61 (2) JUNE CLIN-4 (3)

Invest Clin 61(2): 132 - 141, 2020 https://doi.org/10.22209/IC.v61n2a04


image


Detection of prefrontal lobe blood flow changes in patients with obsessive- compulsive disorder during a verbal fluency task by near-infrared reflectance spectroscopy.


Hong-Ying Yin1, Fan Jiang2, Qi-Hua Peng3 and Yu-Ling Wu1


1Department of Psychology, Jilin Provincial People′Hospital, Changchun, China.

2Department of Psychology, Jilin Normal University,Changchun, China.

3Department of Ultrasonic, Changchun Obstetrics and Gynecology Hospital Changchun, China.


Key words: near-infrared reflectance spectroscopy; obsessive-compulsive disorder; verbal fluency; oxygenate hemoglobin; deoxygenate hemoglobin.


Abstract. The purpose of this work was to use near-infrared reflectance spec- troscopy (NIRS) to investigate the spectral characteristics and frontal lobe activa- tion in patients with obsessive-compulsive disorder (OCD) during a verbal fluency task. Thirty-seven patients with OCD, who met the diagnostic criteria of Interna- tional Classification of Diseases 10 (ICD-10), were recruited. The total score of the Symptom Checklist (SCL-90, grade 1-5) was ≥160 points. The total score of the Yale-Brown Obsessive Compulsive Scale was ≥7 points. The changes in frontal lobe blood flow were measured by NIRS when completing a verbal fluency task. Then, the differences between obsessive-compulsive thoughts and obsessive-compulsive beha- viors in OCD, and the concentration variation of oxygenate hemoglobin (Oxy-Hb) and deoxygenate hemoglobin (Deoxy-Hb) were investigated. The difference was sig- nificant (P<0.03) between obsessive-compulsive behaviors and obsessive-compul- sive thoughts. Obsessive-compulsive behaviors were positively correlated (P<0.01, P<0.05) with depression, anxiety, psychosis and other factors in the SCL-90, and significantly positively correlated (P<0.01, P<0.03) with somatization, hostility and paranoid factor scores in the SCL-90. There was also a trend of partial overlap between the waveform and task period, and the difference was significant (P<0.01) between these. Obsessive-compulsive thoughts were negatively correlated (P<0.05) with channels 9 and 19. The NIRS monitoring spectrum for patients with OCD has certain spectral characteristics of schizophrenia, but there is a repetitive trend bet- ween the recovery period and the task period.


image


Corresponding author: Qi-Hua Peng, Department of Ultrasonic,Changchun Obstetrics and Gynecology Hospital, Nanguan District, Changchun 130021, China, Tel:0431-82904111, E-mail: pengqh62587@163.com


image


Detección de cambios en el flujo sanguíneo del lóbulo prefrontal en pacientes con trastorno obsesivo-compulsivo, durante una tarea de fluidez verbal, mediante espectroscopia de reflectancia en el infrarrojo cercano.

Invest Clin 2020; 61 (2): 132-141


Palabras clave: espectroscopía de reflectancia casi infrarroja; trastorno obsesivo compulsivo; fluidez verbal; hemoglobina oxigenada; hemoglobina desoxigenada.


Resumen. El propósito del presente trabajo fue utilizar la espectroscopia de reflectancia en el infrarrojo cercano (NIRS) para investigar las características espectrales y la activación del lóbulo frontal en el trastorno obsesivo compulsivo (TOC) durante una tarea de fluidez verbal. Se reclutaron 37 pacientes con TOC que cumplían los criterios de diagnóstico de la Clasificación Internacional de Enfermedades 10 (CIE-10). La puntuación total de la Lista de Verificación de Síntomas (SCL-90, grado 1-5) fue de 160 puntos. La puntuación total de la Esca- la Obsesiva-Compulsiva de Yale-Brown fue de ≥7 puntos. Los cambios en el flujo sanguíneo del lóbulo frontal fueron medidos por NIRS al completar la tarea de fluidez verbal. Entonces, se investigaron las diferencias entre los pensamientos obsesivo-compulsivos y los comportamientos obsesivo-compulsivos en TOC, y la variación de las concentraciones de hemoglobina oxigenada (Oxy-Hb) y hemog- lobina desoxigenada (Deoxy-Hb). La diferencia fue significativa (P<0,03) entre los comportamientos obsesivo-compulsivos y los pensamientos obsesivo-compul- sivos. Los comportamientos obsesivo-compulsivos se correlacionaron positiva- mente (P<0,01, P<0,05) con depresión, ansiedad, psicosis y otros factores en el SCL-90, y se correlacionaron positivamente de manera significativa (P<0,01, P<0,03) con la somatización, hostilidad y puntuaciones de factor paranoico en el SCL-90. También hubo una tendencia de superposición parcial entre la forma de onda y el período de tarea, y la diferencia fue significativa (P<0,01) entre estos. Los pensamientos obsesivo-compulsivos estaban correlacionados negativamente (P<0,05) con los canales 9 y 19. El espectro de monitoreo NIRS para pacientes con TOC tiene ciertas características espectrales de la esquizofrenia, pero hay una tendencia repetitiva entre el período de recuperación y el período de tarea.


