Clinical Hemorheology and Microcirculation 60 (2015) 423–435 DOI 10.3233/CH-141898 IOS Press
423
Effect of transcutaneous application of gaseous carbon dioxide on cutaneous microcirculation Miha Finzgar, Ziva Melik and Ksenija Cankar∗ Institute of Physiology, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia
Abstract. BACKGROUND: The inefficient healing of chronic wounds is a result of poor blood perfusion at the wound and surrounding tissues. Artificially applied carbon dioxide (CO2 ) has the potential to improve the perfusion and oxygenation of tissues, hence is useful for the healing of chronic wounds. OBJECTIVE: The aim of the present study was to determine the effect of a transcutaneous application of physiological vasodilator gaseous CO2 on cutaneous blood flow. METHODS: Laser Doppler (LD) flux in cutaneous microcirculation, skin temperature, electrocardiogram and arterial blood pressure were measured simultaneously in a group of 33 healthy men, aged 21–28 years, during rest and a 35-minute CO2 therapy. One lower limb of each subject represented the studied extremity, being exposed to gaseous CO2 . The contralateral limb was the control, being exposed to air. Each limb was sealed in a plastic bag. RESULTS: During CO2 therapy the LD flux in the studied extremity increased from 5.8 PU ± 3.9 PU to 30.3 PU ± 16.7 PU (mean ± standard deviation; paired t-test, p < 0.001), while that in the control extremity did not change significantly. CONCLUSIONS: Our results confirm a local vasodilatory effect of applied CO2 therapy. This finding indicates its potential clinical use. Keywords: Physiological measurements, laser Doppler fluxmetry, CO2 therapy, dry CO2 bath
1. Introduction Chronic wounds are wounds that fail to heal over a period of 3 months [22]. The aetiology of chronic wounds is very diverse: venous insufficiency, insufficient arterial blood supply, diabetes mellitus and various systemic causes, such as nutritional abnormalities, immunosuppression and infection [30]. An important characteristic of these wounds is tissue hypoxia [4]. Carbon dioxide (CO2 ) along with some other metabolites causes vasodilation of blood vessels [1], which in turn increases blood flow. This phenomenon is called active hyperaemia [7]. The exact mechanism of this type of hyperaemia has not yet been discovered. It is most likely a result of an integrated response of multiple vasodilatation mechanisms [7]. Carbon dioxide therapy (CO2 therapy) refers to the transcutaneous or subcutaneous application of CO2 for therapeutic purpose [3, 25]. One example of these therapies is the transcutaneous application of gaseous CO2 , also known as dry CO2 bath. This type of CO2 application is not invasive compared to the ∗ Corresponding author: Ksenija Cankar, Institute of Physiology, Zaloska 4, 1000 Ljubljana Slovenia. Tel.: +386 1 5437500; Fax: +386 1 5437501; E-mail: [email protected].
1386-0291/15/$35.00 © 2015 – IOS Press and the authors. All rights reserved
424
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
subcutaneous method. In comparison to CO2 water baths, it does not involve the risk of inflammation and therefore could be used for patients with chronic wounds. In addition, it does not increase the concentration of CO2 in surrounding air if proper security measures are taken. The effects of all three types of CO2 therapies on skin microcirculation have previously been studied in animal models. Duling [11] observed increased microvascular diameter and increased perivascular pO2 at all sites during the exposure of hamsters’ pouches to the solution, with pCO2 = 32 mmHg and pO2 = 18 ± 2 mmHg. The author reported that increased diameter was attributable to the vasodilatory effect of CO2 , whereas the increased perivascular pO2 was due to the effect of CO2 on oxyhemoglobin dissociation curve [11]. Ito et al. [17] showed that a topical application of CO2 water to a rat’s skin increases skin blood flow about a 100 % at bath temperatures of 23◦ C and 34◦ C. Minamiyama and Yamamoto [21] demonstrated the in vivo effect of a topical application of CO2 water bath on the microcirculation of rats’ skin. By way of intra-vital video microscopy, they observed vasodilatation together with an increase in blood flow; also observed was a decrease in the pH of subcutaneous tissue. In addition to these studies, the work of Diji [10] in 1959 revealed an indirect vasodilatation effect of CO2 : a reported increase in the rate of heat elimination to water at about 40 % (at 29◦ C) due to the immersion of a hand in a CO2 saturated water bath. Diji and Greenfield [9] also showed that a subcutaneous application of gaseous CO2 caused local vasodilatation. By using near-infrared spectroscopy Sakai et al. [25] demonstrated that a transcutaneous CO2 application caused O2 dissociation from oxygenated haemoglobin, which is characteristic of the Bohr effect. To our knowledge, there were no studies investigating the effect of a transcutaneous application of gaseous CO2 on cutaneous blood flow in vivo in humans. The aim of the present study was to determine the effect of a transcutaneous application of gaseous carbon dioxide on cutaneous blood flow as well as possible systemic effects. Also assessed was the vasodilatory effect of elevated temperature and relative humidity on cutaneous blood flow in the sealed bag system, as was used during the CO2 therapy. 2. Subjects and methods 2.1. Subjects The subjects of the present study were 33 male volunteers who were without acute diseases, aged 23.8 ± 2.3 years (from 21 to 28 years) and who had a body mass index of 24.2 ± 2.9 kg/m2 (from 17.3 to 30.1 kg/m2 ). Five of them were smokers, two were asthmatics and two of them had a family history of arterial hypertension (at least one parent of a chosen subject received medical treatment for his/her arterial hypertension before the age of 55). All participating subjects refrained from smoking or drinking coffee or tea for at least 8 hours before the start of the measurements. The present study was carried out with the approval of National Medical Ethics Committee. A written consent was obtained from each subject. 2.2. Electrocardiogram and arterial blood pressure The electrocardiogram (ECG) (Institute of Physiology, Ljubljana, Slovenia) in standard lead II and arterial blood pressure (Ohmeda Finapres 2300, Finapres Medical Systems, Amsterdam, Netherlands) were monitored continuously. Blood pressure was measured on a right middle finger.
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
425
2.3. Laser Doppler flux Laser Doppler (LD) flux sensors (Angled probe 401, Perimed, J¨arf¨alla, Sweden) together with laser light sources at 780 nm (PF 4001 and PF 4002 Satellite, Perimed, J¨arf¨alla, Sweden) were used for a continuous measurement of LD flux. It was monitored simultaneously on both thighs, medial and somewhat proximal to the knee. 2.4. Skin temperature Skin temperature at the site of the LD flux measurement was recorded with a skin spot sensor (Sensor 442, Perimed, J¨arf¨alla, Sweden). It was measured continuously; values were read at 60-second intervals. The temperature of the big toes was measured with an infrared thermometer (Voltcfraft IR 800-20D, Voltcraft, London, UK) three times after acclimatization, just before the start of the CO2 therapy and immediately after its end. 2.5. Temperature and relative humidity The temperature and relative humidity in the bag used for the studied extremity were measured by a high-accuracy, ambient air probe (Testo 0636 9741, Testo, Lenzkirch, Germany) attached to a multifunction measuring instrument (Testo 400, Testo, Lenzkirch, Germany). Both parameters were measured continuously; values were read at 60-second intervals. 2.6. System for the transcutaneous application of gaseous CO2 Gaseous CO2 with 99.995 % purity (CO2 ; Messer Slovenia, Ruˇse, Slovenia) was applied transcutaneously with the CARBOfit® system (DermaArt, Breˇzice, Slovenia). This system consists of a compressor, a CO2 sensor, a conduit tube for CO2 and a drain tube for air/CO2 . CO2 was applied in single-use, low-density, polyethylene bags (thickness 0.4 mm). 2.7. Protocol Before the start of the measurement, each subject filled out a questionnaire for the purpose of obtaining his basic data. During measurements, subjects were in lying position without shoes, socks and pants. They were instructed not to talk or move during the measurements. After installing all the necessary sensors, each lower extremity was packed inside a separate bag and sealed with elastic fastener tape in the upper third part of each thigh. One lower limb represented the studied extremity and the contralateral one served as the control. The studied extremity was randomly chosen for each subject. Temperature and relative humidity sensors were installed in the bag with studied extremity. The time needed for preparation served as an acclimatization time (10 minutes). Acclimatization was followed by a 6-minute-long resting period. After the resting period, the air in the bag with the studied extremity was pumped out and the bag was filled with CO2 . The therapy lasted 35 minutes. Laser Doppler flux and skin temperature in both lower extremities, ECG and arterial blood pressure were measured simultaneously during rest and during the entire CO2 therapy.
