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): Robert J. Schreiner Mark D. Iannettoni Robert H. Bartlett Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lynch, W. R. Articles by Bartlett, R. H. Search for Related Content PubMed PubMed Citation Articles by Lynch, W. R. Articles by Bartlett, R. H. Related Collections Related Article

Ann Thorac Surg 2000;69:351-356
© 2000 The Society of Thoracic Surgeons

---

Original Articles

Hemodynamic effect of a low-resistance artificial lung in series with the native lungs of sheep

^a Department of Surgery, University of Michigan Medical Center, Ann Arbor, Michigan, USA

Address reprint requests to Dr Bartlett, Section of General Surgery, University of Michigan, Taubman 2920, Box 0331, 1500 E Medical Center Dr, Ann Arbor, MI 48109
e-mail: robbar{at}umich.edu

Presented at the Thirty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 25–27, 1999.

	Abstract

Top Abstract Introduction Material and methods Results Comment References

 
Background. An artificial lung with 1 to 6 month work life could act as a bridge to transplantation. A pumpless artificial lung has been developed.

Methods. The artificial lung was placed in series with the native lungs of adult sheep. Hemodynamics were observed, as the right ventricle generated flow through the device. Through a left thoracotomy, two 20-mm grafts were anastomosed in an end-to-side fashion to the pulmonary artery. The grafts were externalized, and directed flow through the chest wall, to the extracorporeal lung. The animals were recovered, weaned from the ventilator, and when standing, flow was diverted through the device.

Results. Five of 7 animals survived 24 hours with 75% to 100% of the cardiac output diverted through the device. All animals were active, with interest in food and water, and able to stand.

Conclusions. The right ventricle perfused the artificial lung with 75% to 100% of the cardiac output for 24 hours. This device demonstrates the feasibility of a pumpless pulmonary assist device relying on the right ventricle for perfusion.

	Introduction

Top Abstract Introduction Material and methods Results Comment References

 
Thousands of patients die annually from acute and chronic respiratory failure. When unsuccessfully treated, the respiratory failure becomes irreversible, and the only viable treatment is lung transplantation. Extracorporeal life support (ECLS) evolved as a treatment option in respiratory failure, providing support while offering time for diseased lungs to recover. ECLS has been accepted as treatment for severe respiratory failure in newborns with survival rates approaching 80%. In children and adults, however, ECLS survival is closer to 50% [1, 2]. Although not widely accepted for these patients, ECLS can be effective therapy for up to 1 month. When ECLS is used beyond this point, it becomes too complicated and expensive, making it an impractical treatment for prolonged support. For children and adults, those who do not recover during this 30 day window, progress to end-stage pulmonary fibrosis without return of native lung function. A possible treatment would be lung transplantation, however, these patients are unattractive candidates because of the infections and bleeding complications that so often are a part of prolonged ECLS. With no other treatment options, these patients die.

Lung transplantation has played a successful role in the treatment of chronic respiratory failure such as emphysema, chronic obstructive pulmonary disease, cystic fibrosis, idiopathic pulmonary fibrosis, ₁-antitrypsin deficiency, and primary pulmonary hypertension. Transplantation is successful for patients who are not acutely ill, and has been reserved for those with an estimated life expectancy of 18 months or less. Donor availability is a major limitation with the average wait for a lung, approaching 2 years. For those who do receive an organ, 1-year survival is about 70% [3]. Some transplant candidates progress to end-stage respiratory failure while waiting for a suitable donor organ. Others acutely decompensate, requiring mechanical ventilation. Infectious complications become prohibitive for these patients, and transplantation becomes less likely. ECLS has been considered as a bridge-to-transplantation for this patient group but, again, complications limit therapy long before a suitable organ becomes available.

A pulmonary assist device, capable of supporting patients for up to 6 months, could sustain both acute and chronic end-stage respiratory failure patients. Prolonged support could offer time for resolution of acute failure as well as provide a window of opportunity for lung transplantation. The concept for such a device is not new, and preliminary reports appeared in the literature in the 1970s [4–7]. Various design strategies began to evolve, as membrane technology improved, making gas exchange more efficient [8–10]. An oxygenator that did not rely on a mechanical pump was an attractive strategy, because the concept was simple. Investigators considered using the right heart or an arteriovenous shunt to generate flow to a pumpless oxygenator [11, 12]. Others tried intravascular oxygenators (IVOX), exposing gas exchange membranes to the cardiac output within the vena cava [13–15]. Some investigators included pumps, incorporating a circulatory assist device with an oxygenator as a form of cardiopulmonary support [16, 17]. Each of these strategies has demonstrated the potential for partial support of respiratory function and investigation continues.

