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

This Article Abstract 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): Guo-Wei He Franklin L. Rosenfeldt James Tatoulis Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by He, G.-W. Articles by Tatoulis, J. Search for Related Content PubMed PubMed Citation Articles by He, G.-W. Articles by Tatoulis, J.

J Thorac Cardiovasc Surg 1994;107:1440-1444
© 1994 Mosby, Inc.

SURGERY FOR ACQUIRED HEART DISEASE

Pharmacologic dilatation of the internal mammary artery during coronary bypass grafting

Melbourne, Australia

From Baker Medical Research Institute and Epworth Hospital, Melbourne, Australia.

Received for publication Jan. 7, 1993. Accepted for publication Oct. 8, 1993. Address for reprints: Franklin L. Rosenfeldt, MD, FRACS, Baker Medical Research Institute, PO Box 348, Prahran, Victoria 3181, Australia.

Abstract

Spasm of the internal mammary artery during coronary bypass grafting is a widely recognized problem during and after mobilization of the IMA. On the basis of previous laboratory studies, we have developed a buffered vasodilator solution containing glyceryl trinitrate and verapamil (pH 7.4). When tested in human internal mammary artery segments in the organ bath, this solution caused full relaxation of the segments with a 1- to 2-minute onset and a duration of action of more than 2 hours. In 31 patients undergoing internal mammary artery grafting, flow through the internal mammary artery was measured immediately after mobilization and 20 minutes later. In 10 untreated patients, flow increased by 13% from 41.8 ± 7.1 to 47.3 ± 7.5 ml/min ( p < 0.025). In 11 patients, intraluminal injection of glyceryl trinitrate–verapamil solution into the internal mammary artery on one side caused an increase in flow of 55 ± 10 ml/min (95%), which was greater than that caused by Ringer's solution, 22 ± 8 ml/min (53%), in the opposite internal mammary artery ( p < 0.025). In another 10 patients intraluminal injection of glyceryl trinitrate–verapamil solution in one internal mammary artery caused an increase in flow of 57.9 ± 8.7 ml/min (107%), which was similar to that caused by papaverine solution (pH 5.2) in the opposite internal mammary artery of 45.0 ± 12.3 ml/min (80%). We conclude that intraluminal injection of vasodilator solution is effective in dilating the IMA graft and that because of its rapid onset, long action, and neutral pH, glyceryl trinitrate–verapamil solution may be preferable to papaverine. (J THORAC CARDIOVASC SURG 1994;107:1440-4)

Spasm of the internal mammary artery (IMA) during mobilization from the chest wall may lead to difficulty in assessing the adequacy of the IMA to supply blood to the targeted coronary bed, as well as making the anastomosis technically difficult. Postoperative spasm of the IMA may cause major hemodynamic problems. ^1,2 In previous studies, we ^3-6 found that isolated human IMA ring segments contracted strongly to a thromboxane A₂ analog (U46619) and also to serotonin and norepinephrine. High levels of thromboxane and catecholamines are known to occur after coronary artery bypass grafting, ^7-10 and these elevations may predispose to spasm of the IMA.

We ¹¹ have previously formulated a new vasodilator solution containing glyceryl trinitrate (GTN) and verapamil (GV solution) to dilate vascular grafts during coronary bypass grafting and applied it successfully to saphenous vein grafts. The aims of the present study were (1) to determine in vitro the efficacy and time course of action of GV solution in isolated ring segments of the human IMA and (2) during coronary artery bypass grafting to evaluate the ability of GV solution to dilate the IMA and to compare the effect of GV solution with that of conventional papaverine solution.

MATERIALS AND METHODS

In vitro study
During IMA graft operations, discarded distal ends of human IMA segments were collected and placed in oxygenated Krebs solution at 4° C, transferred to the laboratory, and placed in organ baths as previously described. ^3-6

After 30 minutes, potassium (25 mmol/L) or U46619 (15 nmol/L) was added to the organ bath to cause a sustained active contraction. After this contraction had stabilized (10 to 20 minutes), verapamil (30 µmol/L, -4.5 log mol/L) and GTN (30 µmol/L, -4.5 log mol/L) were added to the organ bath. The time course of the relaxation caused by verapamil and GTN was measured.

