HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): John A. Kern Irving L. Kron Curtis G. Tribble Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gangemi, J. J. Articles by Tribble, C. G. Search for Related Content PubMed PubMed Citation Articles by Gangemi, J. J. Articles by Tribble, C. G. Related Collections Related Article

Ann Thorac Surg 2000;69:1744-1748
© 2000 The Society of Thoracic Surgeons

---

Original articles: Cardiovascular

Retrograde perfusion with a sodium channel antagonist provides ischemic spinal cord protection

^a Division of Thoracic and Cardiovascular Surgery, Department of Surgery, University of Virginia Health Sciences Center, Charlottesville, Virginia, USA

Address reprint requests to Dr Tribble, Department of Surgery, University of Virginia Health System, PO Box 800679, Charlottesville, VA 22908
e-mail: ctribble{at}virginia.edu

Presented at the Forty-sixth Annual Meeting of the Southern Thoracic Surgical Association, San Juan, Puerto Rico, Nov 4–6, 1999

	Abstract

Top Abstract Introduction Material and methods Results Comment References

 
Background. Neuronal voltage-dependent sodium channel antagonists have been shown to provide neuroprotection in focal and global cerebral ischemic models. We hypothesized that retrograde spinal cord venous perfusion with phenytoin, a neuronal voltage-dependent sodium channel antagonist, would provide protection during prolonged spinal cord ischemia.

Methods. In a rabbit model, spinal cord ischemia was induced for 45 minutes. Six groups of animals were studied. Controls (group I, n = 8) received no intervention during aortic cross-clamping. Group II (n = 8) received systemic phenytoin (100 mg). Group III (n = 4) received systemic phenytoin (200 mg).Group IV (n = 8) received retrograde infusion of room temperature saline (22°C) only. Group V (n = 8) and group VI (n = 9) received retrograde infusion of 50 mg and 100 mg of phenytoin, respectively, (infusion rate: 0.8 mL · kg^-1 · min^-1 during the ischemic period). Mean arterial blood pressure was monitored continuously. Animals were allowed to recover for 24 hours before assessment of neurologic function using the Tarlov scale.

Results. Tarlov scores (0 = complete paraplegia, 1 = slight lower limb movement, 2 = sits with assistance, 3 = sits alone, 4 = weak hop, 5 = normal hop) were as follows (mean ± SEM): group I, 0.50 ± 0.50; group II, 0.25 ± 0.46; group IV, 1.63 ± 0.56; group V, 4.13 ± 0.23; and group VI, 4.22 ± 0.22 (p < 0.0001 V, VI versus I, II, IV by analysis of variance). No differences in mean arterial blood pressure were observed. All animals in group III became profoundly hypotensive and died before the conclusion of the 45-minute ischemic time.

Conclusions. Retrograde venous perfusion of the spinal cord with phenytoin, a voltage-sensitive sodium channel blocker, is safe and provides significant protection during prolonged spinal cord ischemia.

	Introduction

Top Abstract Introduction Material and methods Results Comment References

 
Spinal cord ischemic injury during thoracoabdominal aortic operation remains a potentially devastating outcome despite various methods of protection [1–4]. The exact cellular mechanisms that underlie ischemic spinal cord injury remain to be defined. Extensive research into the pathogenesis of a variety of neurologic diseases has provided more evidence that glutamate neurotoxicity might contribute to neuronal injury during certain acute insults, including ischemia [5, 6]. By binding to the postsynaptic N-methyl-D-aspartate (NMDA) receptors, glutamate causes an influx of sodium, calcium, and secondarily, chloride. Neuronal swelling and cell lysis ensue. Neurons not killed by this rapid osmotic effect can be damaged more slowly by increases in intracellular calcium leading to further receptor activation, proteolysis of neurofilaments, irreversible mitochondrial damage, and ultimately to programmed cell death, or apoptosis [7, 8].