Our strategy is development of a pumpless pulmonary assist device, perfused by the right ventricle. Design goals include total respiratory support, 6-month durability, and the potential to be implanted. Fundamental to the success of this strategy is the ability of the right ventricle to tolerate perfusing an oxygenator for prolonged periods of time and under a variety of conditions. An oxygenator with minimal resistance to blood flow is a critical component of this design strategy. A prototype artificial lung (Michigan Critical Care Consultants, Ann Arbor, MI) has been designed for this purpose. Unique construction of the shell and arrangement of the membrane provides minimal resistance to blood flow. Earlier generations of this prototype were investigated in anesthetized sheep [18], with the device in series with the native lungs. This preliminary work demonstrated that the right ventricle could tolerate perfusing the low resistance device for short periods of time under controlled conditions. This study is an extension of that early work, and evaluates the current version of the artificial lung in series with the native lungs of an active animal model.

	Material and methods

Top Abstract Introduction Material and methods Results Comment References

 
Surgical preparation
Seven adult sheep (55 ± 3 kg) were fasted for 24 hours before surgical preparation. Each animal was sedated with 5 to 10 mg/kg sodium thiopental (Abbott Laboratories, North Chicago, IL), and placed in a supine position upon the operating table. The animal was initially mask ventilated with Isoflurane (Abbott Laboratories, North Chicago, IL) using a gas anesthesia machine (Narkomed 2; North American Drager, Telford, PA). Fifty-thousand µ/kg penicillin G potassium (Marsam Pharmaceuticals, Cherry Hill, NJ) was given intravenously before skin incision. A midline tracheotomy was performed, a cuffed tracheostomy tube placed, and anesthesia continued through the tracheostomy. The left femoral artery and vein were surgically exposed and isolated. Invasive arterial pressure was measured through the femoral artery using a fluid coupled strain-gauge pressure transducer, and the signal was inverted for heart rate display (Hewlett Packard 78304A, 78901A; Hewlett Packard, Andover, MA). The lines were secured and the wounds closed in two layers. Lactated Ringer’s solution was used as maintenance infusion and an orogastric tube was placed.

The animal was repositioned with the right side down and paralyzed with 0.1 mg/kg pancuronium bromide (Elkins-Sinn, Cherry Hill, NJ). A muscle-sparing left lateral thoracotomy was performed through the fourth rib interspace, and the fourth rib was removed. The left lung was collapsed and the pericardium incised, exposing the heart and great vessels. The aorta was isolated, and a transit time ultrasonic flow probe was placed for continuous cardiac output monitoring (Transonic T101; Transonic Systems, Ithaca, NY). The pulmonary artery was isolated, and a 20-mm vascular flow occluder (In Vivo Metric, Healdsburg, CA) was positioned at midlength. A 20-mm Dacron (C.R. Bard, Haverhill, PA) vascular graft, bonded to a 25-cm length of three-quarter inch polyvinylchloride tubing, was used as the connecting conduits to divert blood to and from the prototype lung. The grafts were anastomosed to the pulmonary artery in an end-to-side fashion on either side of the vascular occluder. Heparin sodium (Elkins Sinn, Cherry Hill, NJ) was given as a bolus infusion of 100 µ/kg as the conduits were placed. The two conduits, the aortic flow probe cable and the vascular occluder tubing were tunneled through the subcutaneous tissue and externalized to the chest wall. A chest tube was placed, the lung reexpanded, and the chest was closed in three layers.

