RD

IAC

Dynamic Normal Aortic Root Diameters:Implications for Aortic Root ReconstructionDan Zhu, MD, and Qiang Zhao, MD

Department of Cardiac Surgery of Zhongshan Hospital, The Shanghai Institute of Cardiovascular Disease, Fudan University; andDivision of Cardiac Surgery of Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China

ahat00

hir

AD

ULT

CA

Background. The main aim of this study is to determinethe normal diameter and the relationship of aortic rootcomponents in healthy adults, and to provide a morpho-logic foundation for future clinical applications.

Methods. Echocardiography was performed in 314 nor-mal subjects who were divided into five groups accordingto age. Dynamic aortic root diameters were measured andnormalized to body surface area. Averages of these dy-namic diameters were calculated for each age group and bygender, and differences between them were tested. Corre-lation coefficients were also determined between the dy-namic diameters and age, body surface area, weight, andheight. Aortic root diameters were also tested and com-pared between the end-diastole and the mid-systole.

Results. Normalized diameters for the dynamic aortic

root varied among the age groups. There were apparent

(y

Ruijin Hospital, Shanghai Jiaotong University School of Medicine, 197Ruijin Rd, Shanghai 200025, China; e-mail: zq11607@rjh.com.cn.

© 2011 by The Society of Thoracic SurgeonsPublished by Elsevier Inc

relationships between the dynamic diameters and age,body surface area, weight, and height (p < 0.01). Thenormalized diameters were similar between both gen-ders within each age group (p > 0.05). Each part of theortic root expanded and contracted proportionally andarmoniously during the cardiac cycle. The ratio of theortic valve annulus to the sinus of Valsalva, the sinus-ube joint, and the proximal ascending aorta were 0.70,.85, and 0.78 at the end-diastole, respectively, and 0.71,.85 and 0.78 at the mid-systole, respectively.Conclusions. The dynamic diameters of aortic roots of

ealthy adults were augmented with an increase accord-ng to age, body surface area, weight, and height. Theesults are of applicable value to aortic valve repair.

(Ann Thorac Surg 2011;91:485–90)

© 2011 by The Society of Thoracic Surgeons

The aortic root, first described by Leonardo da Vinciand defined as the junction site between the left

ventricle and the ascending aorta, is composed of fourcomponents, the aortic valve annulus (AVA), cusps, thesinuses of Valsalva (SV) and the sinus-tube junction(STJ). Diseases that result in either changing the diame-ter of the aortic root, or diminishing the coaptation of theaortic cusps, usually cause aortic valve regurgitation.Aortic valve reconstructive surgery has been of limiteduse in correcting this pathology; rather, a more success-ful, effective, and universally accepted procedure is atrio-ventricular valvuloplasty. This is due in part to both alimited understanding of aortic root anatomy and itsextreme complexity. The repair procedure of a primaryleaflet disease showed a limited satisfied midterm result[1]. Recently, attention has been directed toward thereduction of aortic root diameters as an alternative andeffective means to rectify aortic valve regurgitation [2, 3].

Values of normal aortic root dimensions are importantfor aortic valve repair. By having knowledge of normalaortic root values before surgery, the surgeon can reliablydeduce the main pathology causing aortic valve regurgi-tation, classify the valve lesion, and choose the optimalprocedure [4, 5]. Our previous study [6] demonstrated thedimensions and the interrelationship of normal AVA and

Accepted for publication Oct 21, 2010.

Address correspondence to Dr Zhao, Division of Cardiac Surgery of

STJ in the mid-systole. However, the leaflets of the aorticvalve coapt together at the end-diastole. Thus, characteriz-ing the dynamic dimensions may be more relevant. Thepresent study was undertaken to quantitatively assess thedynamic anatomy of the healthy adult aortic root by two-dimensional echocardiography (2DE), and to provide amorphologic foundation for further clinical applications.

