- Email: [email protected]
Transesophageal two-dimensional echocardiographic analysis of right ventricular systolic performance indices during coronary artery bypass grafting
Transesophageal Two-dimensional Echocardiographic Analysis of Right Ventricular Systolic Performance Indices During Coronary Artery Bypass Grafting Terence John
Elefteriades,
Rafferty,
MD, Michael
MD, Roberta
Hines,
Durkin,
Sixteen patients (aged 59 f 14 years) undergoing coronary artery bypass surgery were evaluated to delineate the intraoperative course of transesophageal echocardiographic right ventricular (RV) systolic performance indices. Pre-induction data included thermodilution RV ejection fraction (RVEFrc), 0.43 2 0.13, RV end-diastolic volume index (EDVI), 110 2 33 mL/m*, cardiac index (Cl), 3.4 2 1.0 L/min/mZ, RV enddiastolic pressure (EDP), 7.1 ? 4.2 mmHg, and mean pulmonary artery pressure (PAP), 21 ? 6 mmHg. Eleven patients had significant right coronary artery (RCA) disease (>70% occlusion). Five patients arrived with an ongoing nitroglycerin infusion (I to 3 Pg/kg/min), which was maintained intraoperatively. Echocardiographic measurements included longitudinal-axis (LA) and short-axis (SA) planimetered area excursion fractions (2Dm and 2Ds~. respectively) and LA maximal major and minor axis shortening fractions (max majorL* and max minorm, respectively). Hemodynamic measurements included RVEFro, EDVI, Cl, EDP, andPAP. Measurements were determined following induction/endotracheal intubation, following sternotomylpericardiotomy, and after cardiopulmonary bypass (CPB) with the chest open. All patients were maintained on vasodilator therapy post-CPB (nitroglycerin, 1 to 3 pg/kg/min [N = 161 and nitroprusside, 0.5 to 4.5 Pg/kg/min [N = 41) post-CPB. Two patients re-
T
HE VULNERABILITY of the right ventricle (RV) during coronary artery bypass grafting (CABG) is becoming increasingly recognized. Hines and Barash’ have reported two cases in which the intraoperativc onset of ischemia was associated with isolated right ventricular dysfunction. This was manifested by an elevation in RV end-diastolic pressure and a corresponding decrease in thermodilution ejection fraction (RVEF,ro). Mangano* has reported that transient depression of RV function can occur in the immediate post-cardiopulmonary bypass (CPB) period, evidenced by deterioration in RV stroke work index/central venous pressure relationships. In that series, normalization or gradual recovery of function occurred over the succeeding 24 hours. Boldt et aI3 studied the effects of acute volume loading on RV function following CPB. These authors noted an impairment of RV reserve in patients with prolonged aortic cross-clamp time and native right coronary artery (RCA) disease. Such patients demonstrated volume repletion-associated decreases in RVEFrn concomitant with increases in end-diastolic volume index. Finally, Stein et al4 have reported that hemodynamically meaningful RV depression (RVEFrn measurements) can occur postoperatively despite intraoperative preservation
From the Departments of Anesthesiology, Surgery, and Radiology, Yale University School of Medicine, New Haven, CT. Address reprint request,c to Terence Raffegy, MD, Department of Anesthesiology, Yale University School of Medicine, 333 Cedar St, New Haven, CT 06510. Copyright o 1993 by W.B. Saunders Company IOS3-0770/9310702-0007$03.0010 160
MB, BS, FFARCS,
MD, Edward
Prokop,
Stephen
Harris,
MD, and Teresa
MD,
O’Connor,
MPH
ceived inotropic support (epinephrine, 0.2 to 0.3 Pg/ kg/min). CPB was associated with significant decreases in max major axisu and 2DW (P -z 0.05) as compared to measurements determined prior to CPB. Maximum major axisu values pre-CPB were 0.35 ? 0.06 and 0.33 f 0.08 versus post-CPB values of 0.24 f 0.08.20~ values were 0.50 f 0.16 and 0.47 ir 0.10 versus post-CPB values of 0.37 -C 0.11. 2Dsn was maintained (values 0.43 ? 0.17 and 0.44 + 0.18 versus postCPB 0.52 r 0.18). The pre-CPB maximum major axislA/ maximum minor axism relationship was significant (y = 0.71). There was no correlation between these variables post-CPB. Maximum major axislg, changes were unrelated to CPB duration, aortic cross-clamp time, cardioplegia dose, antecedent RV function, and the presence/absence of RCA disease. Significant changes were also unrelated to changes in heart rate. There was no significant change in EDVI, EDP, and F?f@ measurements. These findings may indicate regional inhomogeneity of RV function or a spectrum of differing sensitivities to changes in RV performance of the respective measurement indices, Copyright b; 1993 by W-8. Saunders Company KEY WORDS: right ventricle, dimensional echocardiography
