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): Brian G. Barratt-Boyes Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wong, S. P. Articles by Raudkivi, P. J. Search for Related Content PubMed PubMed Citation Articles by Wong, S. P. Articles by Raudkivi, P. J.

Ann Thorac Surg 2000;69:1846-1850
© 2000 The Society of Thoracic Surgeons

---

Original articles: Cardiovascular

Early experience with the Mosaic bioprosthesis: a new generation porcine valve

^a Department of Cardiology, Green Lane Hospital, Auckland, New Zealand
^b Department of Cardiothoracic Surgery, Green Lane Hospital, Auckland, New Zealand

Address reprint requests to Dr Legget, Department of Cardiology, Green Lane Hospital, Green Lane West, Auckland 1005, New Zealand
e-mail: malcolml{at}ahsl.co.nz

	Abstract

Top Abstract Introduction Material and methods Results Comment Appendix References

 
Background. The Mosaic bioprosthesis is a new generation stented porcine valve.

Methods. Between May 1995 and April 1998, this valve was implanted in the aortic position in 98 patients (70 men; mean age, 69.2 years [34.2 to 83.6 years]). Preoperatively 35 patients were in New York Heart Association functional class 3 or 4. Fifty-nine patients underwent concomitant procedures. The mean duration at follow-up in January 1999 was 23.7 ± 10.2 months (0.3 to 39.4 months) and totaled 193 patient-years. All but one survivor was in New York Heart Association class 1 or 2.

Results. Early complications included 1 death, 3 reoperations for bleeding, greater than mild regurgitation (paravalvar) in 1 patient and thromboembolism in 4 patients. Late complications included four deaths, study-valve endocarditis in 3 patients, more than mild regurgitation or hemolysis in 2, and thromboembolism in 2 patients. Late follow-up echocardiography in all survivors showed a mean transaortic gradient of 13.6 ± 6.7 mm Hg, and an aortic valve area of 1.80 ± 0.61 cm². Valve replacement was followed by a significant and sustained decrease in left ventricular mass for all valve sizes. There has been no primary structural valve failure.

Conclusions. The early experience with the Mosaic valve in the aortic position has been promising.

	Introduction

Top Abstract Introduction Material and methods Results Comment Appendix References

 
The Medtronic Mosaic bioprosthesis (Medtronic, Minneapolis, MN) is a new generation stented porcine aortic valve. The leaflets are fixed in glutaraldehyde at zero pressure [1], and the root is dilated to a pressure of 40 mm Hg (root pressure or "physiologic" fixation). This is in contrast to the Medtronic Intact valve where the leaflets and the aortic root are both fixed at zero pressure. In this situation, in contrast to the Mosaic valve, the root diameter is small relative to leaflet size and there is the potential for leaflet crowding and suboptimal leaflet motion [2]. The fixed tissue is treated with -aminooleic acid [3] to reduce the potential for calcification and mounted on an appropriately sized Hancock II flexible stent made of acetyl copolymer covered with Dacron fabric. The aortic stent is scalloped to mimic the shape of the aortic annulus. The valve has a lower profile (2 mm across all valve sizes) compared with the Intact valve.

We report our initial experience of implantation of the Mosaic valve in the aortic position in a consecutive series of 98 patients as part of an international multicenter prospective nonrandomized clinical trial to evaluate the efficacy, safety, and clinical performance of this device.

	Material and methods

Top Abstract Introduction Material and methods Results Comment Appendix References

 
Study group
Between May 1995 and April 1998, the Mosaic bioprosthesis was implanted in the aortic position in 98 patients, by one of a group of seven surgeons. Use of the bioprosthesis was according to surgeon preference based on clinical grounds. Patients eligible for inclusion in the study protocol designed by Medtronic Inc (Minneapolis, MN) for use by all centers involved in the clinical trial included those undergoing either first time or redo replacement of the aortic valve and who had not had replacement of the other cardiac valves. Patients with a high potential for noncompliance or logistic problems with follow-up or who were undergoing dialysis, were excluded.