The prototype lung was primed with 240 cc of heparinized saline, and attached to the connecting conduits. The prototype was secured to the lateral chest wall and stabilized with a dressing. A second flow probe was placed on the inlet conduit to the prototype lung to measure continuously the diverted flow rate. The animal was removed from the operating table and placed in a padded pen in the sternal position. The chest tube was placed on -20 cm H₂O continuous suction. Initially, the sweep gas through the artificial lung was oxygen at a rate of 2 L/min. Vacuum was applied to the gas outlet of the lung at -20 cm H₂O. Volume cycled mechanical ventilation (Bennett MA-1; Bennett Respiratory Products, Santa Monica, CA) supported the sheep as it recovered from anesthesia, and the sheep was weaned to room air as directed by blood gas analysis (ABL 505, OSM 3 Hemoximeter; Radiometer, Copenhagen, Denmark). The experimental protocol began once the animal was spontaneously breathing room air and had demonstrated the ability to stand.

Prototype artificial lung
The prototype membrane was a hollow fiber, microporous polypropylene mat, with 300 µm outer diameter fibers and a surface area of 2.25 m². The membrane was wrapped around a mandrill, and housed in a cylindrical shell with two circular endcaps. The rigid housing was made of ABS plastic (Michigan Critical Care Consultants, Ann Arbor, MI). Flow entered the device through a 20-mm diameter inlet at the center of one endcap. Blood entered in the axial direction, through the core, and then proceeded through the membrane in a radial fashion (Fig 1). This manner of flow allowed multiple passes across the wound membrane, providing efficient gas exchange. The blood reached the periphery of the device, and was directed by a gutter to exit the prototype. In vitro studies of gas exchange, using bovine blood, demonstrated delivery of 350 cc O₂/min at 4.0 L/min blood flow, with a pressure gradient across the device of only 7.5 mm Hg. The prototype lung is pictured in Figure 2.

View larger version (31K): [in this window] [in a new window]   Fig 1. Prototype artificial lung. Cross-section through the long axis of the prototype artificial lung depicting membrane, blood path, and gas path.

View larger version (92K): [in this window] [in a new window]   Fig 2. The prototype artificial lung and connecting conduits. Prototype artificial lung with connecting conduits and graft material. Dimensions of housing are 170 mm in length with an 80 mm diameter. The 2.25 m2 membrane is hollow fiber microporous polypropylene with fibers of 300 µm outer diameter. Oxygen transfer rate is 350 cc/min at 4 L/min blood flow rate. The pressure gradient generated at that flow rate is 7.5 mm Hg.

Animals were observed for 24 hours, while maximum flow was diverted to the prototype. Hemodynamic characteristics were recorded hourly. Activated clotting time (Hemochron 800; Technidyne Corporation, Edison, NJ) was measured every 1 to 3 hours and heparin sodium was given to keep clotting times between 200 to 250 seconds. Fifty-thousand µ/kg penicillin G potassium was given every 6 hours, as was 5.0 µg/kg buprenorphine hydrochloride (Reckitt & Colman Pharmaceuticals, Richmond, VA) for analgesia. Hourly measurements of the artificial lung included flow rate, inlet pressure, and outlet pressure. Blood gas analysis of systemic blood, and blood entering and exiting the artificial lung, was done hourly. Sweep gas was adjusted to keep the tension of carbon dioxide exiting the artificial lung within the normal range for mixed venous blood of sheep (pCO₂ = 35 to 40 mm Hg).

Inotropic and vasoactive drugs were not used. Animals that appeared to be in distress were sacrificed. At the conclusion of the experiment, the sheep were euthanized with 300 mg/kg sodium pentobarbitol (Schering Plough, Kenilworth, NJ). The sheep used in these experiments were cared for according to guidelines established by the University of Michigan Unit for Laboratory Animal Medicine and the National Institutes of Health, "Guidelines for the Care and Use of Laboratory Animals."

	Results

Top Abstract Introduction Material and methods Results Comment References

 
Seven sheep underwent and survived surgical placement of the prototype lung. Each survived the immediate postoperative period, and was weaned from the ventilator to room air. Five of these animals tolerated full occlusion of the pulmonary artery to establish flow through the prototype lung of at least 75% of the cardiac output. The hemodynamics are reported as preocclusion and postocclusion of the pulmonary artery, and appear in Table 1. The percentage change of each parameter is based on the preocclusion value. Two animals did not tolerate occlusion of the pulmonary artery. One became hemodynamically unstable as the occluder was inflated. This animal did not recover with deflation of the occluder and was sacrificed. The second animal did not tolerate total occlusion, becoming hypotensive and with a falling cardiac output. Partial occlusion was tolerated, but resulted in less than 50% flow through the prototype. This animal was also sacrificed.

