ISSN: 1524-4539 Copyright © 2010 American Heart Association. All rights reserved. Print ISSN: 0009-7322. Online

72514Circulation is published by the American Heart Association. 7272 Greenville Avenue, Dallas, TX

DOI: 10.1161/CIRCULATIONAHA.110.941310 published online Sep 27, 2010; Circulation

Mahoney and Maurice Enriquez-Sarano Yan Topilsky, Christophe Tribouilloy, Hector I. Michelena, Sorin Pislaru, Douglas W.

Echocardiographic Assessment of Respiratory DependencePathophysiology of Tricuspid Regurgitation. Quantitative Doppler

http://circ.ahajournals.orglocated on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://www.lww.com/reprintsReprints: Information about reprints can be found online at  

journalpermissions@lww.com410-528-8550. E-mail:

Fax:Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters 

http://circ.ahajournals.org/subscriptions/Subscriptions: Information about subscribing to Circulation is online at

at Mayo Clinic Libraries on September 29, 2010 circ.ahajournals.orgDownloaded from

Valvular Heart Disease

Pathophysiology of Tricuspid RegurgitationQuantitative Doppler Echocardiographic Assessment of

Respiratory Dependence

Yan Topilsky, MD; Christophe Tribouilloy, MD; Hector I. Michelena, MD; Sorin Pislaru, MD;Douglas W. Mahoney, MS; Maurice Enriquez-Sarano, MD

Background—Respiratory dependence of tricuspid regurgitation (TR), a long-held concept suggested by murmur variation,remains unproven and of unclear mechanisms.

Methods and Results—In 41 patients with mild or greater TR (median age, 67 years), we performed triple Dopplerechocardiographic quantification (TR severity, right ventricular, and right atrial quantification) with simultaneousrespirometer recording of respiratory phases. Expiration to inspiration changes (median) affected TR peak velocity (�40cm/s; 25th to 75th percentile, �60 to �30 cm/s), duration (�12 milliseconds; 25th to 75th percentile, �45 to 2milliseconds), and time-velocity integral (�17 cm; 25th to 75th percentile, �23.4 to �10 cm; all P�0.001), consistentwith decreased TR driving force. Nevertheless, inspiratory TR augmentation was demonstrated by increased effectiveregurgitant orifice (0.21 cm2; 25th to 75th percentile, 0.09 to 0.34 cm2) and volume (18 mL per beat; 25th to 75thpercentile, 10 to 25 mL per beat; all P�0.001) infrequently detected clinically (2 of 41, 5%). As a result of reduced TRdriving force, regurgitant volume increased less than effective regurgitant orifice (120% [25th to 75th percentile, 78.6%to 169%] versus 169% [ 25th to 75th percentile, 12.9% to 226.1%]; P�0.001). During inspiration, right ventricular areaincreased (diastolic, 27.8 [25th to 75th percentile, 22.6 to 36.3] versus 26.5 [21.1 to 31.9]; P�0.0001) with wideningof right ventricular shape (length-to-width ratio, 1.6 [ 25th to 75th percentile, 1.37 to 1.95] versus 1.7 [1.46 to 2.1];P�0.0001), increased systolic annular diameter (P�0.003), valve tenting height (P�0.0001) and area (P�0.0001), andreduced valvular-to-annular ratio (P�0.006). Effective regurgitant orifice during inspiration was independentlydetermined by inspiratory valvular-to-annular ratio (P�0.026) and inspiratory change in right ventricular length-to-width ratio (P�0.008) and valve tenting area (P�0.015).

Conclusions—TR is dynamic with almost universal respiratory changes of large magnitude and complex pathophysiology.During inspiration, a large increase in effective regurgitant orifice causes, despite a decline in regurgitant gradient, anotable increase in regurgitant volume. Effective regurgitant orifice changes are independently linked to inspiratoryannular enlargement (decreased valvular coverage) and to inspiratory right ventricular shape widening with increasedvalvular tenting. These novel physiological insights into TR respiratory dependence underscore right-side heartplasticity and are important for clinical TR severity evaluation. (Circulation. 2010;122:1505-1513.)

Key Words: echocardiography � regurgitation � surgery � valves function

Tricuspid regurgitation (TR) is a common condition, but itis poorly understood with regard to physiology and

outcome.1 The sketchy pathophysiological knowledge of TR2

emphasized in American Heart Association clinical guide-lines3 and in comprehensive reviews4 probably contributes tocurrent uncertainties about TR management and outcome.5

Clinical Perspective on p 1513

Among TR pathophysiological characteristics, a widelyaccepted fact is the old description of clinical observation ofrespiratory variation of the TR systolic murmur.6 However,mechanistically, such variations have received little explana-

tion, and potential clinical implications are unknown. Theseuncertainties stem in great part from the challenging assess-ment of TR severity, well known for invasive methods,7 butalso remaining often imprecise for Doppler echocardiogra-phy, as underscored by the current guidelines.8 Quantitativemethods of TR assessment have been validated9–12 and offerthe opportunity to obtain physiological insights,12 but thesemethods remain inconsistently used.

Another source of confusion is the variety of potentialtheories on respiratory TR variations that are based on scantfacts. TR systolic murmur respiratory variation has beenvariably assumed to arise from preload-determined right

Received January 26, 2010; accepted July 14, 2010.From the Division of Cardiovascular Diseases and Internal Medicine, Mayo College of Medicine, Mayo Clinic, Rochester, Minn.Correspondence to Maurice Enriquez-Sarano, MD, Division of Cardiovascular Diseases and Internal Medicine, Mayo Clinic, 200 First St SW,

Rochester, MN 55905. E-mail Sarano.maurice@mayo.edu© 2010 American Heart Association, Inc.

