| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
 
|
|
|
||
|
|
Anesth Analg 2005;100:315-320 © 2005 International Anesthesia Research Society doi: 10.1213/01.ANE.0000140243.97570.DE
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
During cardiac surgery the wound is exposed to desiccation, especially as a result of operating room ventilation and the insufflation of dry carbon dioxide (CO2) for de-airing. We compared the gas humidity and desiccation rates in an in vitro model of a cardiothoracic wound during these conditions and during insufflation of humidified CO2. To assess the influence of flow velocity, CO2 was insufflated at 10 L/min via two devices, a standard open-ended tube and a low-velocity gas diffuser. The treatment arms were compared with a control without insufflation. When insufflated via the open-ended tube the humidity in the model was almost equal to the control, both with dry and humidified CO2. However, the total desiccation rate was more rapid than the control (P < 0.001), especially in the area exposed to the gas jet where the desiccation rate was three times more rapid (P < 0.001). With the gas diffuser, dry CO2 caused almost zero humidity and a desiccation rate that was almost equal to the control. Humidified CO2 increased humidity in comparison with the control (P < 0.001) and decreased the desiccation rate by >90% (P < 0.001). Humidified CO2 may be used to avert desiccation of the cardiothoracic wound. The humidified gas is effective only when delivered via a low-velocity outlet device.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
When the thoracic cavity is opened the clinician seldom considers that this will abruptly expose the internal tissues to a totally new environment, ambient air, characterized by less humidity. Although the implications of this sudden change have so far not been studied very extensively, it has become clear that desiccation during surgery leads to tissue damage (1) and that the risk of such damage increases with time (2). Moreover, injury of mesothelial layers may cause postoperative adhesions (1,3), which not only make reoperations more difficult but also may lead to right ventricular dysfunction (4). The effect of desiccation is of special interest in cardiac surgery where gas exchange, i.e., convection, occurs not only as a result of standard operating room (OR) ventilation. CO2 gas is also insufflated into the cardiothoracic wound to prevent arterial air embolism. The question of whether such CO2 de-airing might initiate desiccation effects is relevant because CO2 has to be insufflated continuously throughout the open-heart operation (5), and because the insufflated CO2 is completely dry.
Whereas cell damage and the ensuing adhesion formation may be estimated semi-quantitatively, more direct information about desiccation effects can be obtained by measuring humidity and actual water loss during surgery. To obtain these types of data we must resort to an experimental model. In the present study we investigated whether humidity and rate of water loss in an in vitro model of a cardiothoracic wound are influenced by insufflation of dry and humidified CO2 via a standard open-ended tube (6,7) and a low-velocity outlet device (5,8,9), respectively. We also tested whether cardiotomy suction and hand movements influence water loss rate.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Wound desiccation was studied in a model of a cardiothoracic wound cavity that contained 2 standard 9-cm diameter Petri dishes (Fig. 1). Each plate contained a layer of blood agar representing fresh wound tissue. The dimensions of the model were based on measurements of the open wound cavity of five adult patients undergoing cardiac surgery through a complete median sternotomy and during cardiopulmonary bypass (CPB) with an empty heart (8). The model was elliptically shaped with a length, width, and depth of 20, 12, and 8 cm, respectively. It was placed on the operating table of a fully ventilated OR for cardiac surgery (laminar airflow of approximately 2500 m3/h from the ceiling downward).
|
Two insufflation devices were studied. They were a standard open-ended tube with an inner diameter of 1/4 inch and a gas diffuser (Cardia Innovation AB, Stockholm, Sweden) consisting of a thin fixable tube with a diffuser of polyurethane foam at the end (5). The insufflation devices were positioned at the acute end of the wound cavity model. The open-ended tube was pointed towards the center of the cavity with the orifice positioned a few centimeters inside the brim (Fig. 1), whereas the gas diffuser, which produces a multidirectional gas flow with a low outlet velocity, was pointed downwards and positioned at half the depth of the cavity.
