Anesth Analg 2001;93:151-156
© 2001 International Anesthesia Research Society
OBSTETRIC ANESTHESIA
The Effects of Uterine and Umbilical Blood Flows on the Transfer of Propofol Across the Human Placenta During In Vitro Perfusion
Yan-Ling He, PhD* ,
Hiroshi Seno, MD ,
Saburo Tsujimoto, MD , and
Chikara Tashiro, MD
*Department of Anesthesia and Critical Care, Massachusetts General Hospital, Boston, Massachusetts; and Department of Anaesthesiology, Hyogo College of Medicine, Nishinomiya City, Japan
Address correspondence and reprint requests to Dr. Yan-Ling HE, Department of Anesthesia and Critical Care, Massachusetts General Hospital, 55 Fruit St., Boston, MA 02114. Address e-mail to yhe{at}partners.org
 |
Abstract
|
|---|
The safety of using propofol in parturients is controversial, and little information is available on the factors that influence the placental transfer of propofol. In this study, we investigated the effects of uterine and umbilical blood flows on the placental transfer of propofol by using the dually perfused human placental cotyledon. Placental transfer was evaluated on the basis of the placental clearances at various uterine and umbilical flow rates. The placental transfer of propofol was significantly facilitated by the increased uterine flow rates over the range from 7.5 to 25 mL/min. The placental clearances of propofol were also dependent on the umbilical flow rates over the range from 0.5 to 4.0 mL/min. In contrast, the placental transfer of antipyrine was flow dependent when the umbilical flow rate was <2.0 mL/min and became permeability limited when it was >2.0 mL/min. No differences in either maternal or fetal venous concentrations of propofol were observed as umbilical flow rates varied from 0.5 to 4.0 mL/min, suggesting that an equilibration across the placenta occurs at low flow rates. These results indicate that fetal uptake of propofol can be profoundly altered by the changes in both uterine and umbilical blood flows observed in various pathophysiologic conditions and that lipid solubility greatly influences placental transfer of drugs.
Implications: Uterine and umbilical blood flows are determinant features in controlling theplacental transfer of propofol, and, therefore, changes in these variableswould significantly affect the extent of fetal exposure topropofol.
 |
Introduction
|
|---|
The value and safety of propofol as an induction drug and for the maintenance of anesthesia in parturients undergoing cesarean delivery have been evaluated in comparison with thiopental (19). The safety of using propofol in parturients is controversial, and the fetal/maternal (F/M) ratio of propofol at delivery was reported to be approximately 0.7 (911), indicating that propofol diffuses across the placenta efficiently. The commonly used F/M concentration ratio varies depending on various factors (12), and we have demonstrated that the F/M ratio of propofol can vary from 0.74 to 1.13 over the albumin concentrations from 22 to 44 g/L in the fetal perfusate (13). In addition to protein binding, uterine and umbilical blood flows can be considered alternative important factors for con-trolling the placental transfer of propofol, a small and highly lipophilic molecule. Despite frequently observed changes in fetal and maternal hemodynamics and the fact that these changes alter uterine and placental perfusion, there is little information available regarding how changes in these variables affect the transfer of a drug to the fetus (1419). This may be explained by the practical difficulties in investigating the effects of changes in uterine and umbilical blood flows on the placental transfer of drugs in parturients. The dually perfused human placental cotyledon model is useful for precisely exploring the placental transfer of drugs for changing various physiologic factors such as blood flow (13,2027). Placental diffusion of drugs with high lipid solubility, such as propofol, is expected to be greatly influenced by uterine and umbilical blood flow because most lipophilic anesthetics cross the placenta by a process of flow-dependent passive diffusion (2025). The objective of this study was to investigate the effects of uterine and umbilical blood flows on the placental transfer of propofol by using the dually-perfused human placental cotyledon.