Material and Methods

Study PopulationThree hundred and fourteen healthy adults, chosen fromroutine health examination, were studied. There were133 males and 181 females. The mean age was 37.16 �13.46 years (range, 17 to 60 years). Patients with diseasesthat altered blood flow patterns across the aortic valve orover the anatomy of the aortic root (eg, heart valve andcongenital heart diseases, uncontrolled hypertension,Marfan syndrome) were excluded as participants. Thisstudy was approved by the local Ethics Committee andindividual consent for the study was waived. Enrolledsubjects were divided into five groups according to age:group 1, 17 to 20 years (n � 59); group 2, 21 to 30 yearsn � 58); group 3, 31 to 40 years (n � 59); group 4, 41 to 50ears (n � 71); and group 5, 51 to 60 years (n � 67).

EchocardiographyEchocardiographic examinations were performed with aPhilips SONOS 7500 imaging system (Philips Medical

Systems, Andover, MA) and S4 phase-array transducers

0003-4975/$36.00doi:10.1016/j.athoracsur.2010.10.058

486 ZHU AND ZHAO Ann Thorac SurgDYNAMIC NORMAL AORTIC ROOT DIMENSIONS 2011;91:485–90

AD

ULT

CA

RD

IAC

(Philips). Standard views and techniques were used ac-cording to the guideline published by the AmericanSociety of Echocardiography. All subjects underwent2DE performed by an experienced cardiologist who wasblinded to the nature of this study. The dimensions of theaortic root and ascending aorta were evaluated in detailusing 2DE. Measurements were made by the leading-edge technique in the parasternal long-axis view for thefollowing four diameters [7]: (1) the diameter of AVA; (2)the maximal the diameter of SV; (3) the diameter of STJ;and (4) the maximal diameter of the proximal ascendingaorta (PAA). We defined the anatomic ventricle-aorta junc-tion, not the actual semilunar attachments, as the echocar-diographic AVA, which was generally accepted in 2DE. Thedynamic diameters were measured in two points in thecardiac cycle, the end-diastole (defined as the beginning ofthe QRS) and the mid-systole (when the valve was mostwide open, coinciding with the top of the T wave).

Measurements were taken on up to 4 separate cyclesand the numbers of the parameters were averaged. Thefour dynamic diameters were normalized to body surfacearea (BSA). Averages of these diameters were calculatedfor each age group and for each gender. Differences be-tween the normalized diameters for each age group and forboth genders were then tested. The Pearson product-moment correlation coefficient (r) was used to determineany positive or negative correlations of the four diametersin each point and of the variables between the two points.The r were also determined between the above dynamicdiameters and age, BSA, weight, and height. The relation-ships of aortic root diameters between the end-diastole andthe mid-systole were also tested.

Statistical AnalysisContinuous data were expressed as mean � standarddeviation (SD). The Student t test was used to analyzedata conforming to normal distribution, and the ranksum test was applied to analyze data that did not conformto normal distribution. The Pearson product-moment rwas calculated using the SAS procedure, PROC CORR. Ap value less than 0.05 was considered statistically signif-icant. The SAS version 9.0 (SAS Institute Inc, Cary, NC)was used to perform all statistical analyses.

Results

Normalized diameters of the dynamic aortic root varied

Abbreviations and Acronyms

2DE � two-dimensional echocardiographyAVA � aortic valve annulusBSA � body surface areaPAA � proximal ascending aortar � correlation coefficientSD � standard deviationSTJ � sinus-tube jointSV � sinus of Valsalva

among the age groups (Table 1). The average aortic root

diameters for the whole group were the following: AVA20.35 � 8.67 mm; SV 28.23 � 3.57 mm; STJ 23.45 � 3.14mm; and PAA 25.23 � 3.52 mm at the end-diastole,respectively; and AVA 20.91 � 2.29 mm; SV 29.36 � 3.51mm; STJ 24.64 � 2.93 mm; and PAA 26.65 � 3.32 mm atthe mid-systole, respectively. The dynamic diameters ofaortic root of healthy adults were augmented with anincrease according to age, BSA, weight, and height (Table2). The dynamic diameters were similar in both men andwomen when indexed to BSA (p � 0.05) (Table 3).