Baseline clinical characteristics
As there were 98 patients the figures given are equivalent to percentages. The mean age at implant was 69.2 ± 9.4 years (range, 34.2 to 83.6 years, median 71.7 years). Fifty-five patients were 70 years or older. Seventy were men. The primary indication for operation was aortic valvular stenosis in 72 patients, aortic incompetence in 16, mixed aortic valve disease in 5, and severe coronary disease with concomitant moderate aortic stenosis in 5 patients. Preoperative New York Heart Association functional class was 1 in 2 patients, 2 in 61, 3 in 32, and 4 in 3 patients (mean class, 2.4). The preoperative cardiac rhythm was sinus in 92 patients, atrial fibrillation in 5, and paced in 1 patient. Nine patients had had previous cardiac operation an average of 14.1 years beforehand (range, 4 months to 20.6 years). This group had all had previous aortic valve replacement (AVR)—7 patients with a previous homograft AVR, 1 with a porcine AVR, and 1 patient had undergone aortic valvuloplasty for a congenital lesion. The left ventricular ejection fraction was impaired in 10 patients (35% to 50% in 6 patients, < 35% in 4 patients).

Operative data
The operation was elective in 95 patients, with the remainder being urgent. There were no emergency cases. Fifty-nine patients underwent a concomitant procedure with 49 having coronary procedures, 9 aortic root operations (ascending aortic aneurysm repair/replacement in 2 patients, aortotomy repair in 2, root enlargement in 5 patients), and 1 ventricular aneurysm resection. Those undergoing concomitant coronary operations had an average of 1.9 grafts, and 30 patients had grafting using the left internal thoracic artery. Valve sizes were as follows: 21 mm, 6 patients; 23 mm, 45; 25 mm, 27; 27 mm, 9; and 29 mm, 11 patients. Operation times (in minutes) are as follows: for isolated AVR, 81 ± 20.2 for aortic cross-clamp and 95.5 ± 20.7 for cardiopulmonary bypass; for AVR and concomitant procedure, 106.4 ± 26.9 for aortic cross-clamp and 129.7 ± 30.4 for cardiopulmonary bypass.

Antithrombotic therapy for the early postoperative period was according to each individual surgeon’s preference based on the patient’s clinical situation. Fifty-six patients were anticoagulated with warfarin and 46 received aspirin (5 were treated with both drugs). Fourteen of the patients had new onset atrial fibrillation in the postoperative period.

Follow-up
Patients have been followed with clinical and echocardiographic examinations at 2 weeks, 3 and 12 months, and then annually. Clinical data recorded included New York Heart Association symptomatic class and complications.

Echocardiographic examinations were reviewed by one of two experienced echocardiographers. Studies were carried out using Hewlett Packard Sonos 1000, 1500 or 5500 machines (Hewlett Packard, Andover, MA). Two-dimensional studies examined cusp thickness, calcification, opening and valve integrity, stability, and extraneous echoes or valve bed abnormality. Doppler measurements for each study valve included an assessment of valve area by the continuity equation, measurement of mean and peak instantaneous gradient and quantification of regurgitation by color Doppler imaging, the continuous wave Doppler intensity, aortic flow reversal, and degree of penetration of the regurgitant jet. Left ventricular mass was calculated by the American Society of Echocardiography cube method [4] (see Appendix).

Follow-up was 100% complete and totaled 193 patient-years.

Statistics
Results are presented as the mean ± standard deviation. Continuous data were analyzed using paired and unpaired two-tailed Student’s t test. Statistical significance was assumed for a p value less than 0.05.

	Results

Top Abstract Introduction Material and methods Results Comment Appendix References

 
Complications
Early (< 30 days) complications comprised one death, postoperative bleeding requiring reexploration in 3 patients, thromboembolism in 4 (all transient with subsequent full recovery), and other hemorrhage in 2 patients. There were no in-hospital deaths. The one early death was sudden and occurred 2 weeks after replacement of the aortic valve and ascending aorta. Postmortem examination showed cardiac tamponade despite both the aortic anastomoses appearing intact.

Late complications have included another four deaths (Table 1). Bioprosthesis function was satisfactory in each patient. Three patients had study valve endocarditis (Table 2). Two patients had late thromboembolic events, both while on aspirin therapy. No patient has required study valve reoperation. At latest follow-up all but 1 of the surviving patients were in New York Heart Association functional class 1 or 2 (mean class, 1.3).