In vivo study
Solutions used
The formula for GV solution was as follows: GTN (2.5 mg), verapamil (5 mg), 8.4% sodium bicarbonate (0.2 ml), and Ringer's solution (300 ml). In this solution, the concentrations of verapamil and GTN were each -4.5 log mol/L (30 µmol/L). This concentration of each agent given separately causes full relaxation of human IMA in vitro. ³ The addition of sodium bicarbonate was necessary to bring the pH from about 4.8 to 7.4.

The formula for the papaverine solution (concentration 0.8 mmol/L; pH 4.4 to 4.8) for intraluminal use was 30 mg papaverine in 100 ml Ringer's solution.

Measurements of IMA flow
In 31 patients undergoing IMA graft operations, one or both IMA pedicles were mobilized. After heparinization, the distal end of the IMAs were divided and the blood flow through the IMAs (free flow) at zero distal resistance was recorded by collecting the shed blood into a measuring cylinder. Three measurements were made for 15 seconds each and averaged to calculate the blood flow per minute. The patient was randomized to receive one of three treatments. A Weck hemoclip was then used to occlude the IMA at its distal end. Pericardiotomy and arterial and venous cannulation were then performed. Before cardiopulmonary bypass was commenced, the tip of the IMA, including the hemoclip, was cut off to open the IMA. The systemic blood pressure was maintained at the same mean level ±10% as for the initial measurements by transfusion or withdrawal of blood through the cannulas. The free flow of the IMA was measured again.

TIME CONTROL
In 10 patients undergoing single (left) IMA grafting, no treatment was given between the initial and final measurements. This was the time control group.

COMPARISON I: GTN + VERAPAMIL VERSUS BUFFERED RINGER'S SOLUTION
In 11 patients undergoing bilateral IMA grafting, after the initial measurements had been made, a 25-gauge plastic cannula was inserted into the distal end of the IMA. Then, 3 ml of GV solution was injected slowly into the left or right IMA, chosen at random. A hemoclip was then placed across the distal end of the vessel. GV solution was sprayed over the IMA pedicle and a gauze swab soaked with the same solution was used to wrap the pedicle. The IMA on the opposite side was treated in the same way, except that 3 ml of buffered Ringer's solution (vehicle for GV solution) was injected into the IMA and sprayed over the pedicle.

COMPARISON II: GTN + VERAPAMIL VERSUS PAPAVERINE
In 10 patients undergoing bilateral IMA grafting, after the initial measurements of IMA flow, GV solution was used as in comparison I in the left or right IMA, chosen at random, and the same volume of the papaverine solution was injected into the other IMA. Papaverine solution was sprayed over the IMA pedicle and a gauze swab soaked in the same solution was used to wrap the pedicle.

Data analysis
In the in vitro study, the percentage relaxation of precontraction at each minute was averaged from six rings (taken from six different patients) for either potassium- or U46619-induced contraction.

The Student's paired t test was used to compare the initial and final flow and the initial and final pressure measurements. The Student's unpaired t test was used to compare the initial flow, the increment of flow, and the initial arterial pressure between treatments within each comparison. One-way analysis of variance was used to compare the pretreatment flow and arterial pressure and the time between measurements.

RESULTS

In vitro study
In both potassium- or U46619-contracted IMA rings, verapamil (-4.5 log mol/L) plus GTN (-4.5 log mol/L) caused rapid onset and full relaxation (Fig. 1). In potassium-contracted IMA segments, the time taken to achieve 50% of maximum relaxation was less than 1 minute. In U46619-contracted IMA segments, this time was approximately 2 minutes. The response lasted more than 2 hours.

View larger version (15K): [in this window] [in a new window]   Fig. 1. Average relaxation responses to GTN (30 µmol/L) plus verapamil (30µmol/L) in the human IMA segments precontracted by potassium (K +, 25 µmol/L) or thromboxane A2 mimetic U46619 (15 nmol/L). Force during relaxation was measured at 1-minute intervals and expressed as percentage of the force during precontraction. Symbols represent average data from six vessels from six patients.

Time control
The IMA flow increased by 13% over 21.6 minutes (p < 0.025). The increment of flow was 5.5 ± 1.9 ml/min (Table I).