Received: 09-10-2019 Accepted: 04-05-2020


INTRODUCTION


Multi-channel near-infrared reflectance spectroscopy (NIRS) is a painless and nonin- vasive optical imaging technique (1), which can be used explore the process of internal brain activities in various kinds of human cognitive active procedures as each chan- nel covers a part of the brain area (2-4). The near-infrared light can be absorbed by hemo-

globin through the skull within 2-3 cm, and detect the levels of oxygenate hemoglobin (Oxy-Hb) and deoxygenate hemoglobin (De- oxy-Hb) in the venous blood of brain tissues (5-8). Compared with other neuroimaging techniques, its advantage consists of a high time resolution (0.1 second), which can dis- play the dynamic functional changes of the prefrontal cortex in real time (9).


Herrmann et al. applied the verbal flu- ency test (VFT) in the NIRS test, and con- firmed that in the frontal lobe, Oxy-Hb was significantly elevated and Deoxy-Hb was sig- nificantly decreased, when normal subjects completed the task (10). VFT reflects the complex cognitive activities of subjects, in- cluding choosing the right language from memory, remembering a spoken language, suppressing an inappropriate language, and maintaining attention to cognitive activa- tion (11). Most reports have been based on VFT as an activation task, and NIRS has been used to assist in the diagnosis of schizophre- nia, depression, biphasic affective disorder and other mental diseases at home and abroad (12). However, few studies have been conducted on obsessive-compulsive disor- der. In the present study, the NIRS test was used to investigate the characteristics of the spectrum and frontal lobe activation state in patients with obsessive-compulsive disorder (OCD). The hemodynamic changes of the prefrontal lobe were tested using the 33-lead NIRS system (ETG-4000, Hitachi, Japan) that consisted of 52 channels (14), and the 22 channels that covered the prefrontal lobe were selected as the research area in the cur- rent study.


RESEARCH METHOD


Research object

The present study was comprised of out- patients, who were treated in the Outpatient Department of our hospital between January 2016 and December 2016. Inclusion criteria: patients who meet the OCD diagnostic crite- ria of International Classification of Diseases 10 (ICD-10) (13); patients who were right- handed; patients with a total score of ≥160 points in the Symptom Checklist (SCL -901- 5); patients with a total scores of ≥6 points in the Yale-Brown Obsessive Compulsive Scale. Exclusion criteria: patients who suf- fered from brain organic lesions, mental re- tardation, drug or alcohol abuse history, and other heart, liver, kidney and serious physi-

cal diseases; pregnant or lactating women. A total of 37 patients were enrolled in the pres- ent study. Among these patients, 22 patients were male and 15 patients were female, with the average age of these patients was 25.16

± 0.20 years old. The educational levels of patients were as follow: two patients finished junior high school, nine patients finished senior high school, four patients finished college, 14 patients were undergraduates, and eight patients had an educational back- ground higher than an undergraduate. Mari- tal status: thirty patients were unmarried and seven patients were married. Vocational status: twenty four patients were students, six patients were unemployed, and seven pa- tients had other vocational statuses.

The present study was approved by the Ethics Committee of our hospital, and the patients and guardians were informed. All subjects voluntarily provided a signed in- formed consent.


METHODS


Yale-Brown Obsessive Compulsive Scale and Symptom Checklist (SCL-90)

The Yale-Brown Obsessive Compulsive Scale and the Symptom Checklist (SCL -90, 1-5) were used to evaluate the psychology of the patients.