systolic
performance,
two-
of function. ‘These studies are limited to estimates of RV global performance. It can be speculated that measurement of echocardiographic RV systolic performance indices in multiple planes might be helpful in delineating RV regional dysfunction. Echocardiographic measurements, by definition, assess individual ventricular cross-sections. irrespective of their correlation with independent estimates of global ejection fraction in given situations. A dynamic change in the inter-relationships of the echocardiographic measurements could be interpreted as indicating inhomogeneity of performance, given a constant geometry and image acquisition technique. This concept prompted the following combined hemodynamic and echocardiographic study. METHODS Sixteen consecutive patients undergoing uncomplicated CAUG surgery (Table l), fulfilling institutional Human Investigation Committee criteria, were studied. Anti-angina1 therapy (calcium channel and P-adrenergic blocking drugs) was continued to the time of surgery. Five of the patients presented with an ongoing nitroglycerin infusion, which was maintained throughout the operative period (Table 2). Patients satisfied the criteria of no prior cardiac or thoracic surgical procedure and no bundle-branch block that could interfere with assessment of regional wall motion. Anesthetic and surgical management of all cases were by the same attendinganesthesiologist and surgeon, respectively. Premedication was with morphine, 0.1 mgikg, and scopolamine, 0.4 mg, intramuscularly. Anesthesia was induced and maintained with intravenous fentanyl and pancuronium bromide (93 to 113 kg/kg and 0.1 to 0.15 mgikg, respectively). Controlled ventilation (zero end-expiratory pressure) was maintained with an inspired oxygen
Journalof Cardiorhoracic and VascularAnesthewa,
Vol7, No 2
(April),1993:
pp
160.166
RV SYSTOLIC PERFORMANCE INDICES
161
Table 1. Baseline Characteristics
Thermodilution
Age (mean IT SD; years)
59 ? 14
Sex (male:female)
12:4
Previous myocardial infarction
11
Angiographic disease* Right coronary artery
11
Left main coronary artery
4
Left anterior descending artery
15
Left circumflex artery
10
Pre-induction right ventricular indices RVEF,,
0.43 & 0.13
EDVI (mL/mz)
110 * 33
EDP (mmHg)
7.1 ? 4.2
Cl (Llmin/mz)
3.4 2 1.0
PAP (mmHg) 170%
??
fraction
21 ) 6
stenosis. (F102)
of 1.0. Ventilation
(Controller
2, Air-Shields
was delivered Corporation,
ejection
fraction
(RVEFT~),
output
pulmonary artery, and a VdR (-V,) surface electrode10 (SwanGanz catheter 93A-41H-7.5F, American Edwards Laboratories, Irvine, CA) interfacing with an RVEF computer (REF-1, Baxter Healthcare Corp, Santa Ana, CA) featuring automatic constant adjustment with injectate temperature. RVEFTD and CO were determined at end-expiration using lo-mL injections of iced ( <3”C) 5% dextrose in water. Color-flow Doppler evidence of significant tricuspid regurgitation, as defined by a maximal planimetered regurgitant jet area > 0.3 cm2, was not present during the
via a time-cycled
PA) at 10 to 15 breaths per minute and a tidal volume of 10 to 12 mL/kg to maintain normocarbia as defined by arterial blood gas analysis and end-tidal mass spectrometric capnography (Advantage System, Perkin-Elmer Corporation, Pomona, CA). Cardiopulmonary bypass (CPB) with systemic hypothermia (28°C) was performed with the use of a Cobe heart-lung machine, (Cobe Laboratory Inc, Lakewood, CO). A sump drain was inserted in the left atria1 cavity through the right superior pulmonary vein. Topical hypothermia was afforded by the introduction of iced saline into the pericardial well. The hypothermic (4°C) asanguinous cardioplegia solution was delivered via the aortic root. The cardioplegia solution consisted of 5.6% dextrose in 0.9% normal saline with the addition of mannitol, 12.5 g/L, and potassium chloride, 20 mEq/L. Administration was by roller infusion pump via 14.gauge cannula (Aortic Root Cannula, DLP Inc, Grand Rapids, MI) connected by means of 3132 inch polyvinylchloride tubing and delivered at driving pressures of 150 to 200 mmHg.