426
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
Measurements were performed in a room with an air temperature of 24.8 ± 0.7◦ C, air pressure at 980 ± 2 mbar and relative humidity of 54.1 ± 4.1 % (all these values are expressed as mean ± standard deviation of the mean - SD). 2.8. Analog to digital conversion of data The analog signals of ECG, arterial blood pressure and LD flux were converted (sampling rate of 500 s−1 was used) with an analog-digital converter (DASport, Burr-Brown Corporation, Tucson, USA). 2.9. Statistical analysis Digital signals of LD flux, arterial blood pressure and RR intervals from ECG recording were analyzed with NevrokardTM (Nevrokard Kiauta, Izola, Slovenia), a professional software tool that provides analysis of different data acquisition system file formats. We calculated the percentage change of the mean LD flux in the studied extremity during CO2 therapy in comparison to the mean value of LD flux in the studied extremity during the resting period: LD flux%i =
LD flux2i - LD flux1i · 100%, LD flux1i
(1)
where LD flux%i is an i value of LD flux percentage change; LD flux1i an i mean value of LD flux break during the rest period; LD flux2i an i mean value of LD flux during CO2 therapy; and i = 1, 2, . . . , 33 (index i marks each participating subject). For the assessment of the linear correlation between the chosen pairs of variables a Pearson productmoment correlation coefficient was calculated (results are given as squared values). With the method of least squares, temperature vs. time and relative humidity vs. time curves were approximated with a loharthmic funciton (trend curve) and corresponding coefficients of determination were calculated. Values measured during rest and during CO2 therapy were compared with a paired t-test. A p-value of less than 0.05 was considered as statistically significant. 2.10. Spectral analysis Changes in sympathovagal balance are usually assessed with a spectral analysis of heart rate or RR interval [20]. For spectral analysis of the RR interval, we used autoregression (256 points; autoregressive model order: 13). We defined three frequency bands [18]: VLF band (very low frequency; 0.01–0.04 Hz), LF band (low frequency; 0.04–0.15 Hz) and HF band (high frequency; 0.15–0.4 Hz). For the spectral analysis of the LD flux signal, a Fast Fourier transform (1024 points; Hanning window; normalized power spectral density) was used. Five frequency bands were defined [16]: ULF band (ultra low frequency; 0.001–0.02 Hz), VLF band (very low frequency; 0.02–0.06 Hz), LF band (low frequency; 0.06–0.15 Hz), HF band (high frequency; 0.15–0.6 Hz) and VHF band (very high frequency; 0.6–2 Hz). 3. Results The data of all participating subjects were analyzed.
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
427
Table 1 Values of measured parameters during rest and CO2 therapy Measured quantity ν [beats/min] ps [mmHg] pd [mmHg] LD flux [PU] – control extremity LD flux [PU] – studied extremity T [◦ C] – control extremity T [◦ C] – studied extremity
Last 3 minutes of rest period (period before CO2 therapy) 64.9 ± 8.8 108.0 ± 13.6 63.4 ± 8.6 5.0 ± 2.9 5.8 ± 3.9 32.80 ± 1.15 32.82 ± 0.93
Last 3 minutes of CO2 therapy 62.5 ± 7.3∗ 109.9 ± 12.5 63.4 ± 6.6 7.1 ± 4.9 30.3 ± 16.7†,‡ 33.63 ± 1.18 33.59 ± 0.99
Notes: ν – heart rate; ps – systolic pressure; pd – diastolic pressure; LD flux – laser Doppler flux; T – skin temperature; PU – perfusion units; ∗ – statistically significant difference before and during CO2 therapy (p = 0.003); † – statistically significant difference before and during CO2 therapy for the studied extremity (p < 0.001); ‡ – statistically significant difference between the control and studied extremity during CO2 therapy (p < 0,001). Data are expressed as mean ± SD.
Table 1 shows the values of the measured parameters during the resting period and during the CO2 therapy. All values are given for a 3-minute interval of rest and the last 3 minutes of CO2 therapy, as the most evident effects were expected toward the end of the CO2 therapy. Choosing the 3-minute interval for both periods also enabled the spectral analysis of the LD flux and RR interval. The decrease in heart rate during CO2 therapy compared to the values obtained at rest was statistically significant (paired t-test, p = 0.003). In contrast, the systolic and diastolic pressure did not change significantly. The mean value of LD flux in the cutaneous microcirculation of the studied extremity increased significantly during CO2 therapy in comparison to the mean value during the resting period (paired t-test, p < 0.001). An increase was present in all subjects. In contrast, the change in the mean value of LD flux in cutaneous microcirculation in control extremity was not significant. The difference between LD flux during CO2 therapy in the control and studied extremity was also statistically significant (paired t-test, p < 0.001). Figure 1 shows (A) the LD flux in both extremities during rest and (B) during the entire duration of CO2 therapy. Figure 1 C shows the LD flux percentage change values for the studied extremity during CO2 therapy. Table 2 displays the results of the spectral analysis of the RR interval. The differences between the values before and during the CO2 therapy are not statistically significant. Table 3 provides the results of the spectral analysis of LD flux. Change in the control extremity is not statistically significant, whereas in the studied limb the VLF (paired t-test, p < 0.001), LF (paired t-test, p = 0.015) and HF (paired t-test, p = 0.037) power decreased significantly, while the VHF power increased (paired t-test, p < 0.001). Differences in the values of skin temperature at the site of the LD flux measurements during rest and during CO2 therapy were not statistically significant, neither for the studied nor for the control extremity (except for the point at t = 60 s; paired t-test, p = 0.049; see Fig. 2). There were also no statistically significant differences between the values of skin temperature on plantar side of the big toe on either extremity. Temperature and relative humidity versus time curves (Fig. 3), describing conditions in the sealed bag with the studied extremity, closely follow a logarithmic function.
428
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide A
B
C
Fig. 1. (A) LD flux in the control and studied extremity during rest. (B) LD flux in the control and studied extremity during CO2 therapy. (C) LD flux percent change in the studied extremity during CO2 therapy. All bars represent mean ± standard error (SE); start of rest period at t = 0 s, end of rest period at t = 360 s; start of the CO2 therapy at t = 0 s, end at t = 2100 s; ∗ - p < 0.001.
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
429
Table 2 Spectral analysis of RR interval LF power [n.u.]
HF power [n.u.]
LF/HF [/]
LF/(LF + HF) [/]
Total power [n.u.]
75.9 ± 32.7
48.0 ± 17.6
Rest period (last 3 minutes) 2.2 ± 2.0
0.60 ± 0.18
158.2 ± 54.3
77.0 ± 30.1
52.3 ± 16.4
CO2 therapy (last 3 minutes) 1.8 ± 1.3
0.56 ± 0.19
167.8 ± 52.3
Notes: LF – low frequency; HF – high frequency; n.u. – normalized units. Data are expressed as mean ± SD. Results of very low frequency (VLF) band are not shown. Table 3 Spectral analysis of LD signal ULF power [n.u.]
VLF power [n.u.]
13.3 ± 9.0
HF power [n.u.]
VHF power [n.u.]
Total power [n.u.]
17.6 ± 12.0
Control extremity during rest 26.4 ± 12.4 7.6 ± 4.6
29.6 ± 17.3
94.4 ± 5.4
14.0 ± 11.7
17.6 ± 12.2
Studied extremity during rest 26.8 ± 16.5 7.8 ± 7.8
29.0 ± 17.8
95.2 ± 4.4
13.1 ± 11.1
16.2 ± 12.1
Control extremity during CO2 therapy 21.5 ± 13.9 8.6 ± 8.2
35.6 ± 21.7
95.0 ± 4.8
6.2 ± 5.1∗
Studied extremity during CO2 therapy 19.0 ± 12.6∗ 4.8 ± 3.1∗
57.2 ± 16.9∗
96.3 ± 2.5
9.2 ± 9.1
LF power [n.u.]
Notes: ULF – ultra low frequency; VLF – very low frequency; LF – low frequency; HF – high frequency; VHF – very high frequency; n.u. – normalized units; ∗ – statistically significant difference before and during CO2 therapy for the studied extremity (p < 0.05). Data are expressed as mean ± SD.