Verbal fluency task

The VFT test was carried out in a quiet environment, and the patient was emotion- ally stable before the test. During the test, the patients were instructed to maintain their head posture and look at the cross on the wall, which was placed in the same hori- zontal line as the line of sight, located ap- proximately 1 m away. The test was divided into four parts: The first part was the pre- scanning, and the time was set for 10 sec- onds. The patient was instructed to sit still and look at the cross in front of them. The second part was the waiting time, and the time was set for 30 seconds. The patient re- peated a number from 1 to 5. The third part


was the stimulation time, and the stimulus was induced by presenting three characters: white, heaven and big. Words or idioms were formed using the first given character, and the time for each character to form a word or idiom was 20 seconds. The last part was relaxation, and the time was set for 70 sec- onds. The patient also repeated a number from 1 to 5.


NIRS data detection and analysis

The hemodynamic changes of the pre- frontal lobe were tested using the 33-lead NIRS system (ETG-4000, Hitachi, Japan) during the cognitive activities. The system consisted of 17 optical signal transmitters and 16 detectors, with a total of 52 chan- nels (14). All sensors were fixed to a soft and elastic cloth cover, which was subsequently fixed on the patient’s head. In the present study, 22 channels that covered the prefron- tal lobe were selected as the research area. Oxy-Hb values, Deoxy-Hb values and analy- sis data were collected using the Ver1.61 system software attached to the ETG-4000 brain functional quantitative device. The Gravity value was determined by the mean response amplitudes of the changes during time 15-20 seconds of each individual stimu- lation and coordinates of the channels.


Statistical analysis

Data were processed using SPSS 19.0, and the dose data were expressed as xˉ ± standard deviation (SD). The correlation r- test and paired t-test were used, and P<0.05 was considered statistically significant.


RESULTS


Correlation analysis of the Yale-Brown Obsessive Compulsive Scale and Symptom Checklist (SCL-90)

There was a significant positive correla- tion (P<0.01) between the total score of the Yale-Brown Obsessive Compulsive Scale (Yale- Brown) and the total score of the Symptom Checklist (SCL -90) in patients with OCD.

The difference was statistically significant (P<0.03) between obsessive-compulsive be- haviors and obsessive-compulsive thoughts (t=2.40, P=0.025). Obsessive-compulsive behaviors were positively correlated (P<0.01, P<0.05) with depression, anxiety, psychosis and other factors in the table (SCL -90), and significantly positively correlated (P<0.01, P<0.03; Table I) with somatization, hostility and paranoid factor scores in the table (SCL - 90).


Results for the frontal lobe NIRS data

NIRS spectral characteristics of the frontal lobe in patients with OCD during the verbal fluency task

The score for the verbal fluency task and its integral value in the frontal lobe of patients with OCD were 7.3 ± 0.65 and 6,916 ± 15.97, respectively. The center of gravity (57.44 ± 2.36) was at the last phase. The slope (0.000619 ± 0.000259) suggests that the integral value of the prefrontal lobe was medium in patients with OCD, and that the center of gravity value was at the last phase (the last 20 seconds). The initial ac- tivation was slow, while the subsequent wave fluctuation occurred during the task period (Fig. 1).


Spectral comparisons of the concentra- tions of Oxy-Hb and Deoxy-Hb in patients with OCD during the implementation of the verbal fluency task period and recovery period

When patients with OCD carried out the verbal fluency task, the concentrations of Oxy-Hb and Deoxy-Hb increased again at the end of the task, and this rising trend was slightly higher than the one occurring during the task period. Furthermore, there was an overlapping trend between the wave- form and task period of the project, but the difference was significant between these two groups (P<0.01, Oxy-Hb task period, compared to the recovery period, t=15.07, P=0.000; Deoxy-Hb task period, compared to the recovery period, t=4.52, P=0.000; Figs. 2 and 3).


TABLE I


Items xˉ±SD

Yale-Brown total score

19.83±1.40

Obsessive-compulsive Obsessive-compulsive behaviors thoughts

11.48±0.84 8.43±1.04

r

p

r

p

r

p

CORRELATION ANALYSIS OF THE YALE-BROWN OBSESSIVE-COMPULSIVE SCALE AND SYMPTOM CHECKLIST (SCL -90).