Recent studies have shown that voltage-sensitive sodium channel blockers provide a powerful mechanism of neuroprotection in animal models of focal and global cerebral ischemia. [9, 10] At the initial stage of the ischemic cascade, sodium influx causes cellular depolarization that contributes to a number of undesirable functions, including opening of calcium and potassium channels and calcium-dependent glutamate release. In addition, continued sodium influx elevates cytosolic sodium concentrations to abnormally high levels, which leads to other detrimental changes such as depletion of adenosine triphosphate (ATP) stores, chloride inflow to balance net charge and subsequent cellular swelling, reversal of sodium/calcium transport, and reversal of carrier-mediated glutamate and other neurotransmitter transport. Sodium channel opening occurs at an early stage in the biochemical cascade caused by ischemia and generally accelerates the ischemic cycle.

Sodium channel blockage has been shown to delay the onset of hypoxic depolarization [11], which is a serious load on brain energy stores [12] and might contribute to cellular damage during ischemia. Also during ischemia, blockage of sodium channels reduces sodium loading and cellular depolarization, which in turn reduces calcium influx via the sodium-calcium exchange mechanism. Recently, the antiischemic actions of several sodium channel blockers have been ascribed to reduce glutamate release. Reducing ischemic glutamate release could provide neuroprotection of voltage-sensitive sodium channel blockers. In addition, ATP is depleted early in the cascade of events in brain ischemia [13]. In neurons, ATP is used primarily to fuel the sodium- and potassium-transporting ATPase. Thus, reducing influx would be particularly beneficial because it would spare cellular ATP.

Voltage-sensitive sodium channel blockers are a well known, well tolerated, and frequently used class of drugs. They are often used as local anesthetic agents and antiarrhythmics. In addition, some of these agents, such as carbemazapime, lidocaine, and phenytoin, are also used to treat certain forms of nerve damage, such as trigeminal neuralgia and diabetic neuropathy. However, these agents are used most commonly as anticonvulsants in the treatment of tonic-clonic and partial seizures.

Using a model of prolonged spinal cord ischemia, we hypothesized that retrograde venous perfusion of the ischemic spinal cord with phenytoin, a voltage-sensitive sodium channel blocker, would be hemodynamically safe and provide superior neurologic protection compared with its systemic infusion.

	Material and methods

Top Abstract Introduction Material and methods Results Comment References

 
All protocols in this study were reviewed and approved by the Animal Review Committee of the University of Virginia. All animals received humane care in compliance with the "Principles of Laboratory Animal Care" as described by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" (National Institutes of Health publication 85-23, revised 1985). Adult New Zealand white rabbits of either sex were used throughout this study.

Preparation of experimental animals
Rabbits (2.8 to 3.2 kg) were anesthetized with an intramuscular injection of xylazine (10 mg) and ketamine (100 mg). Once the animals were adequately sedated, an ear vein catheter was placed for administration of additional medications and intravenous fluids. The animals were then intubated, placed supine on a heated operating table, and ventilated (Harvard Rodent Ventilator Model 683; Harvard Apparatus, South Natick, MA) with a mixture of 98% oxygen and 2% halothane. An ear arterial catheter was placed for continuous monitoring of arterial pressure. Heparin sodium (2000 U) was administered intravenously and allowed to circulate for 5 minutes. During this interval, the abdomen was steriley prepared and draped. A midline laparotomy was made and the viscera reflected to the right. After the retroperitoneum was opened, the abdominal aorta and inferior vena cava (IVC) were identified and isolated with soft vessel loops just distal to the left renal artery and vein and just above their bifurcations.

Experimental protocol
We studied six groups of animals. To induce spinal cord ischemia, atraumatic vascular clamps were used to isolate the infrarenal portions of the aorta and IVC proximally and distally. In all groups clamps were applied rapidly to the aorta just distal to the left renal artery and just above the bifurcation, and on the IVC just proximal to the confluence of the iliac veins and just distal to the left renal vein. These clamps were left in place for 45 minutes. The control animals (group I, n = 8) had 45 minutes of aortic cross-clamping with no interventions. The first experimental group (group II, n = 8) received a systemic infusion of phenytoin (100 mg in 250 mL normal saline) through an ear vein catheter. Group III (n = 4) received a 200-mg systemic infusion of phenytoin through the ear vein catheter. In the remaining three experimental groups, a 24-gauge intravenous catheter (Johnson and Johnson Medical Inc, Arlington, TX) was inserted into the midportion of the IVC immediately after application of the clamps. This catheter was used for administration of retrograde saline or saline plus drug during the 45-minute ischemic period. Group IV (n = 8) received retrograde infusion of room temperature saline (22°C). Groups V and VI received retrograde infusion of room temperature saline with phenytoin 50 mg (n = 8) and phenytoin 100 mg (n = 9), respectively. All infusates were delivered at a constant rate of 0.8 mL · kg^-1 · min^-1 using an infusion pump (Syringe infusion pump 22; Harvard Apparatus, South Natick, MA). The infusion was begun immediately after placement of the catheter and continued throughout the 45-minute ischemic period.