Regurgitation
Two patients had more than mild aortic regurgitation (AR). In 1 patient with severe AR and severe left ventricular (LV) impairment preoperatively, the echocardiographic examination at 1 week had shown mild anterior paravalvar regurgitation, which had become moderate 1 week later and remained moderate at 24 months. This is the only patient with any significant (mild) hemolysis. The second patient had Streptoccocus faecalis endocarditis with persistent moderate transvalvar and paravalvar AR that has not progressed during the 27 months of follow-up.

Eighteen patients had evidence of trivial or mild AR. Trivial transvalvar regurgitation was present in 5 of these patients at the early scan with no progression in severity to date. Overall 14 had trivial AR with the origin being transvalvar in 5, paravalvar in 3, and indeterminate in 6, and 4 had mild AR with the origin being transvalvar in 3 and both transvalvar and paravalvar in 1 patient. In patients with mild AR, this was first noted on the 3-month scan in 2, although in 1 of these was subsequently reclassified as trivial, and in the other 2, the AR was first noted at the 2-year study.

Transvalvar velocity, gradient, and valve area
Table 3 shows the peak velocity and mean gradient across the bioprosthetic valves, and the effective orifice area (calculated from the continuity equation) for the aortic study valves at baseline and at 2 years postoperatively. There was no significant change between the measurements for any of the values between the two measurements.

	Comment

Top Abstract Introduction Material and methods Results Comment Appendix References

The 30-day mortality rate of 1% comprised a single event after discharge, which was independent of valve function. This low early mortality occurred despite the presence of factors increasing the risk of the operation (advanced age or New York Heart Association functional class, LV impairment, and redo or concomitant operation). Of the four late deaths, one resulted from osteomyelitis and septicemia with presumed concurrent study-valve endocarditis, whereas valve function was satisfactory in the remainder. The mortality figures compare very favorably to other series of aortic valve replacement by any prosthesis in similar populations [5].

The rates of thromboembolism and endocarditis are comparable to that in other series of AVR [5]. Of the 3 patients with endocarditis, sepsis in 1 patient was very likely from emergency resternotomy in the early postoperative period and a second patient had osteomyelitis suspected clinically before a presumptive diagnosis of endocarditis. Similarly, valve dysfunction rates are low, and at this early stage, there has been neither requirement for reoperation nor any primary structural valve failure. However, it should be noted that trivial regurgitation was seen on the early postoperative study in 5 subjects, although in none of these patients has the regurgitation progressed. Also, mild transvalvar AR has developed early in 1 patient and more than 2 years in 2 other patients. There are 5 patients with paravalvar regurgitation in the absence of known endocarditis. Potential explanations in 3 of these patients include difficulties with an enlarged aortic root in the patient with severe preoperative regurgitation, difficulty in valve sizing with the implantation of an undersized valve, and mild annular and aortic root calcification. In this series, survivors have remained asymptomatic or only mildly symptomatic (New York Heart Association functional class 1 or 2).

Three of the four hemorrhagic complications are directly attributable to poor control of warfarin therapy. Use of routine early short-term anticoagulation was greater early in this series, but subsequent experience with the bioprosthesis, its safety in the absence of warfarin, and complications related to anticoagulation, have seen this practice diminish. However, anticoagulation for a 6-week postoperative period has generally been used after porcine AVR concomitant with aortic operation, postoperative atrial arrhythmias, or severe LV impairment.

Comparison to other series of Mosaic valves
Thompson and colleagues [6] in a series of 227 subjects with Mosaic aortic implantation demonstrated a lower valve gradient, but also a smaller effective orifice across all valve sizes at 6 months. Eichinger and associates [7] in 55 Mosaic aortic implantations, reported similar mean gradients, but larger effective orifice areas than ours in the three smallest sized Mosaic prostheses at 6 months (21 mm, 1.6 cm²; 23 mm, 2.1 cm²; 25 mm, 2.1 cm²), with the larger valve areas persisting at 24 months for the 23-mm and 25-mm prostheses.