Comparison II: GTN + verapamil versus papaverine
GV solution increased IMA flow by 107% (p < 0.001). Intraluminal papaverine increased IMA flow by 80% (p < 0.025) (Table I). No statistically significant difference between these two groups was detected in flow after treatment or in the increment in flow produced. Overall, in the 21 patients receiving GV solution in the two comparisons, IMA flow increased from 57.1 ± 9.8 to 112.9 ± 12.9 ml/min, an increase of 98%.

There was no significant difference in the time interval between the initial and the final IMA flow measurements for each treatment (p > 0.05) or between the groups in the mean arterial pressure during the initial flow measurement (p > 0.05). Also, there was no significant difference between mean arterial pressure recorded during the initial and final flow measurements in any group (p > 0.05).

DISCUSSION

Intraluminal injection of a vasodilator drug into the IMA has been used during IMA operations and provides many advantages. ¹² It produces maximal dilatation of the IMA, which allows the surgeon to evaluate the flow-carrying capacity of the IMA. The maximum flow of an individual IMA may help the surgeon to decide whether or not this vessel should be used and to which artery it should be anastomosed. If the flow from the distal end is poor after pharmacologic dilatation, it may be necessary to discard the terminal portion of the IMA and resort to a free graft or to discard the vessel altogether. Pharmacologic dilatation provides a relaxed dilated distal vessel that allows a precise anastomosis. Also, vasodilatation of the IMA pedicle may unmask small bleeding points at the time of operation and thus improve hemostasis.

We ^3,6 have previously demonstrated that contraction of the human IMA can be readily reversed by a variety of pharmacologic agents including GTN, verapamil, and papaverine. We have also shown that calcium entry blockers are potent in inhibiting the potassium-induced contraction of the human IMA but are less potent when applied to prevent the contraction caused by thromboxane A₂ in the human IMA ³ or by phenylephrine in the canine IMA, ¹³ whereas GTN is a potent vasodilator in these situations against thromboxane A₂ or ₁-adrenoceptor agonists. ^3,6 On the other hand, GTN is effective in reversing but less effective in preventing the contraction (spasm) of the human IMA. ^3,6 Calcium entry blocking drugs act selectively on the voltage-dependent channel whereas GTN acts by releasing nitric oxide in the muscle cell which stimulates guanylate cyclase to raise cyclic guanosine monophosphate. ¹⁴ This in turn leads to calcium removal from the cell. Thus on theoretical grounds the combination of verapamil and GTN with different mechanisms of action should be more effective in preventing or reversing spasm than either agent given alone.

In the canine saphenous vein, we ¹³ showed that the time course of action of commonly used vasodilator agents was different. The time to induce 50% relaxation was 1 minute for GTN, 3 minutes for verapamil and 6 minutes for papaverine. The duration of action of GTN was less than 1 hour, compared with more than 2 hours for verapamil or papaverine. Papaverine relaxes vascular smooth muscle slowly, probably by reducing calcium entry via voltage- or receptor-operated channels and by raising cyclic guanosine monophosphate through inhibition of phosphodiesterase. ^15,16 Also, in an in vitro study of the human saphenous vein, we ¹¹ found that the mixture of verapamil and GTN combined rapid onset and long duration of action in either potassium-contracted or thromboxane A₂–contracted vessels.

The in vitro part of this study showed that the mixture of GTN and verapamil maximally relaxed the IMA with a rapid onset and a long duration of action. This in vitro study provided essential knowledge on which the clinical trial was based.

In the present study, patients receiving bilateral IMA grafts were used for paired comparisons of drugs and solutions, to minimize the influence of other factors such as age, sex, weight, and blood pressure. The side of the IMA used for each treatment was randomized so that any differences between left and right IMAs were excluded. In comparison I, flushing the IMA with Ringer's solution increased IMA flow by 53% (p < 0.025). This increase may be related to mechanical dilatation produced by inserting the plastic cannula into the IMA, by the injection itself, and in part to relaxation with time as was seen in the time course control (13%). GV solution increased IMA flow by 94.8% (p < 0.01), which was nearly twice the increase produced by the vehicle (p < 0.025). This result clearly demonstrated that intraluminal GV solution effectively dilated IMA and increased blood flow.

