Lamotrigine Attenuates Cortical Glutamate Release during Global Cerebral Ischemia in Pigs on Cardiopulmonary Bypass 

Animals were handled according to the guidelines approved by the American Physiological Society and the “Guide for the Care and Use of Laboratory Animals” published by the National Institutes of Health (NIH publication 85-23, revised 1996). The institutional Animal Care and Use Committee approved this study.

General anesthesia was induced using ketamine (20 mg/kg intramuscularly) in 40 mature, female Yorkshire pigs with a mean +/− SEM weight of 32.3 +/− 1.5 kg (range, 24-48 kg). After endotracheal intubation, each animal was ventilated with a tidal volume of 15 ml/kg using a volume-cycled ventilator (Model 607; Harvard Apparatus, Natick, MA). Anesthesia was maintained with a bolus dose of fentanyl (15 [micro sign]g/kg) and diazepam (0.2 mg/kg) administered through a catheter in a lateral auricular vein, followed by a continuous infusion of fentanyl (10 [micro sign]g [middle dot] kg^-1[middle dot] h^-1) and diazepam (0.3 mg [middle dot] kg^-1[middle dot] h^-1). Supplemental bolus doses of fentanyl (10 [micro sign]g/kg) and diazepam (0.1 mg/kg) with isoflurane (0.5-1.0% inspired concentration) were administered as necessary to maintain a sufficient depth of anesthesia during the surgical preparation. Isoflurane (0.25% inspired concentration) was continued throughout CPB in all animals. Muscle paralysis was provided by 0.1 mg/kg pancuronium administered intravenously as necessary to prevent shivering.

Fluid-filled catheters were inserted in the femoral artery for microsphere sampling and arterial blood gas analysis and in the femoral vein for administration of fluid. Core temperature was recorded by a precalibrated thermistor probe (Yellow Springs Instruments, Yellow Springs, OH) placed in the distal esophagus 40 cm from the snout. Via the right cephalic vein, a pulmonary artery catheter was placed for thermodilution cardiac output measurements before and after CPB. The pig was turned prone, and both temporalis muscles were dissected free from the overlying cranium and reflected laterally. Along the midline suture, a 1-cm burr hole was drilled to expose the superior sagittal sinus, which was cannulated. A precalibrated thermistor probe (model MT-29; Physitemp, Clifton, NJ), 0.33 mm in diameter, was inserted into the left midcortex to directly measure brain temperature (Thermalert TH-5; Physitemp). Through a separate burr hole made 1 cm lateral to the midline suture and 0.5 cm caudal to the frontal parietal suture a microdialysis catheter (Model MF-5144; BAS/Carnegie Medicine, West Lafayette, IN) was inserted 6 mm into the lateral gyrus of the right cortex. The temporalis muscles were reapproximated, and the wound was closed. Two electroencephalographic signals were obtained using subcutaneous platinum needle electrodes with the reference electrodes placed anterior to bregma and the active electrodes overlying the midparietal region of each cortex (Model MP 100; Bio Pac Systems, Inc., Goleta, CA). The signals were recorded and stored on computer for later analysis. The animal was turned to the supine position.

Through a median sternotomy incision, transducer-tipped catheters (Model MPC 500; Millar Instruments, Houston, TX) were placed in the aorta via the right internal mammary artery and in the superior vena cava via the right internal mammary vein. The left internal mammary artery was ligated. The innominate and left subclavian arteries, just distal to their origin from the aorta, were isolated and loosely encircled with suture ligatures. A triple-lumen catheter was inserted through the right superior pulmonary vein and positioned with the proximal lumen in the left atrium. Heparin, 400 units/kg, was administered before inserting a 20-French infusion cannula (model 76020; Medtronic DLP, Grand Rapids, MI) in the ascending aortic arch and a 39-French, two-stage venous drainage cannula (model TF2937-0; Research Medical, Inc., Salt Lake City, UT) through the right atrium to harvest venous return. The oxygenator was primed with 700 ml crystalloid solution (Plasmalyte; Baxter Edwards Critical Care, Deerfield, IL) and 500 ml hetastarch, 6%, (Hespan; Dupont Pharmaceuticals, Wilmington, DE). A 40-[micro sign] arterial blood filter and bubble trap (Model SP3840; Pall Biomedical Products, East Hills, NY) was inserted in the arterial infusion line, and a membrane oxygenator with reservoir unit (Model VPCML Plus; Cobe Cardiovascular, Arvada, CO) was used. Non-pulsatile perfusion was provided by a roller pump (Model 5000; Sarns, Inc., Ann Arbor, MI). Normothermia, as measured by the esophageal thermistor, was maintained throughout the experimental procedure.