Circulation is available at http://circ.ahajournals.org DOI: 10.1161/CIRCULATIONAHA.110.941310

1505 at Mayo Clinic Libraries on September 29, 2010 circ.ahajournals.orgDownloaded from

ventricular (RV) contractility changes causing changes in TRturbulence13,14 or from possible changes in TR velocity orventriculoatrial pressure gradient, counterintuitive to ob-served gradient respiratory changes.15–17 An alternative hy-pothesis of TR murmur changes based on respiratory TRseverity changes is doubted because patients with tricuspidvalvulectomy, who obviously have very severe TR, are notreported to have respiratory TR murmur variation.18 Hence,TR physiological characteristics and the presence and mag-nitude of respiratory TR changes remain uncertain. Thesegaps in knowledge affect not only TR physiological compre-hension but also clinical evaluation and assessment of itsimpact on outcome and management.3

Thus, we performed a prospective and comprehensivestudy of TR physiology with the hypothesis that TR displaysphysiological variations during respiration that are due to achange in TR effective regurgitant orifice (ERO). Conse-quently, we conducted a triple quantitative study involvingTR quantification, quantification of RV and right atrial (RA)characteristics, and quantification of tricuspid valvular defor-mation with combined respirometer to assess changes duringthe respiratory cycle to assess pathophysiological changes inTR with respiration.

MethodsPatient PopulationWe prospectively and consecutively enrolled 41 patients with vari-able degrees (mild to severe) of TR in this study of TR pathophys-iology. The inclusion criteria for the present study were (1) thepresence of TR by color-flow Doppler imaging of at least milddegree and holosystolic, (2) comprehensive quantitative Dopplerechocardiography (as detailed in the section on echocardiographicmethods), (3) simultaneous recording of all Doppler echocardio-graphic imaging with respirometer tracing to determine the respira-tory phase of each measurement, and (4) regular cardiac rhythm.Exclusion criteria were absent or trivial TR, poor-quality flowconvergence or flow convergence that was impossible to imageproximal to the TR, and irregular rhythm caused by atrial fibrillationor flutter. The study was powered to detect at least a 30% changewith respiration in one of the variables measuring TR with at least80% power (P�0.05). Clinically, patients had measurement of heartrate and blood pressure and standard clinical assessment by seniorcardiologists with at least 10 years of postfellowship experience.Specifically for TR, we noted by careful detailed cardiac examina-tion if there was a murmur consistent with TR (holosystolic andmaximally projected at the xiphoid or lower left sternum), if themurmur increased with inspiration in the recumbent and sittingpositions, and if there were signs of right-side heart failure (jugularvenous distension and liver enlargement with hepatojugular reflux).

Doppler EchocardiographyAll patients had a complete 2-dimensional and Doppler echocardio-graphic study using multiple windows during the same examination.All echocardiographic examinations were performed with a respi-rometer for precise timing of respiratory cycles. Patients wereinstructed to breathe normally. All measurements of TR weredetermined in midsystole during inspiration and expiration.

The mean of systolic pulmonary pressure was estimated using themean of peak systolic TR velocity in inspiration and expiration.19

RA pressure was estimated by the inferior vena cava diameter and itsresponse to inspiration as previously described.20

From 2-dimensional echocardiographic 4-chamber views allowingcomplete imaging of the cavities examined, the following morpho-logical measurements were performed with inspiration and expira-tion: (1) RV end-systolic and end-diastolic areas, RV length, and

midventricular RV width with calculation of RV area shorteningfraction and RV length-to-width ratio; (2) tricuspid annulus systolicand diastolic diameters with shortening fraction calculation; (3) RAend-systolic area and length measured by calculation of RA volumewith the area-length method; and (4) tricuspid valve tenting area(between the annular line and leaflet contour) and tenting height. Thelength of tricuspid leaflets (sum of the septal and anterior leaflet) wasmeasured, and the ratio to systolic annulus diameter was calculated.

Assessment of TR degree and its changes between inspiration andexpiration used color-flow imaging and quantitative measures.8 TRjet area was measured by color Doppler with a sector allowingvisualization of the entire RA. The images were reviewed frame byframe, and planimetry of the maximal area of the regurgitant jet wasperformed. The ratio of the maximal jet area to the RA areameasured on the same image was also calculated. The width of thevena contracta was measured with inspiration and expiration.9,21 TRquantification relied on the proximal flow convergence (proximalisovelocity surface area) method as validated previously9–12 and wasperformed with all measurements done during inspiration andexpiration and using multiple (at least 3) measurements of eachvariable during each respiratory phase. Practically, color Dopplerimages of the proximal flow convergence of TR were obtained fromapical or para-apical views after maximal depth reduction and zoomof the region of interest. The transducer was positioned to minimizethe angle between the centerline of regurgitant flow and theultrasonic beam. The color-flow velocity scale was maximized, andthe baseline was shifted downward until the flow convergence regionwas clearly visualized. The aliasing velocities selected ranged from21 to 43 cm/s. Radial distance between the first aliasing velocity(red/blue interface) and the center of the tricuspid orifice wasmeasured along the centerline of the flow convergence region inmidsystole during inspiration and expiration. Corrections for theangle of leaflets and for the ratio of aliasing velocity to peak TRvelocity were applied as previously validated,10–12,21 allowing cal-culation of regurgitant flow. The ERO area was then calculated asthe ratio of regurgitant flow to the peak velocity of the TR jet (bycontinuous wave-Doppler), and the regurgitant volume (RVol) wascalculated as the product of ERO by the regurgitant time-velocityintegral of the TR.10 An example of the proximal isovelocity surfacearea measurement during inspiration and expiration is presented inFigure 1.