ventilator
right ventricular
(CO), and intravascular pressure measurements (RV end-diastolic pressure [EDP] and mean pulmonary artery pressure [PAP]) were obtained with a hexalumen pulmonary artery right ventricular ejection fraction catheter.5-9 The catheter was equipped with a 50-msec response thermistor and ventricular,
cardiac
Hatboro,
measurement periods. I1 All measurements were performed in triplicate during sinus rhythm, with the stated result representing an average with less than 10% variation. The washout curve was analyzed by an exponential algorithm.’ The injectate lumen was positioned in the right atrium just above the tricuspid valve in standard fashion by analysis of transduced pressure waveforms as follows: when the transduced injectate port was passed from the right atrium into the right ventricle, as noted by the characteristic change in pressure waveform, the balloon was deflated and the catheter was withdrawn 2 to 3 cm. Values for CO and end-diastolic volume (EDV) were expressed as indices (CI and EDVI, respectively). Two-dimensional echocardiographic images were obtained with a transesophageal 5.0-MHz short-focus phased-array transducer (Imaging System 77020, Hewlett Packard Corp, Andover, MA). Standardized right ventricular views were as follows: (1) The longitudinal-axis (LA) view was provided by a plane transecting the heart from base to apex below the level of the aortic valve with at least the medial portion of the anterior mitral leaflet visible in the far field (Fig 1). This imaging view provided the maximum amplitude of excursion of the lateral border of the tricuspid
Table 2. Pharmacologic Variables PatientNo.
Post-intubation
Pre-CPB
1
0
0
2
0
0
Post-CPB NTG 1 kglkglmin NTG 1 pg/kg/min Epi 0.2 pg/kg/min
3
NTG 1 kg/kg/min
NTG 1 pg/kg/min
NTG 1 pg/kg/min
4
NTG 3 pg/kg/min
NTG 3 kg/kg/min
NTG 3 pg/kg/min
5
NTG 1 pg/kg/min
NTG 1 pg/kg/min
NTG 1 rg/kg/min
6
0
0
NTG 1 pg/kg/min
7
0
0
NTG 1 pg/kg/min
6
0
0
NTG 0.75 kglkgimin
9
NTG 2 pglkglmin
NTG 2 kg/kg/min
NTG 2 kg/kg/min
10
0
0
NTG 1 wg/kg/min
11
0
0
NTG 1 kg/kg/min SNP 4.5 kg/kg/min
12
NTG 1 pglkglmin
NTG 1 pglkglmin
SNP 0.75 pg/kg/min
SNP 0.75 bg/kg/min
NTG 1 pglkgimin
13
0
0
NTG 1 kglkglmin
14
0
0
NTG 1 pg/kg/min
15
0
0
SNP 0.5 pg/kg/min NTG 1 pg/kg/min SNP 0.5 Bg/kg/min Epi 0.5 pg/kg/min 16
0
0
NTG 1 pglkglmin SNP 1 wg/kg/min
Abbreviations: NTG, nitroglycerin; SNP, nitroprusside; Epi, epinephrine.