Medical CO2 was delivered at a flow of 10 L/min from a pressurized gas cylinder consisting of a flow regulator with a backpressure compensated flowmeter (AGA Gas AB, Stockholm, Sweden). The flowmeter and the used calibration procedure have been described in detail (5,8). The CO2 was humidified with a bubble humidifier containing sterile water (Aquapak 340 mL; Hudson Respiratory Care Inc., Temecula, CA). A pilot study showed that for the humidity of the administered gas to remain constant, the humidifier had to be maintained at a constant temperature. Thus, during the experiment the humidifier was warmed in a heat-regulated water bath (Haake D8/L; Haake Mess-Technik GmbH u. CO, Karlsruhe, Germany) that was maintained at room temperature. We took care not to exceed this temperature to prevent condensation in the gas delivery system and the wound model, which would have interfered with our measurements.
Humidity and water loss in the wound model were studied under five different conditions: without insufflation (control) and during insufflation of dry and humidified CO2 that was delivered via the open-ended tube and the gas diffuser, respectively.
After gas insufflation had been started (if used), room temperature (20.2 ± 0.3°C, mean ± sd) and relative humidity inside the model were measured at steady state, i.e., when values fluctuated around a constant value over 30 s. To avoid interference through evaporation from the agar plates, humidity was measured without them. We used a digital hygrometer (HygroPalm 3; Rotronic AG, Bassersdorf, Switzerland) that has a resolution of 0.1% relative humidity and 0.1°C, with an accuracy of ±1.5% relative humidity and ±0.3°C, respectively. The instrument is claimed to be suitable for measurements in both air and CO2. During measurements the sensor (Hygro Clip, Rotronic AG, Bassersdorf, Switzerland) was positioned in the middle of the cavity (just above the bottom). The sensor was kept close to the wall to avoid interference with the CO2 flow.
Two agar plates were then weighed one by one, temporarily without their lids, on a digital precision scale (Sartorius L420S, Satrorius Gmbh, Göttingen, Germany). This instrument has a resolution of 1 mg and an accuracy of ±2 mg. Then, while gas was still flowing (if used), the agar plates were placed into the wound model (Fig. 1), the lids were removed, and a timer was started. After 30 min of exposure the lids were put back on the agar plates while the gas was kept flowing. They were then removed from the model and weighed again (without their lids). In every experiment a fresh pair of agar plates kept at room temperature was used, and the lids were kept free from condensation water. The 5 experiments were repeated 10 times in a random order, resulting in a series of 50 measurements. Random assignment was done with the help of unmarked envelopes, each of which contained a card indicating type of experimental group.
The possible influence of cardiotomy suction and hand movements was investigated in an additional study during insufflation of humidified CO2 with a gas diffuser, the condition found to cause the slowest desiccation rate in the main study. The desiccation rate was measured with the same set-up as in the main study. Three experiments were repeated 10 times, rendering a total of 30 measurements that were performed in random order. One group was assigned to insufflation with continuous cardiotomy suction in combination with hand movements, a second group to insufflation without suction and hand movements, and a third group (control) without either insufflation or suction and hand movements. A standard cardiotomy suction device was placed in the middle of the wound cavity model with the orifice 1 cm above the bottom close to the wall. The suction rate was kept constant at 1.5 L/min. Every 10 seconds an investigator placed his hands into the wound cavity model, simulated surgical suturing and knotting, and removed them again. This was done throughout the 30-min experiment.
Data are presented as medians with ranges. The Mann-Whitney U-test and Wilcoxon’s tests were used when appropriate. The total values (n = 10) of both agar plates combined were used for comparison of desiccation rates among the groups. In the main study we also compared the desiccation rate between the agar plate proximal and distal to the insufflation device. As there were no proximal and distal plates to be defined in the control group, the proximal (n = 10) and distal plates (n = 10) of the insufflation groups were compared with all plates of the control group (n = 20). Differences were considered statistically significant if P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Without insufflation the relative humidity was 45.1% (Fig. 2). When the open-ended tube was used, insufflation of dry CO2 resulted in a decrease of the relative humidity to 42.4% (P = 0.04), whereas insufflation of humidified CO2 increased it to 49.2% (P = 0.02). When the gas diffuser was used, the effect was much more marked. Insufflation of dry CO2 decreased the relative humidity to 1.8%, which was the smallest value of all groups (P < 0.001). By contrast, when humidified CO2 was insufflated via the gas diffuser, the relative humidity increased to 75.8%, which was the largest value of all groups (P < 0.001).