 |
Methods
|
|---|
This study was approved by our IRB, and informed, written consent was obtained. Placentas from healthy parturients (gestational age, 3638 wk) after elective cesarean delivery were collected and transported to the laboratory immediately. Fetal and maternal sides of placenta were perfused by referring to the previously reported approach of the in vitro perfusion model (13,28). An intact placental lobule was selected by inspecting the maternal side, and a fetal chorionic artery and vein supplying a single peripheral cotyledon were cannulated and perfused at a rate of 2.0 mL/min. The cotyledon was then mounted in a specially designed chamber maintained at 37°C by a water bath, with the fetal surface of placenta facing downward. Three 20-gauge needles were gently placed into the decidual plate, and the intervillous space was perfused at 15 mL/min. Both fetal and maternal circuits were perfused at a single-pass mode with a tissue culture Medium M199 modified with Earls salts (ICN Biomedicals, Inc., Aurora, OH). The perfusate also contained human serum albumin (4.4 g/L, Plasma Protein Fraction; Baxter, Tokyo, Japan), heparin (2500 U/L), gentamicin (50 mg/L), and glucose (1.0 g/L). Dextran (molecular weight 40.000; Wako Pure Chemicals Industry, Ltd., Osaka, Japan) was added to adjust the oncotic pressure. Sodium bicarbonate was added to adjust the pH of perfusates within the physiologic range. Fetal flow rates were checked to ensure that arterial inflow equaled venous outflow and that only preparations with <2 mL/h loss were used for the experiments. Fetal and maternal perfusion pressures were measured with a pressure transducer (Yamasu, Tokyo, Japan). The placental clearance (CL) of antipyrine, which is of intermediate solubility and does not bind to protein, was used as a marker of placental transfer in placental cotyledon preparations. A pilot single-pass perfusion experiment with fetal venous samples taken at 2, 5, 10, 15, 20, and 30 min demonstrated that propofol and antipyrine concentrations reached a plateau by 10 min. The CLs of propofol and antipyrine were measured at 30-min intervals.
After a stabilization period of >30 min, propofol (Diprivan; Zeneca Pharmaceuticals, Osaka, Japan) and antipyrine (Wako) were added to the maternal perfusate to target concentrations at 12 and 20 µg/mL for propofol and antipyrine, respectively. Measured concentrations of propofol and antipyrine were used for data analysis. The maternal circuit was perfused at 7.5 mL/min for 30 min with propofol and antipyrine in the perfusate. The uterine flow rate was increased to 15 mL/min for 30 min and then to 25 mL/min for another 30 min. The umbilical flow rate was kept constant at 2.0 mL/min throughout the experiment. The perfusion pressures on the maternal side ranged between 20 and 50 mm Hg, and that on the fetal side was <70 mm Hg.
After a stabilization period of >30 min, the maternal circuit was perfused with propofol and antipyrine in the perfusate at a rate of 15 mL/min. The umbilical flow rates were randomly varied to 0.5, 1.0, 1.5, 2.0, 3.0, and 4.0 mL/min at 30-min intervals while the uterine flow rate was maintained at 15 mL/min. The perfusion pressures on the fetal side ranged between 35 and 70 mm Hg, whereas those on the maternal side were always <30 mm Hg.
Arterial and venous samples from the maternal circuit and venous samples from the fetal circuit were taken at 20 and 30 min during each experimental period for propofol and antipyrine measurement, with duplicate clearance determinations being made at each flow rate. Venous samples were taken from fetal and maternal circuits at 30-min intervals for mea-surement of human chorionic gonadotropin (hCG) concentration. Additional arterial and venous samples were taken for the measurement of glucose and lactate concentrations to estimate glucose consumption and lactate production. At the end of the experiment, perfused placental tissue was dissected and weighed, and the concentrations of propofol and antipyrine were determined.
Propofol and antipyrine were measured with the high-performance liquid chromatography methods reported previously (13). Placental tissue homogenate (10%) was prepared with the mobile phase for propofol measurement. To 100 µL of perfusate or 10% tissue homogenate, 1000 µL of acetonitrile was added and mixed thoroughly on a vortex mixer. The mixture was centrifuged at 15,000g for 10 min at 4°C. Ten microliters of the clear supernatant was injected onto the chromatography. The intra- and interday variability for propofol and antipyrine were <5% over the concentration range studied. The perfusate concentrations of hCG were measured by enzyme immunoassay with IMxTM hCG Dinapack (Dainabot, Tokyo, Japan). Glucose and lactate concentrations were measured with a blood gas, electrolyte, and metabolite analyzer (ABL system 625; Radiometer, Copenhagen, Denmark).