SCL -90 total score


Somatization

218.43±10.53


22.53±1.70

0.62

0.002


0.51


0.013


0.26


0.236

Obsessive- compulsiveness

30.27±1.35

0.40

0.061

0.21

0.332

Interpersonal relation

23.30±1.55

0.35

0.104

0.29

0.181

Depression

36.77±1.97

0.10

0.660

0.64

0.001

Anxiety

25.73±1.30

0.29

0.186

0.58

0.004

Hostility

14.40±1.20

0.66

0.001

0.37

0.079

Horror

13.93±1.06

0.01

0.968

0.56

0.005

Bigotry

13.47±0.87

0.46

0.026

0.29

0.176

Psychosis

21.27±1.43

0.29

0.186

0.44

0.035

Others

17.27±1.09

0.33

0.120

0.59

0.003


image


image

Fig. 1. NIRS spectral characteristics of the frontal lobe in patients with OCD during the verbal fluency task.


image

Fig. 2. Changes of the concentrations of Oxy-Hb in patients with OCD during the task period and recovery period.


image


Fig. 3. Changes of the concentrations of Deoxy-Hb in patients with OCD during the task period and recovery period.


The correlation of hemoglobin con- centrations among obsessive-compulsive thoughts, obsessive-compulsive behaviors and the data obtained from the 22 channels are shown in Fig. 4. It was found that ob- sessive-compulsive thoughts were negatively correlated with channels 9 and 19 (P<0.05), but there was no correlation (P>0.05) be- tween obsessive-compulsive behaviors and channel 22 (Table II).


DISCUSSION


In this Observational-Analytical study, it was found that the mental health of 37 pa- tients with OCD significantly deviated from the normal (SCL -90 total score: 218 points

>160 points). Furthermore, there was a dif- ference between obsessive-compulsive behav- iors and obsessive-compulsive thoughts. Ob- sessive-compulsive behaviors often present symptoms of depression, anxiety, psychosis, diet and sleep disorders (P<0.01, P<0.05), while obsessive-compulsive thoughts often embody somatization, hostility and para- noia (P<0.01, P<0.03). It has been sug- gested that obsessive-compulsive behaviors are partially due to emotional traits and obsessive-compulsive thoughts are biased to mental traits, which confirms the view that obsessive-compulsive disorder patients have thinking disorder (15). Therefore, some ref-

erence for the selection of antidepressants and antipsychotic drugs in different types of clinical treatments can be provided (16, 17).

The NIRS test was performed in 37 pa- tients with OCD using the verbal fluency method. With respect to the achievements of the verbal fluency task, the average num- ber of words formed by OCD patients within

60 seconds was seven, which was signifi- cantly lower than that in normal subjects and schizophrenia patients (18, 19). The in- tegral area of the frontal lobe was medium, and the center of gravity value was at the last phase. The initial activation time was slow and reached the peak again at the end of the task, which was different from the depression integral value area of ≤54, the forward cen- ter of gravity value, and the fast-initial acti- vation. Furthermore, this was similar to the late phase of schizophrenia, irregular chang- es in the task, and the increased reaction at the end of the task, but this did not have the characteristics of the prefrontal lobe of schizophrenia with a small score. This indi- cates that although depression is present in the syndrome of OCD and antidepressant drugs have been used in the treatment of OCD in clinic, the data of the present study suggests that other significantly different diseases still exists between obsessive-com- pulsive disorder and depression, which is more like a kind of disease between depres-


image


Fig. 4. Data obtained from the 22 channels detected by NIRS.


sion and schizophrenia. This also confirms that OCD is separated from the pedigree of neurotic disorder in the DSM-5 diagnostic criteria and becomes the theoretical basis of an independent diagnostic unit.

It is known that OCD presents with symptoms of repetitive thoughts and behav- iors that are not necessary but could not be controlled. OCD has the characteristics of “obsessive-compulsive and counter-obses- sive-compulsive” in clinic. Counter-obses- sive-compulsive means that although the pa- tients do not want to think about or do the

things, their thoughts and behaviors cannot be neglected and the patients cannot help repetitively thinking or doing these things (15). Even when a single thought and behav- ior ends, OCD patients would continue the “obsessive-compulsive repetition”. In the present study, it was found that the atlas of Oxy-Hb and Deoxy-Hb in patients with OCD during VFT task period and recovery period exhibited overlapping signs and characteris- tics of repeatability. From the point of view of NIRS, this proves that the clinical features of “obsessive-compulsive” repeatability was


TABLE II

Obsessive-compulsive thoughts

Obsessive-compulsive behaviors

r

p

r

p

CORRELATION OF HEMOGLOBIN CONCENTRATION AMONG OBSESSIVE-COMPULSIVE THOUGHTS, OBSESSIVE-COMPULSIVE BEHAVIORS AND THE 22 CHANNELS.