The experimental protocol is illustrated in Figure 1. Upon entry into the laboratory, animals were assigned randomly to receive one of four intravascular lamotrigine dosages during CPB; 0 (n = 10), 10 (n = 10), 25 (n = 10), or 50 (n = 10) mg/kg. To ensure concentrations in serum comparable with other experimental studies not using CPB, the dose of lamotrigine was augmented by a pump dilutional factor calculated as where blood volume was estimated as 75 ml/kg x body weight. Baseline hemodynamic and blood flow measurements were acquired before initiation of CPB. Arterial and venous cannulae were inserted, and the animal was placed on total CPB with adjustment of the pump flow rate (80 - 120 ml [middle dot] kg^-1[middle dot] min^-1) to maintain mean aortic pressure at 80 mmHg. After stabilizing for 30 min, repeat measurements were obtained (early CPB). Next, lamotrigine powder (courtesy of Glaxo Wellcome Company, Research Triangle Park, NC) was dissolved in sterile water, and the total dose was administered over 10 min into the venous reservoir (Figure 1). During the ensuing 45-min drug equilibration period, pump flow was increased to maintain mean aortic pressure at 60 mmHg, and the amount of fluid necessary to maintain the venous reservoir level constant was quantified. After repeat hemodynamic and blood flow measurements, each animal was subjected to 15 min of global cerebral ischemia by temporarily ligating the innominate and left subclavian arteries. During ischemia, pump flow was adjusted to maintain mean aortic pressure at 50 mmHg. Next, the animal was reperfused by releasing the ligatures at a mean aortic pressure of 60 mmHg, and hemodynamic and blood measurements were repeated 40 min later (Figure 1). Ventilation was resumed, and the animal was weaned from CPB over 2 to 3 min with the use of a dilute epinephrine infusion (1 mg mixed in 500 ml saline) to prevent right ventricular dilation during volume loading. The infusion of epinephrine was short-lived (30-120 s) and was discontinued in all animals after discontinuation of CPB. Final measurements were acquired 30 min later.

Central venous, sagittal sinus, and aortic pressures were measured using a Gould ES2000 electrostatic recorder (Gould, Valley View, OH). Arterial and sagittal sinus blood gases (Model 1306; Instrumentation Laboratory, Lexington, MA) and concentrations of hemoglobin in serum (CO-Oximeter Model 482; Instrumentation Laboratory) were measured repeatedly. CO-Oximeter saturations were calibrated for animal hemoglobin. After animals were stabilized on CPB, concentrations of glucose in plasma were determined by a glucometer (Lifescan Inc., Milpitas, CA), and plasma osmolality was measured using a vapor pressure osmometer (Wescor, Logan, UT).

Electroencephalographic signals were analyzed using a five-point scale previously validated during porcine cerebral ischemia. [20,21]Each recording was independently reviewed by two investigators blinded to animal treatment group and was assigned a numerical score, where 5 = normal electroencephalogram, indistinguishable from baseline electroencephalogram; 4 = mildly (25%) depressed amplitude and frequency from baseline; 3 = moderately (50%) depressed amplitude and frequency from baseline; 2 = markedly (75%) depressed amplitude and frequently from baseline; and 1 = isoelectric electroencephalogram. Interobserver reliability of the electroencephalographic analysis score was tested using a weighted [small kappa, Greek] statistic of the two observers.