A single-pass experimental design was performed for both maternal and fetal circulation in this study, and the CL was calculated as follows (13,28):
equation
where Qf is the flow rate of fetal circulation and Cfv and Cma are the venous concentration in the fetal circuit and the arterial concentration in the maternal circuit, respectively. Duplicate estimations of CL were performed at 20 and 30 min during each perfusion. The clearance index (CI) of propofol was calculated on the basis of the CLs of propofol (CLpropofol) and antipyrine (CLantipyrine):
equation
All data are expressed as mean ± SEM. SigmaStat for Windows (Version 1.0; Jandel Scientific, Chicago, IL) was used for the statistical analysis. The propofol concentrations, CL, and CIs were analyzed with the one-way repeated-measures analysis of variance. If the analysis of variance was found to be significant, Bonferronis correction was performed to compare the values at various maternal or fetal flow rates. Differences were considered to be significant when P < 0.05.
 |
Results
|
|---|
The characteristics of the perfused cotyledons are summarized in Table 1. Glucose consumption, lactate production, and the production rate of hCG were used to characterize the physiologic integrity and viability of the perfused cotyledons. No hCG was detected in the fetal circulation, suggesting the expected preferential secretion of the hormone and the physical integrity of the placental preparation. Figure 1 illustrates the CLs of propofol and antipyrine, and the CIs at different uterine flow rates while the umbilical flow rate was maintained at 2.0 mL/min. The CLs of propofol and antipyrine at the uterine flow rate of 15 mL/min were significantly larger than those at 7.5 mL/min (P < 0.05). The CLs of propofol and antipyrine were further increased by increasing the uterine flow rate from 15 to 25 mL/min (P < 0.05). Increases in the CLs for propofol and antipyrine were parallel, resulting in unchanged CIs (P = 0.26) at all three uterine flow rates. The propofol concentrations in maternal artery, maternal vein, and fetal vein at various uterine flow rates are illustrated in Figure 2. Both maternal and fetal venous concentrations of propofol were significantly enhanced by increasing the uterine flow rates from 7.5 to 15 mL/min (P < 0.05) and from 15 to 25 mL/min (P < 0.05). Consequently, the CLs of propofol were significantly facilitated by the increased uterine flow rates over the range investigated.
Figure 3 illustrates the effects of umbilical flow rates on the CLs of propofol and antipyrine, and the CIs. The CLs of propofol were 0.38 ± 0.05, 0.88 ± 0.08, 1.13 ± 0.17, 2.01 ± 0.21, 2.42 ± 0.23, and 3.08 ± 0.26 mL · h-1 · g-1 at the umbilical flow rates of 0.5, 1.0, 1.5, 2.0, 3.0, and 4.0 mL/min, respectively. The increases in the CLs of propofol were significant with the sequential increases in umbilical flow rates over the entire range studied (P < 0.05). The CLs of antipyrine at the corresponding umbilical flow rates were 0.97 ± 0.11, 1.81 ± 0.15, 2.12 ± 0.34, 3.31 ± 0.35, 3.41 ± 0.35, and 3.76 ± 0.41 mL · h-1 · g-1, respectively. The CLs of antipyrine also increased significantly with the increases in umbilical flow rates up to 2.0 mL/min but then failed to increase further once umbilical flow rate exceeded 2 mL/min. No significant differences in the CLs of antipyrine at umbilical flow rates between 2.0 and 4.0 mL/min were observed (P = 0.425). The propofol concentrations in the maternal artery, maternal vein, and fetal vein at various umbilical flow rates are illustrated in Figure 4. Propofol concentrations in both fetal and maternal veins were not influenced by umbilical fetal flow rates over the entire range studied.