At the conclusion of the 45-minute ischemic interval, the catheter was withdrawn and the venotomy quickly closed with a figure of eight stitch using 7-0 polypropylene sutures, taking care not to compromise the lumen of the IVC. The clamps were removed rapidly and the abdomen closed. The animals were allowed to recover from anesthesia before being returned to the holding area, where they were permitted to move freely about their cages and were provided food and water ad libitum. After 24 hours, the animals were evaluated for hindleg function by an observer masked to group allocation, and they were graded with the Tarlov scale (0 = complete paralysis, 1 = minimal movement, 2 = stands with assistance, 3 = stands alone, 4 = weak hop, and 5 = normal hop). The animals were then sacrificed using an overdose injection of sodium pentobarbital.

Data acquisition during aortic and caval cross-clamping
Arterial blood pressure data were collected and recorded before application of the clamps, during the ischemic interval, and for 5 minutes after release of the clamps by customized digital data acquisition software (Workbench PC; Strawberry Tree, Inc, Sunnyvale, CA). In 3 of 8 animals in each group, rectal temperature was measured just before release of the clamps to determine the average postprocedure temperature in each group.

Statistical analysis
All results are expressed as the mean ± standard error of the mean. Data were analyzed for between-group differences using analysis of variance. Specific hypotheses were tested using contrast analysis. Significance was defined as a p value less than 0.05. All analyses were done using SPSS Software (SPSS Inc, Chicago, IL).

	Results

Top Abstract Introduction Material and methods Results Comment References

 
Neurologic function using Tarlov scores is shown in Figure 1. The control group that had 45 minutes of spinal cord ischemia had an average Tarlov score (mean ± standard error of the mean) of 0.50 ± 0.50. Group II, which received systemic infusion of phenytoin (100 mg), had an average Tarlov score of 0.25 ± 0.46. All four rabbits in group III that received the systemic infusion of 200 mg of phenytoin became profoundly hypotensive during the infusion and died before the 45-minute ischemic period was completed. We decided not to complete all 8 animals for that portion of the experiment. Group IV, which received retrograde saline, had an average Tarlov score of 1.63 ± 0.56, indicating a small but not significant degree of spinal cord protection. Retrograde infusion of phenytoin provided significant protection from prolonged spinal cord ischemia. Group V, which received retrograde phenytoin (50 mg), had an average Tarlov score of 4.13 ± 0.23, and group VI, which received retrograde phenytoin (100 mg), had an average score of 4.22 ± 0.22. Analysis of variance of these five groups showed that retrograde infusion of both phenytoin doses (groups V and VI) compared with the other groups provided superior protection (p < 0.0001) . There was no significant neurologic difference between the retrograde infusion doses of phenytoin (groups V and VI).

View larger version (33K): [in this window] [in a new window]   Fig 1. Average Tarlov scores after 24 hours of recovery.

View larger version (20K): [in this window] [in a new window]   Fig 2. Mean arterial pressure (MAP) during the 45-minute ischemic period and the first 10 minutes of reperfusion. There were no significant differences with changes of MAP between the groups. (IVC= inferior vena cava.)