During the experiment, artificial CSF (Na⁺147 mM, K⁺4.0 mM, Mg⁺⁺0.9 mM, Ca⁺⁺2.3 mM, and Cl^-157 mM in deionized water) was continually pumped at 2 [micro sign]l/min through the 4-mm length of dialysis tubing (diameter, 500 [micro sign]; molecular weight cutoff, 20,000 d). The effluent was collected during 20-min sampling periods (15 min during ischemia) and analyzed by blinded observers using high-performance liquid chromatography with naphthaldicarboxaldehyde derivatization for amino acid separation and quantification. [23] 

Before CPB, a sagittal sinus blood sample was obtained to determine baseline concentrations of S-100B protein in serum (Figure 1). After CPB was discontinued, a sample of CSF was aspirated via a needle inserted in the cisterna magna, while a venous sample was simultaneously aspirated from the sagittal sinus catheter. All samples were centrifuged, and the supernatant was analyzed for S-100B protein content using a monoclonal two-site immunoradiometric assay (Sangtec 100, Sangtec Medical AB, Bromma, Sweden). Samples were analyzed in duplicate and rejected if there was more than 10% variability. Using this technique, values as low as 0.1 [micro sign]g/l are quantifiable.

At the end of each experiment, a biopsy was performed of the dorsal cerebral cortex surrounding the microdialysis catheter insertion site using a high-speed suction device that deposited 150-250 mg of brain tissue into liquid nitrogen within 0.5 s. [20]The tissue sample was analyzed for protein content using a Lowry modification of the Folin phenol reagent method. [24]The supernatant was filtered and analyzed for high-energy phosphates and lactate. Concentrations of adenosine triphosphate, adenosine diphosphate, and adenosine monophosphate were measured by reverse-phase high-performance liquid chromatography [24]; concentration of lactate was determined enzymatically using a spectrophotometer. [25]The brain energy state was calculated as

energy charge =[ATP +(0.5)(ADP)][divided by][ATP + ADP + AMP]

Blood flow in tissue was measured by injecting 2 to 3 million radioactive microspheres (15 +/− 3 [micro sign]m diameter) labeled with strontium-85, cobalt-57, scandium-46, niobium-95, tin-113, or chromium-51, as previously described. [20,21,25]During spontaneous circulation at baseline and after CPB, microspheres were injected into the left atrium; during CPB, microspheres were injected directly into the aortic cannula. The order of microspheres was randomized, and each dose was selected to ensure that all regional brain tissue samples contained more than 400 microspheres. At the conclusion of each experiment, the heart was arrested using an intravenous injection of saturated potassium solution, and the animal was killed. Coronal sections were acquired from each kidney and the radioactivity of the renal cortex was assessed. The brain was removed, and total radioactivity was counted from the left and right cerebral hemispheres and from the cerebellum. Radioactivity was determined in each tissue specimen using a Packard gamma counter (Packard, Meriden, CT) with a 3-inch sodium iodide crystal; regional blood flow in tissue (ml [middle dot] min^-1[middle dot] 100 g^-1) was calculated using standard formulae. [20]The cerebral metabolic rate for oxygen (CMRO²; ml [middle dot] min^-1[middle dot] 100 g^-1) was calculated as ([arterial oxygen content - sagittal sinus oxygen content) x (cerebral blood flow)][divided by] 100. Systemic vascular resistance (mmHg [middle dot] ml^-1[middle dot] min) was calculated as [mean aortic pressure - central venous pressure][divided by] systemic flow. Cerebral vascular resistance (mmHg [middle dot] ml^-1[middle dot] min [middle dot] 100 g) was calculated as [mean aortic pressure - sagittal sinus pressure][divided by] cerebral blood flow. During cerebral ischemia, distal aortic pressure measured in the right internal mammary artery was substituted for aortic pressure measured in the femoral artery.

Blood samples acquired 45 min after infusion of lamotrigine during CPB were analyzed for concentration of lamotrigine by capillary-zone electrophoresis and expressed as [micro sign]g/ml. [26] 

Outcome variables for repeated measures, such as regional blood flows, hemodynamic variables, excitatory neurotransmitter, and electroencephalographic scores, were analyzed using analysis of variance for a two-factor experiment (lamotrigine dose and time) with repeated measures over time. Although electroencephalographic scores were measured using an ordinal scale, mean readings of the two observers were used for analysis. Other outcome variables without repeated measures (i.e., S-100B protein and brain metabolites) were analyzed by the Kruskal-Wallis test. Fisher's least significant difference procedure was used for multiple comparisons, with P < 0.005 as a comparison-wise error rate. To minimize the incidence of type I errors, an a priori decision was made to limit intragroup comparisons between adjacent values and between baseline and off-CPB values for each variable. Data are expressed as mean +/− SEM. In addition, electroencephalographic data are presented as median and interquartile (25-75%) ranges. Paired organ blood flows in the kidneys and cerebral hemispheres were compared using linear regression analysis.