|
When the open-ended tube was used for insufflation of dry as well as of humidified CO2, the desiccation rate, expressed as water loss in mg/cm2/min, was much more rapid on the distal plate than on the proximal one (P < 0.001), where the rate was slightly faster than the control (P < 0.001, Fig. 3). When the gas diffuser was used the differences were in the opposite direction; the distal plate had the slower desiccation rate (P < 0.001). With dry CO2, the desiccation rate on the proximal plate was just faster than the control (P < 0.001), whereas the desiccation rate on the distal plate was slightly less than the control (P < 0.001). With humidified CO2, the proximal and distal desiccation rates were approximately 87% and 95% less than the control (P < 0.001), respectively.
|
The total desiccation rate in the control group was 0.17 (0.15–0.20) mg/cm2/min. When the open-ended tube was used for insufflation of dry CO2, the desiccation rate was 0.47 (0.46–0.52) mg/cm2/min, almost three times that of the control (P < 0.001). When humidified CO2 was insufflated with the same device, the desiccation rate was still rapid, 0.38 (0.34–0.39) mg/cm2/min, more than twice that of the control (P < 0.001). With the gas diffuser, insufflation of dry CO2 resulted in a slightly more rapid desiccation rate, 0.21 (0.18–0.22) mg/cm2/min, compared with the control (P = 0.001). However, when humidified CO2 was insufflated the rate decreased to 0.015 (0.012–0.021) mg/cm2/min, more than 90% less than the control (P < 0.001).
During the additional study, the temperature and relative humidity in the OR was 20.0 ± 0.2°C and 16.4 ± 0.9 (mean ± sd), respectively. The total desiccation rate in the control group was 0.25 (0.21–0.28) mg/cm2/min. When humidified CO2 was insufflated the desiccation rate decreased to 0.013 (0.010–0.019) mg/cm2/min without suction and hand movements (95% less than the control, P < 0.001). With suction and hand movements the desiccation rate increased to 0.020 (0.015–0.027) mg/cm2/min. This value was more than without suction and hand movements (P < 0.001), but still 92% less than the control (P < 0.001).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In the present study humidified CO2 insufflated through a gas diffuser decreased the desiccation rate more than 90%. Conversely, when humidified CO2 was supplied via an open-ended tube it increased the desiccation rate.
Nkere et al. (2) have shown that during conventional cardiac surgery with a sternotomy the pericardium undergoes inflammatory changes with concomitant damage to the mesothelium. In the presence of shed blood such damage may result in extensive adhesion formation (1,3), which, apart from complicating reoperation, can even lead to right ventricular dysfunction (4). Although the possible desiccation effect of CO2 insufflation for de-airing in open-heart surgery has not yet been addressed, it is a recognized problem in other surgical areas where CO2 is insufflated for other purposes. In laparoscopic surgery, for example, the continuous insufflation of dry CO2 has been found to cause evaporation, manifesting itself by a dramatic increase of viscosity of the peritoneal fluid (10). Moreover, in thoracoscopy (11), as well as in off-pump coronary artery bypass surgery (12), the insufflation of dry CO2 has been shown to cause severe cell damage to the pleural and endothelial surfaces, respectively. In both cases, however, the destructive effects could be alleviated by humidification of the gas (11,12).
When a new technique is to be tested, the logical first step is to study its effect in a controlled experimental setting. It would have been very difficult to analyze the different aspects of desiccation in patients. Desiccation means that water evaporates from wound surfaces and escapes via diffusion and convection. Thus, a high humidity in the wound cavity may paradoxically be attributable to a rapid desiccation rate. Consequently, the humidity in a patient’s wound cavity cannot be used as a measure of the desiccation rate in the wound. As pointed out by Sessler (13) "The contribution of evaporation from within surgical incisions remains to be determined in humans because of technical difficulties."