Statistical AnalysisDescriptive results were expressed as median and 25th to 75thpercentile for continuous variables and as percentages for categoricalvariables. The distribution of continuous variables is depicted infigures through box-and-whisker plots. Comparison between groupsused standard Wilcoxon rank-sums test for continuous variables andthe �2 test for categorical variables. The changes for each variablebetween inspiration and expiration were analyzed by use of thesigned Wilcoxon rank-sums test. The differences between inspira-tory and expiratory measurements were presented as absolute valuesand as percent change from expiration to inspiration. Comparison ofthe magnitude of inspiratory changes between subgroups relied onthe Wilcoxon rank-sums test. The subgroups examined were thosewith mild versus moderate to severe TR based on mean (ofinspiration and expiration) ERO �20 mm2,6 those with versuswithout pulmonary hypertension (based on mean systolic pulmonarypressure �50 mm Hg), and those with organic versus functionaltricuspid valve disease (based on the presence or absence of intrinsictricuspid valve lesions). Multiple linear regression was used to definecorrelates of inspiratory ERO (adjusting for expiratory ERO). Sta-tistical significance was set at P�0.05.

ResultsThe study included 41 patients (age, 67 years [25th to 75thpercentile, 57 to 74 years]; 13 men and 28 women) with TRfulfilling the eligibility criteria. Thirty-five patients were innormal sinus rhythm, and 6 had regular paced rhythm. TRwas of organic cause in 8 patients (2 rheumatic, 2 prolapse, 1

1506 Circulation October 12, 2010

at Mayo Clinic Libraries on September 29, 2010 circ.ahajournals.orgDownloaded from

endocarditis, 2 carcinoid, and 1congenital) and functionalin 33. Calculated systolic pulmonary artery pressure was55 mm Hg (25th to 75th percentile, 40.5 to 75.5 mm Hg),and pulmonary hypertension with systolic pulmonary ar-tery pressure �50 mm Hg was noted in 26 patients.Moderate or severe TR, with ERO �0.20 cm2, was

observed in 24 patients. The baseline characteristics of theoverall population and of patients with and without mod-erate to severe TR are reported in Table 1. Of note, fewpatients had a murmur suggestive of TR, and even fewerhad inspiratory increase or appearance of the systolicmurmur by clinical examination.

Table 1. Baseline Characteristics of Patients With TR Overall and Stratified by Severity of Regurgitation

Variables Entire Sample (n�41) ERO �0.2 cm2 (n�17) ERO �0.2 cm2 (n�24) P

Age, y 67 (57–74) 65.5 (51–70) 68 (57–78) 0.17

Male sex, % 32 29 39 0.53

SBP, mm Hg 129 (119,150) 128 (120,142) 130 (114–151) 0.82

DBP, mm Hg 78 (70–88) 80 (72–86) 77 (70–89) 0.68

Heart rate, bpm 76 (61–85) 71 (60–81) 79 (62–88) 0.1

Systolic murmur of TR, % 20 12 25 0.43

Inspiratory increase of murmur, % 5 0 8 0.50

Right heart failure, % 20 12 25 0.43

Stroke volume, mL 65 (45–78) 74 (49–84) 54 (44–69) 0.056

Cardiac index, L/m2 2.5 (1.9–3.1) 2.6 (2.2–3.5) 2.35 (1.7–3.0) 0.15

LV ejection fraction, % 60 (47.5–65) 60 (57–62.7) 55 (27–65) 0.15

SBP indicates systolic blood pressure; DBP, diastolic blood pressure; and LV, left ventricular. Data are presented as median (25thto 75th percentile) for continuous variables or as percentages for categorical variables.

Figure 1. Example of quantitative measurement of TR with inspiration and expiration. Shown are the measurements of regurgitant flowusing the flow convergence method (top) and peak velocity using continuous-wave Doppler (Vmax; bottom) with calculation of ERO ininspiration (left) and expiration (right). Respirometer curves ascend with inspiration and descend with expiration. During inspiration,Vmax decreases, regurgitant flow increases, and ERO increases considerably.

Topilsky et al Pathophysiology of TR 1507

at Mayo Clinic Libraries on September 29, 2010 circ.ahajournals.orgDownloaded from

Overall Respiratory ChangesChanges in TR driving forces and severity with respirationare reported in Table 2. An example of the measurement ofTR regurgitant flow, peak velocity, and ERO in inspirationand expiration is shown in Figure 1. With inspiration, there isa notable decline in the driving force to TR whether it isexpressed as peak TR velocity (�12%; 25th to 75th percen-tile, �17.8% to �8% from expiration), peak regurgitantgradient (�17.5%; 25th to 75th percentile, �23.1% to�12.5% from expiration), or regurgitant time-velocity inte-gral (�15%; 25th to 75th percentile, �24.4% to �9.5% fromexpiration). There is also a modest decline in TR durationduring inspiration (�2.7%; 25th to 75th percentile, �10.9%to 0% from expiration). Contrasting with this decline in RV toRA gradient during systole, there is a marked increase inproximal flow convergence radius and in regurgitant flowwith inspiration (125%; 25th to 75th percentile, 92.2% to178.7% from expiration) resulting from a considerable in-crease in ERO (169%; 25th to 75th percentile, 129% to 226%

from expiration). RVol also increased but less in percentagethan the ERO (120% [25th to 75th percentile, 78.6% to169%] versus 169% [ 25th to 75th percentile, 129% to226%]; P�0.001) as a result of the decline in the regurgitanttime-velocity integral (the entire systolic integrated drivingforce) for TR. Contrasting with these considerable changes inERO and RVol, the maximum TR jet in the RA and its ratioto RA area showed no significant increase. Color-flowimaging jet extent lack of sensitivity for TR severity changescould be palliated by measuring the vena contracta width ofTR, which demonstrated a significant increase with inspira-tion. All changes were observed coherently in the vastmajority of patients, as shown by Figure 2, which presents TRvariations in individual patients. Percent changes of majormeasures of TR are presented graphically in Figure 3.