 |
Discussion
|
|---|
The placenta is unique in being perfused on both maternal and fetal surfaces, and each of the three components in the maternal/placental/fetal unit exerts an influence on drug disposition independently. The placental barrier consists of the trophoblast, the villus stroma, and the fetal capillary endothelial cells. The trophoblast is bordered on the maternal side by the microvillous membrane, which is in direct contact with the maternal blood in intervillous spaces. Distribution from maternal circulation to the placenta is the first step for a drug to transfer across the placenta and is considered a function of the uterine blood flow and the permeability of the placental membrane (12,20). The placental transfer of propofol was significantly facilitated by increased uterine flow rates, indicating that changes in uterine blood flow will affect the extent of fetal exposure to propofol. Uterine blood flow changes with uterine contractions and maternal cardiac output, which varies in various pathophysiologic conditions and in response to many drugs (1417). Therefore, the fetal uptake of propofol would be expected to decrease when a bolus of propofol coincides with a contraction because uterine contractions impair the placental flow. However, increased cardiac output from volume preloading, medication, or both would result in a greater fetal exposure to propofol because of the proportional increase in the uterine blood flow, which accounts for a large portion of the cardiac output. The maternal venous concentrations of propofol increased significantly with the increasing uterine flow rates, implying a decreased extraction of propofol from the maternal circulation. The decreased extraction may be attributed to the reduced contact time with placental tissue or the saturation of placental tissue binding with propofol when the uterine flow rates increase. The concentration of propofol was significantly enhanced in the fetal vein and was associated with the increased concentration in the maternal vein when maternal perfusion was increased. This is consistent with and provides additional evidence in support of the venous equilibration hypothesis proposed by Wilkening et al. (29). They have demonstrated that uterine and umbilical blood flows form an exchange system that tends to equilibrate the venous concentrations of highly diffusible molecules between the two blood streams.
The CLs of propofol showed significant dependency on the umbilical flow rates over the range from 0.5 to 4.0 mL/min, indicating that the placental transfer of propofol will also be influenced by umbilical blood flow. The unchanged concentrations of propofol in the fetal vein with varying umbilical flow rates suggest that an equilibration across the placenta occurs at a low fetal flow rate. Although no change in the concentrations of propofol in the fetal vein was observed, the placental transfer of propofol in terms of transfer rate that can be calculated from the product of CL and concentration in the fetal vein would be facilitated when umbilical blood flow increases. Obviously, the facilitated placental transfer of propofol by the increased umbilical flow rate was completely caused by the enhanced CLs. Taking the unchanged propofol concentration in the fetal vein into account, the F/M ratio, which is often measured clinically to evaluate the placental transfer of drugs, would not vary at different umbilical blood flows. That is, the umbilical blood flow dependency of the placental transfer of propofol demonstrated in this study cannot be envisaged by the F/M concentration ratio. Even though the umbilical flow rates show no effect on the F/M ratios, the increased CLs would likely result in a significantly larger amount of propofol in the fetal circulation and, therefore, more exposure of the fetus to propofol. This indicates that the amount of propofol received by the fit fetus at elective cesarean delivery would be much more than the amount received by the distressed fetus at emergency cesarean delivery.
In contrast to that of propofol, the CL of antipyrine shows flow dependency when the umbilical flow rate is <2.0 mL/min, and then the CL of antipyrine failed to increase with the umbilical flow rates up to 4.0 mL/min, which is indicative of a permeability-limited diffusion. The divergence of antipyrine and propofol clearances at high umbilical flow rates suggests that the placental transfer of antipyrine is a function of both placenta blood flows and permeability at high rates of placental perfusion. Irrespective of their comparable molecular weight, the greater umbilical flow dependency of propofol than that of antipyrine might be attributed to the much higher lipid solubility of propofol as compared with antipyrine. Antipyrine is of intermediate solubility with a log octanol/water partition coefficient of 0.73 (30), whereas propofol is a highly lipophilic compound with a log octanol/water partition coefficient of 4.33 (31). The higher lipid solubility of propofol may enhance diffusion across the placenta and may also result in a larger affinity to the placental tissue than antipyrine. Consistent with this speculation, large tissue concentrations of propofol (57.1 ± 9.1 µg/g wet tissue) were observed at the end of the experiments, whereas no antipyrine was detected in the placental tissue at that time.
Propofol is a highly lipid-soluble drug that uses intralipid as a vehicle. Investigations on the emulsion formulation of propofol have shown that the emulsion formulation is more than a pharmaceutically elegant means of solubilizing propofol: it is essential for the desirable clinical characteristics of propofol (32). Brain and lung distribution kinetics of propofol in rats are formulation dependent (33). In this study, we used the commercially available formulation of propofol that formulates propofol in a lipid emulsion (Diprivan). Depending on the extent of the intralipid that was translocated across the placenta, the ratio of the concentrations of free propofol in the aqueous phase across the placenta might be different than that of the total. It is, therefore, possible that the intralipid in the emulsion may have altered the placental transfer of propofol. From the results of this study, we cannot determine whether intralipid is potentially important for the placental transfer of propofol incorporated in emulsion. Because it is the emulsion of propofol that is administered for anesthesia, the results from this study should reflect the fetal uptake of propofol observed in obstetric anesthesia.