There was no correlation between the LD flux percentage change in the studied extremity during CO2 therapy and the mean LD flux in the studied extremity during rest (r2 = 0.1452). There was also no correlation between the LD flux percentage change in the studied extremity during CO2 therapy and the temperature inside the sealed bag (r2 = 0.0008). 4. Discussion In our experimental setting, CO2 therapy caused a statistically significant increase in the LD flux of the studied extremity, whereas change in the LD flux of the control extremity was not statistically significant. We did not find any systemic effects except a minor decrease in heart rate. The change in humidity and temperature in the bag containing the studied extremity did not change and therefore could not be the cause of the LD flux increase. The LD flux change in the studied extremity is most likely an indirect sign of successful diffusion of CO2 molecules through the skin into microcirculation and a direct indicator of the vasodilatory effect of CO2 . The results of the LD flux in the control extremity demonstrate that the temperature and humidity
430
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
Fig. 2. (A) Skin temperature at the site of LD flux measurement on both extremities during rest. (B) Skin temperature at the site of LD flux measurement on both extremities during CO2 therapy. All bars represent mean ± SE; start of rest period at t = 0 s, end of rest period at t = 360 s; start of the CO2 therapy at t = 0 s, end at t = 2100 s; ∗ - p = 0.049.
conditions inside the sealed bag did not significantly influence the blood flow in the skin. This fact additionally strengthens the case for CO2 having caused the statistically significant LD flux change in the studied extremity. There were huge variations among the subjects regarding the LD flux percentage change values in the studied extremity during CO2 therapy. In addition, there was no correlation between the LD flux percentage change in the studied extremity during CO2 therapy and the mean LD flux in the studied extremity during rest as well as no correlation between the LD flux percentage change in the studied extremity during CO2 therapy and the temperature inside sealed bag.
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
431
Fig. 3. (A) Relative humidity (RH) in the sealed bag with the studied extremity during CO2 therapy. (B) Temperature (Ts ) in the sealed bag with the studied extremity during CO2 therapy. Logarithmic trend curve, its equation and corresponding r-squared value (r2 ) are shown in upper right corner of each figure. All bars represent mean ± SE; start of the CO2 therapy at t = 0 s, end at t = 2100 s.
The hypotheses upon which we correlated chosen pairs of variables were: (1) the percentage change of LD flux in the studied extremity during CO2 therapy is larger, when mean LD flux in the studied extremity during rest is lower and (2) the percentage change of LD flux in the studied extremity during CO2 therapy is larger, if temperature inside the sealed bag is higher. With calculating the percentage change we wanted to present data for studied extremity from Figs. 1A and 1B on a same Fig. (1 C). This was done in order to stress out the LD change on studied extremity after CO2 therapy in comparison to the rest period. We also used the LD flux percentage change values for the calculation of a correlation between the chosen pairs of variables.
432
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
The variations in the values of the LD flux percentage change are therefore most likely a result of differences in skin hairiness, skin thickness and the amount and composition of sweat. With greater hairiness, the movement of CO2 molecules may be impeded. Increased skin thickness slows the transcutaneous diffusion of CO2 (diffusion rate and total amount of absorbed dose are diminished [31]). A greater amount of sweat production increases the surface of moistened skin, whereas changes in sweat composition influence its pH. In physiological conditions an increased amount of sweat diminishes the transcutaneous CO2 diffusion rate into the surrounding air [28]. The same is most likely true for diffusion in opposite direction. On the other hand, increased hydration of the stratum corneum destroys the barrier function of this skin layer and consequently increases skin permeability [29]. Changes in systolic and diastolic pressure were not statistically significant. In addition, the spectral analysis of the RR interval showed no change in LF/HF and LF / (LF + HF) ratios. The small but statistically significant decrease in heart rate is most likely a result of the prolonged resting condition during CO2 therapy and not that of a systemic CO2 effect. These findings suggest that the implemented CO2 therapy did not cause noticeable systemic effects. Absence of these effects might speak in favour of using this therapy in patients with cardiac diseases without causing adverse side effects. The statistically significant decreases of the VLF (neurogenic activity), LF (myogenic activity) and HF (respiration rate) power values of the LD flux in the studied extremity are most likely a result of vasodilatation, whereas the increased VHF power is a result of a decrease in heart rate during the therapy. There were no statistically significant frequency oscillations present in the LD flux of the control extremity, which speaks in favour of the absence of systemic vasodilatory effects of applied CO2 . Our results are in accordance with results of previous studies showing an increase in measured LD flux during exercise [5, 27], representing active hyperaemia. We analysed spectra of LD flux signal just for the last few minutes of each studied period, when cardiovascular parameters were not changing significantly. This means that transient phenomena were avoided and use of non-linear analysis was not needed. We also believe that a Fast Fourier transform is most commonly used when one analyses LD flux signal. By using it, we therefore enabled researchers an easier potential comparison of their results with ours. Possible reason for the statistically insignificant changes in the skin temperature on the studied extremity is due to the physiologic homeostatic mechanisms. We studied healthy subjects and their skin temperature already rose during rest period. Due to the sweating that occurred during the CO2 therapy on the studied extremity, excessive heat was transferred from a body and that prevented a rise of the skin temperature. With our chosen population the effect of sex and age on the measurement results were eliminated. The effect of sex is expected to be the strongest in fertile female subjects due to a significant oscillation of sex hormone blood levels, which has an important impact on the cardiovascular system [6]. Women’s skin is also thinner than that of men [12]. Sex hormones additionally influence the thickness of dermis and the skin surface pH [8]. With chosen age group we avoided the functional and structural (irreversible) changes in skin microcirculation that occur with aging [19]. Our decision, using the area of the thigh as a measuring spot, enabled us to shorten the accommodation time. We expected that there would be more time needed to ensure the homogeneity among the studied subjects when it comes to skin temperature of more distal parts of the lower extremity. If homogeneity was not ensured, it would have influenced the dispersion of the results. There are some irregularities in our study. Firstly, we integrated the results of two subjects with substantial weekly physical exercise and two subjects with asthma. As a result, our sample involved biological variability. The reasons for the integration of asthmatics are as follow: one subject had not used any medicaments for several years and the second one regularly used glucocorticoide in such low
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
433
dose that no vasoactive effects were expected. Secondly, there were only 14 successful measurements of relative humidity, since it was not possible to properly protect the sensor from water vapour condensate inside a sealed bag. Thirdly, the shape of the LD flux vs. time curve for the studied extremity as in Fig. 1B suggests that it would have ended asymptotically if we prolonged the duration of the CO2 therapy. This means that we could have possibly achieved stationary conditions if the therapies lasted longer. Knowing this is important for defining the optimum CO2 therapy duration in case of treating chronic wounds. During the research an appropriate temperature, illumination and ventilation were maintained; the proliferation of noise from outdoors was prevented; subjects were prohibited to speak and move during measurements; they were instructed not to consume any vasoactive substances at least 8 hours before the start of the measurements. With these measures we wanted to reduce sympathetic nervous system activity in subjects during the measurements. We minimized the biologic variability by using each subject as his own control. With the chosen protocol we eliminated the effect of positive pressure on CO2 diffusion, so the only driving force for diffusion was the concentration gradient. The combination of a significantly increased LD flux in the studied extremity, the absence of systemic effects and the Bohr effect [25] speak in favour of using a transcutaneous application of gaseous CO2 for healing chronic wounds. To the best of our knowledge, there are no documented studies about the effects of transcutaneously applied gaseous CO2 for the healing of chronic wounds. However, Brandi et al. [4] studied the effects of a subcutaneous administration of gaseous CO2 in the treatment of chronic lower limb lesions. They showed a significant increase in transcutaneous pO2 and LD flux in the studied group following treatment. It is to be noted that the studied group was treated using CO2 (therapies were performed twice a week for six weeks) and also advanced dressings. The control group was treated using advanced dressings only. The study lasted 60 days. Hayashi et al. [15] showed that CO2 enriched water (CO2 water bath) prevents the expansion and formation of ischemic ulcers after surgical revascularization in diabetic patients with critical limb ischemia. The free concentration of CO2 in the employed bath was 1000–1200 mg/l. There are also a number of other possible applications for dry CO2 therapies. Previous studies showed positive effects in the treatment of patients with myocardial infarction [2], Raynaud’s phenomenon [26] and intermittent claudication [13]. Onishi et al. [24] demonstrated that a transcutaneous application of gaseous CO2 leads to mitochondria-mediated apoptosis and impaired tumour growth (human malignant histiocytoma) in vivo. The work of Oe et al. [23] revealed a possible therapeutic potential for muscular strength recovery. Harada et al. [14] demonstrated the improvement of hypoxic conditions in metastatic osteosarcoma, the induction of apoptosis in osteosarcoma and the inhibition of pulmonary metastasis of metastatic osteosarcoma.