Morphological measurements of RV, RA, tricuspid valve,and tricuspid annulus are presented in Table 3. RV systolicand diastolic areas increased during inspiration (bothP�0.01), but the RV area shortening showed no change with

Table 2. Variations of TR With Respiration

Variable Expiration Inspiration Change With Inspiration P

TR peak velocity, cm/s 336 (286–406) 290 (222–371) �40 (�30–�60) �0.0001

TR TVI, cm 107 (88–140) 88 (66–125) �17 (�10–�23) �0.0001

TR peak gradient, mm Hg 45 (33–66) 34 (20–55) �11 (�6–�15) �0.0001

TR duration, ms 407 (373–446) 396 (338–442) �12 (�45–2) �0.0001

TR flow convergence radius, cm 0.55 (0.4–0.74) 0.80 (0.6–1.04) 0.25 (0.2–0.37) �0.0001

Regurgitant flow, mL/s 47 (23.5–79.5) 104 (56.5–179.5) 64 (30–112) �0.0001

ERO, cm2 0.13 (0.05–0.24) 0.36 (0.16–0.63) 0.22 (0.09–0.37) �0.0001

RVol, mL per beat 15 (8.5–24) 33 (19.5–47.5) 18 (11–25) �0.0001

Jet area, cm2 4.9 (3–8.6) 4.9 (3.2–9.8) 0 (�0.5–0.8) 0.58

Jet/atrial ratio, % 20.1 (12.3–36.6) 21.6 (12.6–36) 0 (�1.6–3.2) 0.49

TR vena contracta width, mm 5 (3–8.4) 5.1 (3.2–8.6) 0.15 (0–0.63) 0.006

TVI indicates time-velocity integral. Data are presented as median (25th to 75th percentile).

Figure 2. Individual patient changes inTR hemodynamic characteristics withrespiration. Each patient is representedby an individual line symbolizing thechange from expiration (E) to inspiration(I). Top left, ERO (cm2); bottom left, peakTR velocity (Vmax; cm/s); top right, TRduration (seconds); and bottom right,regurgitant (Reg) volume (mL per beat).All changes are significant and affectmost individual patients.

1508 Circulation October 12, 2010

at Mayo Clinic Libraries on September 29, 2010 circ.ahajournals.orgDownloaded from

inspiration. However, there were important RV changes withinspiration. Indeed, with inspiration, the RV “stroke area” (ie,the difference between the diastolic and systolic areas)increased, consistent with increased stroke volume. The RVshape also changed, with RV length showing very littleincrease (P�0.055) and the RV width increasing markedly sothat the length-to-width ratio decreased significantly, denot-ing a marked widening of RV with inspiration. Conversely,RA volume did not change notably. Important changes wereseen at the valvular level. The tricuspid annulus diameterincreased with inspiration in both systole and diastole (bothP�0.01). Thus, with inspiration, there was a significantdecrease in the coverage ratio of the annulus by the combined

anterior and septal leaflet length (P�0.006). Furthermore, thetenting height and tenting area of the tricuspid valve insystole increased notably (Figure 3), further reducing theability of leaflet to coapt in systole.

In multivariable analysis (multiple linear regression), in-spiratory ERO adjusted for expiratory ERO was indepen-dently associated with valvular-to-annular length ratio withinspiration (P�0.026) and by the inspiratory changes in RVlength-to-width ratio (P�0.008) and tricuspid valve tentingarea (P�0.015).

Subgroup Analysis of TR Respiratory ChangesWhen patients were stratified according to TR severity (ERO�0.20 or �0.20 cm2), there was no difference relative to

Table 3. Change in Right-Side Cardiac Cavities and Tricuspid Valvular Morphology With Respiration

Variable Expiration Inspiration Change With Inspiration P

RV diastolic area, cm2 26.5 (21.1–31.9) 27.8 (22.6–36.3) 2.1 (0.1–3.0) �0.0001

RV systolic area, cm2 16 (12.9–26.4) 16.9 (13.1–27.4) 0.8 (0.1–1.8) 0.003

RV area shortening fraction, % 33.4 (17.8–47.5) 33.4 (20.8–49.7) 0.9 (�4–5) 0.37

RV diastolic length, cm 6.9 (6.5–7.7) 6.9 (6.5–7.9) 0 (�0.02–0.31) 0.55

RV diastolic midventricular width, cm 4.1 (3.4–4.7) 4.5 (3.8–5.1) 0.41 (0.26–0.65) �0.0001

RV diastolic length/width ratio 1.7 (1.46–2.1) 1.6 (1.37–1.95) �0.15 (�0.04–�0.24) �0.0001

RA volume, mL 85.8 (52.7–132.3) 89 (51.7–128.5) 1.5 (�3–5) 0.36

Systolic annulus diameter, mm 2.9 (2.7–3.6) 3.1 (2.7–3.8) 0.1 (0–0.2) 0.003

Diastolic annulus diameter, mm 3.7 (3.4–4.3) 3.9 (3.5–4.4) 0.1 (0–0.2) 0.01

Systolic valvular/annular coverage, % 104 (94–110) 100 (93–109) �2 (1–�7) 0.006

Tenting height, cm 0.7 (0.53–0.92) 0.92 (0.74–1.2) 0.2 (0.11–0.42) �0.0001

Tenting area, cm2 1.1 (0.66–1.5) 1.4 (1.15–1.7) 0.29 (0.04–0.54) �0.0001

Systolic valvular/annular coverage is the ratio of the sum of the length of septal and anterior leaflets to the systolic annulusdiameter. Data are presented as median (25th to 75th percentile).