The study was completed successfully in all animals, and CPB lasted 139 +/− 2 min (range, 130-175 min). There were no intergroup differences in the doses of fentanyl, diazepam, or isoflurane administered throughout the experimental procedure. Central venous pressure remained near or below zero during CPB. Values for arterial pH (range, 7.38-7.42), carbon dioxide tension (range, 38-42 mmHg), and oxygen tension (range, 225-300 mmHg) remained comparable among groups throughout the experiment. After administration of lamotrigine, concentrations of drug in serum were 12.5 +/− 2.1 [micro sign]g/ml in the 10-mg/kg group, 32.0 +/− 3.3 [micro sign]g/ml in the 25-mg/kg group, and 49.6 +/− 6.3 [micro sign]g/ml in the 50-mg/kg group.

Hemodynamic and temperature data acquired during the experimental procedure are shown in Table 1. Administration of lamotrigine caused a dose-dependent reduction in systemic vascular resistance, with a 50% decrease after the 50-mg/kg dose. Accordingly, pump flow rates were significantly higher after administration of lamotrigine. In contrast, lamotrigine produced no direct effects on cerebral vascular resistance. Administration of lamotrigine required greater amounts of fluid to be added to the venous reservoir to maintain a constant level (Table 2). As a consequence, concentrations of hemoglobin in serum decreased in the 25- and 50-mg/kg groups.

Adequate microsphere mixing during the study was confirmed by comparing paired regional blood flows in the cerebral hemispheres (right cerebral blood flow = 0.98 left cerebral blood flow + 0.43 ml [middle dot] min^-1[middle dot] 100 g^-1; n = 40 animals; r²= 0.941; P < 0.001) and kidneys (right renal blood flow = 0.99 left renal blood flow + 3.88 ml [middle dot] min^-1[middle dot] 100 g^-1; n = 40 animals; r (²)= 0.960; P < 0.001). Cerebral cortical and cerebellar blood flows remained comparable between groups throughout the experiment (Table 3). CMRO²was reduced 10-25% by CPB (Table 3), with further decreases after administration of lamotrigine. CMRO²after administration of 50 mg/kg lamotrigine was nearly 30% lower than in the 0- and 10-mg/kg groups. After cerebral ischemia, CMRO²was restored in all groups, with significant but comparable decreases when baseline and off-CPB values were compared.

Lamotrigine 50 mg/kg decreased electroencephalographic scores 15-20% more than the other groups (Table 4). The time interval between onset of cerebral ischemia and an isoelectric electroencaphalogram was similar in all groups (0 mg/kg, 14.2 +/− 0.8 s; 10 mg/kg, 14.6 +/− 0.8 s; 25 mg/kg, 14.9 +/− 0.6 s; 50 mg/kg, 14.7 +/− 1.0 s). Electroencephalographic scores were restored with reperfusion but remained lower than baseline in all groups after CPB was terminated. The weighted [small kappa, Greek] statistic between the two electroencephalographic graders was 0.83, indicating excellent interobserver agreement.

Concentrations of glutamate and aspartate in the brain from microdialysis are shown in Figure 2A and Figure 2B, respectively. Cerebral ischemia significantly but comparably increased concentrations of glutamate in the brain in the 0-, 10-, and 25-mg/kg groups, with further augmentation during reperfusion so that periischemic concentrations were elevated fivefold. In contrast, concentrations of glutamate in the brain were unchanged in the 50-mg/kg group and remained significantly lower than the other three groups during reperfusion (Figure 2A). Concentrations of aspartate in the brain showed similar although less dramatic changes (Figure 2B). Concentration of glutamate and aspartate returned to baseline values after CPB was terminated.

Concentrations of S-100B protein in sagittal sinus samples at baseline before CPB were not detectable in all groups. After ischemia and discontinuation of CPB, concentrations of S-100B protein in the sagittal sinus were elevated without intergroup differences (Table 5). Samples of CSF obtained simultaneously showed comparable concentrations of S-100B protein among groups, although concentrations in CSF were significantly greater than those in blood (P < 0.005). Overall, there was a weak but significant correlation between the concentrations of S-100B protein in the sagittal sinus and CSF after CPB (r²= 0.137; n = 40 animals; P < 0.02). In cortical biopsy samples taken at the conclusion of each experiment, there were no differences in concentrations of high-energy phosphate or brain lactate (Table 5). Accordingly, brain energy charges were similar among groups: 0.879 +/− 0.013 (0 mg/kg), 0.884 +/− 0.007 (10 mg/kg), 0.877 +/− 0.009 (25 mg/kg), and 0.892 +/− 0.009 (50 mg/kg; P = not significant).