Comparison to other stented bioprostheses
Studies of older generation bioprosthetic valves, and even of more recent valves, lack the standardized detail in this study [8–11]. However, some comparisons can be made with published data. The reported effective valve orifice area for both the Hancock II (Medtronic Inc, Minneapolis, MN) and Carpentier-Edwards bioprosthetic valves (Baxter Healthcare Corp, Irvine, CA) is 0.2 to 0.3 cm² less than in this series at 1 year [8]. Jaffe and colleagues [9] reported a similar peak aortic velocity to our group for the Intact valve, but the mean gradient found in our group is less by 2 to 3 mm Hg at similar postoperative times for all valve sizes. The mean gradient for the 21- and 23-mm valves in our group is similarly lower compared to a second Intact series, and for similar sized valves, our series shows larger effective orifice areas by 0.3 to 0.4 cm² at 1-year for both small and large sizes [10].

Comparison to stentless bioprostheses
In comparing individual valve sizes, the Mosaic size (taken from the external diameter) is compared to a stentless prosthesis one "size" smaller, which would be appropriate for the same sized aortic root [12]. The Toronto stentless valve (Toronto SPV; St. Jude Medical, St. Paul, MN) was reported to have an overall lower mean gradient compared to this series at 3 to 6 months of 2 to 3 mm Hg [8]. A second series of the Toronto SPV showed a lower mean gradient of 4 to 5 mm Hg for equivalent valve sizes, but a comparable valve area to this series [13]. The Freestyle valve (Medtronic Inc, Minneapolis, MN) has been demonstrated to have a larger valve area, in the order of 0.2 to 0.3 cm², compared to this series [14]. The better hemodynamic profile is proposed as an advantage of stentless bioprostheses [15], although in this series of stented valves, the larger valve sizes also have an excellent hemodynamic profile without the disadvantage of the more complex implantation technique associated with stentless valves.

Left ventricular mass
The regression of left ventricular hypertrophy, and consequently left ventricular mass, has been considered an important accompaniment and goal of aortic valve replacement [16]. Persistent hypertrophy has been linked to reduction in long-term survival and an increase in late events because of increased myocardial oxygen demand, a poorly compliant chamber with diastolic dysfunction [17, 18], and to sudden death [19]. In this series there is a significant reduction in left ventricular wall thickness and mass over time. The reduction in LV mass starts early and is sustained for at least 24 months. A reduction of 10% in the left ventricular mass index (LVMI) was noted at the initial postoperative echocardiogram and at 24 months the LVMI was reduced by 25%. This reduction in LVMI is similar to that shown for stentless valves over a similar follow-up period [8, 14]. Table 4 shows that this reduction in LVMI is not limited to the larger sized valves, with the LVMI for the small valves not significantly different at equivalent postoperative times. However, a significant reduction in LVMI compared to preoperative values occurs earlier for the larger sized valves than for the smaller valves (3 months versus 12 months). The lack of a significant difference for the smaller valves at 24 months compared to preoperative LVMI may reflect the small number of values at that time.

Our initial results with the stented Mosaic valve in the aortic position are promising. Stented valves appeal as prostheses because of the relative ease of implantation and shorter operative times. The stent mounting does contribute unfavorably to hemodynamic performance particularly in the smaller valve sizes, but less so in this series than in some previous reports. Left ventricular mass does show a significant and sustained regression for all valve sizes. The stentless valve allows greater implant options and a potential hemodynamic profile more closely resembling the native aortic valve, but has a substantial learning curve associated with its implantation because of the more difficult implantation technique [8]. This is of particular relevance given its current limited use in a smaller number of centers generally with either larger volumes of operations or particular expertise. Although leaflet fixation and preservation techniques can be duplicated between stented and nonstented bioprosthetic valves, it remains to be seen whether the long-term results of valve-related mortality and morbidity will be different. The long-term effect of the newer antimineralization treatment used in the Mosaic valve is also to be determined.

Although the Mosaic valve combines the new design features of root pressure fixation and -aminooleic acid treatment, it is appropriate to compare its clinical behavior with that of the Intact valve as the leaflets are fixed at zero pressure in both bioprostheses. Follow-up of the Intact valve [20] reveals a 10-year actuarial freedom from structural valve degeneration of 92% ± 8% in patients more than 40 years of age, and 100% in patients more than 60 years old. The new design features of the Mosaic valve are expected to improve those results further.

In conclusion, our initial experience of the clinical and hemodynamic performance of the Mosaic valve in the aortic position is gratifying with low perioperative and medium-term mortality and morbidity. Whether this initially favorable performance is sustained will require ongoing follow-up of the cohort.