Because desiccation results from superficial water loss, we quantified the water loss rate under different surgical conditions. As part of our efforts to reproduce conditions in practice, the experiments were performed on the same operating table in the same fully ventilated OR used for cardiac surgery. Moreover, the measurements were performed in a cardiothoracic wound model (5,14,15) of adult patients undergoing cardiac surgery with complete median sternotomy and during CPB with a collapsed heart (8). We used 2 standard 9-cm blood agar plates in the wound model, one close to and one farther away from the insufflation source. This enabled us to detect any covariation that might exist between desiccation rate and distance from the insufflation device. As the size of the agar plate is standardized, its weight loss could easily be converted into water loss per cm2 of exposed surface. Blood agar provides a wet surface just like a fresh surgical wound. An open water surface has been used to represent evaporation from an open wound (16) on the basis that "a wound after full-thickness excision of skin transmits water vapor at a rate equal to 91% of that for an open water surface" (17). In control experiments under the same conditions as in the main study, we compared the water loss from 12 Petri dishes containing blood agar and water, respectively, and found that the water loss rate from blood agar (103%) was almost equal to that of water (100%) during a period of 30 minutes. In comparison, agar plates were easier to handle, thus allowing for the detection of delicate water losses without the risk of spill. The wound model used conditions of room temperature. The temperature in a wound during cardiac surgery may be influenced by many variables, including cold or warm cardioplegia and temperature of CPB.
The studied standard 1/4 inch tube is commonly used for de-airing (6,7,14). It was pointed into the cavity model as recommended (7,18) to achieve a central CO2 supply (Fig. 1). The gas diffuser was positioned below the brim, a position that has been shown to be suitable for efficient CO2 de-airing (15). Both insufflation devices were positioned at the caudal end of the simulated wound because this is a position that causes little interference with surgery (15). The CO2 was insufflated at a constant flow of 10 L/min as has been used with both studied insufflation devices (6,9) and has also been found necessary for efficient air displacement in the cardiothoracic wound during surgical conditions (5,14,15).
As seen in Figure 2, we chose to perform the main study when the humidity was approximately 45% (control). This is approximately the middle of the normal range of variation that occurs over the latitudes and seasons. When CO2 was insufflated with a gas diffuser the humidity in the model was close to 0% and reached more than 75% after humidification. Because the gas diffuser can provide CO2 concentrations larger than 99.5% in a cardiothoracic wound (5,8,14,15), these values most likely represent the true humidity of the insufflated gas. The fact that the model’s relative humidity did not reach 0% with dry CO2 we attribute to the limitation of the measuring instrument. The fact that 100% was not reached with humidified CO2 was possibly attributable to a suboptimal humidification technique. In sharp contrast, when insufflated via the open-ended tube, dry CO2 decreased the model’s median relative humidity by only 3%, whereas the humidified gas increased it by 4%. Admittedly, the differences are statistically significant but they can hardly be claimed to be of any clinical value.
When looking at the resulting desiccation rates (Fig. 3), the finding that blowing dry CO2 through an open-ended tube increased the desiccation was not surprising. The desiccation rate was most marked on the distal plate towards which the jet was directed (Fig. 1), where the desiccation rate was more than four times the control value. It was also not surprising that when CO2 was delivered via the same tube, humidification of the gas did not help very much to reduce the desiccation rate. The desiccation rate was again most rapid in the area exposed to the gas jet, where the rate was somewhat slower but still more than three times the control value. The probable explanation for this lack of effect is that the CO2 jet drags down ambient air by ejector effects and also causes turbulence inside the model that mixes the supplied gas with air (8).
The results obtained with the gas diffuser are, however, more difficult to explain. Although insufflation of untreated CO2 via the gas diffuser produced almost zero humidity in the model, the total desiccation rate was only slightly more than the control with 45% relative humidity. In fact, on the agar plate furthest from the gas diffuser the desiccation rate was even less than the control. A potential explanation is that water loss from a surface occurs through diffusion (i.e., molecular movements from the surface) in the direction towards less humidity, and the diffusion rate from the surface may increase in the presence of convection (i.e., gas movements at the surface). When the relative humidity is <100%, the latter process is the dominant factor for desiccation, and when the gas diffuser is used, its very low outflow velocity reduces turbulence to a minimum (8). Under those conditions CO2, which is heavier than air, will gravitate in the cavity and cover it almost like a protective layer. This also explains the dramatic decrease in the desiccation rate to approximately 1/10 of the control when the humidified CO2 was insufflated. Again, the CO2 gravitated in the cavity but this time the gas layer contained water. Thus, if the insufflated CO2 would have had a humidity close to 100% the desiccation rate could probably have been averted completely.