162
Fig 1. Two-dimensional echocardiographic longitudinal (top) and short-axis (bottom) images.
annulus consistent with simultaneous visualization of the free wall-septal junction; (2) short-axis (SA) images were obtained at the previously described transesophageal left ventricular papillary muscle level view, with the transducer oriented to position the RV ventricle close to the center of the field to optimize lateral resolution12~‘” (Fig 1). Endocardial image clarity was optimized by gray scale adjustment. Videotapes, recorded in real-time simultaneously with calibration grids and an ECG, were reviewed on an off-line analysis system providing forward, reverse. slow motion, and freeze viewing formats with the areas of interest traced by a hand-held digitizer. Adequate image clarity was obtained in all data sets. Special attention was paid to image right ventricular dimensions at their maximal excursion (minimum three cardiac cycles) and at zero end-expiratory pressure to obviate the effects of positive-pressure ventilation and cardiac motion on measured values. End-diastolic and end-systolic measurements included LA and SA planimetered area excursion fractions (2DrA and 2DsA, respectively)‘4.‘5 (Fig 2). and LA maximum major and minor axes shortening fractions (max major axisrA and minor axist,A, respectively) (Fig 3).*
*Maximal major axis shortening fraction rationale: The longitudinal major axis of the ventricle bisects the mid-annular plane of the tricuspid valve.1h The maximal major axis connects the apex of the ventricle to the lateral border of the annulus. The amplitude of motion from end-diastole to end-systole represents the excursion of this latter annular border. This measurement embodies the concept of Kaul et al” (tricuspid annular plane systolic excursion [TAPSE]), modified to include specific localization of the free wall-septal junction during both systole and diastole to confer validity under conditions of post-CPB ventricular translocation,
Data wcrc obtained following endotracheal intubation, after sternotomyipericardiotomy. and at the conclusion of the procedure following removal of the arterial and venous drainage cannulae and resumption of sinus rhythm (prior to chest closure). Pre-CPB echocardiographic measurements were analyzed in terms of max major axist.A/2Dr ,\ and max minor axis&2DLA relationships, respectively, to determine whether sternotomy and pericardiotomy-induced lack of restraint might preclude valid data comparison between post-intubation (closed chest) and open chest measurements. There was no evidence to reject the null hypothesis that the respective regressions were parallel and coincident (P > 0.05). Further statistical analyses were as follows: data sets were compared by repeated measures analysis of variance: interrelationships were analyzed by correlation analysis. All patients completed the protocol. Echocardiographic measurement variability was assessed by selecting 257~ of profiles for re-analysis by means of a random
and with the expression of the systolic excursion as a ratio rather than in terms of absolute values to facilitate data intercomparison and interpretation. Maximal minor axis shortening fraction rationale: The longitudinal minor axis of the ventricle is the distance between the septal and free wall endocardium in a plane perpendicular to the major axis, with the precise supero-inferior position of the maximal minor axis varying according to the intravascular volume status and the presence or absence of ventricular hypertrophy.rh.rn The maximal minor axis shortening fraction has previously been proposed as a right ventricular contractility index,ry and determinations are comparable in accuracy/inaccuracy to planimetered area excursion fraction measurements.?”
RV SYSTOLIC
PERFORMANCE
163
INDICES
~DLA
Diastole
MAX MAJOR AXIQ
Systole
Diastole
2DSA
Diastole
Systole
MAX MINOR AXIS,,
Systole
Fig 2. Representation of longitudinal- and short-axis plenimetered area excursion fraction (2Du and 2DSA, respectively).
number table. Observer
variabilities were calculated as estimate 1 minus estimate 2 divided by mean estimate 1 and 2, expressed in percent. Mean intra- and inter-observer variabilities did not exceed 8% and lo%, respectively.
Diastole
Systole
Fig 3. Representation of longitudinal axis maximal major and minor axis shortening fractions (max major axisu and mex minor exisw, respectively).