	Acknowledgments

 
We gratefully acknowledge the permission to include patients operated on by Drs Alan R. Kerr, Kenneth Graham, David A. Haydock, Kirsten Finucane, and Xhao Chen. We also acknowledge the support of Medtronic Inc for patient follow-up.

	Appendix

Top Abstract Introduction Material and methods Results Comment Appendix References

 
Calculations
Effective orifice area
The effective orifice area (EOA) was calculated using the continuity equation as follows: EOA = (ALVOT x TVILVOT)/TVIA, ALVOT, area of left ventricular outflow tract (in centimeters squared). If the LVOT diameter cannot be confidently measured, the bioprosthetic sewing ring is substituted; TVILVOT, TVI of the left ventricular outflow tract (in centimeters); and TVIA, TVI across the aortic bioprosthesis (in centimeters).

Left ventricular mass index
The left ventricular mass (LVM) was calculated using the American Society of Echocardiography cube method: LVM = 0.000832 x [(LVEDD + IVS + PW)³ - (LVEDD)³] + 0.6, where LVM, left ventricular mass (in grams); LVEDD, left ventricular end-diastolic dimension (in centimeters); IVS, interventricular septal wall thickness (in centimeters); and PW, left ventricular posterior wall thickness (in centimeters).

The left ventricular mass was indexed to body surface area (BSA) in square meters: BSA = [(height in cm x body weight in kg)/3,600]^1/2.

	References

Top Abstract Introduction Material and methods Results Comment Appendix References

 

1. Broom N., Thomsom E.J. Influence of fixation conditions on the performance of gluteraldehyde-treated porcine aortic valves. Thorax 1979;34:166-176.[Abstract/Free Full Text]
2. Lockie K.J., Fisher J., Juster N.P., Davies G.A., Watterson K. Biomechanics of gluteraldehyde-treated porcine aortic roots and valves. J Thorac Cardiovasc Surg 1994;108:1037-1042.[Abstract/Free Full Text]
3. Gott J.P., Pan-Chih, Dorsey L.M.A., et al. Calcification of porcine valves. Ann Thorac Surg 1992;53:207-216.
4. Devereux R.B., Alonso D., Lutas E., et al. Echocardiographic assessment of left ventricular hypertrophy. Am J Cardiol 1986;57:450-458.[Medline]
5. Kirklin JW, Barratt-Boyes BG. Aortic valve disease. In: Kirklin JW, Barratt-Boyes BG, eds. Cardiac surgery, 2nd ed. New York: Churchill Livingston, 1993:491–571.
6. Thompson D.J., Jamieson W.R.E., Dumesnil J.G., et al. Medtronic Mosaic porcine bioprosthesis. Ann Thorac Surg 1998;66:S122-S125.
7. Eichinger W.B., Schutz A., Simmerl D., et al. The Mosaic bioprosthesis in aortic position. Ann Thorac Surg 1998;66:S126-S130.
8. Cohen G., Christakis G.T., Buth K.J., et al. Early experience with stentless versus stented valves. Circulation 1997;96(suppl 2):76-82.[Abstract/Free Full Text]
9. Jaffe W.M., Barratt-Boyes B.G., Sadri A., Gavin J.B., Coverdale H.A., Neutze J.M. Early follow-up of patients with the Medtronic Intact porcine valve. J Thorac Cardiovasc Surg 1989;98:181-192.[Abstract]
10. Lemieux M.D., Jamieson W.R.E., Landymore R.W., et al. Medtronic Intact porcine bioprosthesis. Ann Thorac Surg 1995;60:S258-S263.
11. David T.E., Armstrong S., Sun Z. Clinical and hemodynamic assessment of the Hancock II bioprosthesis. Ann Thorac Surg 1992;54:661-668.[Abstract]
12. Jaffe W.M., Coverdale H.A., Roche A.H.G., Whitlock R.M.L., Neutze J.M., Barratt-Boyes B.G. Rest and exercise hemodynamics of 20 to 23 mm allograft, Medtronic Intact (porcine), and St. Jude Medical valves in the aortic position. J Thorac Cardiovasc Surg 1990;100:167-174.[Abstract]
13. Wong K., Shad S., Waterworth P.D., Khaghani A., Pepper J.R., Yacoub M.H. Early experience with the Toronto stentless porcine valve. Ann Thorac Surg 1995;60:S402-S405.
14. Westaby S., Jin X.Y., Katsumata T., Arifi A., Braidley P. Valve replacement with a stentless bioprosthesis. J Thorac Cardiovasc Surg 1998;116:477-484.[Abstract/Free Full Text]
15. Westaby S., Huysmans H.A., David T.E. Stentless aortic bioprostheses. Ann Thorac Surg 1998;65:235-240.[Abstract/Free Full Text]
16. Christakis G.T., Campbell D.J., Morgan C.D., et al. Left ventricular mass regression early after aortic valve replacement. Ann Thorac Surg 1996;62:1084-1089.[Abstract/Free Full Text]
17. Lund O., Pilegaard H.K., Magnussen K., Knudsen M.A., Nielsen T.T., Albrechtsen O.K. Long-term prosthesis-related and sudden cardiac-related complications after aortic valve replacement for aortic stenosis. Ann Thorac Surg 1990;50:396-406.[Abstract]
18. Lytle B.W., Cosgrove D.M., Taylor P.C., et al. Primary isolated aortic valve replacement. J Thorac Cardiovasc Surg 1989;97:675-694.[Abstract]
19. Bikkina M., Larson M.G., Levy D. Asymptomatic ventricular arrythmias and mortality risk in subjects with left ventricular hypertrophy. J Am Coll Cardiol 1993;22:1111-1116.[Abstract]
20. Barratt-Boyes B.G., Jaffe W.M., Whitlock R.M.L. The Medtronic Intact porcine valve. J Thorac Cardiovasc Surg 1998;116:1005-1014.[Abstract/Free Full Text]

