SoLEAH

tvrs

pfslsdipag4

Tahaarm(c[rfag

ptara

A

PS

ACO

©P

CA

RD

IAC

implified David Reimplantation With Reductionf Anular Size and Creation of Artificial Sinuses

ars G. Svensson, MD, PhD, Maxwell Cooper, Lillian H. Batizy, MS, anddward R. Nowicki, MD, MSorta Center, Department of Thoracic and Cardiovascular Surgery, Heart and Vascular Institute, and Department of Quantitative

ealth Sciences, Research Institute, Cleveland Clinic, Cleveland, Ohio

dmctgwaap

apa

AD

ULT

Background. The David reimplantation procedure ishe preferred method of preserving tricuspid aorticalves during aortic root replacement. We report theesults of a simplified approach to the David valve-paring root reimplantation.

Methods. Of 234 patients who underwent David reim-lantation or some modification thereof, 129 operated onrom January 2001 to June 2008 formed a consecutiveingle-surgeon series for midterm evaluation. Aortic anu-us-left ventricular outflow tract and proximal tube graftize were reduced over a Hegar dilator to mean normaliameter based on body surface area, in the process creat-

ng neo-sinuses to accommodate cusp opening. Sixty-oneatients (47%) had Marfan syndrome. Mean body surfacerea was 2.1 � 0.27 m2. Preoperative aortic regurgitationrade was 1� or less in 46%, 2� in 26%, 3� in 24%, and

� in 4.3%.

P

Os2toIPsdp1drnw

aewd

SIart

onnective Tissue Disorder Clinic, 9500 Euclid Ave/Desk J4-1, Cleveland,H 44195; e-mail: svenssl@ccf.org.

2010 by The Society of Thoracic Surgeonsublished by Elsevier Inc

Results. Left ventricular outflow tract sizing by Hegarilator was 17 mm in 9.5% of patients, 19 mm in 18%, 21m in 56%, and 23 mm in 16%. Fifty-five (43%) had

oncomitant cusp repair. Postoperative aortic regurgita-ion grade was 0 in 98%, and none of the remaining hadreater than 2� AR. Postoperative mean aortic gradientas 9.0 � 3.5 mm Hg. No patient had intraoperative

bandonment of the repair, and there were no postoper-tive deaths or strokes. Five-year survival was 99%, and 4atients (3%) required late valve replacement.Conclusions. A simple modification of the David oper-

tion, reducing anular size, and creating neo-sinusesreserves the aortic valve, eliminates aortic regurgitation,voids aortic stenosis, and has favorable midterm results.

(Ann Thorac Surg 2010;89:1443–7)

© 2010 by The Society of Thoracic Surgeons

he David reimplantation procedure is the preferredmethod of preserving tricuspid aortic valves during

ortic root replacement [1–27]. Several modificationsave been proposed, however, to simplify the operationnd make it more predictable; namely, more science thanrt. Previously, we recommended a simplified Davideimplantation procedure that reduced the anulus to aore normal size for the patient’s body surface area

BSA) and created a new sinotubular junction and neworonary sinuses to accommodate the aortic valve cusps10]. It was our conviction that these measures wouldesult in a more durable preservation of the aortic valveree of residual aortic regurgitation (AR), with minimalortic anulus-left ventricular outflow tract (LVOT)radient.The purpose of this study of an expanded series of

atients undergoing a modified David reimplantationechnique was to focus on the valve reimplantationspect of this operation and to report the followingesults: (1) postoperative residual AR; (2) LVOT gradient;nd (3) midterm valve-related reoperation.

ccepted for publication Jan 20, 2010.

resented at the Fifty-sixth Annual Meeting of the Southern Thoracicurgical Association, Marco Island, FL, Nov 4–7, 2009.

ddress correspondence to Dr Svensson, Aorta Center, and Marfan &

atients and Methods

f 234 patients undergoing David reimplantation, orome modification thereof, from January 1, 2002 to June 1,008, 129 formed a single-surgeon series for the evalua-ion of midterm results. Data from supplemental reviewf medical records was approved for research by thenstitutional Review Board, with patient consent waived.atient characteristics and operative details are pre-ented in Table 1. Sixty-one patients had Marfan syn-rome (47%). Mean BSA was 2.1 � 0.27 m2, and meanredicted normal size of the aortic anulus-LVOT was 23 �.4 mm. The Z value, defined as the number of stan-ard deviations by which the intended diameter of theepaired anulus (Hegar size, mm) deviated from meanormal LVOT diameter (mm) for a patient with that BSA,as calculated and is shown in Figure 1.Preoperative in-house echocardiographic records were

vailable in 116 patients and postoperative predischargechocardiographic records in 101 patients (Table 1). Weere particularly interested in postoperative LVOT gra-ients because size had been reduced in most patients.

urgical Indications and Techniquendications for surgery were combinations of symptom-tic AR with or without left ventricular dysfunction, andoot and ascending aortic dilatation (cross-sectional aor-

ic area to height ratio exceeding 10) [28, 29].