5. Conclusions Transcutaneously applied gaseous CO2 caused a statistically significant increase in the LD flux of the studied extremity only; increased temperature and relative humidity had negligible effect on that phenomenon. The reduction of heart rate was statistically significant but most likely due to the additional rest effect during CO2 therapy. Changes in systolic and diastolic pressure were not statistically significant. We did not find any systemic effects due to the employed CO2 therapy. We conclude that a transcutaneous application of gaseous CO2 has a significant effect on skin microcirculation and has local effects.
434
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
Acknowledgments The transcutaneous system for administration of gaseous CO2 CARBOfit® was provided by DermaArt (Breˇzice, Slovenia). The company was not involved in the study, the writing of the manuscript or the decision for the submission of the manuscript for publication. None of the authors have competing financial or non-financial interests in the work presented in this manuscript. References [1] W. Alhejily, A. Aleksi, B.J. Martin and T.J. Anderson, The effect of ischemia-reperfusion injury on measures of vascular function, Clinical Hemorheology and Microcirculation 56 (2014), 265–271. [2] N.L. Barashkova, N.L. Kartamysheva, V.P. Krasnova, L.N. Kriuchkova and E.S. Miasoedova, “Dry” carbon dioxide baths in treating patients with myocardial infarction at the sanatorium stage of rehabilitation, Kliniceskaia Meditsina (Moskva) 67 (1989), 38–41. [3] D.A. Blair, W.E. Glover, L. McArrdle and I.C. Roddie, The mechanism of the peripheral vasodilatation following carbon dioxide inhalation in man, Clinical Science 19 (1960), 407–413. [4] C. Brandi, L. Grimaldi, G. Nisi, A. Brafa, A. Campa, M. Calabr`o, M. Campana and C. D’Aniello, The role of carbon dioxide therapy in the treatment of chronic wounds, In Vivo 24 (2010), 223–226. [5] R. Brown, U. Kemp and V. Macefield, Increases in muscle sympathetic nerve activity, heart rate, respiration, and skin blood flow during passive viewing of exercise, Frontiers in Neuroscience 7 (2013), 102. [6] K. Cankar, Z. Finderle and M. Strucl, Gender differences in cutaneous laser doppler flow response to local direct and contralateral cooling, Jorunal of Vascular Research 37 (2000), 183–188. [7] P.S. Clifford and Y. Hellsten, Vasodilatory mechanisms in contracting skeletal muscle, Journal of Applied Physiology 97 (2004), 393–403. [8] H. Dao Jr. and R.A. Kazin, Gender differences in skin: A review of the literature, Gender Medicine 4 (2007), 308–328. [9] A. Diji and A.D. Greenfield, The local effect of carbon dioxide on human blood vessels, American Heart Journal 60 (1960), 907–914. [10] A. Diji, Local vasodilator action of carbon dioxide on blood vessels of the hand, Journal of Applied Physiology 14 (1959), 414–416. [11] B.R. Duling, Changes in microvascular diameter and oxygen tension induced by carbon dioxide, Circulation Research 32 (1973), 370–376. [12] C. Escoffier, J. de Rigal, A. Rochefort, R. Vasselet, J.L. L´evˆeque and P.G. Agache, Age-related mechanical properties of human skin: An in vivo study, Journal of Investigative Dermatology 93 (1989), 353–357. [13] R. Fabry, P. Monnet, J. Schmidt, J.R. Lusson, P.H. Carpentier, J.C. Baguet and C. Dubray, Clinical and microcirculatory effects of transcutaneous CO2 therapy in intermittent claudication. Randomized double-blind clinical trial with a parallel design, Vasa 38 (2009), 213–224. [14] R. Harada, T. Kawamoto, T. Ueha, M. Minoda, M. Toda, Y. Onishi, N. Fukase, H. Hara, Y. Sakai, M. Miwa, R. Kuroda, M. Kurosaka and T. Akisue, Reoxygenation using a novel CO2 therapy decreases the metastatic potential of osteosarcoma cells, Experimental Cell Research 319 (2013), 1988–1997. [15] H. Hayashi, S. Yamada, Y. Kumada, H. Matsuo, T. Toriyama and H. Kawahara, Immersing feet in carbon dioxide-enriched water prevents expansion and formation of ischemic ulcers after surgical revascularization in diabetic patients with critical limb ischemia, Annals of Vascular Diseases 1 (2008), 111–117. [16] H.W. Huang, I.M. Jou, C.K. Wang, P.Y. Chen, W.C. Wang and C.C. Lin, Power spectral analyses of index finger skin blood perfusion in carpal tunnel syndrome and diabetic polyneuropathy, Experimental Diabetes Research (2011), 465910. [17] T. Ito, J.I. Moore and M.C. Koss, Topical application of CO2 increases skin blood flow, Journal of Investigative Dermatology 93 (1989), 259–262. [18] M. Klemenc, Aktivnost avtonomnega zˇ ivˇcnega sistema in diastoliˇcna disfunkcija levega prekata pri mladih hipertonikih, Zdravniˇski Vestnik 70 (2001), 627–635. [19] L. Li, S. Mac-Mary, J.M. Sainthillier, S. Nouveau, O. de Lacharriere and P. Humbert, Age-related changes of the cutaneous microcirculation in vivo, Gerontology 52 (2006), 142–153.
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
435
[20] A. Malliani, F. Lombardi and M. Pagani, Power spectrum analysis of heart rate variability: A tool to explore neural regulatory mechanisms, British Heart Journal 71 (1994), 1–2. [21] M. Minamiyama and A. Yamamoto, Direct evidence of the vasodilator action of carbon dioxide on subcutaneous microvasculature in rats by use of intra-vital video-microscopy, Journal of Biorheology 24 (2010), 42–46. [22] T.A. Mustoe, K. O’Shaughnessy and O. Kloeters, Chronic wound pathogenesis and current treatment strategies: A unifying hypothesis, Plastic and Reconstructive Surgery 117 (2006), 35S–41S. [23] K. Oe, T. Ueha, Y. Sakai, T. Niikura, S.Y. Lee, A. Koh, T. Hasegawa, M. Tanaka, M. Miwa and M. Kurosaka, The effect of transcutaenous application of carbon dioxide (CO2) on skeletal muscle, Biochemical and Biophysical Research Communications 407 (2011), 148–152. [24] Y. Onishi, T. Kawamoto, T. Ueha, H. Hara, N. Fukase, M. Toda, R. Harada, Y. Sakai, M. Miwa, K. Nishida, M. Kurosaka and T. Akisue, Transcutaneous application of carbon dioxide (CO2) enhances chemosensitivity by reducing hypoxic conditions in human malignant fibrous histiocytoma, Journal of Cancer Science and Therapy 4 (2012), 174–181. [25] Y. Sakai, M. Miwa, K. Oe, T. Ueha, A. Koh, T. Niikura, T. Iwakura, S.Y. Lee, M. Tanaka and M. Kurosaka, A novel system for transcutaneous application of carbon dioxide causing an »artificial Bohr effect« in the human body, PloS One 6 (2011), e24137. [26] J. Schmidt, P. Monnet, B. Normand and R. Fabry, Microcirculatory and clinical effects of serial percutaneous application of carbon dioxide in primary and secondary Reynaud’s phenomenon, Vasa 34 (2005), 93–100. [27] J. Smolander and P. Kolari, Laser-Doppler and plethysmographic skin blood flow during exercise and during acute heat stress in the sauna, European Journal of Applied Physiology and Occupational Physiology 54 (1985), 371–377. [28] A. Szakall, Die Ver¨anderungen der obersten Hautschichten durch den Dauergebrauch einiger Handwaschmittel, Arbeitsphysiologie 13 (1943), 49–56. [29] G. Tan, P. Xu and L.B. Lawson, J. He, L.C. Freytag, J.D. Clements and V.T. John, Hydration effects on skin microstructure as probed by high-resolution cyro-scanning electron microscopy and mechanistic implications to enhanced transcutaneous delivery of biomacromolecules, Jorunal of Pharmaceutical Sciences 99 (2010), 730–740. [30] F. Werdin, M. Tennenhaus, H.E. Schaller and H.O. Rennekampff, Evidence-based management strategies for treatment of chronic wounds, ePlasty 9 (2009), e19. [31] S.C. Wilkinson, W.J. Maas, J.B. Nielsen, L.C. Greaves, J.J. van de Sandt and F.M. Williams, Interactions of skin thickness and physicochemical properties of test compounds in percutaneous penetration studies, International Archives of Occupational and Environmental Health 79 (2006), 405–413.