Between the end-diastole and the mid-systole, the r ofAVA diameter was 0.75 (p � 0.0001) and the absoluteproportion was 0.95; the r of SV diameter was 0.85 (p �0.0001) and the absolute proportion was 0.96; the r ofSTJ diameter was 0.86 (p � 0.0001) and the absoluteproportion was 0.95; the r of PAA diameter was 0.89(p � 0.0001) and the absolute proportion was 0.95.Diameter rates of AVA divided by SV, STJ, and PAAwere 0.70, 0.85, and 0.78; and 0.71, 0.85, and 0.78 at themid-systole, respectively. The results showed that eachpart of aortic root expanded and contracted propor-tionally and harmoniously during the cardiac cycle.The diameter of the aortic valve annulus was shorter(about 15%) than that of the STJ. The AVA was thenarrowest part of the aortic root in both points.

Comment

The thorough evaluation and measurement of the aorticroot components are very crucial to the aortic root recon-struction. The preoperative echocardiography study pro-vides more precise dynamic information of the aortic rootthan direct measurement during the procedure. Thediameters of the aortic valve annulus and the sino-tubular junction are the two most important parametersto determine the size of the graft for aortic root recon-struction, and consequently for the aortic valve compe-tence. On the other hand, the aortic valve leaflets aremorphologically normal and are not prolapsed in thesepatients. Therefore, we only measured the diameter ofthe aorta from the annulus to the ascending aorta, ratherthan the perimeter and the area of the aortic valve, for thereason of simple and effective application. The end-diastole and mid-systole are two points in the cardiaccycle when the diameters of the aorta are minimal andmaximal, respectively.

There were three key findings of this study, which areof important implications to understand the pathogene-sis and surgically repair the aortic valve regurgitation.First, dimensions of the aortic root in normal adults atfour levels were obtained. Second, the interrelationshipof AVA, SV, STJ, and PAA dimensions was illustrated.Third, the rule of dynamic dimensions of the four aorticroot components was demonstrated.

Dimensions of the Aortic Root in Normal AdultsAs an integral anatomic and functional unit, the aorticroot is a complicated structure comprised of four frac-tions: AVA, cusps, SV, and STJ. Each part plays an

important role in maintaining proper function of the

TJ �

487Ann Thorac Surg ZHU AND ZHAO2011;91:485–90 DYNAMIC NORMAL AORTIC ROOT DIMENSIONS

AD

ULT

CA

RD

IAC

aortic root and effective leaflet coaptation. Furthermore,each part directly or indirectly impacts the other partswhen aortic root disease developed. Appropriate ana-tomic proportion of these four parts, therefore, is vital tomaintain proper function. Any change in the above fourfractions may lead to damage of the leaflet coaptationand (or) lead to aortic valve insufficiency.

The present study shows the dynamic dimensions ofthe aortic root of healthy adults augmented with theincreasing of age, BSA, weight, and height. The data aremore exact than that of Svensson [8] confined to BSA.

Table 1. Normalized Diameters of Dynamic Aortic Root (mm

Group Number Point Variable

1 59 End-diastole AVASVSTJPAA

End-systole AVASVSTJPAA

2 58 End-diastole AVASVSTJPAA

End-systole AVASVSTJPAA

3 59 End-diastole AVASVSTJPAA

End-systole AVASVSTJPAA

4 71 End-diastole AVASVSTJPAA

End-systole AVASVSTJPAA

5 67 End-diastole AVASVSTJPAA

End-systole AVASVSTJPAA

AVA � aortic valve annulus; PAA � proximal ascending aorta; S

The relation between the diameters and age may be due

to the degeneration of the aorta wall. The dynamicdiameters were similar in both men and women whenindexed to BSA. Using these values as references, thelesion classification, surgical strategy, and graft size canbe easily decided before surgery. For example, a patientwas diagnosed with aortic valve regurgitation secondaryto an aortic root aneurysm. Comparing the aortic rootdiameter with the given reference values, the surgeoncan deduce that the main pathogenesis of the aorticregurgitation is the whole aortic root dilatation from AVAto PAA. According to the lesion classification, the sur-

