Mitral valve resistance as a hemodynamic indicator in mitral stenosis

VALVULAR HEAR7 DtSAsE

Mitral Valve Resistance as a Hemodynamic Indicator in Mitral Stenosis Reinaldo W. Beyer, MD, Alfonso Olmos, MD, Ruben F. Bermtidez, MD, and H. Elizabeth Nell, MD

Variability of the valve area calculated by the Gorlin fonnuia has beon noted in bioprosthetic and aortic valves, but few data are available for native stenotic mitral vatves. Valve resistance has been proposed as an alternattve hemodynamic indicator; however, its value in mitraf stenods has not been assessed. Thirty-four patients had simultaneous recordings of teft atrial and ventricular pressures, 26 after percutaneous batioon mitral dilatation (PBMD). Patients with shunt or mitral regurgitation were excluded. Mitral vaive resistance correlated exponentially with Got-tin mitral area (y = 133*[area]-1*s; p <0.0601). Both Goriin mitral area and mitral resistance improved after PBMD (0.89 f 0.07 cm2 to 2.22 f 0.15 cm2; p
l

l

l

he valve area derived from the Gorlin formula has been used clinically for decadesas an index of severity in the assessmentof valve stenoses.*-3 The Gorlin formula was derived from principles valid for steady nonpulsatile flow, with a Newtonian fluid forced through a diaphragm with a circular orifice.1+4 Several assumptionsare accepted when using the formula in the assessmentof the intact heart with valve stenosis.3*4 The most clinically relevant assumption is that the calculated valve area is a constant characteristic of each particular valve being assessed,and depends only on the anatomic severity of the obstruction to flow, so that differences in calculated areas can reliably be assignedto a different degreeof stenosis.This is of particular interest in obtaining accurate assessmentduring interventions such as percutaneousballoon mitral dilatation (PBMD). However, studies in aortic stenosis5 and mitral bioprostheses637 have shown that Gorlin area varies if the hemodynamic conditions during measurement are changed.Low flow states,among others, have beenproposedas a condition in which the Gorlin formula is inaccurate, explaining discrepanciesbetween anatomic and hemodynamically estimated Gorlin mitral areas in native valve mitral stenosis.8 Valve resistance,as a variable in the assessmentof valve stenoses,was theoretically dismissed before the development of techniques to measure left ventricular and left atria1 pressuresdirectly.9 Initial studies with modem techniques,limited to aortic valve stenosis,have renewedinterest in valve resistance.10Our purposewas to assessmitral valve resistanceas a hemodynamicindicator by comparing it to Gorlin mitral area under different hemodynamic conditions in patients who were candidatesfor PBMD.

T

METHOD8 Patisnt chersckfsticu

A total of 34 patients (8 men, 26 women) with pure rheumatic mitral stenosis betweenthe agesof 17 to 56 years (mean 37) were entered into this study. The baselinerhythm was sinus in 25 patients and atria1 fibrillation in 9. New York Heart Association functional classification was II in 4 patients, III in 24 and IV in 6 patients. Gorlin mitral area ranged from 0.36 to 1.61 cm2 (mean 0.97) and mean left atria1 From the Cardiology Division, Wadsworth Veterans Affairs Medical pressure ranged from 12 to 40 mm Hg (mean 23). Center/UCLA Sc&ol of Medicine, Los Angeles, California, and PBMD was undertaken in 32 patients. Symptoms were Facultad de Medicina, Universidad de Concqxk%t, Chile. This study assessedagain 4 weeks after PBMD, with the patients wassupportedby Grant 2 1.0509 from the ResearchDivision, Universi- performing normal and usual activity. In the follow-up dad de Concepc%n,Chile. Manuscript received August 26, 1991;re- evaluation after PBMD, 21 patients were in New York vised manuscript receivedand acceptedNovember 12,1991. Address for reprints: Reinaldo W. Reyer, MD, Cardiology 691/ Heart Association classI, 3 in classII and 1 in III. One 11lE, Wadsworth Veterans Affairs Medical Center, Wilshire & Saw- patient was lost to follow-up. Caseswith >l+ on a scale telle Boulevards,Los Angeles, California 90073. of 1 to 4+ ventriculographic mitral regurgitation’ 1 or MITRAL RESISTANCE 775