Figure 3. Relative changes for variables characterizing TR, right-side cavities, and tricuspid valve represented by box plots (middle hash ofthe box indicates the median; 25th to 75th percentiles are represented by end caps of the box; whiskers extend to the last observed valuestill within 1.5 times the interquartile range [difference between the 25th and 75th percentiles] above or below the 25th and 75th percentiles;extreme values are represented by circles that lie beyond the whiskers). The variables are presented as follows: Left, TR driving force (peakvelocity [TR Vmax], time-velocity integral of jet [TR TVI], and TR duration). Middle left, TR severity (ERO and RVol). Middle right, Right-sidedcavities (RV area diastolic area, RV systolic area, RV length-to-width ratio [RV L{D], and RA systolic area). Right, Annular and valvular defor-mation (annulus systolic diameter [annulus systolic], valve tenting height [tricuspid tenting height], and valve tenting area [tricuspid tentingarea]). Changes are positive for inspiratory increases and negative for inspiratory decreases and are expressed as percentage (Note thehigher scale for TR severity than for other measures). With inspiration, TR driving force declines, TR severity increases markedly, RV enlargesand widens (decrease in length-to-width ratio), and annular diameter and valve tenting increase.

Topilsky et al Pathophysiology of TR 1509

at Mayo Clinic Libraries on September 29, 2010 circ.ahajournals.orgDownloaded from

inspiratory change in TR peak velocity (P�0.24) or time-velocity integral (P�0.33) between groups. There was a trendfor a larger decline in TR duration (P�0.07) and a definitelylarger increase in ERO (0.32 cm2 [25th to 75th percentile,0.24 to 0.66 cm2] versus 0.08 cm2 [ 25th to 75th percentile,0.06 to 0.13 cm2]; P�0.0001) in patients with mean ERO�0.20 cm2. However, expressed in relative terms, the percentchange in ERO was similar (156% [ 25th to 75th percentile,129% to 185%] versus 200% [ 25th to 75th percentile, 122%to 283%]; P�0.30) in both groups (ERO �0.20 or �0.20cm2). Thus, with inspiration, a relatively fixed proportionalincrease in ERO is observed regardless of TR severity.Relative changes in TR peak velocity, duration, ERO, andRVol are stratified by TR pathogenesis (organic versusfunctional TR; top row) and by pulmonary hypertension(systolic pulmonary artery pressure �50 versus �50 mm Hg;bottom row) in Figure 4. There was no difference in anyinspiratory change according to pathogenesis of TR. Withpulmonary hypertension, there was a smaller change in TRpeak velocity, but the ERO change with inspiration wasidentical to that without pulmonary hypertension. Thus, therelative change in ERO with inspiration was similar regard-less of the severity, pathogenesis, or association of pulmonaryhypertension with TR.

DiscussionThe present study, the first to analyze quantitatively thepathophysiology of TR changes with respiration, reveals newinsights into TR physiological characteristics based on instan-taneous proximal isovelocity surface area quantification ofTR severity during respiration phases timed by simultaneousrespirometer recording. Indeed, inspiratory TR accentuationis not just an auscultatory phenomenon but also involves anincrease in volumetric TR severity. However, although TRERO accentuation is almost universal, it is infrequently

audible clinically and is undetected by color jet extent.Respiratory TR variations are complex, with TR driving force(RV-RA gradient) and TR duration declining with inspira-tion, which is more than compensated for by a large EROincrease. These changes in opposite directions compound theinspiratory RVol increase, less impressive than the EROincrease, but more than doubling the TR volume on average.ERO magnitude of change is higher in those with largeexpiratory ERO, but expressed as percent change, the inspira-tory ERO increase is similar across categories of TR severity,pathogenesis, or pulmonary hypertension. Mechanistically,we observed marked RV (not RA) plasticity with inspiratorychanges in size and shape, particularly RV widening. RVwidening is associated with valvular changes and inspiratoryannular enlargement, leading to less systolic annular coverageby tricuspid leaflets and increased valvular tenting, both ofwhich contribute to coaptation loss. These new observationshave important clinical implications. The association withloading changes of marked RV plasticity and TR variationssuggests that clinically optimization of loading conditionsmay have beneficial effects on TR severity. Other implica-tions regard TR evaluation, which can be difficult because jetsize changes do not clearly reveal even marked changes in TRseverity, and measurement of vena contracta width is moresensitive. Additionally, because large TR variations can beobserved over the respiratory cycle, it is essential that we donot rely on a single TR measurement but on multiplemeasurements averaged over the respiratory cycle. Theseconcepts applied in clinical practice should enhance compre-hension of this “forgotten valve” pathophysiology and clini-cal management.4

Respiratory Dependence of TRA classic feature of TR is respiratory variation of systolicmurmur,6 described decades ago but still poorly understood.13

TR pathophysiology remains elusive,4 in part because of the

Figure 4. Inspiratory TR changes strati-fied by pathogenesis and pulmonaryhypertension. Percent changes (repre-sented by box plots as in Figure 3) withinspiration in TR peak velocity (Vmax;top), duration (TRDur; middle), ERO (bot-tom middle), and RVol (bottom) are pre-sented stratified by TR pathogenesis(top, organic vs functional TR; open vssolid bars) and by pulmonary hyperten-sion (bottom, systolic pulmonary arterypressure �50 vs �50 mm Hg; open vssolid bars). Although there is less per-cent decline in TR velocity in patientswith pulmonary hypertension, inspiratorypercent changes in ERO are similar in allsubgroups.