ean Median SD Maximum Minimum

0.88 10.72 1.11 12.98 8.694.75 14.68 1.61 18.27 11.092.05 11.91 1.26 14.82 9.642.72 12.50 1.47 17.00 10.191.53 11.59 1.05 14.33 9.565.58 15.68 1.56 19.78 13.002.85 12.69 1.27 15.71 10.613.76 13.47 1.44 17.66 11.761.23 11.15 1.02 14.12 8.855.60 15.74 1.62 19.94 11.722.75 12.77 1.33 15.72 9.873.54 13.44 1.52 17.13 10.521.86 11.94 1.11 15.24 8.986.39 16.48 1.73 20.51 12.923.64 13.54 1.25 16.14 10.224.45 14.32 1.53 17.63 10.521.31 11.43 1.19 13.97 8.546.05 16.09 1.41 19.72 12.783.47 13.54 1.55 16.32 9.134.49 14.45 1.63 17.60 10.311.74 11.71 1.02 14.55 9.866.51 16.50 1.61 20.98 10.664.45 14.59 1.64 18.81 10.665.23 15.26 1.73 18.76 10.011.13 11.06 1.30 14.12 7.966.42 16.57 2.05 20.89 9.613.77 13.69 1.64 17.79 8.954.94 14.96 1.76 19.27 9.281.88 12.04 1.26 15.94 8.297.11 17.43 2.11 21.10 9.954.21 14.04 1.56 17.86 9.615.68 15.77 1.73 19.27 9.611.77 12.02 1.05 13.65 9.267.07 17.07 1.97 21.25 11.354.29 14.60 1.67 17.54 9.735.65 15.52 2.09 21.08 10.542.28 12.20 1.16 17.55 10.277.56 17.44 1.87 22.12 12.574.68 15.07 1.64 17.63 10.546.33 16.25 2.01 21.68 10.54

sinus-tube joint; SV � sinus of Valsalva.

/m2)

M

1111111111111111111111111111111111111111

geon can optimize the procedure accordingly to restore

E

roxim

488 ZHU AND ZHAO Ann Thorac SurgDYNAMIC NORMAL AORTIC ROOT DIMENSIONS 2011;91:485–90

AD

ULT

CA

RD

IAC

the diseased aortic root to normal. In this case, the patientneeds a modified David-V procedure [9]. The size of alarge Dacron graft is chosen according to the normal sizeof the SV, calculated according to age and BSA. This isnot the same as the methods generally used to select graftsize, which is usually according to intraoperative mea-surements [10].The graft is then implanted by using 12 to14 braided polyester 2-0 mattress sutures placed in theleft ventricular outflow tract. A Hegar dilator is chosenaccording to the normal AVA size. Next, the Hegardilator is placed in the left ventricular outflow throughthe aortic valve and the sutures are then tied downaround the Hegar dilator. This technique has the effect ofcrimping the proximal polyester graft down to the appro-

Table 2. The r Between the Dynamic Diameters and Age, BSA

Point Variable Age

End-diastole AVAr 0.2137p 0.0001

SVr 0.4071p �0.0001

STJr 0.4686p �0.0001

PAAr 0.5179p �0.0001

nd-systole AVAr 0.2098p 0.0002

SVr 0.3557p �0.0001

STJr 0.3772p �0.0001

PAAr 0.4749p �0.0001

AVA � aortic valve annulus; BSA � body surface area; PAA � pjoint; SV � sinus of Valsalva.