423
Effect of transcutaneous application of gaseous carbon dioxide on cutaneous microcirculation Miha Finzgar, Ziva Melik and Ksenija Cankar∗ Institute of Physiology, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia
Abstract. BACKGROUND: The inefficient healing of chronic wounds is a result of poor blood perfusion at the wound and surrounding tissues. Artificially applied carbon dioxide (CO2 ) has the potential to improve the perfusion and oxygenation of tissues, hence is useful for the healing of chronic wounds. OBJECTIVE: The aim of the present study was to determine the effect of a transcutaneous application of physiological vasodilator gaseous CO2 on cutaneous blood flow. METHODS: Laser Doppler (LD) flux in cutaneous microcirculation, skin temperature, electrocardiogram and arterial blood pressure were measured simultaneously in a group of 33 healthy men, aged 21–28 years, during rest and a 35-minute CO2 therapy. One lower limb of each subject represented the studied extremity, being exposed to gaseous CO2 . The contralateral limb was the control, being exposed to air. Each limb was sealed in a plastic bag. RESULTS: During CO2 therapy the LD flux in the studied extremity increased from 5.8 PU ± 3.9 PU to 30.3 PU ± 16.7 PU (mean ± standard deviation; paired t-test, p < 0.001), while that in the control extremity did not change significantly. CONCLUSIONS: Our results confirm a local vasodilatory effect of applied CO2 therapy. This finding indicates its potential clinical use. Keywords: Physiological measurements, laser Doppler fluxmetry, CO2 therapy, dry CO2 bath
1. Introduction Chronic wounds are wounds that fail to heal over a period of 3 months [22]. The aetiology of chronic wounds is very diverse: venous insufficiency, insufficient arterial blood supply, diabetes mellitus and various systemic causes, such as nutritional abnormalities, immunosuppression and infection [30]. An important characteristic of these wounds is tissue hypoxia [4]. Carbon dioxide (CO2 ) along with some other metabolites causes vasodilation of blood vessels [1], which in turn increases blood flow. This phenomenon is called active hyperaemia [7]. The exact mechanism of this type of hyperaemia has not yet been discovered. It is most likely a result of an integrated response of multiple vasodilatation mechanisms [7]. Carbon dioxide therapy (CO2 therapy) refers to the transcutaneous or subcutaneous application of CO2 for therapeutic purpose [3, 25]. One example of these therapies is the transcutaneous application of gaseous CO2 , also known as dry CO2 bath. This type of CO2 application is not invasive compared to the ∗ Corresponding author: Ksenija Cankar, Institute of Physiology, Zaloska 4, 1000 Ljubljana Slovenia. Tel.: +386 1 5437500; Fax: +386 1 5437501; E-mail: [email protected].
1386-0291/15/$35.00 © 2015 – IOS Press and the authors. All rights reserved
424
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
subcutaneous method. In comparison to CO2 water baths, it does not involve the risk of inflammation and therefore could be used for patients with chronic wounds. In addition, it does not increase the concentration of CO2 in surrounding air if proper security measures are taken. The effects of all three types of CO2 therapies on skin microcirculation have previously been studied in animal models. Duling [11] observed increased microvascular diameter and increased perivascular pO2 at all sites during the exposure of hamsters’ pouches to the solution, with pCO2 = 32 mmHg and pO2 = 18 ± 2 mmHg. The author reported that increased diameter was attributable to the vasodilatory effect of CO2 , whereas the increased perivascular pO2 was due to the effect of CO2 on oxyhemoglobin dissociation curve [11]. Ito et al. [17] showed that a topical application of CO2 water to a rat’s skin increases skin blood flow about a 100 % at bath temperatures of 23◦ C and 34◦ C. Minamiyama and Yamamoto [21] demonstrated the in vivo effect of a topical application of CO2 water bath on the microcirculation of rats’ skin. By way of intra-vital video microscopy, they observed vasodilatation together with an increase in blood flow; also observed was a decrease in the pH of subcutaneous tissue. In addition to these studies, the work of Diji [10] in 1959 revealed an indirect vasodilatation effect of CO2 : a reported increase in the rate of heat elimination to water at about 40 % (at 29◦ C) due to the immersion of a hand in a CO2 saturated water bath. Diji and Greenfield [9] also showed that a subcutaneous application of gaseous CO2 caused local vasodilatation. By using near-infrared spectroscopy Sakai et al. [25] demonstrated that a transcutaneous CO2 application caused O2 dissociation from oxygenated haemoglobin, which is characteristic of the Bohr effect. To our knowledge, there were no studies investigating the effect of a transcutaneous application of gaseous CO2 on cutaneous blood flow in vivo in humans. The aim of the present study was to determine the effect of a transcutaneous application of gaseous carbon dioxide on cutaneous blood flow as well as possible systemic effects. Also assessed was the vasodilatory effect of elevated temperature and relative humidity on cutaneous blood flow in the sealed bag system, as was used during the CO2 therapy. 2. Subjects and methods 2.1. Subjects The subjects of the present study were 33 male volunteers who were without acute diseases, aged 23.8 ± 2.3 years (from 21 to 28 years) and who had a body mass index of 24.2 ± 2.9 kg/m2 (from 17.3 to 30.1 kg/m2 ). Five of them were smokers, two were asthmatics and two of them had a family history of arterial hypertension (at least one parent of a chosen subject received medical treatment for his/her arterial hypertension before the age of 55). All participating subjects refrained from smoking or drinking coffee or tea for at least 8 hours before the start of the measurements. The present study was carried out with the approval of National Medical Ethics Committee. A written consent was obtained from each subject. 2.2. Electrocardiogram and arterial blood pressure The electrocardiogram (ECG) (Institute of Physiology, Ljubljana, Slovenia) in standard lead II and arterial blood pressure (Ohmeda Finapres 2300, Finapres Medical Systems, Amsterdam, Netherlands) were monitored continuously. Blood pressure was measured on a right middle finger.
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
425
2.3. Laser Doppler flux Laser Doppler (LD) flux sensors (Angled probe 401, Perimed, J¨arf¨alla, Sweden) together with laser light sources at 780 nm (PF 4001 and PF 4002 Satellite, Perimed, J¨arf¨alla, Sweden) were used for a continuous measurement of LD flux. It was monitored simultaneously on both thighs, medial and somewhat proximal to the knee. 2.4. Skin temperature Skin temperature at the site of the LD flux measurement was recorded with a skin spot sensor (Sensor 442, Perimed, J¨arf¨alla, Sweden). It was measured continuously; values were read at 60-second intervals. The temperature of the big toes was measured with an infrared thermometer (Voltcfraft IR 800-20D, Voltcraft, London, UK) three times after acclimatization, just before the start of the CO2 therapy and immediately after its end. 2.5. Temperature and relative humidity The temperature and relative humidity in the bag used for the studied extremity were measured by a high-accuracy, ambient air probe (Testo 0636 9741, Testo, Lenzkirch, Germany) attached to a multifunction measuring instrument (Testo 400, Testo, Lenzkirch, Germany). Both parameters were measured continuously; values were read at 60-second intervals. 2.6. System for the transcutaneous application of gaseous CO2 Gaseous CO2 with 99.995 % purity (CO2 ; Messer Slovenia, Ruˇse, Slovenia) was applied transcutaneously with the CARBOfit® system (DermaArt, Breˇzice, Slovenia). This system consists of a compressor, a CO2 sensor, a conduit tube for CO2 and a drain tube for air/CO2 . CO2 was applied in single-use, low-density, polyethylene bags (thickness 0.4 mm). 2.7. Protocol Before the start of the measurement, each subject filled out a questionnaire for the purpose of obtaining his basic data. During measurements, subjects were in lying position without shoes, socks and pants. They were instructed not to talk or move during the measurements. After installing all the necessary sensors, each lower extremity was packed inside a separate bag and sealed with elastic fastener tape in the upper third part of each thigh. One lower limb represented the studied extremity and the contralateral one served as the control. The studied extremity was randomly chosen for each subject. Temperature and relative humidity sensors were installed in the bag with studied extremity. The time needed for preparation served as an acclimatization time (10 minutes). Acclimatization was followed by a 6-minute-long resting period. After the resting period, the air in the bag with the studied extremity was pumped out and the bag was filled with CO2 . The therapy lasted 35 minutes. Laser Doppler flux and skin temperature in both lower extremities, ECG and arterial blood pressure were measured simultaneously during rest and during the entire CO2 therapy.