Although in the present study papaverine is as effective as GV solution in increasing IMA flow, we ¹³ have previously found that the onset of papaverine-induced relaxation had a slower onset than that of verapamil and GTN. Also, papaverine solution is highly acidic (pH 4.4 at to 2.6 log mol/L and pH 4.8 at -4.5 log mol/L as measured in our laboratory). Papaverine hydrochloride is relatively unstable in nonacidic solutions and a white precipitate sometimes forms when papaverine is added to electrolyte solution (Plasma-Lyte; pH approximately 7.4). ¹⁷ Acidic solutions have been shown to damage the endothelium. ¹⁸ A recent study ¹⁹ has suggested that intraluminal papaverine injection does not impair the IMA endothelium, but the endothelium was exposed to papaverine for less than 5 minutes. Also, others have suggested that the use of papaverine initially stimulates release of 6-keto-prostaglandin F₁ from veins, followed by reduction in release, and this may be associated with subsequent graft closure. ²⁰ In addition, we ¹¹ have previously shown that GV solution is useful in preventing spasm of veins during harvesting of saphenous vein grafts. Because most patients need saphenous vein grafts in addition to IMA grafts, it is convenient to have one solution that can be used for both the IMA and the saphenous vein.

Acknowledgments

We acknowledge the statistical advice of Dr S. Farrish, the assistance and cooperation provided by the operating theater nursing staff at the Epworth Hospital, and the assistance of Clara Chan and Lesley Langley.

Footnotes

*Present address: Medical City Dallas Hospital, 7777 Forest Lane, Dallas, TX 75230.

References

1. Sarabu MR, McClung JA, Fass A, Reed GE. Early postoperative spasm in left internal mammary artery bypass grafts. Ann Thorac Surg 1987;44:199-200. [Abstract]
   
2. Houghton JL, Callaghan WE, Frank MJ. Disappearance of high-grade left anterior descending stenosis after revascularization. Cathet Cardiovasc Diagn 1988;14:169-71. [Medline]
   
3. He G-W, Buxton BF, Rosenfeldt FL, Angus JA. Reactivity of the human isolated internal mammary artery to constrictor and dilator agents with implications for the treatment of IMA spasm. Circulation 1989;80(Suppl):I141-50.
   
4. He G-W, Buxton BF, Rosenfeldt FL, Wilson AC, Angus JA. Weak ß-adrenoceptor–mediated relaxation in the human internal mammary artery. J THORAC CARDIOVASC SURG 1989;97:259-66. [Abstract]
   
5. He G-W, Shaw J, Hughes CF, et al. Predominant function of ₁-adrenoceptors in the human internal mammary artery. J Cardiovasc Pharmacol 1993;21:256-63. [Medline]
   
6. He G-W, Shaw J, Yang C-Q, et al. Inhibitory effects of glyceryl trinitrate on -adrenoceptor–mediated contraction in the human internal mammary artery. Br J Clin Pharmacol 1992;34:236-43. [Medline]
   
7. Davis GC, Sobel M, Salzman EW. Elevated plasma fibrinopeptide A and thromboxane B₂ levels during cardiopulmonary bypass. Circulation 1980;61:808-13. [Abstract/Free Full Text]
   
8. Ylikorkala O, Saarela E, Viinikka L. Increased prostacyclin and thromboxane production in man during cardiopulmonary bypass. J THORAC CARDIOVASC SURG 1981;82:245-7. [Abstract]
   
9. Addonizio VP, Smith JB, Strauss JF, Cohnan RW, Edmunds LH. Thromboxane synthesis and platelet secretion during cardiopulmonary bypass with bubble oxygenator. J THORAC CARDIOVASC SURG 1980;79:91-6. [Medline]
   
10. Teoh KH, Fremes SE, Weisel RD, et al. Cardiac release of prostacyclin and thromboxane A₂ during coronary revascularization. J THORAC CARDIOVASC SURG 1987;93:120-6. [Abstract]
    
11. He G-W, Rosenfeldt FL, Angus JA. Pharmacological relaxation of the human saphenous vein during harvesting for coronary bypass grafting. Ann Thorac Surg 1993;55:1210-7. [Abstract]
    
12. Green GE. Internal mammary–coronary artery anastomosis for myocardial ischemia. In: Gibbon's surgery of the chest. 4th ed. volume I. Philadelphia: WB Saunders, 1983;1451-7.
    