RESULTS
Durations of CPB, aortic cross-clamp time, and cardioplegia dose were 83 2 32 minutes, 48 +- 19 minutes, and 1.6 ? 0.4 L, respectively. AI1 patients were maintained on a nitroglycerin infusion (range, 1 to 3 pg/kg/min) following CPB. Five patients received a nitroprusside infusion (range, 0.75 to 3 pg/kg/min). Two patients received inotropic support (epinephrine, 0.2 and 0.5 kg/kg/min, respectively) (Table 2). CPB was associated with significant decreases in maximum major axisLA and 2DLA (P < 0.05) as compared to post-intubation and pre-CPB measurement sets (Table 3). Pre- and post-CPB maximum major axisLA and 2DLA measurements were significantly related (y = 0.64 and 0.65, respectively) (Fig 4). Pre-CPB maximum major axisLA and maximum minor axisLA measurements were significantly related (y = 0.71). There was no significant relationship between these latter variables following CPB (Fig 5). Heart rate (HR) increased significantly (P < 0.05) following CPB. Echocardiographic changes were unrelated to changes in heart rate. There were no significant changes
in RVEFro, CI, EDVI, EDP, and PAP (Table 3). Significant changes were unrelated to CPB duration, aortic cross-clamp time, cardioplegia dose, antecedent RV function, and the presence or absence of RCA disease. DISCUSSION
To the authors’ knowledge, exclusive of interventricular septum data, this report represents the first echocardiographic study of RV systolic function during CABG. The echocardiographic measurements depict the performance of individual RV cross-sections. Significant decreases occurred following CPB. These decreases were restricted to longitudinal axis measurements; specifically, maximal major axis and planimetered area values. The subsequent course of these changes was not studied. Accordingly, it can only be speculated whether spontaneous recovery occurred or whether changes were premonitory to further deterioration. In this series only two patients received inotropic support
RAFFEHY
164
i i .i,
04c
05s No lr. pcoc1 T = 0 64
0 50 r
4 m
045
j
.
2 G '5 I
04oc
z I
035
z! 0
.
0 . ?? e
. .
.
.
0 30.-
.
.
0
. .
s! CL 025
.
--
. a
.
0.15
0
e
0
0201_
I
02
??
.
??
i___-.a
03
04
05
I..I___1____,_i 02 03
0 10
06
04
05
Post-CPB 2D,,
Pre-CPB 2D,,
absence of RCA. Finally, the data suggest that differences in antecedent RV function, HR, and loading conditions could not explain the results. Accordingly, a precise etiology for the findings could not be documented. However, the RV has been shown to be susceptible to ambient rewarming the myocardial protection afduring CPB,27 mitigating forded by hypothermia, and it seems reasonable to postulate such an etiology. The constancy of the association between the longitudinal axis maximal major axis area values suggests that decreases were predominantly due to changes along the maximal major axis. The short-axis measurement plane transects that of the maximal major axis. Short-axis planimetered area values were maintained following CPB. This could be interpreted to suggest that the decreases were
following CPB. The involved dosages were small and this variable was unlikely to have played a major role. Vasodilator therapy was a facet of management. This is reflected by the modest end-diastolic volume indices in this population. Such therapy was relatively uniform. It should be noted that this factor may have had a direct impact on image quality by maintaining the ventricular wall close to the high-frequency transesophageal transducer and, thus, optimizing lateral resolution. This study was unable to define a relationship between the echocardiographic changes and either aortic crossclamp time or cardioplegia dose, variables commonly associated with ventricular dysfunction.3J-24 Similarly, though preexisting RCA disease can limit adequate cardioplegia delivery,3J5,2h changes were unrelated to the presence or
N = 16 p< 0003 r:oi1
0504 .?? 2
-l J .z
045.
040.
E a
-I 5 'm' I
. .
. .
g 0
030
.
035
.
. .
01
??
’
02
. 0
.
s
0.15 .
.
’
. .
LL
.
iI-
0.3
0
04
Pre-CPB Max Minor AxisLA
05
.
025-
J .
. .
1 2 kt! 0.20"
030-
020’
I
.
. .
025.
??
I 035
s I
N=16 NS
.
z '5 zs
.
040
0 55
Fig 5. Pre-cardiopulmonary and post-cardiopulmonary bypass longitudinal axis maximal major axis shortening fraction (max major ax&J versus longitudinal axis maximal minor axis shortening fraction (max minor axis&
06
Fig 4. Pm-cardiopulmonary and post-cardiopulmonary bypass longitudinal axis maximal major axis shortening fraction (max major axisJ verws longitudinal axis planimetered area excursion fraction (P&J.
