- Email: [email protected]
Avulsion of a tricuspid valve leaflet during traction on an infected, entrapped endocardial pacemaker electrode
Avulsion of a tricuspid valve leaflet during traction on an infected, entrapped endocardial pacemaker electrode The role of electrode design Endocardial pacemaker electrodes rely in part upon endocardial fixation proximal to the electrode tip to prevent ejection of the tip from the ventricular apex. Fixation of these electrodes to the superior vena cava and tricuspid valve, in particular, has been reported. Infection of endocardial electrodes necessitates their removal. This report concerns avulsion of a tricuspid valve leaflet during traction on an infected electrode. With the availability of new methods of apical fixation, the utilization of electrode sheathing materials which discourage endocardial fixation would increase the safety of their removal under the circumstances reported herein.
Myles Edwin Lee, M.D., Aurelio Chaux, M.D., and Jack M. Matloff, M.D., Los Angeles, Calif.
Entrapment of endocardial pacemaker electrodes is an infrequent sequel of permanent transvenous endocardial cardiac pacing. Infection involving the electrode may necessitate removal. Early reports of removal of entrapped, infected electrodes by traction have suggested that this may be a safe method of retrieval. The present report of avulsion of a tricuspid valve leaflet during traction on an infected, entrapped endocardial pacemaker electrode supplements the opinion expressed in other reports in the literature that this approach is not without potential serious complications. Case report A 75-year-old man was admitted to the Medical Service because of chills, fever, and a nonproductive cough. Eight years prior to admission, a permanent, transvenous endocardial pacemaker was inserted to control syncopal episodes caused by complete heart block. Eight months later, the pulse
From the Department of Thoracic, Cardiac, and Vascular Surgery, Cedars-Sinai Medical Center, Los Angeles, Calif. Received for publication March 15, 1977. Accepted for publication April 25, 1977. Address for reprints: Myles Edwin Lee, M.D., Department of Thoracic, Cardiac, and VascularSurgery, Cedars-Sinai Medical Center, Los Angeles, Calif. 90054.
generator was removed because of extrusion through the skin. The pacing electrode could not be removed and was left in place. A new pulse generator and endocardial electrode were inserted on the opposite side with the use of the external jugular vein. During the ensuing 7 years, the patient had had an intermittently draining sinus tract in the right supraclavicular fossa. The past history was remarkable for pulmonary tuberculosis with pleural effusion, for which he had been treated with isonicotinic acid hydrazide and para-aminosalicylic acid in 1954 and 1955. On admission, the temperature was 104° F., the blood pressure 110/60 mm. Hg, and the pulse rate 72 beats per minute. Physical examination was unremarkable, except for a cough and the draining sinus tract in the right supraclavicular fossa. Blood cultures were positive for Staphylococcus aureus, coagulase positive. The same organism was cultured from the sinus tract. Antibiotic therapy with gentamicin sulfate, 80 mg. intramuscularlyevery 12 hours, and cefazolin sodium, I Gm. intravenously every 6 hours, was begun. On the fifth hospital day, dicloxacillin sodium monohydrate, 0.5 Gm. every 6 hours, was started orally. By the ninth hospital day, the patient was afebrile. On the tenth hospital day, 5 days after the patient started taking dicloxacillin, bloody stools developed. Sigmoidoscopy was performed which revealed blood coming from above 20 cm. The hematocrit value dropped from 40 to 36 percent. The bleeding was initially felt to have resulted from an antibiotic-induced colitis, and it subsided with cessation of dicloxacillin. Results of a subsequent barium enema were negative except for a questionable 3 mm. polyp seen at the
433
434
The Journal of Thoracic and Cardiovascular Surgery
Lee, Chaux, Mat/off
tient was seen at follow-up 2 months later. There had been no recurrence of chills or fever and no evidence of tricuspid insufficiency on physical examination.