These data indicate that pretreatment with lamotrigine 50 mg/kg administered during CPB significantly attenuated the release of excitatory neurotransmitters from normothermic cerebral ischemia. Other indices of neurologic morbidity, however, such as electroencephalographic and cerebral metabolic recovery, brain lactate and energy change, and concentrations of S-100B protein in CSF, were not improved by high-dose lamotrigine. Compared with our previous study of hypothermia, [20]better results were found with 28 [degree sign]C and 31 [degree sign]C management during experimental cerebral ischemia on CPB than with pretreatment with lamotrigine. Further, dose-dependent arterial and venous dilation produced by lamotrigine could complicate its prophylactic administration in high-risk patients, owing to a decrease in systemic vascular resistance, which required higher pump flow rates, and secondary hemodilution from supplemental administration of fluid.

In focal ischemia models, lamotrigine and other related compounds, such as BW1003C87 and BW619C87, block the presynaptic release of glutamate, causing a dose-related reduction in ischemic cortical injury, as assessed by infarct size and neurobehavioral outcome. [13,15]By acting on voltage-sensitive sodium channels, veratrine-induced but not potassium-induced excitatory amino acid release is reduced by lamotrigine without affecting other neurotransmitters, such as dopamine, adenosine, and adrenergic agonists. [13]These drugs produce multiple neuronal effects, including inhibiting high-frequency firing, delaying ischemia-induced anoxic depolarization, maintaining intracellular and extracellular sodium gradients to reduce the transcellular movement of glutamate and calcium, reducing postischemic cytotoxic edema, and inhibiting spreading depolarization [13]while not affecting core temperature or the production of heat shock protein in the brain. [14]Other alternative methods to attenuate excitatory neurotransmitter release with ischemia include adenosine analogues to block presynaptic A1 receptors, opioid antagonists, and hypothermia. [13]In clinical practice, adenosine agonists may produce significant hypotension, [27]and opioid antagonists may confound the anesthetic management of cardiac surgical patients.

In an attempt to reduce the need for systemic hypothermia, a study of phenyltriazine compounds such as lamotrigine to reduce cerebral excitotoxicity during ischemia on normothermic CPB clinical relevance. Although BW619C87 and BW1003C87 are more potent glutamate release inhibitors than lamotrigine, the clinical use of BW1003C87, in particular, is limited by antifolate side effects that cause anemia and thrombocytopenia and by teratogenicity. We chose lamotrigine because it is commercially available in an oral form, which permits preoperative loading, and because it has rapid absorption into the bloodstream, with 98% bioavailability, dose-linear pharmacokinetics, [17,18]and rapid penetration of brain tissue. [17]In the recommended therapeutic range of 4-10 [micro sign]g/ml concentration in serum, side effects from lamotrigine in patients, which include transient ataxia, impaired locomotion, dizziness, and skin rash, are rare.

Several experimental studies without extracorporeal circulation have demostrated the neuroprotective properties of lamotrigine. In a rat focal stroke model, Smith and Meldrum [13]reported dose-dependent reductions in cortical infarct size over a dose range of 3-20 mg/kg, whereas 50 mg/kg caused arterial hypotension sufficient to limit neuroprotection. In the current study, in which perfusion was maintained by increasing pump flow, this higher dose of lamotrigine was tolerated with greater suppression of excitatory neurotransmitter release. Lamotrigine 10 mg/kg administered 15 min after cardiac arrest in rats reduced CA1 hippocampal damage by 50%. [17]In gerbils with transient global cerebral ischemia, lamotrigine 25 mg/kg ablated excitatory neurotransmitter release, and 50 mg/kg significantly improved histologic and behavioral outcomes. [16]In a rabbit model of repetitive global cerebral ischemic insults, pretreatment with lamotrigine 50 mg/kg was superior to 20 mg/kg and equally efficacious as 30 [degree sign]C hypothermia to attenuate glutamate release. [28]Although concentrations in serum between 13 and 27 [micro sign]g/ml reduced global ischemic neuronal injury in rats [17]and gerbils, [14]much higher concentrations (45-52 [micro sign]g/ml) were required to attenuate excitatory neurotransmitter release during porcine cerebral ischemia during CPB, even when potential dilutional effects were prevented. The lack of benefit from lamotrigine in the current study may reflect reduced collateral oxygen delivery from the hemodilution, nonpulsatile blood flow associated with extracorporeal perfusion, [21,22]and potential interspecies variability. The concentrations in serum required to achieve this benefit greatly exceed the therapeutic range recommended as an anticonvulsant in patients [18,19]and may produce greater toxicity.