I
.
. . .
,oL_I-.-J 01
02
0.3
0.4
Post-CPB Max Minor AxisLA
05
RV SYSTOLIC PERFORMANCE INDICES
165
Table 3. Hemodynamic and Echocardiagraphic
Values
Postintubation
RVEF,, Cl (L/minim*) HR (beats/min)
PWCPB
Post-CPB
0.41 ‘- 0.14
0.38 k 0.12
0.37 k 0.12
3.0 k 0.9
2.6 2 0.5
2.8 2 0.6
67k
92 k 12*
742
17
13
Diastolic max major axisLA (cm) Max major axisLA
7.4 + 1.2
7.8 + 1.3
7.3 f 1.2
0.35 + 0.06
0.33 2 0.08
0.24 r 0.08’
Diastolic max minor axisLA(cm) Max minor axisLA
4.2 + 1.0
4.4 -c 0.7
4.1 f 1.1
0.34 + 0.13
0.32 f 0.11
0.27 + 0.10
Diastolic 2D,, (cm21
25.0 k 6.9
24.8 + 6.0
23.1 + 6.7
~DLA
0.50 + 0.16
0.47 + 0.10
0.37 + 0.11*
Diastolic 2Dsn (cm2)
19.8 ? 10.4
16.7 + 7.3
15.9 + 6.9
2Dsa
0.43 5 0.17
0.44 + 0.18
0.52 + 0.18
EDVI (mL/m*)
100 2 25
105 5 25
88 t 33
EDP (mmHg)
6.0 2 4.1
4.5 2 3.2
5.4 2 3.2
PAP (mmHg)
17.8 -t 4.8
16.5 + 6.0
16.8 2 5.1
*Significantly different (P < 0.05) than post-intubation and pre-CPB values.
restricted to the upper portion of the RV. Further interpretation is theoretically possible because of the orthogonal relationship of the maximal major and minor axes. The CPB-associated dissociation of shortening fraction measurements determined in these axes suggests that decreases were confined to the base of the ventricle. A more definitive localization might have been afforded by centerline motion analysis. However, inadequate epicardial definition precluded such an evaluation in this series. The above interpretation is, of course, indirect and dependent on the absence of widely separated areas of local dysfunction. In addition, it should be borne in mind that the maximal minor axis measurements would be expected to be affected by dyskinesia of the interventricular septum. Paradoxical septal motion, left ventricular systolic correction of diastolic right ventricular displacement of the septum into the left ventricular cavity, would be anticipated to result in exaggeration of systolic excursion of the maximal minor axis. This would be expected to generate a shortening value directly reflecting right and left ventricular performance.
There are no published data relating to this issue. There is a precedent for preoperative echocardiographic estimates of cardiac systolic function to delineate spurious findings. Lehmann et al** have demonstrated that the apparently high prevalence of dyskinetic interventricular septal motion following CABG determined by echocardiography may be artifactual and, at least in part, due to cardiac translocation. Accordingly, it is important to address measurement technique limitations specifically referable to this study. Technical considerations pertaining to the echocardiographic measurements include the possibility that postCPB foreshortening of right ventricular images may have been responsible for the results. However, pre-CPB and post-CPB measured loading conditions were comparable. Further, systolic shortening measurements were expressed as a function of corresponding end-diastolic dimensions, and foreshortening would be expected to minimize decreases in these measurements. The complex geometry of the RV must also be considered. Inconsistencies in shape could conceivably furnish differing regions for analysis despite meticulous standardization of image acquisition technique. Certainly, at least in terms of echocardiographic estimation of RV volumes, Levine et alz9 have noted that findings derived from even multiple planes of interrogation may be suspect in individual cases. Again, the loading conditions during each of the measurement periods were not disparate, suggesting constancy of shape. A further issue that must be addressed is that the echocardiographic .and RVEFro findings, superficially, appear to be contradictory. Decreases in longitudinal axis performance indices occurred in the face of maintenance of RVEFm values. Though short-axis indices increased, this increase was not statistically significant, invoking the question as to how overall RVEF was maintained. It can only be assumed that further compensation occurred in areas that were not interrogated, such as the outflow tract, or in regions that were not accessible to transverse plane analysis. Equally, it may well be that the thermodilution technique estimates were less sensitive indicators of RV global performance than the echocardiographic measurements.