1510 Circulation October 12, 2010

at Mayo Clinic Libraries on September 29, 2010 circ.ahajournals.orgDownloaded from

lack of adequate methods to measure all specific componentsof the condition.22 However, Doppler echocardiographicprogress now allows us to instantaneously measure TRregurgitant gradient,19 duration, and regurgitant flow.10 Al-though it is well established that venous return to the heartincreases with inspiration, hypotheses about TR murmurvariation range from increased turbulence to increased gradi-ent to increased regurgitant flow.13,14 In TR, as in allregurgitations, RVol depends on regurgitant flow and dura-tion. In turn, regurgitant flow is determined directly by theERO area through which it is forced by the systolic RV-RApressure gradient.23,24 The present study shows that TR isundeniably dynamic with respiration, demonstrating in-creased RVol during inspiration. Clinically, auscultatorychanges are infrequent whereas TR almost universally in-creases in volume during inspiration. Inspiratory increase inRVol is not due to increased TR pressure gradient or durationbut to inspiratory ERO increase. In fact, the RV-RA gradientdecreases with inspiration, probably owing to decreasedpulmonary vascular resistance with decreased intrathoracicpressure.25 An alternative explanation for decreased TRvelocity with inspiration would be a “ventricularization” ofatrial pressure during inspiration15 secondary to increasedTR.25 Previous reports mentioning increased TR velocityduring inspiration17 pondered that it may affect patients withpulmonary hypertension, but we observed inspiratory declinein TR velocity with and without pulmonary hypertension.Regardless, because of the decline in TR gradient andduration, inspiratory RVol augmentation is of lesser magni-tude than ERO. This complex counterbalancing of TR com-ponents may explain the lack of TR inspiratory color-flow jetincrease. Indeed, color-flow jet extent is determined by itsmomentum26,27 and complex interaction with the receivingchamber.28 Jet momentum is proportional to regurgitant flowand velocity, and with inspiration, the TR flow increase iscompensated for by a TR velocity decrease, so the jet extentand jet RA-to-area ratio do not significantly change. Con-versely, vena contracta width,9 directly related to TR severity,shows a definite increase during inspiration coherently toERO and RVol. Thus, TR dynamic nature may be missed bycasual examination of color-flow imaging and requires com-prehensive quantitative assessment. Importantly, TR dynamicchanges are similar in percentage across ranges of TRseverity, pathogenesis, and pulmonary hypertension.

Morphological Alterations Associated WithTR ChangesDetailed information on the annular-valvular complex andRV by comprehensive high-resolution prospectively recordedmultiview 2-dimensional imaging with simultaneous respi-rometry sheds unique light on RV and TR pathophysiology.With inspiration, the RV enlarges with increased strokevolume, in contrast to the unchanged RA. End-systolic RVsize increases and the RV becomes wider without elongation.Such changes affect the tricuspid valve through 2 mechanisticpathways. First, an enlarged systolic annulus accompanies thelarger RV,29 leading to decreased valve coverage of theannulus.30 Second, RV enlargement and widening centrifu-gally displace RV walls and tricuspid papillary muscles

attached to each wall31 away from the tricuspid valve, causingincreased tethering and tenting height and area, which,similar to functional mitral regurgitation,32 yield decreasedleaflet coaptation. These annular and valvular mechanisms ofdecreased coaptation independently determine larger EROwith inspiration. Thus, respiratory changes provide uniqueinsights into the plasticity of the right-side ventriculo-valvu-lar complex, particularly the RV,33 that may have importantclinical implications.

Clinical ImplicationsFor TR assessment, clinical appraisal is infrequently typical,even to experienced cardiologists, and suspicion of signifi-cant TR should be verified by Doppler echocardiography.Comprehensive TR grading is recommended by guidelines,8

but in clinical practice, color flow is often the main method ofassessing TR. Because jet assessment is insensitive tochanges in TR severity, we believe that focusing on measur-ing the vena contracta and/or quantifying the TR should bepreferred. The dynamic nature of regurgitation should notcontraindicate quantitative methods use,24 but measurementshould be cognizant of large respiratory TR variations,avoiding measurement of largest values and instead obtainingmeasurements covering the entire respiratory cycle.10–12,21 Inaddition, a large RV pressure inspiratory decrease suggeststhat measuring only the highest velocities (in expiration) mayoverestimate pulmonary hypertension. Thus, for all measure-ments relative to TR, it is essential to account for both phasesof the respiratory cycle.

Marked plasticity of the right-side ventriculo-valvularcomplex33 leads to marked inspiratory TR accentuation. Suchplasticity may explain the load sensitivity of TR, which mayrapidly regress with medical treatment.4 Although aggressivesurgical TR management has been advocated,34 optimizingRV unloading before proceeding with TR surgery may bedesirable.

LimitationsThe study was powered to detect notable respiratory TRchanges and did not aim to define all issues relative toTR,especially outcome, which will require further longitudinalstudies. We observed no trend of difference between TRpathogeneses, but specific pathophysiology of each TR causeshould be established by future studies now that theseimportant pathophysiological insights have been clarified.

To instantaneously quantify TR completely, one shouldmeasure color-flow convergence and continuous-wave Dopp-ler simultaneously. Such technology does not exist, so weperformed optimally each measurement timed to the cardiacand respiratory cycle with the patient in a steady state andstable position. Future developments of 3-dimensional ultra-sonic technology may provide comprehensive imaging tools35

and quantitative software36 that are currently unavailable. It isimportant that physiological concepts revealed by currenttechnology be incorporated into practice.