In summary, fetal uptake of propofol can be greatly altered by the changes in uterine and umbilical blood flows observed in various pathophysiologic conditions. Furthermore, lipid solubility is one of the important physicochemical properties that influences placental transfer.
 |
Acknowledgments
|
|---|
Supported, in part, by Grants-in-Aid 10770774 and 11470330 for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan.
 |
References
|
|---|
-
Moore J, Bill KM, Flynn RJ, et al. A comparison between propofol and thiopentone as induction agents in obstetric anaesthesia. Anaesthesia 1989; 44: 7537.[Web of Science][Medline]
-
Valtonen M, Kanto J, Rosenberg P. Comparison of propofol and thiopentone for induction of anaesthesia for elective caesarean section. Anaesthesia 1989; 44: 75862.[Web of Science][Medline]
-
Celleno D, Capogna G, Tomassetti M, et al. Neurobehavioral effects of propofol on the neonate following elective caesarean section. Br J Anaesth 1989; 62: 64954.[Abstract/Free Full Text]
-
Gin T, Gregory MA, Oh TE. The haemodynamic effects of propofol and thiopentone for induction of caesarean section. Anaesth Intensive Care 1990; 18: 1759.[Web of Science][Medline]
-
Siafaka I, Vadalouca A, Gatziou B, et al. A comparative study of propofol and thiopental as induction agents for elective caesarean section. Clin Exp Obstet Gynecol 1992; 19: 936.[Medline]
-
Celleno D, Capogna G, Emanuelli M, et al. Which induction drug for caesarean section? A comparison of thiopental sodium, propofol, and midazolam. J Clin Anesth 1993; 5: 2848.[Web of Science][Medline]
-
Gin T, OMeara ME, Kan AF, et al. Plasma catecholamines and neonatal condition after induction of anaesthesia with propofol or thiopentone at caesarean section. Br J Anaesth 1993; 70: 3116.[Abstract/Free Full Text]
-
Abboud TK, Zhu J, Richardson M, et al. Intravenous propofol vs thiamylal-isoflurane for caesarean section, comparative maternal and neonatal effects. Acta Anaesthesiol Scand 1995; 39: 2059.[Web of Science][Medline]
-
Dailland P, Cockshott ID, Lirzin JD, et al. Intravenous propofol during caesarean section: placental transfer, concentrations in breast milk, and neonatal effectsa preliminary study. Anesthesiology 1989; 71: 82734.[Web of Science][Medline]
-
Gin T, Gregory MA, Chan K, Oh TE. Maternal and fetal levels of propofol at caesarean section. Anaesth Intensive Care 1990; 18: 1804.[Web of Science][Medline]
-
Gin T, Yau G, Jong W, Tan P, et al. Disposition of propofol at caesarean section and in the postpartum period. Br J Anaesth 1991; 67: 4953.[Abstract/Free Full Text]
-
Morgan DJ. Drug disposition in mother and foetus. Clin Exp Pharmacol Physiol 1997; 24: 86973.[Web of Science][Medline]
-
He YL, Tsujimoto S, Tanimoto M, et al. Effects of protein binding on the placental transfer of propofol in the human dually-perfused cotyledon. Br J Anaesth 2000; 85: 2816.[Abstract/Free Full Text]
-
Robson SC, Hunter S, Boys R, et al. Changes in cardiac output during epidural anaesthesia for caesarean section. Anaesthesia 1989; 44: 4759.[Web of Science][Medline]
-
Robson SC, Boys R, Rodeck C, Morgan B. Maternal and fetal hemodynamic effects of spinal and extradural anaesthesia for elective caesarean section. Br J Anaesth 1992; 68: 549.[Abstract/Free Full Text]
-
Karinen J, Rasanen J, Paavilainen T, et al. Uteroplacental and fetal haemodynamics and cardiac function of the fetus and newborn after crystalloid and colloid preloading for extradural caesarean section anaesthesia. Br J Anaesth 1994; 73: 7517.[Abstract/Free Full Text]
-
Rasanen J, Alahuhta S, Kangas-Saarela T, et al. The effects of ephedrine and etilefrine on uterine and fetal blood flow and on fetal myocardial function during spinal anesthesia for caesarean section. Int J Obstet Anesth 1991; 1: 38.