REFERENCES
1. Hines R, Barash P: Intraoperative right ventricular dysfunction detected by a right ventricular ejection fraction catheter. J Clin Monitoring 2:206-208,1986 2. Mangano D: Biventricular function after myocardial revascularization in humans: Deterioration and recovery patterns during the first 24 hours. Anesthesiology 62:571-577,1985 3. Boldt J, Kling D, Moosdorf R, et al: Influence of acute volume loading on right ventricular function after cardiopulmonary bypass. Crit Care Med 17:518-525,1989 4. Stein K, Breisblatt W, Wolfe C, et al: Depression and recovery of right ventricular function after cardiopulmonary bypass. Crit Care Med 18:1197-1200, 1990 5. Kay H, Afshari M, Barash P, et al: Measurement of ejection fraction by thermal dilution techniques. J Surg Res 34:337-346, 1983 6. Biondi J, Hines R, Rafferty T, et al: The effect of high-
frequency positive-pressure ventilation on right and left ventricular function. Anesth Analg 65:679-682, 1986 7. Dhainaut F-F, Brunet F, Monsallier J, et al: Bedside evaluation of right ventricular performance using a rapid computerized thermodilution method. Crit Care Med 15:148-152,1987 8. Urban P, Scheidegger D, Gabathuler J, et al: Thermodilution determination of right ventricular volume and ejection fraction: A comparison with biplane angiography. Crit Care Med 15:652-655, 1987 9. Voelker W, Gruber H, Ickrath 0, et al: Determination of right ventricular ejection fraction by thermodilution technique-A comparison to biplane cineventriculography. Intensive Care Med 14:461-466,1988 10. Klein H, Tordjman T, Ninio R, et al: The early recognition of right ventricular infarction: Diagnostic accuracy of the electrocardiographic V4R lead. Circulation 67:558-565, 1983
RAFFEHiV
166
11. Rafferty T, Durkin M, Mathew regurgitation: Color-flow Dopplerversus ography. Anesth Analg 72:S219, 1991
J: Estimation of tricuspid saline contrast echocardi-
12. Schluter M, Hinrichs A, Thier W, et al: ‘Transesophageal two-dimensional echocardiography: Comparison of ultrasonic and anatomic sections. Am J Cardiol 53:1173-1178, 19X4 13. Seward J, Khanderia echocardiography: Technique, tion, and clinical applications.
B, Oh J, et al: Transesophageal anatomic correlations. implementaMayo Clin Proc 63649-680, 1988
14. Jardin F, Gueret P, Dubourg 0, et al: Right ventricular volumes by thermodilution in the adult respiratory distress syndrome: A comparative study using two-dimensional echocardiography as a reference. Chest 88:34-39, 1985 15. Jardin F, Brun-Ney D, Hardy A, et al: Combined thermodilution and two-dimensional echocardiographic evaluation of right ventricular function during respiratory support with PEEP. Chest 99:162-168, 1991 16. Weyman H: Cross-sectional Echocardiography. phia, PA, Lea and Febiger, 1982. pp 382-395 17. Kaul S, Tei C, Hopkins J. Shah ventricular function using two-dimensional Heart J 107:526-531, 1984
Philadel-
P: Assessment of right echocardiography. Am
18. Bommer W, Weinert L, Neumann A, et al: Determination of right atrial and right ventricular size by two-dimensional echocardiography. Circulation 60:91-100, 1979 19. Feigenbaum H: Echocardiography. and Febiger. 1986, pp 158-167 20. Durkin
M, Rafferty
Philadelphia,
T: Two-dimensional
PA, Lea
echocardiographic
ti Al