Continuous cardiotomy suction and surgical hand movements in the wound model had a statistically significant but marginal influence on the desiccation rate, during insufflation of humidified CO2 with the gas diffuser. The small increase (3%) was most likely caused by a slightly increased convection in the model and not by introduction of ambient air. We have previously found that the CO2 concentration in the cardiothoracic wound remains close to 100% during active surgery with CO2 insufflation at 10 L/min through a gas diffuser (14,15).
Assuming that evaporation from blood agar can be extrapolated to tissue in a surgical wound, the results have several implications. In an OR with standard ventilation the surgical wound is subjected to desiccation, which starts immediately after exposure to ambient air. CO2 insufflation with an open-ended tube dramatically increases the desiccation rate in the region that is exposed to the jet, even if the CO2 is humidified. Accordingly, humidified CO2 still causes endothelial damage during high-flow gas insufflation to facilitate the suturing of a precise coronary anastomosis (12). By contrast, when being insufflated via a low-velocity outlet device, use of dry CO2 will, contrary to expectation, not lead to an increased desiccation in the cardiothoracic wound.
The optimal way to protect tissue against desiccation would be to enclose it in a plastic bag (19), thus providing it with a fully humidified environment without convection. This is possible only with tissues that are easily externalized such as intestines during abdominal surgery. Furthermore, it cannot be done in areas where the surgery takes place. Instead, some surgeons irrigate the open wound to protect it against desiccation. This can of course only be done intermittently to avoid interfering with surgery. As a result, the wound will be subjected to desiccation between the irrigations. Moreover, even as a spray, the irrigation will not reach hidden recesses. In contrast, humidified CO2 can be insufflated continuously into a cardiothoracic wound cavity without interfering with surgery, thus preventing evaporation from all surfaces of the wound throughout the operation.
The present study showed that this could be achieved by insufflating the wound with humidified CO2 through a gas diffuser. Because the gas diffuser can provide a cardiothoracic surgical wound with a 100% CO2 atmosphere (15), the slow desiccation rate will be independent of the OR’s air humidity, as indicated by the almost equal desiccation rates in the main and additional studies, despite a large difference in ambient air humidity. However, the implications of humidified CO2 insufflation will be most evident in a dry climate or when air conditioning is used. Future clinical studies will show whether humidified CO2 insufflation via a gas diffuser prevents tissue damage and postoperative adhesions caused by desiccation and whether solutions other than water are useful for wound protection via this system (1,20,21).
In conclusion, humidified CO2 may be used to avert desiccation of the cardiothoracic wound. The humidified gas is effective only when delivered via a low-velocity outlet device.
We thank Mattias Öhman, MSc in mathematical statistics at Umeå University, Umeå, Sweden, for reviewing the statistical analysis and Prof. Em. Willem van der Linden for help with the preparation of the manuscript.
|
Footnotes |
|---|
Supported, in part, by Karolinska Institute, and Cardia Innovation AB, Stockholm, Sweden.
Accepted for publication July 9, 2004.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Seeger JM, Kaelin LD, Staples EM, et al. Prevention of postoperative pericardial adhesions using tissue-protective solutions. J Surg Res 1997;68:63–6.[Web of Science][Medline]
- Nkere UU, Whawell SA, Sarraf CE et al. Perioperative histologic and ultrastructural changes in the pericardium and adhesions. Ann Thorac Surg 1994;58:437–44.[Abstract]
- Ryan GB, Grobety J, Majno G. Mesothelial injury and recovery. Am J Pathol 1973;71:93–112.[Web of Science][Medline]
- Bailey LL, Ze-jian L, Schulz E, et al. A cause of right ventricular dysfunction after cardiac operations. J Thorac Cardiovasc Surg 1984;87:539–42.[Abstract]
- Svenarud P, Persson M, van der Linden J. Intermittent or continuous carbon dioxide insufflation for de-airing of the cardiothoracic wound cavity? An experimental study with a new gas-diffuser. Anesth Analg 2003;96:321–7.