-
Karinen J, Rasanen J, Alahuhta S, et al. Effect of crystalloid and colloid preloading on uteroplacental and maternal haemodynamic state during spinal anaesthesia for caesarean section. Br J Anaesth 1995; 75: 5315.[Abstract/Free Full Text]
-
Park GE, Hauch MA, Curlin F, et al. The effects of varying volumes of crystalloid administration before cesarean delivery on maternal hemodynamics and colloid osmotic pressure. Anesth Analg 1996; 83: 299303.[Abstract]
-
Vella LM, Knott C, Reynolds F. Transfer of fentanyl across the rabbit placenta: effect of umbilical flow and concurrent drug administration. Br J Anaesth 1986; 58: 4954.[Abstract/Free Full Text]
-
Zakowski M, Ham AA, Grant GJ. Transfer and uptake of alfentanil in the human placenta during in vitro perfusion. Anesth Analg 1994; 79: 108993.[Abstract/Free Full Text]
-
Johnson RF, Herman N, Arney TL, et al. Bupivacaine transfer across the human term placenta. Anesthesiology 1995; 82: 45968.[Web of Science][Medline]
-
Ala-Kokko TI, Pienimaki P, Herva R, et al. Transfer of lidocaine and bupivacaine across the isolated perfused human placenta. Pharmacol Toxicol 1995; 77: 1428.[Web of Science][Medline]
-
Johnson RF, Herman N, Johnson HV, et al. Effects of fetal pH on local anesthetic transfer across the human placenta. Anesthesiology 1996; 85: 60815.[Web of Science][Medline]
-
Johnson RF, Herman N, Arney TL, et al. The placental transfer of sufentanil: effects of fetal pH, protein binding, and sufentanil concentration. Anesth Analg 1997; 84: 12628.[Abstract]
-
Schneider H, Dancis J. edsIn vitro perfusion of human placental tissue: international workshop. Zurich, March 80-9,1984. Contrib Gynecol Obstet 1985; 13: 1189.
-
Schneider H, Proegler M, Duft R. Transfer of antipyrine, H2O, L-glucose and a aminoisobutyric acid across in vitro perfused human placental lobe. Trophoblast Res 1987; 2: 52537.
-
Schneider H, Panigel M, Dancis J. Transfer across the perfused human placenta of antipyrine, sodium, and leucine. Am J Obstet Gynecol 1972; 114: 8228.[Web of Science][Medline]
-
Wilkening RB, Anderson S, Martensson L, Meschia G. Placental transfer as a function of uterine blood flow. Am J Physiol 1982; 242: H42936.
-
Giroux M, Teixera MG, Dumas JC, et al. Influence of maternal blood flow on the placental transfer of three opioids: fentanyl, alfentanil, sufentanil. Biol Neonate 1997; 72: 13341.[Web of Science][Medline]
-
Chou JT, Jurs PC. Computation of partition coefficients from molecular structures by a fragment addition method. In: Yalkowsky SH, Sinkula AA, Valvani SC, eds. Physical chemical properties of drugs. New York: Marcel Dekker, 1980: 16399.
-
Dutta S, Ebling WF. Emulsion formulation reduces propofols dose requirements and enhances safety. Anesthesiology 1997; 87: 1394405.[Web of Science][Medline]
-
Dutta S, Ebling WF. Formulation-dependent brain and lung distribution kinetics of propofol in rats. Anesthesiology 1998; 89: 67885.[Web of Science][Medline]
Accepted for publication February 23, 2001.
This article has been cited by other articles:

|
 |

|
 |
 
N. Germann, F. Goffinet, and F. Goldwasser
Anthracyclines during pregnancy: embryo-fetal outcome in 160 patients
Ann. Onc.,
January 1, 2004;
15(1):
146 - 150.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
Y.-L. He, H. Seno, K. Sasaki, and C. Tashiro
The Influences of Maternal Albumin Concentrations on the Placental Transfer of Propofol in Human Dually Perfused Cotyledon In Vitro
Anesth. Analg.,
May 1, 2002;
94(5):
1312 - 1314.
[Abstract]
[Full Text]
[PDF]
|
 |
|
|