Discussion
12 Fig. 1. Tricuspid valve leaflet incorporated in tissue reaction around the pacemaker electrode. junction of the sigmoid and descending colon. An upper gastrointestinal series revealed coarsened rugae in the prepyloric area and deformity of the duodenal bulb. The patient was continued on a regimen of cefazolin until the fifteenth hospital day, when this was discontinued and he was switched to cephalexin, having been afebrile for 6 days. The temperature again rose to 102° F. on the nineteenth hospital day and consultation with the Cardiac Surgical Service was requested. Physical examination revealed a 2.5 by 2.0 cm. ulcer in the right supraclavicular fossa. The lungs were clear. A Grade 3/6 systolic ejection murmur was heard at the left sternal border with no radiation and no changes with respiration. Our impression was recurrent bacteremia, most likely caused by endocarditis, involving one or both endocardial pacemaker electrodes. A blood culture at this time grew Escherichia coli and Staphylococcus epidermidis. The pacemaker electrode had been on traction for 3 weeks, the pressure having been maintained with a hemostat. Initial consideration was therefore given to extracting the extruded electrode by cardiotomy during cardiopulmonarybypass. The open technique was not used because of the patient's persistent cough and episode of rectal bleeding. Accordingly, 3 pounds of traction was applied to the electrode with orthopedic pulleys. The patient tolerated this without arrhythmias and without hemodynamic changes. Eleven hours later, the electrode suddenly became free. Attached to the electrode tip was a piece of valve tissue with chordae tendineae (Fig. I). Because of a recrudescence of fever, antibiotics were resumed. Subsequently, the patient underwent removal of the functioning pacemaker and the remaining endocardial electrode. It was replaced with a Cordis OmniStanicor pacemaker, with Cordis epicardial screw-in leads. He was taken off antibiotics 4 days postoperatively and remained afebrile until discharged from the hospital. The pa-
Entrapment of endocardial pacemaker electrodes is an infrequent complication of permanent transvenous endocardial pacing. Earlier descriptions of the traction method of removal suggested this technique to be a safe one. Imparato and Kim' in 1972 reported a series of 130 endocardial pacemakers in which II (9 percent) cases presented the problem of electrode entrapment. Six of these cases were complicated by septicemia. Of these six cases, three were managed by graded skin traction. The authors felt that graded skin traction resulted in ischemia of the fibrous bands trapping the electrode, resulting in necrosis and subsequent rupture. No complications were noted. Two earlier reports, one by Bilgutay and associates- in 1969 and the other by Wallace, Sherafat, and Blakemore" in 1970, described methods of continuous traction using orthopedic pulleys and weights with no complications. The potential risks of the traction method were suggested by the study of Robboy and associates," who in 1969 presented an autopsy series of seven patients who had undergone transvenous endocardial pacing. These authors described an intensive endocardial reaction to the electrode including a fibrin sheath, multiple points of attachment of the sheath to the endocardium, fibrosis about the electrode tip and, in four cases, adherence of the electrode to the chordae tendineae of the tricuspid valve. Further evidence of adherence of endocardial pacemaker electrodes to the tricuspid valve was presented in the papers of Friedberg and D'Cunha" and Huang and Baba." Our case illustrates the potential risks of the traction method of electrode retrieval when the electrode is attached to the tricuspid valve. Risks of the traction method were further illustrated in Furman's? case report of an instance of continuous traction applied to an infected, entrapped electrode which; upon removal, was found to have a 2 by 3 em. piece of right ventricular myocardium attached to the electrode tip. Conventional silicone-coated pacemaker electrodes rely in part upon fixation by a fibrous reaction to endocardial surfaces to prevent dislodgment. Smyth and associates," in a recent publication addressing the problem of dislodgment of atrial pacing electrodes, described a J-shaped electrode with three rows of silicone rubber tines at the tip. In II patients in whom atrial pacing was established, there was no case of lead displacement in an 18 month follow-up period. Numerous other methods of endocardial electrode fixation, includ-
Volume 74 Number 3 September, 1977
ing self-affixing hooks, barbs, and screws, have been described." If such methods prove effective, it should be possible to utilize an electrode sheathing material which would discourage endocardial fixation proximal to the electrode tip. This point was emphasized by Furman and Escher'? in 1968. The work of Sawyer and associates, II, 12 extending over two decades, has demonstrated that a nonconduction prosthetic surface should possess a uniformly negative surface charge density free of contaminants if it is to remain thrombus-resistant during long-term implantation in the cardiovascular system. These considerations have been echoed by Wilkes-" in his recent discussion of considerations for selecting polymers (polyurethanes) for clinical implantation. In his discussion of polyurethane chemistry, he cites the role played by trace catalysts or other surface contaminants in triggering thrombus formation, protein deposition, giant cell growth, or other tissue responses. Erythrocytes, leukocytes, and platelets will deposit on an electrode at a potential of + 250 to +400 mv. with respect to the normal hydrogen electrode (NHE) at physiological pH'S.14 Bioelectric polyurethane has been shown to have a negative resting potential in blood of - 140 to - 160 mv. NHE.15 This finding correlates with Furman's observation of a polyurethane-coated electrode which demonstrated no thrombogenesis nor tissue fixation after 15 months of implantation in a patient. 10 Further research and ultimate clinical application of a satisfactory biomedical grade of polyurethane or other prosthetic material in conjunction with a satisfactory method of apical fixation as described by Smyth could result in an endocardial pacing electrode with the implantation characteristics envisioned by Furman in 1968.