View larger version (22K): [in this window] [in a new window]   Fig 3. Oxygen transfer rate versus flow rate of blood through the prototype artificial lung.

	Comment

Top Abstract Introduction Material and methods Results Comment References

The first question addressed was if an awake and active sheep would hemodynamically tolerate diversion of the cardiac output through the prototype. Prior to diverting flow, the animal had to be spontaneously breathing room air and able to stand. These requirements reflected anesthetic recovery. Flow through the device was established by slowly inflating the vascular occluder. Preliminary investigations demonstrated that flows of 75% to 100% could be reproducibly achieved, if the anastomoses were patent and the pulmonary artery adequately occluded. The ability to establish at least 75% flow through the prototype was arbitrarily chosen to reflect successful surgical preparation. Diverting 75% of cardiac output through the prototype was felt to be a sufficient challenge to the right ventricle for 24 hours.

Two of the 7 animals failed to meet this standard. These animals were sacrificed, and at necropsy the anastomoses were noted to be widely patent. It was observed that each animal had relatively short pulmonary arteries, which resulted in close graft positioning. With inflation of the vascular occluder, the back wall of the artery was drawn within close proximity to each arteriotomy. It is possible that this resulted in obstruction to flow, increasing resistance, and acute right heart failure.

Five animals survived diversion of flow through the prototype lung, and the results appear in Table 1. The percentage change in these parameters as flow was established is listed. The cardiac output for animal 4 is not reported because an acoustical error was occurring as flow was diverted. The animal remained in the study because flow through the prototype equaled expected cardiac output for a sheep this size. The acoustical error subsequently resolved, revealing prototype flow to be 95% of cardiac output. Only the heart rate in animal 4, and the mean blood pressure in animal 3 changed by greater than 10% as flow was established (12.3% and 11.5% respectively). All other changes were less than 10%, with most less than 5%. Hemodynamically, these 5 animals tolerated diversion of flow through the prototype lung.

The 5 animals were observed for 24 hours, with flow diverted through the prototype. All animals survived and remained awake, active, with interest in food and water for the duration of the study. Table 2 displays hemodynamic data summarized as the mean ± standard deviation over 24 hours. Mixed venous saturation is also included in this table. The data in Table 1 represent baseline hemodynamics for the 24 hour study.

Reviewing Table 2, the standard deviation for most entries reflects wide variation in each characteristic. There is also a large range of means when comparing the 5 animals. Comparing mean 24 hour heart rate, blood pressure, and cardiac output to baseline values from Table 1, some observations can be made. The mean 24 hour heart rate is elevated in each animal, when compared to baseline values. All rhythms were sinus tachycardias. Mean 24 hour blood pressure was similar to baseline for each animals. Mean cardiac output was increased from baseline in animals 2, 3, and 5, but decreased in animal 1 (baseline not available for 4).

Continuing review of Table 2, animal 2 is noticeable for a persistent hypotension comparable to the baseline pressure. This animal remained awake and active with good urine output for the 24 hour study. Cardiac output was generous, and mixed venous saturation indicated adequate perfusion. Drawing attention to animal 2, the standard deviation for percent cardiac output (Table 2) is large. A leak in the vascular occluder, allowing intermittent deflation, was responsible for this spread.

Mixed venous saturation data are included in both Tables 1 and 2. Mean values for the 24 hour preparation are comparable to the baseline values listed in Table 1. The mixed venous saturations were stable for the duration of the experiment, without drifting far from baseline values. In active, awake animals, the stable mixed venous saturations remained within the normal limits for sheep, suggesting the cardiopulmonary performance was satisfactory to support metabolic demands.

The wide variability is difficult to interpret. There are many factors in the immediate postoperative period that could explain such variation. A partial list includes normal physiology, pneumothorax, hemorrhage, and sepsis. At necropsy, there was no evidence of significant hemorrhage or pneumothorax to the side opposite the thoracotomy. Chest tubes appeared to function during the 24 hour preparation, but partial or total collapse on the surgical side could not be ruled out. Hemodynamic performance did not suggest sepsis. Another cause of hemodynamic variation could have been right ventricular dysfunction related to the work of perfusing the prototype. Specific measures of right ventricular function were not made, and this too, could not be ruled out. Whatever the cause, mixed venous saturations, as an indicator of integrated cardiopulmonary performance, suggested hemodynamics compensated adequately to support metabolism and activity.