426
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
Measurements were performed in a room with an air temperature of 24.8 ± 0.7◦ C, air pressure at 980 ± 2 mbar and relative humidity of 54.1 ± 4.1 % (all these values are expressed as mean ± standard deviation of the mean - SD). 2.8. Analog to digital conversion of data The analog signals of ECG, arterial blood pressure and LD flux were converted (sampling rate of 500 s−1 was used) with an analog-digital converter (DASport, Burr-Brown Corporation, Tucson, USA). 2.9. Statistical analysis Digital signals of LD flux, arterial blood pressure and RR intervals from ECG recording were analyzed with NevrokardTM (Nevrokard Kiauta, Izola, Slovenia), a professional software tool that provides analysis of different data acquisition system file formats. We calculated the percentage change of the mean LD flux in the studied extremity during CO2 therapy in comparison to the mean value of LD flux in the studied extremity during the resting period: LD flux%i =
LD flux2i - LD flux1i · 100%, LD flux1i
(1)
where LD flux%i is an i value of LD flux percentage change; LD flux1i an i mean value of LD flux break during the rest period; LD flux2i an i mean value of LD flux during CO2 therapy; and i = 1, 2, . . . , 33 (index i marks each participating subject). For the assessment of the linear correlation between the chosen pairs of variables a Pearson productmoment correlation coefficient was calculated (results are given as squared values). With the method of least squares, temperature vs. time and relative humidity vs. time curves were approximated with a loharthmic funciton (trend curve) and corresponding coefficients of determination were calculated. Values measured during rest and during CO2 therapy were compared with a paired t-test. A p-value of less than 0.05 was considered as statistically significant. 2.10. Spectral analysis Changes in sympathovagal balance are usually assessed with a spectral analysis of heart rate or RR interval [20]. For spectral analysis of the RR interval, we used autoregression (256 points; autoregressive model order: 13). We defined three frequency bands [18]: VLF band (very low frequency; 0.01–0.04 Hz), LF band (low frequency; 0.04–0.15 Hz) and HF band (high frequency; 0.15–0.4 Hz). For the spectral analysis of the LD flux signal, a Fast Fourier transform (1024 points; Hanning window; normalized power spectral density) was used. Five frequency bands were defined [16]: ULF band (ultra low frequency; 0.001–0.02 Hz), VLF band (very low frequency; 0.02–0.06 Hz), LF band (low frequency; 0.06–0.15 Hz), HF band (high frequency; 0.15–0.6 Hz) and VHF band (very high frequency; 0.6–2 Hz). 3. Results The data of all participating subjects were analyzed.
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
427
Table 1 Values of measured parameters during rest and CO2 therapy Measured quantity ν [beats/min] ps [mmHg] pd [mmHg] LD flux [PU] – control extremity LD flux [PU] – studied extremity T [◦ C] – control extremity T [◦ C] – studied extremity
Last 3 minutes of rest period (period before CO2 therapy) 64.9 ± 8.8 108.0 ± 13.6 63.4 ± 8.6 5.0 ± 2.9 5.8 ± 3.9 32.80 ± 1.15 32.82 ± 0.93
Last 3 minutes of CO2 therapy 62.5 ± 7.3∗ 109.9 ± 12.5 63.4 ± 6.6 7.1 ± 4.9 30.3 ± 16.7†,‡ 33.63 ± 1.18 33.59 ± 0.99
Notes: ν – heart rate; ps – systolic pressure; pd – diastolic pressure; LD flux – laser Doppler flux; T – skin temperature; PU – perfusion units; ∗ – statistically significant difference before and during CO2 therapy (p = 0.003); † – statistically significant difference before and during CO2 therapy for the studied extremity (p < 0.001); ‡ – statistically significant difference between the control and studied extremity during CO2 therapy (p < 0,001). Data are expressed as mean ± SD.
Table 1 shows the values of the measured parameters during the resting period and during the CO2 therapy. All values are given for a 3-minute interval of rest and the last 3 minutes of CO2 therapy, as the most evident effects were expected toward the end of the CO2 therapy. Choosing the 3-minute interval for both periods also enabled the spectral analysis of the LD flux and RR interval. The decrease in heart rate during CO2 therapy compared to the values obtained at rest was statistically significant (paired t-test, p = 0.003). In contrast, the systolic and diastolic pressure did not change significantly. The mean value of LD flux in the cutaneous microcirculation of the studied extremity increased significantly during CO2 therapy in comparison to the mean value during the resting period (paired t-test, p < 0.001). An increase was present in all subjects. In contrast, the change in the mean value of LD flux in cutaneous microcirculation in control extremity was not significant. The difference between LD flux during CO2 therapy in the control and studied extremity was also statistically significant (paired t-test, p < 0.001). Figure 1 shows (A) the LD flux in both extremities during rest and (B) during the entire duration of CO2 therapy. Figure 1 C shows the LD flux percentage change values for the studied extremity during CO2 therapy. Table 2 displays the results of the spectral analysis of the RR interval. The differences between the values before and during the CO2 therapy are not statistically significant. Table 3 provides the results of the spectral analysis of LD flux. Change in the control extremity is not statistically significant, whereas in the studied limb the VLF (paired t-test, p < 0.001), LF (paired t-test, p = 0.015) and HF (paired t-test, p = 0.037) power decreased significantly, while the VHF power increased (paired t-test, p < 0.001). Differences in the values of skin temperature at the site of the LD flux measurements during rest and during CO2 therapy were not statistically significant, neither for the studied nor for the control extremity (except for the point at t = 60 s; paired t-test, p = 0.049; see Fig. 2). There were also no statistically significant differences between the values of skin temperature on plantar side of the big toe on either extremity. Temperature and relative humidity versus time curves (Fig. 3), describing conditions in the sealed bag with the studied extremity, closely follow a logarithmic function.
428
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide A
B
C
Fig. 1. (A) LD flux in the control and studied extremity during rest. (B) LD flux in the control and studied extremity during CO2 therapy. (C) LD flux percent change in the studied extremity during CO2 therapy. All bars represent mean ± standard error (SE); start of rest period at t = 0 s, end of rest period at t = 360 s; start of the CO2 therapy at t = 0 s, end at t = 2100 s; ∗ - p < 0.001.
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
429
Table 2 Spectral analysis of RR interval LF power [n.u.]
HF power [n.u.]
LF/HF [/]
LF/(LF + HF) [/]
Total power [n.u.]
75.9 ± 32.7
48.0 ± 17.6
Rest period (last 3 minutes) 2.2 ± 2.0
0.60 ± 0.18
158.2 ± 54.3
77.0 ± 30.1
52.3 ± 16.4
CO2 therapy (last 3 minutes) 1.8 ± 1.3
0.56 ± 0.19
167.8 ± 52.3
Notes: LF – low frequency; HF – high frequency; n.u. – normalized units. Data are expressed as mean ± SD. Results of very low frequency (VLF) band are not shown. Table 3 Spectral analysis of LD signal ULF power [n.u.]
VLF power [n.u.]
13.3 ± 9.0
HF power [n.u.]
VHF power [n.u.]
Total power [n.u.]
17.6 ± 12.0
Control extremity during rest 26.4 ± 12.4 7.6 ± 4.6
29.6 ± 17.3
94.4 ± 5.4
14.0 ± 11.7
17.6 ± 12.2
Studied extremity during rest 26.8 ± 16.5 7.8 ± 7.8
29.0 ± 17.8
95.2 ± 4.4
13.1 ± 11.1
16.2 ± 12.1
Control extremity during CO2 therapy 21.5 ± 13.9 8.6 ± 8.2
35.6 ± 21.7
95.0 ± 4.8
6.2 ± 5.1∗
Studied extremity during CO2 therapy 19.0 ± 12.6∗ 4.8 ± 3.1∗
57.2 ± 16.9∗
96.3 ± 2.5
9.2 ± 9.1
LF power [n.u.]
Notes: ULF – ultra low frequency; VLF – very low frequency; LF – low frequency; HF – high frequency; VHF – very high frequency; n.u. – normalized units; ∗ – statistically significant difference before and during CO2 therapy for the studied extremity (p < 0.05). Data are expressed as mean ± SD.