[Abstract/Free Full Text] - Nadolny EM, Svensson LG. Carbon dioxide field flooding techniques for open heart surgery: monitoring and minimizing potential adverse effects. Perfusion 2000;15:151–3.
[Abstract/Free Full Text] - Ng SW, Rosen M. Carbon dioxide in the prevention of air embolism during open-heart surgery. Thorax 1968;23:194–6.
[Abstract/Free Full Text] - Persson M, van der Linden J. De-airing of a cardiothoracic wound cavity model with carbon dioxide: theory and comparison of a gas diffuser with conventional tubes. J Cardiothorac Vasc Anesth 2003;17:329–35.[Web of Science][Medline]
- Svenarud P, Persson M, van der Linden J. Effect of CO2 insufflation on the number and behavior of air microemboli in open-heart surgery: a randomized clinical trial. Circulation 2004;109:1127–32.
[Abstract/Free Full Text] - Ott DE. Laparoscopy and tribology: the effect of laparoscopic gas on peritoneal fluid. J Am Assoc Gynecol Laparosc 2001;8:117–23.[Web of Science][Medline]
- Mouton WG, Bessell JR, Pfitzner J, et al. A randomized controlled trial to determine the effects of humidified carbon dioxide insufflation during thoracoscopy. Surg Endosc 1999;13:382–5.[Web of Science][Medline]
- Okazaki Y, Takarabe K, Murayama J, et al. Coronary endothelial damage during off-pump CABG related to coronary-clamping and gas insufflation. Eur J Cardiothorac Surg 2001;19:834–9.
[Abstract/Free Full Text] - Sessler DI. Perioperative heat balance. Anesthesiology 2000;92:578–96.[Web of Science][Medline]
- Svenarud P, Persson M, van der Linden J. Efficiency of a gas-diffuser and influence of suction in carbon dioxide de-airing of a cardiothoracic wound cavity model. J Thorac Cardiovasc Surg 2003;125:1043–9.
[Abstract/Free Full Text] - Persson M, Svenarud P, van der Linden J. What is the optimal device for carbon dioxide de-airing of the cardiothoracic wound and how should it be positioned? J Cardiothorac Vasc Anesth 2004;18:180–4.[Web of Science][Medline]
- Bowker P, Martin CJ, Fulton D, Muir IF. The use of infrared radiation to reduce heat loss in burned patients: experiments with a phantom. Clin Phys Physiol Meas 1981;2:257–70.[Medline]
- Larkin JM, Moylan JA. Evaporative water losses in patients treated in a laminar flow burn unit. J Surg Res 1978;24:65–9.[Web of Science][Medline]
- Rosen M. Replacement of air by carbon dioxide. Anesthesiology 1998;89:1036.
- Rowe MI, Taylor M, Sheehan K. Prevention of water and heat losses from the exposed intestine. J Pediatr Surg 1982;17:608–10.[Web of Science][Medline]
- Persson M, Flock JI, van der Linden J. Antiseptic wound ventilation with a gas diffuser: a new intraoperative method to prevent surgical wound infection? J Hosp Infect 2003;54:294–9.[Web of Science][Medline]
- Persson M, van der Linden J. A simple system for intraoperative antiseptic wound ventilation. J Hosp Infect 2003;55:152–3.[Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  J. van der Linden and M. Persson CO2 Field Flooding May Also Reduce Oxidative Stress in Open Surgery Anesth. Analg., August 1, 2009; 109(2): 683 - 683. [Full Text] [PDF]
|
|
|
|
|
|
M. Tsuchiya, E. F. Sato, M. Inoue, and A. Asada CO2 Field Flooding May Also Reduce Oxidative Stress in Open Surgery Anesth. Analg., August 1, 2009; 109(2): 683 - 684. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