This article has been cited by other articles:

				J. S. Lawton, N. Moazami, M. K. Pasque, M. R. Moon, and R. J. Damiano Jr. Early stenosis of Medtronic Mosaic porcine valves in the aortic position. J. Thorac. Cardiovasc. Surg., June 1, 2009; 137(6): 1556 - 1557. [Full Text] [PDF]

				J. B. Chambers, R. Rajani, D. Parkin, H. M. Rimington, C. I. Blauth, G. E. Venn, C. P. Young, and J. C. Roxburgh Bovine pericardial versus porcine stented replacement aortic valves: Early results of a randomized comparison of the Perimount and the Mosaic valves. J. Thorac. Cardiovasc. Surg., November 1, 2008; 136(5): 1142 - 1148. [Abstract] [Full Text] [PDF]

				F.-C. Riess, R. Bader, E. Cramer, L. Hansen, B. Kleijnen, G. Wahl, J. Wallrath, S. Winkel, and N. Bleese Hemodynamic Performance of the Medtronic Mosaic Porcine Bioprosthesis Up to Ten Years Ann. Thorac. Surg., April 1, 2007; 83(4): 1310 - 1318. [Abstract] [Full Text] [PDF]

				W. B. Eichinger, F. Botzenhardt, R. Gunzinger, B. M. Kemkes, A. Sosnowski, D. Maiza, E. O. Coto, and N. Bleese European experience with the Mosaic bioprosthesis J. Thorac. Cardiovasc. Surg., August 1, 2002; 124(2): 333 - 339. [Abstract] [Full Text] [PDF]

				T. Aberg, X. Y. Jin, and P. Deverall Post operative assessment of a bioprosthetic aortic valve (Labcor) Eur. J. Cardiothorac. Surg., September 1, 2001; 20(3): 658 - 658. [Full Text] [PDF]

				D. J. Thomson, W.R. E. Jamieson, J. G. Dumesnil, J. J. Burgess, C. M. Peniston, J. Metras, J. A. Sullivan, J. C.W. Parrott, A. Maitland, and I. J. Cybulsky Medtronic mosaic porcine bioprosthesis: midterm investigational trial results Ann. Thorac. Surg., May 1, 2001; 71 (2007): S269 - S272. [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): Brian G. Barratt-Boyes Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wong, S. P. Articles by Raudkivi, P. J. Search for Related Content PubMed PubMed Citation Articles by Wong, S. P. Articles by Raudkivi, P. J.