There was no correlation between the LD flux percentage change in the studied extremity during CO2 therapy and the mean LD flux in the studied extremity during rest (r2 = 0.1452). There was also no correlation between the LD flux percentage change in the studied extremity during CO2 therapy and the temperature inside the sealed bag (r2 = 0.0008). 4. Discussion In our experimental setting, CO2 therapy caused a statistically significant increase in the LD flux of the studied extremity, whereas change in the LD flux of the control extremity was not statistically significant. We did not find any systemic effects except a minor decrease in heart rate. The change in humidity and temperature in the bag containing the studied extremity did not change and therefore could not be the cause of the LD flux increase. The LD flux change in the studied extremity is most likely an indirect sign of successful diffusion of CO2 molecules through the skin into microcirculation and a direct indicator of the vasodilatory effect of CO2 . The results of the LD flux in the control extremity demonstrate that the temperature and humidity
430
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
Fig. 2. (A) Skin temperature at the site of LD flux measurement on both extremities during rest. (B) Skin temperature at the site of LD flux measurement on both extremities during CO2 therapy. All bars represent mean ± SE; start of rest period at t = 0 s, end of rest period at t = 360 s; start of the CO2 therapy at t = 0 s, end at t = 2100 s; ∗ - p = 0.049.
conditions inside the sealed bag did not significantly influence the blood flow in the skin. This fact additionally strengthens the case for CO2 having caused the statistically significant LD flux change in the studied extremity. There were huge variations among the subjects regarding the LD flux percentage change values in the studied extremity during CO2 therapy. In addition, there was no correlation between the LD flux percentage change in the studied extremity during CO2 therapy and the mean LD flux in the studied extremity during rest as well as no correlation between the LD flux percentage change in the studied extremity during CO2 therapy and the temperature inside sealed bag.
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
431
Fig. 3. (A) Relative humidity (RH) in the sealed bag with the studied extremity during CO2 therapy. (B) Temperature (Ts ) in the sealed bag with the studied extremity during CO2 therapy. Logarithmic trend curve, its equation and corresponding r-squared value (r2 ) are shown in upper right corner of each figure. All bars represent mean ± SE; start of the CO2 therapy at t = 0 s, end at t = 2100 s.
The hypotheses upon which we correlated chosen pairs of variables were: (1) the percentage change of LD flux in the studied extremity during CO2 therapy is larger, when mean LD flux in the studied extremity during rest is lower and (2) the percentage change of LD flux in the studied extremity during CO2 therapy is larger, if temperature inside the sealed bag is higher. With calculating the percentage change we wanted to present data for studied extremity from Figs. 1A and 1B on a same Fig. (1 C). This was done in order to stress out the LD change on studied extremity after CO2 therapy in comparison to the rest period. We also used the LD flux percentage change values for the calculation of a correlation between the chosen pairs of variables.
432
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
The variations in the values of the LD flux percentage change are therefore most likely a result of differences in skin hairiness, skin thickness and the amount and composition of sweat. With greater hairiness, the movement of CO2 molecules may be impeded. Increased skin thickness slows the transcutaneous diffusion of CO2 (diffusion rate and total amount of absorbed dose are diminished [31]). A greater amount of sweat production increases the surface of moistened skin, whereas changes in sweat composition influence its pH. In physiological conditions an increased amount of sweat diminishes the transcutaneous CO2 diffusion rate into the surrounding air [28]. The same is most likely true for diffusion in opposite direction. On the other hand, increased hydration of the stratum corneum destroys the barrier function of this skin layer and consequently increases skin permeability [29]. Changes in systolic and diastolic pressure were not statistically significant. In addition, the spectral analysis of the RR interval showed no change in LF/HF and LF / (LF + HF) ratios. The small but statistically significant decrease in heart rate is most likely a result of the prolonged resting condition during CO2 therapy and not that of a systemic CO2 effect. These findings suggest that the implemented CO2 therapy did not cause noticeable systemic effects. Absence of these effects might speak in favour of using this therapy in patients with cardiac diseases without causing adverse side effects. The statistically significant decreases of the VLF (neurogenic activity), LF (myogenic activity) and HF (respiration rate) power values of the LD flux in the studied extremity are most likely a result of vasodilatation, whereas the increased VHF power is a result of a decrease in heart rate during the therapy. There were no statistically significant frequency oscillations present in the LD flux of the control extremity, which speaks in favour of the absence of systemic vasodilatory effects of applied CO2 . Our results are in accordance with results of previous studies showing an increase in measured LD flux during exercise [5, 27], representing active hyperaemia. We analysed spectra of LD flux signal just for the last few minutes of each studied period, when cardiovascular parameters were not changing significantly. This means that transient phenomena were avoided and use of non-linear analysis was not needed. We also believe that a Fast Fourier transform is most commonly used when one analyses LD flux signal. By using it, we therefore enabled researchers an easier potential comparison of their results with ours. Possible reason for the statistically insignificant changes in the skin temperature on the studied extremity is due to the physiologic homeostatic mechanisms. We studied healthy subjects and their skin temperature already rose during rest period. Due to the sweating that occurred during the CO2 therapy on the studied extremity, excessive heat was transferred from a body and that prevented a rise of the skin temperature. With our chosen population the effect of sex and age on the measurement results were eliminated. The effect of sex is expected to be the strongest in fertile female subjects due to a significant oscillation of sex hormone blood levels, which has an important impact on the cardiovascular system [6]. Women’s skin is also thinner than that of men [12]. Sex hormones additionally influence the thickness of dermis and the skin surface pH [8]. With chosen age group we avoided the functional and structural (irreversible) changes in skin microcirculation that occur with aging [19]. Our decision, using the area of the thigh as a measuring spot, enabled us to shorten the accommodation time. We expected that there would be more time needed to ensure the homogeneity among the studied subjects when it comes to skin temperature of more distal parts of the lower extremity. If homogeneity was not ensured, it would have influenced the dispersion of the results. There are some irregularities in our study. Firstly, we integrated the results of two subjects with substantial weekly physical exercise and two subjects with asthma. As a result, our sample involved biological variability. The reasons for the integration of asthmatics are as follow: one subject had not used any medicaments for several years and the second one regularly used glucocorticoide in such low
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
433
dose that no vasoactive effects were expected. Secondly, there were only 14 successful measurements of relative humidity, since it was not possible to properly protect the sensor from water vapour condensate inside a sealed bag. Thirdly, the shape of the LD flux vs. time curve for the studied extremity as in Fig. 1B suggests that it would have ended asymptotically if we prolonged the duration of the CO2 therapy. This means that we could have possibly achieved stationary conditions if the therapies lasted longer. Knowing this is important for defining the optimum CO2 therapy duration in case of treating chronic wounds. During the research an appropriate temperature, illumination and ventilation were maintained; the proliferation of noise from outdoors was prevented; subjects were prohibited to speak and move during measurements; they were instructed not to consume any vasoactive substances at least 8 hours before the start of the measurements. With these measures we wanted to reduce sympathetic nervous system activity in subjects during the measurements. We minimized the biologic variability by using each subject as his own control. With the chosen protocol we eliminated the effect of positive pressure on CO2 diffusion, so the only driving force for diffusion was the concentration gradient. The combination of a significantly increased LD flux in the studied extremity, the absence of systemic effects and the Bohr effect [25] speak in favour of using a transcutaneous application of gaseous CO2 for healing chronic wounds. To the best of our knowledge, there are no documented studies about the effects of transcutaneously applied gaseous CO2 for the healing of chronic wounds. However, Brandi et al. [4] studied the effects of a subcutaneous administration of gaseous CO2 in the treatment of chronic lower limb lesions. They showed a significant increase in transcutaneous pO2 and LD flux in the studied group following treatment. It is to be noted that the studied group was treated using CO2 (therapies were performed twice a week for six weeks) and also advanced dressings. The control group was treated using advanced dressings only. The study lasted 60 days. Hayashi et al. [15] showed that CO2 enriched water (CO2 water bath) prevents the expansion and formation of ischemic ulcers after surgical revascularization in diabetic patients with critical limb ischemia. The free concentration of CO2 in the employed bath was 1000–1200 mg/l. There are also a number of other possible applications for dry CO2 therapies. Previous studies showed positive effects in the treatment of patients with myocardial infarction [2], Raynaud’s phenomenon [26] and intermittent claudication [13]. Onishi et al. [24] demonstrated that a transcutaneous application of gaseous CO2 leads to mitochondria-mediated apoptosis and impaired tumour growth (human malignant histiocytoma) in vivo. The work of Oe et al. [23] revealed a possible therapeutic potential for muscular strength recovery. Harada et al. [14] demonstrated the improvement of hypoxic conditions in metastatic osteosarcoma, the induction of apoptosis in osteosarcoma and the inhibition of pulmonary metastasis of metastatic osteosarcoma.