0003-4975/10/$36.00doi:10.1016/j.athoracsur.2010.01.058

pat3u3ttfTsB2f2rm

dotp

SCdatLfiK

R

Nmitps(gpw

Fo

T

V

P

P

O

a

At

TP

P

01234T

1444 SVENSSON ET AL Ann Thorac SurgSIMPLIFIED DAVID REIMPLANTATION 2010;89:1443–7

AD

ULT

CA

RD

IAC

Modifications simplifying this operation have beenreviously described in detail [10]. Briefly, both coronaryrtery buttons were mobilized and the LVOT-aortic junc-ion dissected out to below the aortic valve anulus. A0-mm polyester tube graft was selected for implantationnless the patient’s BSA exceeded 2.5 m2, in which case a2-mm graft was used. If BSA was less than 1.5 m2 andhe cusp height small, a 28-mm graft was chosen. Nine toen pledgeted valve sutures were placed in the LVOT,rom inside out, and passed through the outer graft.hese were then tied down around a Hegar dilator,elected based on patient BSA. Thus, for males with aSA of 2.5 m2, a 23-mm Hegar was chosen, for 2.0 m2 a1-mm Hegar, and for 1.5 m2 a 19-mm Hegar. Foremales, these were reduced by one size; eg, for a BSA of.0 m2, a 19-mm Hegar was used. The valve was theneimplanted with the sutures at the apices of the com-

able 1. Patient Characteristics, Procedures, and Outcomes

ariable naNo. (% of n)

or Mean � SD

atient:Connective tissue disorder 129

No connective tissue disease 45 (35)Marfan syndrome 61 (47)Possible Marfan syndrome 4 (3.1)Uncertain 19 (15)

Body surface area (m2) 129 2.08 � 0.27Predicted average normal diameter

of aortic valve/LVOT (mm)129 22.7 � 1.41

Preoperative AR grade 1160 36 (31)1� 17 (15)2� 30 (26)3� 28 (24)4� 5 (4.3)

rocedure:Hegar size (mm)

17 126 12 (9.5)19 23 (18)21 71 (56)23 20 (16)

Z value (average normal size) 126 �1.09 � 0.6Cusp repair 127 55 (43)utcomes:Stroke 128 0 (0)Transient ischemic attack 128 2 (1.6)Hospital death 129 0 (0)Predischarge AR grade 101

0 91 (90)1� 8 (8)2�2 (2)3� 0 (0)

AV mean gradient (mmHg) 76 9.0 � 3.5

Patients with data available.

R � aortic regurgitation; AV � aortic valve; LVOT � left ven-ricular outflow tract; SD � standard deviation.

issures placed 4 to 5 mm apart, so that when tying them A

own, a neo-sinotubular junction was created. The cor-nary buttons were then reattached and the distal anas-omosis completed. Cusp repair was performed in 55atients (43%) to correct redundant or prolapsing cusps.

tatisticsontinuous variables are summarized by mean � stan-ard deviation and categoric variables by frequenciesnd percentages. Correlation between Z value for in-ended LVOT and postoperative predischarge meanVOT gradient was analyzed (Pearson correlation coef-cient). Unadjusted survival was determined by theaplan-Meier method.

esults

o patient had intraoperative or in-hospital abandon-ent or reversal of the repair. No fistulae into surround-

ng chambers, ventricular septal defects, or anterior mi-ral valve perforations were documented. There were noostoperative deaths or strokes; however, 2 patientsuffered transient ischemic attacks. Of 101 patients, 9998%) had 1� or less AR postoperatively, and none hadreater than 2� AR (Table 1). Of 94 patients with bothreoperative and postoperative data, 24 (26%) remainedithout AR and 66 (70%) had improvement in AR; only 4

ig 1. Cumulative distribution of Z value of intended left ventricularutflow tract diameter.

able 2. Aortic Valve Regurgitation Before and After Davidrocedure

reoperative AR Grade

Postoperative ARGrade

Total0 1� 2�

24 3 1 28� 12 0 0 12� 23 2 0 25� 22 3 0 25� 3 0 1 4otal 84 8 2 94

R � aortic regurgitation.