ConclusionsThe present quantitative physiological study shows thatinspiratory accentuation of TR severity is almost universal

Topilsky et al Pathophysiology of TR 1511

at Mayo Clinic Libraries on September 29, 2010 circ.ahajournals.orgDownloaded from

but clinically infrequently audible. Respiratory variations ofTR characteristics are complex, with decreased TR drivingforce and duration with inspiration that are more thancompensated for by a large ERO increase, compounding themarked inspiratory RVol increase. Mechanistically, we ob-served marked RV plasticity with striking inspiratory widen-ing. Inspiratory valvular changes and annular enlargementleading to less systolic annular coverage by tricuspid leafletsand increased valvular tenting contribute to coaptation loss.These new observations have important clinical implicationsfor TR severity assessment in clinical practice that should becomprehensive throughout the respiratory cycle but also forRV plasticity associated with large TR severity variations.These concepts should enhance TR pathophysiological com-prehension and clinical management.4

DisclosuresDr Enriquez-Sarano is a consultant for Valtec Inc, Tel Aviv, Israel.The other authors report no conflicts.

References1. Singh JP, Evans JC, Levy D, Larson MG, Freed LA, Fuller DL, Lehman

B, Benjamin EJ. Prevalence and clinical determinants of mitral, tricuspid,and aortic regurgitation (the Framingham Heart Study). Am J Cardiol.1999;83:897–902.

2. Hansing C, Rowe G. Tricuspid insufficiency: a study of hemodynamicsand pathogenesis. Circulation. 1972;45:793–799.

3. Bonow RO, Carabello BA, Kanu C, de Leon AC Jr, Faxon DP, FreedMD, Gaasch WH, Lytle BW, Nishimura RA, O’Gara PT, O’Rourke RA,Otto CM, Shah PM, Shanewise JS, Smith SC Jr, Jacobs AK, Adams CD,Anderson JL, Antman EM, Fuster V, Halperin JL, Hiratzka LF, Hunt SA,Nishimura R, Page RL, Riegel B. ACC/AHA 2006 guidelines for themanagement of patients with valvular heart disease: a report of theAmerican College of Cardiology/American Heart Association Task Forceon Practice Guidelines (Writing Committee to Revise the 1998Guidelines for the Management of Patients With Valvular Heart Disease):developed in collaboration with the Society of Cardiovascular Anesthe-siologists: endorsed by the Society for Cardiovascular Angiography andInterventions and the Society of Thoracic Surgeons. Circulation. 2006;114:e84–e231.

4. Shah PM, Raney AA. Tricuspid valve disease. Curr Probl Cardiol.2008;33:47–84.

5. Song H, Kim MJ, Chung CH, Choo SJ, Song MG, Song JM, Kang DH,Lee JW, Song JK. Factors associated with development of late significanttricuspid regurgitation after successful left-sided valve surgery. Heart.2009;95:931–936.

6. Rivero-Carvallo JM. Signo para el diagnostico de las insufficienciastricuspideas. Arch Inst Cardiol Mex. 1946;16:531.

7. Lingamnemi R, Cha S, Maranhao V, Gooch A, Goldberg H. Tricuspidregurgitation: clinical and angiographic assessment. Cathet CardiovascDiagn. 1979;5:7–14.

8. Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD,Levine RA, Nihoyannopoulos P, Otto CM, Quinones MA, Rakowski H,Stewart WJ, Waggoner A, Weissman NJ. Recommendations for eval-uation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr.2003;16:777–802.

9. Tribouilloy CM, Enriquez-Sarano M, Bailey KR, Tajik AJ, Seward JB.Quantification of tricuspid regurgitation by measuring the width of thevena contracta with Doppler color flow imaging: a clinical study. J AmColl Cardiol. 2000;36:472–478.

10. Rivera JM, Mele D, Vandervoort PM, Morris E, Weyman AE, ThomasJD. Effective regurgitant orifice area in tricuspid regurgitation: clinicalimplementation and follow-up study. Am Heart J. 1994;128:927–933.

11. Rivera JM, Vandervoort PM, Mele D, Siu S, Morris E, Weyman AE,Thomas JD. Quantification of tricuspid regurgitation by means of theproximal flow convergence method: a clinical study. Am Heart J. 1994;127:1354–1362.

12. Tribouilloy CM, Enriquez-Sarano M, Capps MA, Bailey KR, Tajik AJ.Contrasting effect of similar effective regurgitant orifice area in mitral

and tricuspid regurgitation: a quantitative Doppler echocardiographicstudy. J Am Soc Echocardiogr. 2002;15:958–965.

13. Leon DF, Leonard JJ, Lancaster JF, Kroetz FW, Shaver JA. Effect ofrespiration on pansystolic regurgitant murmurs as studied by biatrialintracardiac phonocardiography. Am J Med. 1965;39:429–441.

14. Braunwald E, Zipes D, Libby P, Bonow R, eds. Braunwald’s HeartDisease: A Textbook of Cardiovascular Medicine. 6th ed. New York, NY:W.B. Saunders Co; 2001.

15. Cha SD, Desai RS, Gooch AS, Maranhao V, Goldberg H. Diagnosis ofsevere tricuspid regurgitation. Chest. 1982;82:726–731.

16. Cha SD, Gooch AS, Maranhao V. Intracardiac phonocardiography intricuspid regurgitation: relation to clinical and angiographic findings.Am J Cardiol. 1981;48:578–583.

17. Klodas E, Nishimura RA, Appleton CP, Redfield MM, Oh JK. Dopplerevaluation of patients with constrictive pericarditis: use of tricuspid re-gurgitation velocity curves to determine enhanced ventricular interaction.J Am Coll Cardiol. 1996;28:652–657.

18. Nihoyannopoulos P. Tricuspid valvectomy following tricuspid valveendocarditis on an intravenous drug addict. Heart. 2001;86:144.

19. Currie P, Seward J, Chan K. Continuous wave Doppler determination ofright ventricular pressure: a simultaneous Doppler-catheterization studyin 127 patients. J Am Coll Cardiol. 1985;6:750–756.

