Skin Pharmacol Physiol 24, 238–244(2011).

[18] N. I. Krinsky, Nutrition 17, 815–817 (2001).[19] S. De Spirt, H. Sies, H. Tronnier, U. Heinrich, Skin

Pharmacol Physiol 25, 2–8 (2012).[20] J. Lademann, M. C. Meinke, W. Sterry, M. E. Darvin,

Exp Dermatol 20, 377–382 (2011).[21] P. Di Mascio, S. Kaiser, H. Sies, Arch Biochem Bio-

phys 274, 532–538 (1989).[22] S. F. Haag, B. Taskoparan, M. E. Darvin, N. Groth,

J. Lademann, W. Sterry, M. C. Meinke, Exp Dermatol20, 483–487 (2011).

[23] P. Palozza, N. I. Krinsky, Arch Biochem Biophys 297,184–187 (1992).

[24] M. Wrona, W. Korytowski, M. Rozanowska, T. Sarna,T. G. Truscott, Free Radic Biol Med 35, 1319–1329(2003).

[25] W. Stahl, U. Heinrich, H. Jungmann, J. von Laar,M. Schietzel, H. Sies, H. Tronnier, J Nutr 128, 903–907 (1998).

[26] W. Stahl, U. Heinrich, S. Wiseman, O. Eichler, H. Sies,H. Tronnier, J Nutr 131, 1449–1451 (2001).

[27] S. T. Mayne, B. Cartmel, S. Scarmo, H. Lin, D. J. Lef-fell, E. Welch, I. Ermakov, P. Bhosale, P. S. Bernstein,W. Gellermann, Am J Clin Nutr 92, 794–800 (2010).

[28] J. Lademann, A. Patzelt, S. Schanzer, H. Richter,M. C. Meinke, W. Sterry, L. Zastrow, O. Doucet,T. Vergou, M. E. Darvin, Skin Pharmacology andPhysiology 24, 269–273 (2011).

[29] A. Herman, A. P. Herman, Skin Pharmacol Physiol26, 8–14 (2013).

[30] G. Park, H. G. Kim, Y. O. Kim, S. H. Park, S. Y. Kim,M. S. Oh, Skin Pharmacol Physiol 25, 93–99 (2012).

[31] M. E. Darvin, A. Patzelt, S. Gehse, S. Schanzer,C. Benderoth, W. Sterry, J. Lademann, Eur J PharmBiopharm 69, 943–947 (2008).

[32] D. Segger, F. Schonlau, J Dermatolog Treat 15, 222–226 (2004).

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758




Galley P

roofs

[33] J. Terao, Y. Minami, N. Bando, J Clin Biochem Nutr48, 57–62 (2011).

[34] T. R. Hata, T. A. Scholz, I. V. Ermakov, R. W.McClane, F. Khachik, W. Gellermann, L. K. Pershing,J Invest Dermatol 115, 441–448 (2000).

[35] C. S. Sander, H. Chang, F. Hamm, P. Elsner, J. J.Thiele, Int J Dermatol 43, 326–335 (2004).

[36] L. Steinberg, S. Graham, L. O’Brien, J. Woolard,E. Cauffman, M. Banich, Child Dev 80, 28–44 (2009).

[37] F. Muller-Riemenschneider, M. Nocon, S. N. Will-ich, Eur J Cardiovasc Prev Rehabil 17, 204–210(2010).

[38] S. Schneider, S. Zimmermann, K. Diehl, E. W. Breit-bart, R. Greinert, Acta Derm Venereol 89, 470–475(2009).

[39] C. B. Yucha, D. Montgomery, Evidence-based prac-tice in biofeedback and neurofeedback, (2008).

[40] M. E. Darvin, A. Patzelt, F. Knorr, U. Blume-Peytavi,W. Sterry, J. Lademann, J Biomed Opt 13, 044028(2008).

[41] T. B. Fitzpatrick, Arch Dermatol 124, 869–871 (1988).[42] M. E. Darvin, C. Sandhagen, W. Koecher, W. Sterry,

J. Lademann, M. C. Meinke, J Biophotonics 5, 550–558 (2012).

[43] M. E. Darvin, J. W. Fluhr, P. Caspers, A. van der Pool,H. Richter, A. Patzelt, W. Sterry, J. Lademann, ExpDermatol 18, 1060–1063 (2009).

[44] N. Luther, M. E. Darvin, W. Sterry, J. Lademann,A. Patzelt, Skin Pharmacol Physiol 25, 182–191(2012).

[45] M. C. Meinke, A. Lauer, B. Taskoparan, I. Gersonde,J. Lademann, M. E. Darvin, Free Radicals and Anti-oxidants 1, 15–20 (2011).

[46] D. Talwar, T. K. Ha, J. Cooney, C. Brownlee, D. S.O’Reilly, Clin Chim Acta 270, 85–100 (1998).

[47] S. Scarmo, B. Cartmel, H. Lin, D. J. Leffell, E. Welch,P. Bhosale, P. S. Bernstein, S. T. Mayne, Arch Bio-chem Biophys 504, 34–39 (2010).

[48] M. E. Darvin, M. C. Meinke, W. Sterry, J. Lademann,J Biomed Opt 18, 61230 (2013).

[49] M. E. Darvin, I. Gersonde, M. C. Meinke, W. Sterry,J. Lademann, Journal of Physics D Applied Physics38, 2696–2700 (2005).

[50] E. Boelsma, L. P. van de Vijver, R. A. Goldbohm,I. A. Klopping-Ketelaars, H. F. Hendriks, L. Roza,Am J Clin Nutr 77, 348–355 (2003).

[51] F. Muller-Riemenschneider, A. Rasch, A. Bockel-brink, C. Vauth, S. N. Willich, W. Greiner, GMSHealth Technol Assess 4, Doc10 (2008).

[52] S. Scarmo, K. Henebery, H. Peracchio, B. Cartmel,H. Lin, I. V. Ermakov, W. Gellermann, P. S. Bern-stein, V. B. Duffy, S. T. Mayne, Eur J Clin Nutr 66,555–560 (2012).

[53] F. Anders, Hautscanner sorgt fur neuen Lebensstil,Berliner Morgenpost, 2011.

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758


Journal of

BIOPHOTONICS


Documents

DOI BIOPHOTONICS - MyBiozoommybiozoom.com/.../2016/04/Journal-of-Biophotonics_2013_Schulstudi… · BIOPHOTONICS The cutaneous ... abstain from smoking, ... bility of the measured