((Ht�of

TiloStpedsppmt

oa

C

TpvL

viAltiti

wPlbem

Ft

Fep

F

T

IIt

A

S

A

1445Ann Thorac Surg SVENSSON ET AL2010;89:1443–7 SIMPLIFIED DAVID REIMPLANTATION

AD

ULT

CA

RD

IAC

2%) had a slight increase in AR after the procedureTable 2). Mean postoperative gradient was 9.0 � 3.5 mm

g. There was no correlation between Z value of in-ended LVOT diameter and mean gradient (Pearson r �

0.03, p � 0.8; Fig 2). Patients with residual regurgitationr gradients were treated medically as indicated withouturther surgical intervention, except as noted in Table 3.

Postoperative valve-related reoperations are listed inable 3 and illustrated in Figure 3. All patients undergo-

ng reoperation had aortic valve replacement, three re-ated to endocarditis-inflammation. One patient devel-ped an infected toenail after swimming in the sea inouth America, which resulted in endocarditis that was

reated with an allograft, a second patient was treated forresumed endocarditis and fevers and underwent reop-ration for scarred valve leaflets, and a third had Crohnisease with a pelvic abscess and colostomy, and thecarred down valve leaflets and inflamed root wereresumed to be related to endocarditis. In the remainingatient, the anulus below the left main coronary ostiumay have slipped below the graft, allowing prolapse of

he left cusp and AR. Survival was 99% at 6 years, with

ig 2. Scatter plot of Z value of intended left ventricular outflowract (LVOT) diameter and postoperative LVOT mean gradient.

ig 3. Aortic valve-related reoperation. Open circles represent reop-rations, vertical bars are 68% confidence limits, and numbers in

arentheses are patients remaining at risk. r

ne late postoperative death from alcoholic cardiomyop-thy and sepsis (Fig 4).

omment

his study demonstrates that the David reimplantationrocedure can be simplified while maintaining normalalve function, with minimal postoperative AR and lowVOT gradient.As the David reimplantation procedure has evolved,

arious methods have been employed to preserve andmprove valve function. In our series, the near absence ofR after operation, low LVOT gradient, and low preva-

ence of valve-related reoperation testify to the success ofhis evolutionary process. Experience with cusp repair insolated aortic valve repair has allowed us to incorporatehis procedural component in 43% of patients undergo-ng reimplantation to further improve results.

Critical to success is ensuring that the entire valve isell seated within the tube graft, as demonstrated byethig and colleagues [13] and Kallenbach and col-

eagues [22]. It is also noteworthy that the tube graft muste large enough to accommodate the cusps. In thexperience of David and colleagues [5, 6, 11, 19, 20, 23],ost grafts have been 30 mm in diameter. The higher

ig 4. Unadjusted survival. Numbers in parentheses are patients

able 3. Reoperations After Discharge

nterval Fromndex Operationo Reoperation

Cusp Repairat Index

Operation Type of Reoperation

ortic valve-relatedprocedures

126 days Yes AVR � descending aortarepair

1.9 years Yes AVR for endocarditis �aorta repair

2.2 years No AVR4.5 years No AVR � aorta repair

ee text for more details on reoperations.

VR � aortic valve repair.

emaining at risk.

fht“gahnh[ri

dnibwvTdo

tipcgmttq

Dmwrwiu

R

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

1446 SVENSSON ET AL Ann Thorac SurgSIMPLIFIED DAVID REIMPLANTATION 2010;89:1443–7

AD

ULT

CA

RD

IAC

ailure rate reported by Burkhart and colleagues [30] mayave been due to use of smaller diameter grafts. The

rend by some surgeons is to use grafts with preformedsinuses” to accommodate the cusps [12, 15, 20]. We haveenerally not used this graft because of the native vari-bility of cusps and the height at which the commissuresave to be attached. Whether or not recreating a neweo-sinotubular junction is important is unclear, but mayelp in improving cusp opposition. The data from Miller

16] and the Stanford groups [21, 24, 25] suggest thatecreating a “normal” morphologic root and sinuses ismportant for laminar flow and proper valve function.