In this model with reduced collateral perfusion of 3-5 ml [middle dot] min^-1[middle dot] 100 g^-1, pretreatment with high-dose lamotrigine completely attenuated the release of excitatory neurotransmitters. Nevertheless, short-term neurologic recovery, as evidenced by S-100B protein release, cerebral metabolic and electroencephalographic recovery, and brain lactate, was not improved. Severe ischemia can cause intracellular calcium to reach a threshold level for neuronal toxicity independent of glutamate receptor activation. For example, with severe depletion of adenosine triphosphate and intracellular acidosis, activation of membrane-bound phospholipase C can induce inositol triphosphate-triggered release of intracellular calcium from the endoplasmic reticulum. [8]Furthermore, incomplete ischemia with collateral perfusion may produce greater neuronal injury because of a persistent low-flow delivery of exogenous calcium. [29]These findings are compatible with those reported for glutamate receptor antagonists that demonstrate greater protection in penumbral regions during focal ischemia than with severe forebrain or global cerebral ischemia. [8] 

Several complex issues in this study should be addressed. Brain high-energy metabolites were normalized at the time of biopsy, which likely reflects the reparative time interval between reperfusion and discontinuation of the experiment. Earlier biopsy specimens may have revealed differences but were not performed, owing to confounding influences on microdialysis sampling. Second, although electroencephalographic scores did not differ after CPB, the failure of the 50-mg/kg group to show greater recovery may reflect direct depressant effects of lamotrigine on cortical electrical activity, presumably secondary to sodium channel inhibition and antiepileptic properties. [17]With similar concentrations of brain lactate and S-100B protein in the sagittal sinus and CSF, however, it would be difficult to attribute short-term recovery properties to lamotrigine despite attenuation of brain excitotoxicity. Relative to the profound benefits from systemic hypothermia found in this model, [20]cooling to 31 [degree sign]C or 28 [degree sign]C provides similar suppression of glutamate and aspartate while reducing other indices of acute neurologic morbidity. These findings underscore the beneficial effects of preischemic cooling other than excitotoxicity alone. The levels of excitatory amino acids observed by microdialysis were not linked to neurologic outcome, however, and further examination of these therapies in surviving animals using neurohistopathology as a therapeutic end point is required to demonstrate neuroprotection. Finally, mean aortic blood pressure after administration of lamotrigine was maintained at 60 mmHg by adjusting pump flow, which required additional administration of fluid to the pump reservoir to maintain a safe level. Consequently, hemodilution occurred in the 25- and 50-mg/kg groups. Another approach would be the titration of a systemic vasoconstrictor, such as phenylephrine, that has no direct cerebral effects [30]but can alter splanchnic blood flow distribution during normothermic CPB. [31] 

A pharmacologic approach using lamotrigine to reduce excitatory neurotransmitter release during transient global cerebral ischemia on normothermic CPB was studied. Lamotrigine 50 mg/kg significantly attenuated glutamate and aspartate recovery in microdialysis samples without improving other neurologic markers. The clinical usefulness of lamotrigine may be limited by the high doses required and the associated vaso- and venodilatory responses during CPB.

The authors thank Greg Asimakis, Ph.D., Galveston, Texas, for brain metabolic analysis; Z. K. Shihabi, Ph.D., Departments of Neurology and Pathology, Wake Forest University School of Medicine, Winston-Salem, North Carolina, for serum lamotrigine analysis; David McAdoo, Ph.D., Galveston, Texas, for excitatory neurotransmitter analysis; Tatsuo Uchida, M.S., Galveston, Texas, for statistical analysis; and Faith McLellan, Ph.D., Galveston, Texas, for editorial review.