Although there was hemodynamic variability, none of the animals suffered catastrophic hemodynamic collapse. Anticipated causes of hemodynamic collapse from this surgical preparation could be exsanguination, pulmonary embolus, tension pneumothorax, and sepsis. Another reasonable cause could be right ventricular failure related to the prototype lung, with an active animal model, exaggerating this cardiovascular stress. Because hemodynamic collapse did not occur during the 24 hours for any of the animals, we concluded the prototype lung did not cause right ventricular failure.

Table 3 presents data related to performance of the prototype lung. Oxygen transfer and the pressure gradient are presented for each trial as the mean ± standard deviation over the 24 hour duration. Carbon dioxide transfer is not reported because the sheep continued to breathe, regulating ventilation. Sweep gas was adjusted to maintain mixed venous pCO₂ in normal ranges (35 to 40 mm Hg). The ability of earlier versions of this prototype to clear CO₂ has been previously demonstrated and reported [18]. Oxygen transfer was sufficient to support oxygen demands of animals this size. Oxygen transfer data were pooled from all 5 animal trials, and plotted against device blood flow in Figure 3. The best linear fit yields the expected oxygen transfer rate for this prototype. Compared to in vitro testing, oxygen transfer was approximately 66% predicted. This difference is because the hemoglobin content of sheep blood is normally 8 to 10 g/dL. Bovine blood was used for the in vitro testing with normal hemoglobin values of 12 to 15 g/dL.

Hemodynamic collapse, as a marker of right ventricular failure, although crude, is easy to recognize in an active animal. While the many hemodynamic variations in this experiment are unexplained, hemodynamic collapse did not occur. The hemodynamic variations could be the result of the prototype device, but the absence of hemodynamic collapse suggests the right ventricle can tolerate perfusing the prototype in an active animal. The flow rate generated through the prototype was sufficient for oxygen exchange capable of total support. A better understanding of the cardiovascular impact of the prototype will result from specific measures of right ventricular function. Echocardiography and invasive pressure monitoring would be useful adjuncts when evaluating right ventricular performance as it perfuses this artificial lung. This knowledge will become important as the design iteration continues.

In summary, the prototype artificial lung was tolerated as an addition to the pulmonary circulation for 24 hours in an active animal model. These results suggest that the right ventricle has the ability to perfuse this prototype for extended periods of time, while accommodating the additional stress of activity. The flow rates generated through the prototype were sufficient to maintain metabolic needs during activity. This pilot study of this prototype artificial lung suggests the concept of a pumpless pulmonary assist device could be a practical method of prolonged respiratory support.

	Acknowledgments

 
We thank Edward Salah and Shigeki Sawada for their surgical and technical assistance. We also appreciate the financial support of the National Institutes of Health (HL 53168-01), making this work possible.

	References

Top Abstract Introduction Material and methods Results Comment References

 