13. He G-W, Angus JA, Rosenfeldt FL. Reactivity of the canine isolated internal mammary artery, saphenous vein and coronary artery to constrictor and dilator substances: relevance to coronary bypass graft surgery. J Cardiovasc Pharmacol 1988;12:12-25. [Medline]
    
14. Rapoport RM, Murad F. Endothelium-dependent and nitrovasodilator-induced relaxation of vascular smooth muscle: role of cyclic GMP. J Cyclic Nucl Prot Phosphoryl Res 1983;9:281-96.
    
15. Martin W, Furchgott RF, Villani GM, Jothianandan D. Phosphodiesterase inhibitors induce endothelium-dependent relaxation of rat and rabbit aorta by potentiating the effects of spontaneously released endothelium-derived relaxing factor. J Pharmacol Exp Ther 1986;237:539-47. [Abstract/Free Full Text]
    
16. Huddart H, Saad KHM. Papaverine-induced inhibition of electrical and mechanical activity and calcium movements of rat ileal smooth muscle. J Exp Biol 1980;86:99-114. [Abstract/Free Full Text]
    
17. Cunningham JN. Papaverine hydrochloride preservation of vein grafts. J THORAC CARDIOVASC SURG 1982;84:933-4.
    
18. Constantinides P, Rohmson M. Ultrastructural injury of arterial endothelium: 1. Effects of pH, osmolarity, anoxia and temperature. Arch Pathol 1969;88:99-105. [Medline]
    
19. Hillier C, Watt PAC, Spyt TJ, Thurston H. Contraction and relaxation of human internal mammary artery after intraluminal administration of papaverine. Ann Thorac Surg 1992;53:1033-7. [Abstract]
    
20. Roberts AJ, Hay DA, Mehta JL, et al. Biochemical and ultrastructural integrity of the saphenous vein conduit during coronary artery bypass grafting: preliminary results of effect of papaverine. J THORAC CARDIOVASC SURG 1984;88:39-48.[Abstract]
    

This article has been cited by other articles:

				G.-W. He, L. Fan, A. Furnary, and Q. Yang A new antispastic solution for arterial grafting: Nicardipine and nitroglycerin cocktail in preparation of internal thoracic and radial arteries for coronary surgery J. Thorac. Cardiovasc. Surg., September 1, 2008; 136(3): 673 - 680. [Abstract] [Full Text] [PDF]

				M. Bahcivan, F. Kolbakir, H. Karamustafa, and H T. Keceligil Endothoracic Papaverine Application for Internal Thoracic Artery Harvest Asian Cardiovasc Thorac Ann, June 1, 2007; 15(3): 234 - 237. [Abstract] [Full Text] [PDF]

				S. Na, Y. J. Oh, Y. H. Shim, Y. W. Hong, S. O. Bang, and Y. L. Kwak Effects of milrinone on blood flow of the Y-graft composed with the radial and the internal thoracic artery in patients with coronary artery disease. Eur. J. Cardiothorac. Surg., August 1, 2006; 30(2): 324 - 328. [Abstract] [Full Text] [PDF]

				F. Formica, O. Ferro, M. Brustia, F. Corti, L. Colagrande, E. Bosisio, and G. Paolini Effects of Papaverine and Glycerilnitrate-Verapamil Solution as Topical and Intraluminal Vasodilators for Internal Thoracic Artery Ann. Thorac. Surg., January 1, 2006; 81(1): 120 - 124. [Abstract] [Full Text] [PDF]

				K. Takeuchi, S. Sakamoto, Y. Nagayoshi, H. Nishizawa, and J. Matsubara Reactivity of the human internal thoracic artery to vasodilators in coronary artery bypass grafting Eur. J. Cardiothorac. Surg., November 1, 2004; 26(5): 956 - 959. [Abstract] [Full Text] [PDF]