CH1

-0.01

0.975

0.32

0.142

CH2

-0.17

0.452

0.38

0.076

CH3

0.27

0.215

-0.29

0.179

CH4

0.33

0.119

0.24

0.272

CH5

-0.19

0.397

0.13

0.570

CH6

-0.18

0.411

-0.06

0.794

CH7

-0.32

0.132

0.27

0.219

CH8

-0.14

0.511

-0.11

0.623

CH9

-0.43

0.041

0.01

0.980

CH10

-0.11

0.623

-0.01

0.965

CH11

-0.09

0.682

-0.14

0.529

CH12

-0.05

0.810

0.08

0.734

CH13

-0.00

0.984

-0.04

0.874

CH14

-0.26

0.224

-0.01

0.984

CH15

0.09

0.675

-0.09

0.682

CH16

-0.29

0.182

-0.13

0.562

CH17

-0.10

0.657

-0.14

0.521

CH18

0.01

0.976

0.15

0.505

CH19

-0.42

0.047

0.09

0.682

CH20

-0.19

0.374

-0.19

0.384

CH21

-0.36

0.093

-0.22

0.317

CH22

-0.17

0.436

-0.13

0.543


image

persistent, and that the difference continued to be significant (P<0.01) between these.

Obsessive-compulsive thoughts were negatively correlated with channels 9 and 19 of the NIRS. This suggests that the clinical symptoms of obsessive-compulsive thoughts result in the decrease in Oxy-Hb concentra- tion in channels 9 and 19 in a low activation state. This area is part of bilateral dorsolat- eral frontal lobe cortex and is correlated to continuous attention and working memory. The brain dysfunction of attention and work- ing memory occurs in obsessive-compulsive

thoughts during VFT, which is similar to the characteristics of schizophrenia (20).

In summary, the activation time and spectrum tested by NIRS in patients with OCD have certain spectral characteristics of schizophrenia. However, the recovery pe- riod and task period of Oxy-Hb and Deoxy- Hb have repetitive characteristics, which is different from the peak of schizophrenia, or even higher than the task period at the end of the task.

In the present study, the frontal lobe blood flow was investigated in patients with


OCD during the completion of the speech fluency task. However, the effect of drugs on this approach has not yet been investigated. In future researches, the characteristics of the mesencephalic blood flow spectrum in the treatment of OCD would be further in- vestigated. Furthermore, there is a need to further expand the sample study and clarify the spectral characteristics of the NIRS test in other mental disorders. This would be an important research direction in the future for the application of the NIRS test to as- sist in the differential diagnosis of mental ill- nesses, providing objective and reliable elec- trophysiological indexes for their diagnosis.


REFERENCES


  1. Yin L, Zhou J, Chen D, Han T, Zheng B, You- nis A, Shao Q. A review of the application of near-infrared spectroscopy to rare tradi- tional Chinese medicine. Spectrochim Acta A Mol Biomol Spectrosc 2019;221:117208.

  2. Rashnavadi T, Macnab A Cheung A, Shad- gan A, Kwon BK, Shadgan B. Monito- ring spinal cord hemodynamics and tissue oxygenation: a review of the literature with special focus on the near-infrared spectros- copy technique. Spinal Cord 2019;57:617- 625.

  3. AlZubaidi AK, Ethawi Y, Schmölzer GM, Sherif S, Narvey M, Seshia M. Review of biomedical applications of contactless imaging of neonates using infrared ther- mography and beyond. Methods Protoc 2018;1(4):39. doi: 10.3390/mps1040039.

  4. Hou L, Liu Y, Qian L, Zheng Y, Gao J, Cao W, Shang Y. Portable near-infrared te- chnologies and devices for noninvasive as- sessment of tissue hemodynamics. J Health Eng 2019;2019:3750495.