oximetric step-up from the mixed venous blood to pulmonary artery 24% were excluded, thus, only 26 patients were included in the analysis after PBMD. All patients gave informed consent before catheterization and PBMD. Catheteddiom technique and measurements: Leftand right-sided cardiac catheterization were performed by conventional percutaneous techniques with fluidfilled catheters. At least 4 thermodilution curve values with
let and microcomputer to obtain diastolic times and mean gradients. Calculations were performed using the following formulas: Gorlin mitral area [cm’] =

[ImQ;

HR)

39.9 x JpApLv

Mitral resistanceI”y;:s]=l,333xi;~;~R)

where PLA - PLVis mean transvalvular mitral pressure gradient [mm Hg], Qp is cardiac output [ml/mm], DFTb is diastolic filling time per beat [seconds] and HR is heart rate [beat/min]. A constant of 39.9 was used in the Gorlin formula since direct left atria1 pressureswere measured.15The number 1,333usedin equation 2 only provides for unit conversionfrom mm Hg to dynes/cm2 (1 mm Hg = 1,333 dynes/cm2). Statistics: Results are expressedas mean f SEM. PBMD and isoproterenol-inducedchangeswere evaluated by Student’s paired t test. Correlation was sought by the least-squares regression method. Significance was acceptedat p <0.05. Analysis was performed using the BMDP statistical software (BMDP Statistical Software, Inc., Los Angeles, California) on a microcomputer. RESULTS In all, 111 hemodynamic valve area observations were obtained. These consistedof 34 baseline measurements, 14 after an initial but not final mitral balloon dilatation, 26 immediately after final PBMD, 20 at follow-up after PBMD, and 17 during isoproterenol infusion. Mitral resistancehad a negative exponential relation with Gorlin mitral area (Figure 1).

600 27 'E : Z 5 -0 Y.

400

t 5 G ZJ

FtGURE l.Re4athofGodinmitrdarea wltb mitrat valve resistance. A sigdhnt exponedddinverve -iSMen. SmdcbaqgesinGerlbtmitrdareabave bnperht ehcts en mitral valve reststanceatsmdodtkeareamnges.AtartNceweas>2.Scm2,mhimatefhctonmitralvatveresistancebobtahdbytull#r incr8ase in 8wa.

n = 111 r = -0.96 p<0.0001 y=133x45

200

: a F 5 0 0

1 GORLIN

776

2

3

MITRAL

4

AREA

5

[cm*1

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 69

MARCH 15, 1992

6

Functbnal class: Both Gorlin mitral area and mitral resistancecorrelated roughly with New York Heart Association functional class (Figure 2).

TABLE

I Hemodynamic

Effects of lsoproterenol

in Mitral

Stenosis Basal (mean 2 SEM)

EffoctsofpareutanaorubaUoonmitraldilatation

lsoproterenol (mean 2 SEM)

Paired t p Value

and iroproteranol intudon: After PBMD the Gorlin mi99 +I 4 123 f 5 2.5 cm2 only minor mitral resistanceeffects can be expectedby further increasesin orifice area (Figure 1). Discrimination of mitral stenosis severity: Both Gorlin mitral area and mitral resistance correlated roughly with functional classand improved significantly after PBMD. 61

I

I Elw I

Ill

IV

NYHA FUNCTIONAL

CLASS

II

I--PCO.05

BASAL

1

ISOPROTERENOL

FIGURE 3. Gadn mitral mea chmge dving hophem flwh.Eacb~andgrap

in-

meansareshown.Thechtmge

issigdhntbypahdttest.