Summary Entrapment of endocardial pacemaker electrodes is an infrequent occurrence. Removal by traction may be dangerous because of fixation of the electrode to the endocardium and to the tricuspid valve apparatus. Our patient survived avulsion of a tricuspid valve leaflet during traction on an infected, entrapped endocardial pacemaker electrode. A review of pacemaker electrode tip design and polymer chemistry suggests that it should be possible to produce an electrode which will not require tissue entrapment along its entire length to ensure long-term fixation in the ventricular apex. REFERENCES Imparato, A. M., and Kim, G. E.: Electrode Complications in Patients With Permanent Cardiac Pacemakers, Arch. Surg. 105: 705, 1972.
Entrapment of endocardial pacemaker electrodes
43 5
2 Bilgutay, A. M., Jensen, N. K., Schmidt, W. R.,
Garamella, J. J., and Lynch, M. F.: Incarceration of Transvenous Pacemaker Electrode. Removal by Traction, Am. Heart J. 77: 377, 1969. 3 Wallace, H. W., Sherafat, M., and Blakemore, W. S.: The Stubborn Pacemaker Catheter, Surgery 68: 914, 1970. 4 Robboy, S. J., Harthorne, J. W., Leinbach, R. C., Sanders, C. A., and Austen, W. G.: Autopsy Findings With Permanent Pervenous Pacemakers, Circulation 39: 495, 1969. 5 Friedberg, H. D., and D'Cunha, G. F.: Adhesions of Pacing Catheter to Tricuspid Valve: Adhesive Endocarditis, Thorax 24: 498,1969. 6 Huang, T. Y., and Baba, N.: Cardiac Pathology of Transvenous Pacemakers, Am. Heart 1. 83: 469, 1972. 7 Furman, S.: Removal of a Pacemaker Electrode Containing a Myocardial Fragment, Ann. Thorac. Surg. 19: 716, 1975. 8 Smyth, N. P. D., Citron, P., Keshishian, J. M., Garcia, J. M., and Kelly, L. c.: Permanent Pervenous Atrial Sensing and Pacing With a New J-Shaped Lead, J. THORAC. CARDIOVASC. SURG. 72: 565, 1976. 9 Parsonnet V., and Manhardt, M.: Permanent Pacing of the Heart: 1952 to 1976, Am. 1. Cardiol. 39: 250, 1977. IO Furman, S., and Escher, D. J. W.: Retained Endocardial Pacemaker Electrodes, J. THORAc. CARDIOVASC. SURG. 55: 737, 1968. I I Sawyer, P. N., and Srinivasan, S.: Essential Criteria for the Selection of Metallic and Non-Metallic Vascular Prostheses, Adv. Biomed. Eng. Med. Phys. 3: 273, 1970. 12 Sawyer, P. N., Stanczewski, B., Ramsey, W. S., Jr., Ramasamy, N., and Srinivasan, S.: Electrochemical Interactions at the Endothelial Surface, J. Supramol. Struct. 1: 417, 1973. 13 Wilkes, G. L.: Necessary Considerations for Selecting a Polymeric Material for Implantation With Emphasis on Polyurethanes, in Kronenthal, R. L., et aI., editors: Polymers in Medicine and Surgery, New York, 1975, Plenum Press, pp. 45-75. 14 Boddy, P. J., Brattain, W. H., and Sawyer, P. N.: Some Electrochemical Properties of Solid-Liquid Interfaces and the Electrode Behavior of Erythrocytes, in Sawyer, P. N., editor: Biophysical Mechanisms in Vascular Homeostasis and Intravascular Thrombosis, New York, 1965, Appleton-Century-Crofts, p. 30. 15 Sawyer, P. N., Stanczewski, B., Srinivasan, S., Stempack, J. c., and Kammlott, G. W.: Implantation Characteristics of Metal-Backed Polymer-Coated Heart Valves: Biophysical Scanning and Transmission Electron Microscopic Studies, Trans. Am. Soc. Artif. Intern. Organs 20: 692, 1974.