1. Bartlett R.H. Extracorporeal life support for cardiopulmonary failure. Curr Prob Surg 1990;27:623-705.
2. Tracy T.F., DeLosh T., Stolar C.J.H. The registry of the extracorporeal life support organization. In: Zwischenberger J.B., Bartlett R.H., eds. ECMO. Ann Arbor: ELSO, 1995:251-260.
3. Davis R.D., Pasque M.K. Pulmonary transplantation. Ann Surg 1995;221:14-28.[Medline]
4. Bodell B.R., Head J.M., Head L.R., Formolo A.J. An implantable artificial lung. JAMA 1965;191:125-127.
5. Shah-Mirany J., Head L.R., Ghetzler R., Formolo A.J., Palmer A.S., Bodell B.R. An implantable artificial lung. Ann Thorac Surg 1972;13:381-387.[Medline]
6. Palmer A.S., Collins J., Head L.R. Development of an implantable artificial lung. J Thorac Cardiovasc Surg 1973;66:521-525.[Medline]
7. Mortensen J.D. Feasibility of total lung replacement by a pumpless prosthetic lung. Artif Organs 1978;2:141-145.[Medline]
8. Clowes G.H.A., Jr, Hopkins A.L. Further studies with plastic films and their use in oxygenating blood. Trans ASAIO 1956;2:6-12.
9. Bodell B.R., Head J.M., Head L.R., Formolo A.J., Head J.R. A capillary membrane oxygenator. J Thorac Cardiovasc Surg 1963;46:639-650.
10. Keller K.H., Shultis K.L. Oxygen permeability in ultrathin and microporous membranes during gas-liquid transfer. Trans ASAIO 1979;25:469-472.
11. Ohtake S., Kawashima Y., Hirose H., et al. Experimental evaluation of pumpless arteriovenous ECMO with polypropylene hollow fiber membrane oxygenator for partial respiratory support. Trans ASAIO 1983;29:237-241.
12. Kostopoulos C., Rasidakis A., Vardas P., et al. Pumpless extracorporeal oxygenation of the blood through a hollow fiber oxygenator. Trans ASAIO 1985;31:363-366.
13. Mortensen J.D. An intravenacaval blood gas exchange (IVCBGE) device. ASAIO J 1987;33:570-573.
14. High K.M., Snider M.T., Richard R., et al. Clinical trials of an intravenous oxygenator in patients with adult respiratory distress syndrome. Anesthesiology 1992;77:856-863.[Medline]
15. Hattler B.G., Johnson P.C., Sawzik P.J., et al. "Respiratory dialysis". ASAIO J 1992;38:M296-M300.[Medline]
16. Tatsumi E., Takano H., Taenaka Y., et al. A novel method for long-term respiratory support. Trans ASAIO 1989;35:671-673.[Medline]
17. Tatsumi E., Taenaka Y., Nakatani T., et al. A VAD and novel high performance compact oxygenator for long-term ECMO with local anticoagulation. Trans ASAIO 1990;36:M480-M483.[Medline]
18. Fazzalari F.L., Montoya J.P., Bonnell M.R., Bliss D.W., Hirschl R.B., Bartlett R.H. The development of an implantable artificial lung. ASAIO J 1994;40:M728-M731.[Medline]

This article has been cited by other articles:

				Z. J. Wu, M. Gartner, K. N. Litwak, and B. P. Griffith Progress toward an ambulatory pump-lung J. Thorac. Cardiovasc. Surg., October 1, 2005; 130(4): 973 - 978. [Abstract] [Full Text] [PDF]

				M. Czerny, A. Salat, T. Fleck, W. Hofmann, D. Zimpfer, F. Eckersberger, W. Klepetko, E. Wolner, and M.-R. Mueller Lung wedge resection improves outcome in stage I primary spontaneous pneumothorax Ann. Thorac. Surg., May 1, 2004; 77(5): 1802 - 1805. [Abstract] [Full Text] [PDF]

				J. B. Zwischenberger, D. Wang, S. D. Lick, D. J. Deyo, S. K. Alpard, and S. D. Chambers The paracorporeal artificial lung improves 5-day outcomes from lethal smoke/burn-induced acute respiratory distress syndrome in sheep Ann. Thorac. Surg., October 1, 2002; 74(4): 1011 - 1018. [Abstract] [Full Text] [PDF]

				J. W. Haft, P. Montoya, O. Alnajjar, S. R. Posner, J. L. Bull, M. D. Iannettoni, R. H. Bartlett, and R. B. Hirschl An artificial lung reduces pulmonary impedance and improves right ventricular efficiency in pulmonary hypertension J. Thorac. Cardiovasc. Surg., December 1, 2001; 122(6): 1094 - 1100. [Abstract] [Full Text] [PDF]

				J. B. Zwischenberger, S. K. Alpard, and A. Bidani Severe respiratory failure Ann. Thorac. Surg., January 1, 2001; 71(1): 394 - 396. [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): Robert J. Schreiner Mark D. Iannettoni Robert H. Bartlett Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lynch, W. R. Articles by Bartlett, R. H. Search for Related Content PubMed PubMed Citation Articles by Lynch, W. R. Articles by Bartlett, R. H. Related Collections Related Article