MITRAL

RESISTANCE

777

-P=NS-

O-

BASAL

I ISOPROTERENO

FlGURE4.MitrdvdvemistmcedmmeMngisopmbm lleiinfudewEadlpalisntandgMupmeansareshewn.Re dgdkUltCbbgS!S~CiEhbd.N5=llOt~ IndqmhmeofmRralresistancead~y

of Gorlin mitral area on flow changes: Although mitral resistancedid not change during an isoproterenol infusion that induced an increasein mitral valve flow, Gorlin mitral area did increasesignificantly. This diminishes the clinical usefulnessof the Gorlin mitral area, because this value varies depending on the particular hemodynamic conditions during measurement (Figure 5). A fundamental question is whether the orifice size actually changeswith increased flow or this is a factitious change in calculated Gorlin mitral area due to the assumptionsof the formula. In a pulsatile model using bioprosthetic valves, the planimetered valve orifice area varied <0.05 cm2, whereas the Gorlin calculated area increased with valve flow.16Accepting that the Gorlin

mitral area increaseseenwith increasedflow in our patients is not paralleled by an increase in the anatomic orifice area as suggestedby the previously mentioned study, it follows that the Gorlin formula is not completely reliable in the accurate hemodynamic assessment of mitral stenosisunder changing hemodynamic conditions. Our data do not allow for elucidation of the reasonfor the inaccuracy, although the fact that mitral resistance was more constant than Gorlin mitral area under changing hemodynamic conditions suggeststhat the pressure/flow relation to be used in an empirical formula for hemodynamic assessmentmay be approximately linear in the type of patients we studied. The behavior of mitral resistance in our study is in accordancewith the finding of a linear pressure/flow relation that predicts area better than the Gorlin formula.16This can be hypothetically explained by the fact that the Gorlin formula does not account for actual changesin the Gorlin constant. Hydraulic theory and experiments in vitro have shown that the Gorlin constant dependson the ratio of orifice area to the inlet area, and that it is constant only above a high “threshold” Reynolds number, below which it decreases.“J8It has been suggested that Reynolds numbers at this threshold level can be found in aortic stenosiswith low output.*8 The Reynolds number is directly proportional to the entry velocity and the hydraulic depth, which is the quotient of cross-sectional area divided by perimeter.19Becausethe entry velocities in a stenotic mitral valve are lower than in the stenotic aortic valve, and the hydraulic depth of an ellip tical or slit-like orifice is smaller than that for a circular orifice, it is possiblethat Reynolds numbers are low in mitral stenosis,so that the Gorlin constant operatesbelow its maximal steady value and therefore varies with flow. To calculate valve area accurately at this range, a Gorlin constant for each flow would have to be known using a larger constant with a higher flow. Study limitationr: It could be argued that the Gorlin mitral area in our study increasedartificially while un-

Ii/ 2 E E z 7 a 8

FIGURE 5. Go& dtrd m dutwe varsusmiirdvdveflowch8qp.Thedtmges ucexpmssedasapwwntageefths ferdmretll bfidlWV~tO0 iMdlKJV~.llWQorlnlnibrlrveCr

changeciowd&swithmHrdvdvetkw

0

-201 -20

I 0

1 20

1 40

MITRAL FLOW CHANGE

778

0

n =21 r = 0.82 p
I 60

I 80

[o/o]

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 69

I

MARCH 15. 1992

der the effects of isoproterenol due to an augmentation of atria1 left to right shunting, i.e., without an actual increase in mitral valve flow. However, in our study only 6 of 17 patients with isoproterenol infusion were studied after PBMD, so that most patients were studied before the creation of an anatomic interatrial communication. In addition, our criterion for shunt was stricter than previous reports,20and we could not identify a qualitative difference in response between the group before and the group after PBMD. It is unlikely that the findings of our study represent variable shunting through an interatrial septal defect after PBMD. A study with a larger number of patients may show significant changesof mitral resistanceunder increased flow conditions. It is conceivable that under increased flow conditions the hemodynamic efficiency of the stenosis improves.4J7The size and shapeof the inflow portion before a stenosisdetermine the rate at which pressure energy is transformed into flow velocity, a quantity known as coefficient of discharge.tJ7J8T21 With higher flow, part of the increasedhydraulic energy may favorably alter the inflow and orifice shape,so that the pressure gradient is now able to drive a much higher transvalvular flow. Our results indicate that, if this changein mitral resistance occurs, it is less marked than the changesin calculated Gorlin mitral area.