				D. S. Girard, J. P. Sutton III, T. H. Williams, A. J. Crumbley III, J. L. Zellner, J. M. Kratz, and F. A. Crawford Papaverine Delivery to the Internal Mammary Artery Pedicle Effectively Treats Spasm Ann. Thorac. Surg., October 1, 2004; 78(4): 1295 - 1298. [Abstract] [Full Text] [PDF]

				S. Mussa, T. J. Guzik, E. Black, M. A. Dipp, K. M. Channon, and D. P. Taggart Comparative efficacies and durations of action of phenoxybenzamine, verapamil/nitroglycerin solution, and papaverine as topical antispasmodics for radial artery coronary bypass grafting J. Thorac. Cardiovasc. Surg., December 1, 2003; 126(6): 1798 - 1805. [Abstract] [Full Text] [PDF]

				M. J. Flynn, D. Winters, P. Breen, G. O'Sullivan, G. Shorten, D. O'Connell, A. O'Donnell, and T. Aherne Dopexamine increases internal mammary artery blood flow following coronary artery bypass grafting Eur. J. Cardiothorac. Surg., October 1, 2003; 24(4): 547 - 551. [Abstract] [Full Text] [PDF]

				M.-H. Liu, H. Jin, H. S. Floten, Z. Ren, A. P.C. Yim, and G.-W. He Vascular endothelial growth factor-mediated, endothelium-dependent relaxation in human internal mammary artery Ann. Thorac. Surg., March 1, 2002; 73(3): 819 - 824. [Abstract] [Full Text] [PDF]

				G.-W. He Invited commentary Ann. Thorac. Surg., September 1, 2001; 72(3): 816 - 816. [Full Text] [PDF]

				J. Zhang, J. E. Lipa, C. E. Black, N. Huang, P. C. Neligan, F. T. K. Ling, R. H. Levine, J. L. Semple, and C. Y. Pang Pharmacological characterization of vasomotor activity of human musculocutaneous perforator artery and vein J Appl Physiol, December 1, 2000; 89(6): 2268 - 2275. [Abstract] [Full Text] [PDF]

				W. F. Turner Jr ""Off-pump"" coronary artery bypass grafting: the first one hundred cases of the Rose City experience Ann. Thorac. Surg., October 1, 1999; 68(4): 1482 - 1485. [Abstract] [Full Text] [PDF]

				F. L. Rosenfeldt, G.-W. He, B. F. Buxton, and J. A. Angus Pharmacology of coronary artery bypass grafts Ann. Thorac. Surg., March 1, 1999; 67(3): 878 - 888. [Abstract] [Full Text] [PDF]

				J.-L. Cracowski, O. Chavanon, M. Durand, E. Borrel, P. Devillier, J.-M. Mallion, and D. Blin Effect of Low-Dose Positive Inotropic Drugs on Human Internal Mammary Artery Flow Ann. Thorac. Surg., December 1, 1997; 64(6): 1742 - 1746. [Abstract] [Full Text]

				D. Arnaudov, A. J. Cohen, D. Zabeeda, E. Hauptman, L. Sasson, A. Schachner, and S. Ezra Effect of Systemic Vasodilators on Internal Mammary Flow During Coronary Bypass Grafting Ann. Thorac. Surg., December 1, 1996; 62(6): 1816 - 1819. [Abstract] [Full Text]

				G.-W. He and C.-Q. Yang Use of Verapamil and Nitroglycerin Solution in Preparation of Radial Artery for Coronary Grafting Ann. Thorac. Surg., February 1, 1996; 61(2): 610 - 614. [Abstract] [Full Text]

				G.-W. He, T. E. Acuff, C.-Q. Yang, W. H. Ryan, and M. J. Mack Middle and proximal sections of the human internal mammary artery are not "passive conduits" J. Thorac. Cardiovasc. Surg., October 1, 1994; 108(4): 741 - 746. [Abstract] [Full Text]

This Article Abstract 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): Guo-Wei He Franklin L. Rosenfeldt James Tatoulis Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by He, G.-W. Articles by Tatoulis, J. Search for Related Content PubMed PubMed Citation Articles by He, G.-W. Articles by Tatoulis, J.