5. Conclusions Transcutaneously applied gaseous CO2 caused a statistically significant increase in the LD flux of the studied extremity only; increased temperature and relative humidity had negligible effect on that phenomenon. The reduction of heart rate was statistically significant but most likely due to the additional rest effect during CO2 therapy. Changes in systolic and diastolic pressure were not statistically significant. We did not find any systemic effects due to the employed CO2 therapy. We conclude that a transcutaneous application of gaseous CO2 has a significant effect on skin microcirculation and has local effects.
434
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
Acknowledgments The transcutaneous system for administration of gaseous CO2 CARBOfit® was provided by DermaArt (Breˇzice, Slovenia). The company was not involved in the study, the writing of the manuscript or the decision for the submission of the manuscript for publication. None of the authors have competing financial or non-financial interests in the work presented in this manuscript. References [1] W. Alhejily, A. Aleksi, B.J. Martin and T.J. Anderson, The effect of ischemia-reperfusion injury on measures of vascular function, Clinical Hemorheology and Microcirculation 56 (2014), 265–271. [2] N.L. Barashkova, N.L. Kartamysheva, V.P. Krasnova, L.N. Kriuchkova and E.S. Miasoedova, “Dry” carbon dioxide baths in treating patients with myocardial infarction at the sanatorium stage of rehabilitation, Kliniceskaia Meditsina (Moskva) 67 (1989), 38–41. [3] D.A. Blair, W.E. Glover, L. McArrdle and I.C. Roddie, The mechanism of the peripheral vasodilatation following carbon dioxide inhalation in man, Clinical Science 19 (1960), 407–413. [4] C. Brandi, L. Grimaldi, G. Nisi, A. Brafa, A. Campa, M. Calabr`o, M. Campana and C. D’Aniello, The role of carbon dioxide therapy in the treatment of chronic wounds, In Vivo 24 (2010), 223–226. [5] R. Brown, U. Kemp and V. Macefield, Increases in muscle sympathetic nerve activity, heart rate, respiration, and skin blood flow during passive viewing of exercise, Frontiers in Neuroscience 7 (2013), 102. [6] K. Cankar, Z. Finderle and M. Strucl, Gender differences in cutaneous laser doppler flow response to local direct and contralateral cooling, Jorunal of Vascular Research 37 (2000), 183–188. [7] P.S. Clifford and Y. Hellsten, Vasodilatory mechanisms in contracting skeletal muscle, Journal of Applied Physiology 97 (2004), 393–403. [8] H. Dao Jr. and R.A. Kazin, Gender differences in skin: A review of the literature, Gender Medicine 4 (2007), 308–328. [9] A. Diji and A.D. Greenfield, The local effect of carbon dioxide on human blood vessels, American Heart Journal 60 (1960), 907–914. [10] A. Diji, Local vasodilator action of carbon dioxide on blood vessels of the hand, Journal of Applied Physiology 14 (1959), 414–416. [11] B.R. Duling, Changes in microvascular diameter and oxygen tension induced by carbon dioxide, Circulation Research 32 (1973), 370–376. [12] C. Escoffier, J. de Rigal, A. Rochefort, R. Vasselet, J.L. L´evˆeque and P.G. Agache, Age-related mechanical properties of human skin: An in vivo study, Journal of Investigative Dermatology 93 (1989), 353–357. [13] R. Fabry, P. Monnet, J. Schmidt, J.R. Lusson, P.H. Carpentier, J.C. Baguet and C. Dubray, Clinical and microcirculatory effects of transcutaneous CO2 therapy in intermittent claudication. Randomized double-blind clinical trial with a parallel design, Vasa 38 (2009), 213–224. [14] R. Harada, T. Kawamoto, T. Ueha, M. Minoda, M. Toda, Y. Onishi, N. Fukase, H. Hara, Y. Sakai, M. Miwa, R. Kuroda, M. Kurosaka and T. Akisue, Reoxygenation using a novel CO2 therapy decreases the metastatic potential of osteosarcoma cells, Experimental Cell Research 319 (2013), 1988–1997. [15] H. Hayashi, S. Yamada, Y. Kumada, H. Matsuo, T. Toriyama and H. Kawahara, Immersing feet in carbon dioxide-enriched water prevents expansion and formation of ischemic ulcers after surgical revascularization in diabetic patients with critical limb ischemia, Annals of Vascular Diseases 1 (2008), 111–117. [16] H.W. Huang, I.M. Jou, C.K. Wang, P.Y. Chen, W.C. Wang and C.C. Lin, Power spectral analyses of index finger skin blood perfusion in carpal tunnel syndrome and diabetic polyneuropathy, Experimental Diabetes Research (2011), 465910. [17] T. Ito, J.I. Moore and M.C. Koss, Topical application of CO2 increases skin blood flow, Journal of Investigative Dermatology 93 (1989), 259–262. [18] M. Klemenc, Aktivnost avtonomnega zˇ ivˇcnega sistema in diastoliˇcna disfunkcija levega prekata pri mladih hipertonikih, Zdravniˇski Vestnik 70 (2001), 627–635. [19] L. Li, S. Mac-Mary, J.M. Sainthillier, S. Nouveau, O. de Lacharriere and P. Humbert, Age-related changes of the cutaneous microcirculation in vivo, Gerontology 52 (2006), 142–153.
M. Finzgar et al. / Effect of transcutaneous application of gaseous carbon dioxide
435
[20] A. Malliani, F. Lombardi and M. Pagani, Power spectrum analysis of heart rate variability: A tool to explore neural regulatory mechanisms, British Heart Journal 71 (1994), 1–2. [21] M. Minamiyama and A. Yamamoto, Direct evidence of the vasodilator action of carbon dioxide on subcutaneous microvasculature in rats by use of intra-vital video-microscopy, Journal of Biorheology 24 (2010), 42–46. [22] T.A. Mustoe, K. O’Shaughnessy and O. Kloeters, Chronic wound pathogenesis and current treatment strategies: A unifying hypothesis, Plastic and Reconstructive Surgery 117 (2006), 35S–41S. [23] K. Oe, T. Ueha, Y. Sakai, T. Niikura, S.Y. Lee, A. Koh, T. Hasegawa, M. Tanaka, M. Miwa and M. Kurosaka, The effect of transcutaenous application of carbon dioxide (CO2) on skeletal muscle, Biochemical and Biophysical Research Communications 407 (2011), 148–152. [24] Y. Onishi, T. Kawamoto, T. Ueha, H. Hara, N. Fukase, M. Toda, R. Harada, Y. Sakai, M. Miwa, K. Nishida, M. Kurosaka and T. Akisue, Transcutaneous application of carbon dioxide (CO2) enhances chemosensitivity by reducing hypoxic conditions in human malignant fibrous histiocytoma, Journal of Cancer Science and Therapy 4 (2012), 174–181. [25] Y. Sakai, M. Miwa, K. Oe, T. Ueha, A. Koh, T. Niikura, T. Iwakura, S.Y. Lee, M. Tanaka and M. Kurosaka, A novel system for transcutaneous application of carbon dioxide causing an »artificial Bohr effect« in the human body, PloS One 6 (2011), e24137. [26] J. Schmidt, P. Monnet, B. Normand and R. Fabry, Microcirculatory and clinical effects of serial percutaneous application of carbon dioxide in primary and secondary Reynaud’s phenomenon, Vasa 34 (2005), 93–100. [27] J. Smolander and P. Kolari, Laser-Doppler and plethysmographic skin blood flow during exercise and during acute heat stress in the sauna, European Journal of Applied Physiology and Occupational Physiology 54 (1985), 371–377. [28] A. Szakall, Die Ver¨anderungen der obersten Hautschichten durch den Dauergebrauch einiger Handwaschmittel, Arbeitsphysiologie 13 (1943), 49–56. [29] G. Tan, P. Xu and L.B. Lawson, J. He, L.C. Freytag, J.D. Clements and V.T. John, Hydration effects on skin microstructure as probed by high-resolution cyro-scanning electron microscopy and mechanistic implications to enhanced transcutaneous delivery of biomacromolecules, Jorunal of Pharmaceutical Sciences 99 (2010), 730–740. [30] F. Werdin, M. Tennenhaus, H.E. Schaller and H.O. Rennekampff, Evidence-based management strategies for treatment of chronic wounds, ePlasty 9 (2009), e19. [31] S.C. Wilkinson, W.J. Maas, J.B. Nielsen, L.C. Greaves, J.J. van de Sandt and F.M. Williams, Interactions of skin thickness and physicochemical properties of test compounds in percutaneous penetration studies, International Archives of Occupational and Environmental Health 79 (2006), 405–413.