  5. Hoshi Y, Tamura M. Dynamic multichan- nel near-infrared optical imaging of human brain activity. J Appl Physiol 1993;75:1842- 1846.

  6. Badurdeen S, Roberts C, Blank D, Miller S, Stojanovska V, Davis P, Hooper S, Pol- glase G. Haemodynamic instability and bra- in injury in neonates exposed to hypoxia-is- chaemia. Brain Sci 2019;9(3): pii E49.

  7. Tinga AM, de Back TT, Louwerse MM. Non-invasive neurophysiological measu- res of learning: a meta-analysis. Neurosci Biobehav Rev 2019;99:59-89.

  8. Andersen AV, Simonsen SA, Schytz HW, Iversen HK. Assessing low-frequency os- cillations in cerebrovascular diseases and related conditions with near-infrared spec- troscopy: a plausible method for evaluating cerebral autoregulation? Neurophotonics 2018;5:030901.

  9. Akashi H, Tsujii N, Mikawa W, Adachi T, Kirime E, Shirakawa O. Prefrontal cortex activation is associated with a discrepancy beween self-and observer-rated depression severities of major depressive disorder: A multichannel near-infrared spectroscopy study. J Affect Disord 2015;174:165-172.

  10. Herrmann MJ, Ehlis AC, Fallgatter AJ. Frontal activation during a verbal-fluency task as measured by near-infrared spectros- copy. Brain Res Bull 2003;61:51-56.

  11. Henry JD, Crawford JR. A meta-analytic review of verbal fluency performance fo- llowing focal cortical lesions. Neuropsycho- logy 2004;18:284-295.

  12. Kinou M, Takizawa R, Marumo K, Kawa- saki S, Kawakubo Y, Fukuda M, Kasai K. Differential spatiotemporal characteristics of the prefrontal hemodynamic response and their association with functional im- pairment in schizophrenia and major de- pression. Schizophr Res 2013;150:459-467.

13. Hon SKH, Siu BWM, Cheng CW, Wong WCW, Foa EB. Validation of the Chinese Version of Obsessive-Compulsive Inven- tory-Revised. East Asian Arch Psychiatry 2019;29(4):103-111.

  1. Quan WX, Bai YF, Dong WT, Tian J, Yang SZ. Normal frontotemporal oxygenated he- moglobin level detection by near-infrared imaging. Journal of China Medical Univer- sity 2012;41:1044-1046.

  2. Yin HY, Zhao MS, Jiang HB, Yang LH, Zhong YF. Preliminary Evidences of Ob- sessive Compulsive Disorder as Thinking Disorder. International Conference on E- Commerce and Contemporary Economic Development (ECED 2014) 2014:343-347.

  3. Gregory ST, Kay B, Riemann BC, Good- man WK, Storch EA. Cost-effectiveness of treatment alternatives for treatment-


    refractory pediatric obsessive-compulsive disorder. J Anxiety Disord 2020;69:102151.

  4. Marazziti D, Mucci F, Avella MT, Dell’Oste V, Baroni S, Dell’Osso L. Treatment of the obsessive-compulsive and bipolar disorders comorbidity: pharmacodynamic and phar- macokinetic evaluation. Expert Opin Drug Metab Toxicol 2019;15(8):619-631.

  5. Houghton DC, Stein DJ, Cortese BM. Re- view. Exteroceptive sensory abnormalities in childhood and adolescent anxiety and obsessive-compulsive disorder: A critical re- view. J Am Acad Child Adolesc Psychiatry 2020;59(1):78-87.

  6. Tezenas du Montcel C, Pelissolo A, Schür- hoff F, Pignon B. Obsessive-compulsive symptoms in schizophrenia: an up-to-date review of literature. Curr Psychiatry Rep 2019;21(8):64.

  7. de Boer JN, Vingerhoets C, Hirdes M, McAlonan GM, Amelsvoort TV, Zinkstok JR. Efficacy and tolerability of riluzole in psychiatric disorders: A systematic review and preliminary meta-analysis. Psychiatry Res 2019;278:294-302.



Universidad del Zulia / Venezuela / Investigación Clínica /riclinicas@gmail.com / p-ISSN: 0535-5133 / e-ISSN: 2477-9393

ReviCyHLUZ

Licencia de Creative Commons
Este obra está bajo una licencia de Creative Commons Reconocimiento-NoComercial-CompartirIgual 3.0 Unported.