	Comment

Top Abstract Introduction Material and methods Results Comment References

Our results supported our hypothesis. All animals were allowed to recover 24 hours before evaluation, to allow for full recovery from the surgical procedure. The control group, which had 45 minutes of spinal cord ischemia with no intervention, had a mean Tarlov score of 0.50 ± 0.50 (a Tarlov score of 0 represents complete paraplegia). Animals in the groups that received retrograde venous perfusion of phenytoin had significantly better neurologic function than other groups in the study. The mean Tarlov scores of the groups receiving retrograde spinal cord perfusion with 50 mg phenytoin and 100 mg phenytoin were 4.14 ± 0.46 and 4.22 ± 0.55, respectively (a Tarlov score of 5 indicates a normal hop). There was no statistically significant difference between groups V and VI, suggesting that there is no advantage to a higher dose of this neuroprotective agent. Concentrated, localized delivery using retrograde venous perfusion does not require high systemic doses of the agent to provide protection during prolonged ischemia.

Another advantage of localized retrograde perfusion of a sodium channel blocker is its hemodynamic safety. There were no significant changes in blood pressure in any group throughout the experiment. Although there was no statistical difference in mean arterial pressures between the retrograde perfusion groups and the systemic (100 mg) infusion group, there was an obvious difference with respect to mean arterial pressure in the group that received a 200-mg systemic infusion of phenytoin, which was clearly a toxic dose. Although animals tolerated the 100-mg systemic infusion of phenytoin, it failed to protect the spinal cord during prolonged periods of spinal cord ischemia.

Previously, our lab has found significant neurologic protection using retrograde venous perfusion during spinal cord ischemia [14, 15]. Specifically, retrograde perfusion of the spinal cord with a profoundly hypothermic (4°C) adenosine solution preserved neurologic function during spinal cord ischemia. Hypothermia alone reduces glutamate accumulation and excitotoxic damage. The solutions used in this study, which were at room temperature (22°C), were neuroprotective (groups V and VI) without profound hypothermia. Avoiding the potential hazards of systemic hypothermia (coagulopathy, arrhythmia, myocardial ischemia, and rewarming metabolic demands) while providing neurologic protection from glutamate toxicity is clinically appealing.

The rabbit model of spinal cord ischemia used in this experiment has been a consistent model in our laboratory. In rabbits the blood supply to the spinal cord is different than in humans because it is segmentally distributed. However, we have found similar results when we applied these techniques of neuroprotection to a swine model of spinal cord ischemia (pigs have a vascular anatomy similar to that of humans) [14, 15].

Most research studying the reduction of ischemia-induced neurotoxic effects of glutamate has been done in models of focal and global cerebral ischemia. In addition to sodium channel blockers, NMDA receptor antagonists, adenosine receptor activators, calcium-channel blockers, and hypothermia have been shown to reduce glutamate accumulation and neurotoxic effects in cerebral models [9].

Spinal cord ischemic injury secondary to the neurotoxic effects of glutamate has been confirmed previously [16, 17]. Research in the area of reducing glutamate toxicity, although limited compared with cerebral models, has focused on NMDA receptor antagonism [18–20]. The involvement of the NMDA receptors in the pathogenesis of glutamate neurotoxicity provides a novel approach to pharmacotherapy directed at preventing or ameliorating ischemic neuronal injury. Using our retrograde technique, we recently showed that retrograde venous perfusion of MK-801, a noncompetitive NMDA receptor antagonist, provided significant neurologic protection compared with systemic infusion during prolonged spinal cord ischemia [21]. The highly concentrated, localized delivery of the neuroprotective agent that is possible with retrograde venous perfusion offers a safe and effective means of spinal cord protection.

Although the neuroprotective effects of NMDA receptor antagonists are encouraging in experimental models in terms of preventing the accumulation of glutamate and subsequent neurotoxic effects, the clinical experience with competitive and noncompetitive NMDA receptor antagonists has been disappointing. Clinical trials of noncompetitive NMDA receptor antagonists at moderate dosages have been associated with light-headedness, dizziness, and paresthesias that progress to disinhibition, nystagmus, and diplopia. Slightly higher doses have been associated with paranoid ideation, hallucinations, and significant dose-dependent hypertension in healthy volunteers, with increases in mean arterial blood pressure of up to 30 mm Hg [22, 23]. Similar studies with competitive NMDA receptor antagonists caused delayed central nervous system effects such as severe anxiety, agitation, nightmares for several nights, acute paranoid psychosis, and hallucinations [24, 25].