20. Kircher BJ, Himelman RB, Schiller NB. Noninvasive estimation of rightatrial pressure from the inspiratory collapse of the inferior vena cava.Am J Cardiol. 1990;66:493–496.

21. Rivera JM, Vandervoort P, Mele D, Weyman A, Thomas JD. Value ofproximal regurgitant jet size in tricuspid regurgitation. Am Heart J.1996;131:742–747.

22. Chan KM, Zakkar M, Amirak E, Punjabi PP. Tricuspid valve disease:pathophysiology and optimal management. Prog Cardiovasc Diss. 2009;51:482–486.

23. Enriquez-Sarano M, Bailey KR, Seward JB, Tajik AJ, Krohn MJ, Mays JM.Quantitative Doppler assessment of valvular regurgitation. Circulation. 1993;87:841–848.

24. Enriquez-Sarano M, Sinak LJ, Tajik AJ, Bailey KR, Seward JB. Changesin effective regurgitant orifice throughout systole in patients with mitralvalve prolapse: a clinical study using the proximal isovelocity surfacearea method. Circulation. 1995;92:2951–2958.

25. Jaber WA, Sorajja P, Borlaug BA, Nishimura RA. Differentiation oftricuspid regurgitation from constrictive pericarditis: novel criteria fordiagnosis in the cardiac catheterisation laboratory. Heart. 2009;95:1449–1454.

26. Rivera J, Vandervoort P, Vazquez de Prada J, Mele D, Karson T,Morehead A, Morris E, Weyman A, Thomas J. Which physical factorsdetermine tricuspid regurgitation jet area in the clinical setting? Am JCardiol. 1993;72:1305–1309.

27. Thomas J, Liu C, Flachskampf F, O’Shea J, Davidoff R, Weyman A.Quantification of jet flow by momentum analysis: an in vitro colorDoppler flow study. Circulation. 1990;81:247–259.

28. Grimes RY, Pulido GA, Levine RA, Yoganathan AP. Quasisteadybehavior of pulsatile, confined, counterflowing jets: implications for theassessment of mitral and tricuspid regurgitation. J Biomech Eng. 1996;118:498–505.

29. Mutlak D, Lessick J, Reisner SA, Aronson D, Dabbah S, Agmon Y.Echocardiography-based spectrum of severe tricuspid regurgitation: thefrequency of apparently idiopathic tricuspid regurgitation. J Am SocEchocardiogr. 2007;20:405–408.

30. Colombo T, Russo C, Ciliberto GR, Lanfranconi M, Bruschi G, Agati S,Vitali E. Tricuspid regurgitation secondary to mitral valve disease: tri-cuspid annulus function as guide to tricuspid valve repair. CardiovascSurg. 2001;9:369–377.

31. Rogers JH, Bolling SF. The tricuspid valve: current perspective andevolving management of tricuspid regurgitation. Circulation. 2009;119:2718–2725.

32. Yiu SF, Enriquez-Sarano M, Tribouilloy C, Seward JB, Tajik AJ. Deter-minants of the degree of functional mitral regurgitation in patients withsystolic left ventricular dysfunction: a quantitative clinical study. Circulation.2000;102:1400–1406.

33. Hill JA, Olson EN. Cardiac plasticity. N Engl J Med. 2008;358:1370–1380.

34. Dreyfus GD, Chan KM. Functional tricuspid regurgitation: a morecomplex entity than it appears. Heart. 2009;95:868–869.

35. Seo Y, Ishizu T, Nakajima H, Sekiguchi Y, Watanabe S, Aonuma K.Clinical utility of 3-dimensional echocardiography in the evaluation of

1512 Circulation October 12, 2010

at Mayo Clinic Libraries on September 29, 2010 circ.ahajournals.orgDownloaded from

tricuspid regurgitation caused by pacemaker leads. Circ J. 2008;72:1465–1470.

36. Roes SD, Hammer S, van der Geest RJ, Marsan NA, Bax JJ, Lamb HJ,Reiber JH, de Roos A, Westenberg JJ. Flow assessment through four

heart valves simultaneously using 3-dimensional 3-directional velocity-encoded magnetic resonance imaging with retrospective valve tracking inhealthy volunteers and patients with valvular regurgitation. Invest Radiol.2009;44:669–675.

CLINICAL PERSPECTIVETricuspid regurgitation (TR) is present in most humans in variable degrees, but its pathophysiology is poorly understood.Previous clinical observations suggested that the murmur of TR may vary with respiration, but the reality, magnitude,pathophysiology, and clinical implications of this potential variation with respiration have not been investigated. However,progress in Doppler echocardiographic assessment of TR with instantaneous quantification of its degree and quantificationof cardiac cavity changes, coupled with the ability to record the respiratory phases with a respirometer, allowed us to obtainunique insights into the pathophysiology of TR. Applying these methods, we first observed that TR is indeed markedlydynamic but with complex changes of opposite directions. Indeed, during inspiration, TR driving force and durationdecrease, but because this decline is more than compensated for by a considerable increase in effective regurgitant orifice,TR volumetric severity increases with inspiration by �100%. The marked effective regurgitant orifice inspiratory increaseis in turn due to reduced valve coverage of an enlarging annulus and increased valve tenting, resulting in reduced valvecoaptation. These valve changes causing TR inspiratory augmentation are in turn linked to right ventricular shape changeswith inspiratory right ventricular widening. These profound pathophysiological respiratory changes, poorly detectedclinically, underscore the TR respiratory dependence linked to valve deformation and right ventricular plasticity. Inpractice, the major respiratory variation of TR requires multiplying measurements throughout the cardiac cycle toappropriately assess TR and right ventricular pressure.

Topilsky et al Pathophysiology of TR 1513

at Mayo Clinic Libraries on September 29, 2010 circ.ahajournals.orgDownloaded from