Myles Edwin Lee, M.D., Aurelio Chaux, M.D., and Jack M. Matloff, M.D., Los Angeles, Calif.
Entrapment of endocardial pacemaker electrodes is an infrequent sequel of permanent transvenous endocardial cardiac pacing. Infection involving the electrode may necessitate removal. Early reports of removal of entrapped, infected electrodes by traction have suggested that this may be a safe method of retrieval. The present report of avulsion of a tricuspid valve leaflet during traction on an infected, entrapped endocardial pacemaker electrode supplements the opinion expressed in other reports in the literature that this approach is not without potential serious complications. Case report A 75-year-old man was admitted to the Medical Service because of chills, fever, and a nonproductive cough. Eight years prior to admission, a permanent, transvenous endocardial pacemaker was inserted to control syncopal episodes caused by complete heart block. Eight months later, the pulse
From the Department of Thoracic, Cardiac, and Vascular Surgery, Cedars-Sinai Medical Center, Los Angeles, Calif. Received for publication March 15, 1977. Accepted for publication April 25, 1977. Address for reprints: Myles Edwin Lee, M.D., Department of Thoracic, Cardiac, and VascularSurgery, Cedars-Sinai Medical Center, Los Angeles, Calif. 90054.
generator was removed because of extrusion through the skin. The pacing electrode could not be removed and was left in place. A new pulse generator and endocardial electrode were inserted on the opposite side with the use of the external jugular vein. During the ensuing 7 years, the patient had had an intermittently draining sinus tract in the right supraclavicular fossa. The past history was remarkable for pulmonary tuberculosis with pleural effusion, for which he had been treated with isonicotinic acid hydrazide and para-aminosalicylic acid in 1954 and 1955. On admission, the temperature was 104° F., the blood pressure 110/60 mm. Hg, and the pulse rate 72 beats per minute. Physical examination was unremarkable, except for a cough and the draining sinus tract in the right supraclavicular fossa. Blood cultures were positive for Staphylococcus aureus, coagulase positive. The same organism was cultured from the sinus tract. Antibiotic therapy with gentamicin sulfate, 80 mg. intramuscularlyevery 12 hours, and cefazolin sodium, I Gm. intravenously every 6 hours, was begun. On the fifth hospital day, dicloxacillin sodium monohydrate, 0.5 Gm. every 6 hours, was started orally. By the ninth hospital day, the patient was afebrile. On the tenth hospital day, 5 days after the patient started taking dicloxacillin, bloody stools developed. Sigmoidoscopy was performed which revealed blood coming from above 20 cm. The hematocrit value dropped from 40 to 36 percent. The bleeding was initially felt to have resulted from an antibiotic-induced colitis, and it subsided with cessation of dicloxacillin. Results of a subsequent barium enema were negative except for a questionable 3 mm. polyp seen at the
433
434
The Journal of Thoracic and Cardiovascular Surgery
Lee, Chaux, Mat/off
tient was seen at follow-up 2 months later. There had been no recurrence of chills or fever and no evidence of tricuspid insufficiency on physical examination.