Voltage-sensitive sodium channel blockers are used frequently as anticonvulsants, antiarrhythmics, and local anesthetic agents. This class of drug is well tolerated both orally and intravenously. The potential for applying this type of neuroprotective agent using a technique of retrograde spinal cord perfusion during prolonged thoracoabdominal aneurysm operation has exciting and potential clinical applications.

	Acknowledgments

 
We are grateful to Mr Anthony J. Herring for invaluable technical assistance.

	References

Top Abstract Introduction Material and methods Results Comment References

 

1. Crawford E.S., Svensson L.G., Hess K.R., et al. A prospective randomized study of cerebrospinal fluid drainage to prevent paraplegia after high-risk surgery on the thoracoabdominal aorta. J Vasc Surg 1991;13:36-46.[Medline]
2. Frank S.M., Parker S.D., Rock P., et al. Moderate hypothermia, with partial bypass and segmental sequential repair for thoracoabdominal aortic aneurysm. J Vasc Surg 1994;19:687-697.[Medline]
3. Forbes A.D., Slimp J.C., Winn R.K., Verrier E.D. Inhibition of neutrophil adhesion does not prevent ischemic spinal cord injury. Ann Thorac Surg 1994;58:1064-1068.[Abstract]
4. Agee J.M., Flanagan T., Blackbourne L.H., Kron I.L., Tribble C.G. Reducing postischemic paraplegia using conjugated superoxide dismutase. Ann Thorac Surg 1991;51:911-915.[Abstract]
5. Rothman S.M., Olney J.W. Glutamate and the pathophysiology of hypoxic-ischemic brain damage. Ann Neurol 1986;19:105-111.[Medline]
6. Choi D.W., Rothman S.M. The role of glutamate neurotoxicity in hypoxic-ischemic neuronal death. Annu Rev Neurosci 1990;13:171-182.[Medline]
7. Baudry M., Bundman M.C., Smith E.K., Lynch G.S. Micromolar calcium stimulates proteolysis and glutamate binding in rat brain synaptic membranes. Science 1981;212:937-938.[Abstract/Free Full Text]
8. Gilbert D.S., Newby B.J. Neurofilament disguise, destruction and discipline. Nature 1975;256:586-589.[Medline]
9. Owen A.J., Ijaz S., Miyashita H., Wishart T., Howlett W., Shuaib A. Zonisamide as a neuroprotective agent in an adult gerbil model of global forebrain ischemia. Brain Res 1997;770:115-122.[Medline]
10. Graham S.H., Chen J., Lan J., Leach M.J., Simon R.P. Neuroprotective effects of a use-dependent blocker of voltage-dependent sodium channels, BW619C89, in rat middle cerebral artery occlusion. J Pharmacol Exp Ther 1994;269:854-859.[Abstract/Free Full Text]
11. Weber M.L., Taylor C.P. Damage from oxygen and glucose deprivation in hippocampal slices is prevented by tetrodotoxin, lidocaine and phenytoin without blockade of action potentials. Brain Res 1994;664:167-177.[Medline]
12. Nedergaard M., Astrup J. Infarct rim. J Cereb Blood Flow Metab 1986;6:607-615.[Medline]
13. Kiedrowski L., Wroblewski J.T., Costa E. Intracellular sodium concentration in cultured cerebellar granule cells challenged with glutamate. Mol Pharmacol 1994;45:1050-1054.[Abstract]
14. Parrino P.E., Kron I.L., Ross S.D., et al. Spinal cord protection during aortic cross-clamping using retrograde venous perfusion. Ann Thorac Surg 1999;67:1589-1595.[Abstract/Free Full Text]
15. Ross S., Gangemi J., Kern J., et al. Hypothermic retrograde venous perfusion cools the spinal cord, and reduces paraplegia during thoracic aortic surgery. J Thorac Cardiovasc Surg 2000;119:588-595.[Abstract/Free Full Text]
16. Mori A., Ueda T., Nakamichi T., et al. Detrimental effects of exogenous glutamate on spinal cord neurons during brief ischemia in vivo. Ann Thorac Surg 1997;63:1057-1062.[Abstract/Free Full Text]
17. Choi D.W. Glutamate receptors and the induction of excitotoxic neuronal death. Prog Brain Res 1994;100:47-51.[Medline]
18. Yum S.W., Faden A.I. Comparison of the neuroprotective effects of the N-methyl-D-aspartate antagonist MK-801 and the opiate-receptor antagonist nalmefene in experimental spinal cord ischemia. Arch Neurol 1990;47:277-281.[Abstract]
19. Rokkas C.K., Helfrich L.R., Jr, Lobner D.C., Choi D.W., Kouchoukos N.T. Dextrorphan inhibits the release of excitatory amino acids during spinal cord ischemia. Ann Thorac Surg 1994;58:312-320.[Abstract]
20. Madden K.P., Clark W.M., Zivin J.A. Delayed therapy of experimental ischemia with competitive N-methyl-D- aspartate antagonists in rabbits. Stroke 1993;24:1068-1071.[Abstract/Free Full Text]
21. Gangemi J, Kern J, Kron I, Ross S, Shockey K, Tribble C. Retrograde venous perfusion of an NMDA receptor antagonist provides spinal cord protection during aortic crossclamping. Surg Forum Vol L, 1999:183–4.
22. Schmitt B., Netzer R., Fanconi S., Baumann P., Boltshauser E. Drug refractory epilepsy in brain damage. J Neurol Neurosurg Psychiatry 1994;57:333-339.[Abstract/Free Full Text]
23. Albers G.W., Atkinson R.P., Kelley R.E., Rosenbaum D.M. Safety, tolerability, and pharmacokinetics of the N-methyl-D-aspartate antagonist dextrorphan in patients with acute stroke. Dextrorphan Study Group. Stroke 1995;26:254-258.[Abstract/Free Full Text]
24. Clark W., Coull B. Randomised trial of CGS19755, a glutamate antagonist, in acute ischemic stroke treatment. Neurology 1994;44(Suppl 2):A270.
25. Kristensen J.D., Svensson B., Gordh T., Jr The NMDA-receptor antagonist CPP abolishes neurogenic ‘wind-up pain’ after intrathecal administration in humans. Pain 1992;51:249-253.[Medline]