Table 3. Normalized Diameters of Dynamic Aortic Root in Bo

Point Variable Male Mean SD (n

End-diastole AVA 11.15 1SV 15.83 1STJ 13.21 1PAA 14.27 1

End-systole AVA 11.80 1SV 16.59 1STJ 13.91 1PAA 15.01 1

AVA � aortic valve annulus; PAA � proximal ascending aorta; STJ �

priate AVA size. The valve is reimplanted inside the largegraft by using 4-0 polypropylene sutures in a runningfashion. Another smaller graft, sized according to thenormal STJ, connects the distal aorta and the previousartificial aortic root above the commissures to create aneo-STJ. Theoretically, creation of pseudosinuses pro-duces more natural leaflet motion, minimizes systoliccontact of the valve cusps to the Dacron graft, andreduces diastolic closing cusp stresses, all of which mayenhance long-term valve durability [11]. With themethod described above, the aortic root abnormality canbe easily corrected and pseudosinuses are created. Wehave applied these principles in almost 30 cases withsatisfactory results.

eight, and Height

BSA Height Weight

0.5484 0.4020 0.5376�0.0001 �.0001 �0.0001

0.4560 0.3424 0.4446�0.0001 �.0001 �0.0001

0.4440 0.2437 0.4607�0.0001 �.0001 �0.0001

0.3801 0.1615 0.4090�0.0001 0.0041 �0.0001

0.5688 0.4299 0.5537�0.0001 �.0001 �0.0001

0.4591 0.3256 0.4535�0.0001 �.0001 �0.0001

0.4549 0.2613 0.4685�0.0001 �.0001 �0.0001

0.3736 0.1744 0.3972�0.0001 0.0019 �0.0001

al ascending aorta; r � correlation coefficient; STJ � sinus-tube

enders (mm/m2)

33) Female Mean SD (n � 181) p Value

11.36 1.14 0.111416.17 1.93 0.115613.39 1.69 0.121314.38 2.07 0.126311.92 1.01 0.094116.74 1.84 0.108814.06 1.56 0.110215.24 1.89 0.0873

, W

th G

� 1

.21

.90

.61

.75

.30

.97

.58

.74

sinus-tube joint; SV � sinus of Valsalva.

489Ann Thorac Surg ZHU AND ZHAO2011;91:485–90 DYNAMIC NORMAL AORTIC ROOT DIMENSIONS

AD

ULT

CA

RD

IAC

Interrelationships of Normal Adult Aortic RootDimensionsIt is popularly accepted that the diameter of the AVA was15% shorter than that of the STJ, contrary to formerlyaccepted values derived from experiments with cadavers.The present study showed that the diameter rate of AVAdivided by SV, STJ, and PAA were 0.70, 0.85, and 0.78,respectively, in the end-diastole, while 0.71, 0.85, and 0.78 inthe mid-systole. Therefore, we should have no great con-cern about the periodical change in the cardiac cycle. Theserelations may achieve significant relevance in the future.Although a deformed and (or) enlarged aortic root cannotalways be retailored to achieve normal size, it may still beimportant to restore normal relationships and the surgeoncan easily calculate the suitable revised size according tothe normal part of the diseased root. For example, aorticvalve regurgitation, which is mostly moderate and merelycaused by STJ distension as a result of an ascending aorticaneurysm, may be corrected with an ascending aorta re-placement. The graft size can be selected on the base ofnormal AVA diameters and the proportion to the STJ.

Changes of Normal Adult Aortic Root Dimensions inthe Cardiac CycleIn the past, studies have focused on the opening andclosing characteristics of the aortic valve, the anatomy of theaortic root [12], and leaflet stress and the aortic sinus.However, few investigations have been conducted on themotion of the aortic root during the cardiac cycle in hu-mans. Kazui and associates [13] reported that by multide-tector computed tomography, the lengths of the AVA andSV were unchanged during the cardiac cycle and the lengthof the STJ in systole was significantly greater than that indiastole. Dagum and associates [14] reported that the changein the shape and torsion force of AVA and SV caused amorphologic change in the aortic root. With a sheep model,using three-dimensional sonomicrometry, Lansac and associ-ates [15] showed that the aortic root expanded during systole.The present study shows the aortic root expanded and con-tracted proportionally and harmoniously. The differences be-tween these studies may be due to many aspects, includingbut not limited to the number of objects, different methods,and different evaluation points. The SV showed a fairly lesschange during the cardiac cycle than the other parts. Weregarded the phenomenon as the result of the blood pressurechange in the aortic root during the cycle, and in the end-diastole the vortex in SV forbidding it from contracting asmuch as other parts. Our study found that there was a 5%difference in these diameters between the end-diastole andthe mid-systole. The surgeon may neglect the variance duringthe procedure of aortic root reconstruction. An unstretchablegraft can meet the physiologic need. Therefore, the sinus of DePaulis Valsalva graft may have no superiority to pseudosi-nuses created by the David-V procedure.