Our data reveal that the mean intended LVOT (anular)iameter by Hegar dilator is 1 standard deviation belowormal for our patients. We originally undersized the

ntended LVOT because size in normalized tables wasased on aortic allograft sizes adjusted for BSA [10]. Itas believed that the ex vivo size was larger than the in

ivo size because of the relaxation of postmortem tissues.his study thus confirms that slightly smaller LVOTiameter (2 mm below a table size) is associated withnly a minimal gradient.Another reason for downsizing the anulus was a belief

hat a smaller valve is more likely to have better durabil-ty, allowing for some stretching over time. It was notossible to measure anular diameter after repair in allases by echocardiography because of the surroundingraft. However, in those patients for whom it could beeasured, anular diameter remained at Hegar size for

hat patient. We believe that use of a Hegar dilator withinhe anulus when tying down the sutures maintains ade-uate LVOT diameter, preventing excessive gradient.In conclusion, this study demonstrates that a simplifiedavid technique results in good aortic valve function withinimal gradients. The midterm results are encouraging,ith only four failures (3%). Long-term follow-up will be

equired in patients with this operative approach, althoughe have obtained 94% freedom from reoperation at 9 years

n our total experience with our larger series of patientsndergoing traditional David reimplantation.

eferences

1. Svensson LG, Blackstone EH, Cosgrove DM 3rd. Surgicaloptions in young adults with aortic valve disease. Curr ProblCardiol 2003;28:417–80.

2. Miller DC. Valve-sparing aortic root replacement in patientswith the Marfan syndrome. J Thorac Cardiovasc Surg 2003;125:773–8.

3. Karck M, Kallenbach K, Hagl C, Rhein C, Leyh R, HaverichA. Aortic root surgery in Marfan syndrome: comparison ofaortic valve-sparing reimplantation versus composite graft-ing. J Thorac Cardiovasc Surg 2004;127:391–8.

4. Kallenbach K, Pethig K, Schwarz M, Milz A, Haverich A,Harringer W. Valve sparing aortic root reconstruction versuscomposite replacement—perioperative course and earlycomplications. Eur J Cardiothorac Surg 2001;20:77–81.

5. David TE, Feindel CM. An aortic valve-sparing operation forpatients with aortic incompetence and aneurysm of theascending aorta. J Thorac Cardiovasc Surg 1992;103:617–21.

6. David TE, Feindel CM, Bos J. Repair of the aortic valve inpatients with aortic insufficiency and aortic root aneurysm.

J Thorac Cardiovasc Surg 1995;109:345–52.

7. Svensson LG, Longoria J, Kimmel WA, Nadolny E. Manage-ment of aortic valve disease during aortic surgery. AnnThorac Surg 2000;69:778–84.

8. Schafers H, Fries R, Langer F, Nikoloudakis N, Graeter T,Grundmann U. Valve-preserving replacement of the as-cending aorta: remodeling versus reimplantation. J ThoracCardiovasc Surg 1998;116:990–6.

9. Leyh RG, Fischer S, Kallenbach K, et al. High failure rateafter valve-sparing aortic root replacement using the “re-modeling technique” in acute type A aortic dissection.Circulation 2002;106:I229–33.

0. Svensson LG. Sizing for modified David’s reimplantationprocedure. Ann Thorac Surg 2003;76:1751–3.

1. Kunzelman KS, Grande KJ, David TE, Cochran RP, VerrierED. Aortic root and valve relationships. Impact on surgicalrepair. J Thorac Cardiovasc Surg 1994;107:162–70.

2. De Paulis R, De Matteis GM, Nardi P, et al. One-yearappraisal of a new aortic root conduit with sinuses ofValsalva. J Thorac Cardiovasc Surg 2002;123:33–9.

3. Pethig K, Milz A, Hagl C, Harringer W, Haverich A. Aorticvalve reimplantation in ascending aortic aneurysm: risk factorsfor early valve failure. Ann Thorac Surg 2002;73:29–33.

4. Svensson LG, Blackstone EH, Feng J, et al. Are Marfansyndrome and marfanoid patients distinguishable on long-term follow-up? Ann Thorac Surg 2007;83:1067–74.

5. Patel ND, Williams JA, Barreiro CJ, et al. Valve-sparing aorticroot replacement: early experience with the De Paulis Valsalvagraft in 51 patients. Ann Thorac Surg 2006;82:548–53.

6. Miller DC. Valve-sparing aortic root replacement: currentstate of the art and where are we headed? Ann Thorac Surg2007;83:S736–9.

7. Svensson LG. Aortic valve stenosis and regurgitation: anoverview of management. J Cardiovasc Surg (Torino) 2008;49:297–303.