Discussion
12 Fig. 1. Tricuspid valve leaflet incorporated in tissue reaction around the pacemaker electrode. junction of the sigmoid and descending colon. An upper gastrointestinal series revealed coarsened rugae in the prepyloric area and deformity of the duodenal bulb. The patient was continued on a regimen of cefazolin until the fifteenth hospital day, when this was discontinued and he was switched to cephalexin, having been afebrile for 6 days. The temperature again rose to 102° F. on the nineteenth hospital day and consultation with the Cardiac Surgical Service was requested. Physical examination revealed a 2.5 by 2.0 cm. ulcer in the right supraclavicular fossa. The lungs were clear. A Grade 3/6 systolic ejection murmur was heard at the left sternal border with no radiation and no changes with respiration. Our impression was recurrent bacteremia, most likely caused by endocarditis, involving one or both endocardial pacemaker electrodes. A blood culture at this time grew Escherichia coli and Staphylococcus epidermidis. The pacemaker electrode had been on traction for 3 weeks, the pressure having been maintained with a hemostat. Initial consideration was therefore given to extracting the extruded electrode by cardiotomy during cardiopulmonarybypass. The open technique was not used because of the patient's persistent cough and episode of rectal bleeding. Accordingly, 3 pounds of traction was applied to the electrode with orthopedic pulleys. The patient tolerated this without arrhythmias and without hemodynamic changes. Eleven hours later, the electrode suddenly became free. Attached to the electrode tip was a piece of valve tissue with chordae tendineae (Fig. I). Because of a recrudescence of fever, antibiotics were resumed. Subsequently, the patient underwent removal of the functioning pacemaker and the remaining endocardial electrode. It was replaced with a Cordis OmniStanicor pacemaker, with Cordis epicardial screw-in leads. He was taken off antibiotics 4 days postoperatively and remained afebrile until discharged from the hospital. The pa-
Entrapment of endocardial pacemaker electrodes is an infrequent complication of permanent transvenous endocardial pacing. Earlier descriptions of the traction method of removal suggested this technique to be a safe one. Imparato and Kim' in 1972 reported a series of 130 endocardial pacemakers in which II (9 percent) cases presented the problem of electrode entrapment. Six of these cases were complicated by septicemia. Of these six cases, three were managed by graded skin traction. The authors felt that graded skin traction resulted in ischemia of the fibrous bands trapping the electrode, resulting in necrosis and subsequent rupture. No complications were noted. Two earlier reports, one by Bilgutay and associates- in 1969 and the other by Wallace, Sherafat, and Blakemore" in 1970, described methods of continuous traction using orthopedic pulleys and weights with no complications. The potential risks of the traction method were suggested by the study of Robboy and associates," who in 1969 presented an autopsy series of seven patients who had undergone transvenous endocardial pacing. These authors described an intensive endocardial reaction to the electrode including a fibrin sheath, multiple points of attachment of the sheath to the endocardium, fibrosis about the electrode tip and, in four cases, adherence of the electrode to the chordae tendineae of the tricuspid valve. Further evidence of adherence of endocardial pacemaker electrodes to the tricuspid valve was presented in the papers of Friedberg and D'Cunha" and Huang and Baba." Our case illustrates the potential risks of the traction method of electrode retrieval when the electrode is attached to the tricuspid valve. Risks of the traction method were further illustrated in Furman's? case report of an instance of continuous traction applied to an infected, entrapped electrode which; upon removal, was found to have a 2 by 3 em. piece of right ventricular myocardium attached to the electrode tip. Conventional silicone-coated pacemaker electrodes rely in part upon fixation by a fibrous reaction to endocardial surfaces to prevent dislodgment. Smyth and associates," in a recent publication addressing the problem of dislodgment of atrial pacing electrodes, described a J-shaped electrode with three rows of silicone rubber tines at the tip. In II patients in whom atrial pacing was established, there was no case of lead displacement in an 18 month follow-up period. Numerous other methods of endocardial electrode fixation, includ-
Volume 74 Number 3 September, 1977
ing self-affixing hooks, barbs, and screws, have been described." If such methods prove effective, it should be possible to utilize an electrode sheathing material which would discourage endocardial fixation proximal to the electrode tip. This point was emphasized by Furman and Escher'? in 1968. The work of Sawyer and associates, II, 12 extending over two decades, has demonstrated that a nonconduction prosthetic surface should possess a uniformly negative surface charge density free of contaminants if it is to remain thrombus-resistant during long-term implantation in the cardiovascular system. These considerations have been echoed by Wilkes-" in his recent discussion of considerations for selecting polymers (polyurethanes) for clinical implantation. In his discussion of polyurethane chemistry, he cites the role played by trace catalysts or other surface contaminants in triggering thrombus formation, protein deposition, giant cell growth, or other tissue responses. Erythrocytes, leukocytes, and platelets will deposit on an electrode at a potential of + 250 to +400 mv. with respect to the normal hydrogen electrode (NHE) at physiological pH'S.14 Bioelectric polyurethane has been shown to have a negative resting potential in blood of - 140 to - 160 mv. NHE.15 This finding correlates with Furman's observation of a polyurethane-coated electrode which demonstrated no thrombogenesis nor tissue fixation after 15 months of implantation in a patient. 10 Further research and ultimate clinical application of a satisfactory biomedical grade of polyurethane or other prosthetic material in conjunction with a satisfactory method of apical fixation as described by Smyth could result in an endocardial pacing electrode with the implantation characteristics envisioned by Furman in 1968.