LimitationsThis study was focused only on the diameters of the aorticroot for the root reconstruction. If the aortic valve is pro-

lapsed and valve leaflet repair is mandatory, the morphol-

ogy and the measurement of the valve leaflet including theheight and free edge length of the cusps, the coaptationlength, and the height of commissures should be evaluated.On the other hand, the difference of the maximum and theminimum aortic root diameters needs to be determined.The study chose two points only rather than the wholesequential period of the cardiac cycle.

ConclusionThe dynamic diameters of aortic roots of healthy adults hadrelationships with age, BSA, weight, and height. There wasno difference in the dynamic diameters between genderswhen indexed to BSA. Each part of aortic root expandedand contracted proportionally and harmoniously duringthe cardiac cycle. The interrelationships of the dynamicdimensions are mostly constant. The result has the vastvalue of optimizing the procedure of aortic root reconstruc-tion. Moreover, findings from this study may help to im-prove techniques for restoring normal functional anatomicrelations in pathologically altered aortic roots.

References

1. Alsoufi B, Karamlou T, Bradley T, et al. Short and midtermresults of aortic valve cusp extension in the treatment ofchildren with congenital aortic valve disease. Ann ThoracSurg 2006;82:1292–300.

2. Hahm SY, Choo SJ, Lee JW, et al. Novel technique of aorticvalvuloplasty. Eur J Cardiothorac Surg 2006;29:530–6.

3. Lansac E, Di Centa I, Bonnet N, et al. Aortic prosthetic ringannuloplasty:a useful adjunct to a standardized aortic valve-sparing procedure? Eur J Cardiothorac Surg 2006;29:537–44.

4. Boodhwani M, de Kerchove L, Glineur D, et al. Repair-oriented classification of aortic insufficiency: Impact on sur-gical techniques and clinical outcomes. J Thorac CardiovascSurg 2009;137:286–94.

5. El Khoury G, Boodhwani M. Aortic valve repair. OperativeTechniques in Thoracic and Cardiovascular Surgery 2009;14:266–80.

6. Zhu D, Zhao Q. Aortic valve annulus and sinus-tube jointdiameters in normal adults of Chinese Han ethnic group.Chin Med J 2008;121:1093–5.

7. Roman MJ, Devereux RB, Kramer-Fox R, et al. Two-dimensional echocardiographic aortic root dimensions innormal children and adults. Am J Cardiol 1989;64:507–12.

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

9. 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.

10. 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.

11. Leyh RG, Schmidtke C, Sievers HH, et al. Opening andclosing characteristics of the aortic valve after different typesof valve-preserving surgery. Circulation 1999;100:2153–60.

12. 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.

13. Kazui T, Izumoto H, Yoshioka K, Kawazoe K. Dynamicmorphologic changes in the normal aortic annulus duringsystole and diastole. J Heart Valve Dis 2006;15:617–21.

14. Dagum P, Green GR, Nistal FJ, et al. Deformational dynam-ics of the aortic root: modes and physiologic determinants.Circulation 1999;100(19 Suppl):II54–62.

15. Lansac E, Lim HS, Shomura Y, et al. A four-dimensional

study of the aortic root dynamics. Eur J Cardiothorac Surg2002;22:497–503.