REFERENCES 1. Gorlin R, Gorlin SG. Hydraulic formula for calculation of the area of the stenotic mitral valve, other cardiac valves, and central circulatory shunts.I. Am Heart J 1951;41:1-29. 2. Cohen MV, Gorlin R. Modified orifice equation for the calculation of mitral valve area. Am Heart J 1972;84:839-840. 3. Carabello BA. Advancesin the hemodynamicassessmentof stenotic cardiac

valves. J Am Co/l Cardiol 1987;10:912-919. 4. Gorlin R. Calculations of cardiac valve stenosis:restoring an old conceptfor advancedapplications.J Am Co/l Cardiol 1987;10:920-922. 5. Bathe RJ, Wang Y, JorgensscnCR. Hemodynamic effects of exercise in isolated valvular aortic stenosis.Circulation 1971;44:1003-1013. 6. Ubago JL, Figueroa A, Colman T, Ckhoteco A, Durln C. Hemodynamic factors that affect calculatedorifice areasin the mitral Hancock xenograft valve. Circulation 1980;61:388-394. 7. Czer LS, Gray RJ, Bateman TM, DeRobertis MA, Resser K, Chaux A, Matloff JM. Hemodynamicdifferentiation of pathologicand physiologicstenosis in mitral porcine bioprostheses.J Am Coil Cardiol 1986;7:284-294. 8. Richter HS. Mitral valve area:measurementscanafter catheterization.Report of a case. Circulation 1963;28:451-454. 9. Rodrigo FA. Estimation of valve area and “‘valvular resistance.” A critical study of the physical basisof the methodsemployed.Am Heart J 1953;45:1-12. 10. Ford LE, Feldman T, Chiu C, Carroll JD. Hemodynamic resistanceas a measureof functional impairment in aortic valvular stenosis.Circ Res 1990;66: 1-7. 11. GrossmanW. Profiles in valvular heart disease.In: GrossmanW, Bairn DS, eds.Cardiac Catheterization,Angiographyand Intervention. Philadelphia:Lea & Febiger, 1991:565. 12. Ross Jr J, Braunwald E, Morrow A. Left heart catheterization by the transseptalroute. A descriptionof the techniqueand its applications.Circulation 1960;22:927-934. 13. Croft CH, LipscombK. Modified techniqueof transseptalleft heart catheterization J Am Coil Cardiol 1985;5:904-910. 14. Palacios I, Block PC, Brandi S, Blanco P, Casal H, Pulido JI, Munoz S, D’Empaire G, Ortega MA, JacobsM, VlahakesG. Percutaneousballoon valvotomy for patients with severemitral stenosis.Circulation 1987;75:778-784. 15. HammermeisterKE, Murray JA, BlackmanJR. Revisionof Gorlin constant for calculation of mitral valve area from left heart pressures.Br Heart J 1973;35:392-396. 16. Cannon SR, Richards KL, Crawford M. Hydraulic estimation of stenotic orifice area: a correction of the Gorlin formula. Circulation 1985;71:1170-1178. 17. Burger HC, Van BrummelenAGW, Dannenburg FJ. Theory and experimerits on schematizedmodelsof stenosis.Circ Res 1956;4:425-429. 18. Segal J, Lerner J, Miller C, Mitchell RS, Alderman EA. Popp RL. When shouldDoppler-determinedvalve area be better than the Gorlin formula?: variation in hydraulic constants in low flow states. J Am Co11 Cardiol 1987;9: 1294-1305. 19. Milnor WR. Nonlaminar flow. In: Milnor WR, ed. Hemodynamics.Baltimore: Williams & Wilkins, 198934. 20. PetrossianGA, Tuzcu M, Ziskind AA, Block PC, Palacios I. Atrial septal occlusion improves the accuracy of mitral valve area determination following percutaneousmitral balloonvalvotomy. Cathet Cardiouax Diagn 199 1;22:2l-24. 21. Milnor WR. Hydraulic energies.In: Milnor WR, ed. Hemodynamics.Baltimore: Williams & Wilkins, 1989:24.

MITRAL RESISTANCE 779