This article has been cited by other articles:

				T. B. Reece, D. O. Okonkwo, P. I. Ellman, T. S. Maxey, C. Tache-Leon, P. S. Warren, J. J. Laurent, J. Linden, I. L. Kron, C. G. Tribble, et al. Comparison of Systemic and Retrograde Delivery of Adenosine A(2A) Agonist for Attenuation of Spinal Cord Injury After Thoracic Aortic Cross-Clamping. Ann. Thorac. Surg., March 1, 2006; 81(3): 902 - 909. [Abstract] [Full Text] [PDF]

				E. Shi, T. Kazui, X. Jiang, N. Washiyama, K. Suzuki, K. Yamashita, and H. Terada NS-7, a novel Na+/Ca2+ channel blocker, prevents neurologic injury after spinal cord ischemia in rabbits J. Thorac. Cardiovasc. Surg., February 1, 2005; 129(2): 364 - 371. [Abstract] [Full Text] [PDF]

				T.-A. Miyamoto and K.-J. Miyamoto Is retrograde perfusion necessary for spinal cord protection? Ann. Thorac. Surg., April 1, 2002; 73(4): 1360 - 1360. [Full Text] [PDF]

				J. J. Gangemi, J. A. Kern, I. L. Kron, and C. G. Tribble Reply Ann. Thorac. Surg., April 1, 2002; 73(4): 1360 - 1361. [Full Text] [PDF]

				I. Y.P. Wan, G. D. Angelini, A. J. Bryan, I. Ryder, and M. J. Underwood Prevention of spinal cord ischaemia during descending thoracic and thoracoabdominal aortic surgery Eur. J. Cardiothorac. Surg., February 1, 2001; 19(2): 203 - 213. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): John A. Kern Irving L. Kron Curtis G. Tribble Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gangemi, J. J. Articles by Tribble, C. G. Search for Related Content PubMed PubMed Citation Articles by Gangemi, J. J. Articles by Tribble, C. G. Related Collections Related Article