8. Svensson LG, Deglurkar I, Ung J, et al. Aortic valve repairand root preservation by remodeling, reimplantation, andtailoring: technical aspects and early outcome. J Card Surg2007;22:473–9.

9. David TE, Armstrong S, Maganti M, Colman J, Bradley TJ.Long-term results of aortic valve-sparing operations in pa-tients with Marfan syndrome. J Thorac Cardiovasc Surg2009;138:859–64.

0. Cameron DE, Alejo DE, Patel ND, et al. Aortic root replace-ment in 372 Marfan patients: evolution of operative repairover 30 years. Ann Thorac Surg 2009;87:1344–9.

1. Fleischmann D, Liang DH, Mitchell RS, Miller DC. Pre- andpostoperative imaging of the aortic root for valve sparing aorticroot repair. Semin Thorac Cardiovasc Surg 2008;20:365–73.

2. Kallenbach K, Baraki H, Khaladj N, et al. Aortic valve-sparing operation in Marfan syndrome: what do we knowafter a decade? Ann Thorac Surg 2007;83:S764–8.

3. David TE. Sizing and tailoring the Dacron graft for reim-plantation of the aortic valve. J Thorac Cardiovasc Surg2005;130:243–4.

4. Demers P, Miller DC. Simple modification of “T. David-V”valve-sparing aortic root replacement to create graft pseudo-sinuses. Ann Thorac Surg 2004;78:1479–81.

5. Markl M, Draney MT, Miller DC, et al. Time-resolvedthree-dimensional magnetic resonance velocity mapping ofaortic flow in healthy volunteers and patients after valve-sparing aortic root replacement. J Thorac Cardiovasc Surg2005;130:456–63.

6. Volguina IV, Miller DC, LeMaire SA, et al. Valve-sparing andvalve-replacing techniques for aortic root replacement in pa-tients with Marfan syndrome: analysis of early outcome. AorticValve Operative Outcomes in Marfan Patients study group.J Thorac Cardiovasc Surg 2009;137:1124–32.

7. Hess PJ Jr, Harman PK, Klodell CT, et al. Early outcomes usingthe Florida sleeve repair for correction of aortic insufficiencydue to root aneurysms. Ann Thorac Surg 2009;87:1161–8.

8. Svensson LG, Kim KH, Lytle BW, Cosgrove DM. Relation-ship of aortic cross-sectional area to height ratio and the risk

of aortic dissection in patients with bicuspid aortic valves.J Thorac Cardiovasc Surg 2003;126:892–3.

2 3

D

Dgtitpyt

DistAss

Dp

DlI1ptar

Dtt

DsWsNcdWBa

Djyeyrwytort

Dt

1447Ann Thorac Surg SVENSSON ET AL2010;89:1443–7 SIMPLIFIED DAVID REIMPLANTATION

9. Svensson LG, Khitin L. Aortic cross-sectional area/heightratio timing of aortic surgery in asymptomatic patients with

Marfan syndrome. J Thorac Cardiovasc Surg 2002;123:360–1.

hat when you do the reimplantation the height of the commissures

imisssilt

rpvst

DtovtEgagwaoauc

Dmd

D

DdHdilbrc

sb

Dbs

Dmob

0. Burkhart HM, Zehr KJ, Schaff HV, Daly RC, Dearani JA,Orszulak TA. Valve-preserving aortic root reconstruction: a

comparison of techniques. J Heart Valve Dis 2003;12:62–7.

AC

ISCUSSION

AD

ULT

CA

RD

I

R JOHN S. IKONOMIDIS (Charleston, SC): That was a reallyreat talk and I enjoyed your review of your series. The modifica-ions of the T. David operation, the reimplantation operation,nclude use of a larger graft when the root is reconstructed in ordero recapitulate sinuses. I noticed that with your subanularledgeted suture line you don’t actually plicate the graft first beforeou insert it, but you rely just on the sutures themselves to plicatehe graft. Do you have any concerns about late failures with that?

R SVENSSON: That is a good question. I think what happenss when you plicate around the Hegars, it clearly reduced to theize of the Hegars, and we have seen that on the echo, and if youry and put the Hegars back in again, it is very difficult to do that.s far as does that affect durability, I don’t think so, because the

tress is taken by the polyester graft and that is not going totretch over time.

R IKONOMIDIS: Next question. You used a 30 mm graft in allatients?