Summary Entrapment of endocardial pacemaker electrodes is an infrequent occurrence. Removal by traction may be dangerous because of fixation of the electrode to the endocardium and to the tricuspid valve apparatus. Our patient survived avulsion of a tricuspid valve leaflet during traction on an infected, entrapped endocardial pacemaker electrode. A review of pacemaker electrode tip design and polymer chemistry suggests that it should be possible to produce an electrode which will not require tissue entrapment along its entire length to ensure long-term fixation in the ventricular apex. REFERENCES Imparato, A. M., and Kim, G. E.: Electrode Complications in Patients With Permanent Cardiac Pacemakers, Arch. Surg. 105: 705, 1972.
Entrapment of endocardial pacemaker electrodes
43 5
2 Bilgutay, A. M., Jensen, N. K., Schmidt, W. R.,
Garamella, J. J., and Lynch, M. F.: Incarceration of Transvenous Pacemaker Electrode. Removal by Traction, Am. Heart J. 77: 377, 1969. 3 Wallace, H. W., Sherafat, M., and Blakemore, W. S.: The Stubborn Pacemaker Catheter, Surgery 68: 914, 1970. 4 Robboy, S. J., Harthorne, J. W., Leinbach, R. C., Sanders, C. A., and Austen, W. G.: Autopsy Findings With Permanent Pervenous Pacemakers, Circulation 39: 495, 1969. 5 Friedberg, H. D., and D'Cunha, G. F.: Adhesions of Pacing Catheter to Tricuspid Valve: Adhesive Endocarditis, Thorax 24: 498,1969. 6 Huang, T. Y., and Baba, N.: Cardiac Pathology of Transvenous Pacemakers, Am. Heart 1. 83: 469, 1972. 7 Furman, S.: Removal of a Pacemaker Electrode Containing a Myocardial Fragment, Ann. Thorac. Surg. 19: 716, 1975. 8 Smyth, N. P. D., Citron, P., Keshishian, J. M., Garcia, J. M., and Kelly, L. c.: Permanent Pervenous Atrial Sensing and Pacing With a New J-Shaped Lead, J. THORAC. CARDIOVASC. SURG. 72: 565, 1976. 9 Parsonnet V., and Manhardt, M.: Permanent Pacing of the Heart: 1952 to 1976, Am. 1. Cardiol. 39: 250, 1977. IO Furman, S., and Escher, D. J. W.: Retained Endocardial Pacemaker Electrodes, J. THORAc. CARDIOVASC. SURG. 55: 737, 1968. I I Sawyer, P. N., and Srinivasan, S.: Essential Criteria for the Selection of Metallic and Non-Metallic Vascular Prostheses, Adv. Biomed. Eng. Med. Phys. 3: 273, 1970. 12 Sawyer, P. N., Stanczewski, B., Ramsey, W. S., Jr., Ramasamy, N., and Srinivasan, S.: Electrochemical Interactions at the Endothelial Surface, J. Supramol. Struct. 1: 417, 1973. 13 Wilkes, G. L.: Necessary Considerations for Selecting a Polymeric Material for Implantation With Emphasis on Polyurethanes, in Kronenthal, R. L., et aI., editors: Polymers in Medicine and Surgery, New York, 1975, Plenum Press, pp. 45-75. 14 Boddy, P. J., Brattain, W. H., and Sawyer, P. N.: Some Electrochemical Properties of Solid-Liquid Interfaces and the Electrode Behavior of Erythrocytes, in Sawyer, P. N., editor: Biophysical Mechanisms in Vascular Homeostasis and Intravascular Thrombosis, New York, 1965, Appleton-Century-Crofts, p. 30. 15 Sawyer, P. N., Stanczewski, B., Srinivasan, S., Stempack, J. c., and Kammlott, G. W.: Implantation Characteristics of Metal-Backed Polymer-Coated Heart Valves: Biophysical Scanning and Transmission Electron Microscopic Studies, Trans. Am. Soc. Artif. Intern. Organs 20: 692, 1974.