R SVENSSON: The majority of patients we use a 30. If it is aarge male with a body surface area over 2.5, we will use a 32 mm.f it is a young, small female with a body surface area of less than.5, I might go to a 28. But I think, as you basically imply andointed out, it is critical to have large area of a root for the leaflets

o be accommodated so they don’t bang up against the polyester,nd I think that was one of the reasons there have been someeports of early failures with the David reimplantation operation.

R IKONOMIDIS: The next question is, what is your approacho the neo-sinotubular junction? How do you remodel that withhis repair?

R SVENSSON: The way we have done that is that we space theutures approximately 4 to 5 mm apart at the commissures.

hat that does in effect is plicate once again at the neo-inotubular junction area and that reduces the circumference.ow, I think the Stanford data, particularly Craig Miller’s work,

learly shows that that probably does result in better flowynamics across the aortic valve and closure of the leaflets.hether it is going to affect long-term durability, I am not sure.

ut if we are trying to replicate as much as possible the normalnatomy, I think it is useful to do.

R IKONOMIDIS: The last thing I would like to pose to you isust a basic patient question. I noticed that 40% of the patients inour series had Marfan’s syndrome. One thing that I havencountered operating on these patients is that when you take aoung patient with Marfan’s syndrome and a 6 to 6.5 cm aorticoot to surgery, and reduce the aortic root down to 3 cm as youould with a 30 or 32 mm graft, upon suturing the valve inside,

ou notice that now the valve leaflets are redundant and willend to prolapse past the anulus, which we know is a predictorf failure. What is your approach to those patients? How do youestore the coaptation of these leaflets and bring the bellies ofhe valves up to the level of the anulus?

R SVENSSON: There are two things that are important. One is

s adequate to take some of the body out of the leaflets. That usuallyeans a fairly high implantation. And the second is once I have

mplanted the valve, I will look at all three leaflets, see if they areymmetrical, and look at my area of apposition. If I then still haveome prolapse, my first choice is to do what is a modified Trusslertitch, namely a figure-of-eight suture on the leaflet that is prolaps-ng. If that still is not sufficient and if I have good strength in theeaflets, then I will do a mid-leaflet placation. With that combina-ion you can usually get a valve to function very well.

I have a couple of times used Tirone David’s technique of aunning suture along the edge of the leaflet, however, and inarticular, as you mentioned, in Marfan patients, the leaflets areery thin walled and fragile, and thus in those patients I amomewhat reluctant to do a running suture because I think longerm that may pose a problem.

R JASON SPERLING (Ridgewood, NJ): I very much enjoyedhe presentation. Thanks, Dr Svensson. Two questions. Numberne, I didn’t see how you were sizing the aortic anulus, the actualentriculo-aortic junction. I find that somewhat difficult to do byransesophageal echo. We seem to have good accuracy usingCG-gated CT [electrocardiographic-gated computed tomo-raphic] angiograms, and that may influence selection of theppropriate size for the Hegar. I noticed in your presentation withreater than 100 patients that most of the Hegars that you usedere 19, 21, 23. In my own much smaller experience, our native

ortic anulus sizes oftentimes are 26, 28, 30, or higher. So, numberne, I wanted to know if you are routinely undersizing the anulusnd, number two, if you have any rules on how much you canndersize an anulus to your Hegar if you were attempting to ac-omplish a reduction anuloplasty to correct some aortic insufficiency?

R SVENSSON: What I didn’t really go into much, other thanentioning our Z scores, is that we downsized by one standard

eviation from the normalized diameter of the anulus.

R SPERLING: So it is always undersized?

R SVENSSON: Basically it is undersized by one standardeviation. So, for example, a 2.0 m2 male I will use a size 21egar. If it is a 2.5 m2, I will use a 23 mm. Now, for females I willownsize by one more size. My reasoning being that if you make

t as small as possible as can be tolerated, and that is why weooked at the gradient pressures, you are probably going to haveetter long-term durability in the sense that you have noweduced the anulus to a smaller size that is reasonable withoutausing severe gradients.

So, although we measured the anuli on echo, the actualelection of Hegars size is entirely dependent on the patient’sody surface area.

R SPERLING: So is there sometimes quite a large discrepancyetween your calculated target for them based on their bodyurface area and their actual anulus size?

R SVENSSON: Yes, and that is why the plication can be quitearked in these patients, particularly in the Marfan patients. Some

f those patients will have quite a big left ventricular outflow tract,

ut you downsize it on the basis of the body surface area.