determination of rtght ventricular- maximum minor- axis duncrrsrori fractional shortening. Anesthesiology ?3:AS42. 1990 2 I. Braimhridge M: Myocardial protection. in Taylor I(. (c(i) C‘ardtopulmonary Bypass-Principle and Management. London: Chapman and Ilall. 19X6, pp 375-39X 22. Takach 1‘. Glassman I Milewicz A. et al: C’ontmuous measurement of intramyocardial pH: relative importance of hypothermia and cardioplegic perfusion pressure and temperature Ann Thorac Surg 42:365-371. I%h 2.1.Behrendt D. Jochim K: Effect of temperature of cardiople gic solution. J Thorac C‘ardiovasc Surg 76:3.53-356. 1978 24. Chiu R. Blundell P, Scott J. Cain S: The importance ot monitoring intramyocardial temperature during hypothermic myocardial protection. Ann Thorac Surg 28:317-322, 1978 25. Boldt J. Kling D. Dapper I;. Hempelmann G: Myocardial temperature during cardiac operations: Influence on right ventricular function. J Thorac Cardiovasc Surg 100:562-568, 1990 26. Becker H, Vinter-Johansen .I. Buckberg G: Critical importance of ensuring cardioplegic delivery with coronary stenoses. .I Thorac Cardiovasc SurgXl:507-515, 1981 27. Gonzalez A. Brandon T. Fortune R: Acute right ventricular failure is caused by inadequate right ventricular hypothermia. J Thorac Cardiovasc Surg X9:38&39X. 19X5 2X. Lehmann K. Lee F. McKenzie W. et al: Onset of altered interventricular septal motion during cardiac surgery. Circulation 82:1325-13.34. 1990 29. Levine R, Gibson T. Aretz T. et al: Echocardiographic measurement of right ventricular volume. Circulation 69:497-SOS. 1984
Elefteriades,
Rafferty,
MD, Michael
MD, Roberta
Hines,
Durkin,
Sixteen patients (aged 59 f 14 years) undergoing coronary artery bypass surgery were evaluated to delineate the intraoperative course of transesophageal echocardiographic right ventricular (RV) systolic performance indices. Pre-induction data included thermodilution RV ejection fraction (RVEFrc), 0.43 2 0.13, RV end-diastolic volume index (EDVI), 110 2 33 mL/m*, cardiac index (Cl), 3.4 2 1.0 L/min/mZ, RV enddiastolic pressure (EDP), 7.1 ? 4.2 mmHg, and mean pulmonary artery pressure (PAP), 21 ? 6 mmHg. Eleven patients had significant right coronary artery (RCA) disease (>70% occlusion). Five patients arrived with an ongoing nitroglycerin infusion (I to 3 Pg/kg/min), which was maintained intraoperatively. Echocardiographic measurements included longitudinal-axis (LA) and short-axis (SA) planimetered area excursion fractions (2Dm and 2Ds~. respectively) and LA maximal major and minor axis shortening fractions (max majorL* and max minorm, respectively). Hemodynamic measurements included RVEFro, EDVI, Cl, EDP, andPAP. Measurements were determined following induction/endotracheal intubation, following sternotomylpericardiotomy, and after cardiopulmonary bypass (CPB) with the chest open. All patients were maintained on vasodilator therapy post-CPB (nitroglycerin, 1 to 3 pg/kg/min [N = 161 and nitroprusside, 0.5 to 4.5 Pg/kg/min [N = 41) post-CPB. Two patients re-
T
HE VULNERABILITY of the right ventricle (RV) during coronary artery bypass grafting (CABG) is becoming increasingly recognized. Hines and Barash’ have reported two cases in which the intraoperativc onset of ischemia was associated with isolated right ventricular dysfunction. This was manifested by an elevation in RV end-diastolic pressure and a corresponding decrease in thermodilution ejection fraction (RVEF,ro). Mangano* has reported that transient depression of RV function can occur in the immediate post-cardiopulmonary bypass (CPB) period, evidenced by deterioration in RV stroke work index/central venous pressure relationships. In that series, normalization or gradual recovery of function occurred over the succeeding 24 hours. Boldt et aI3 studied the effects of acute volume loading on RV function following CPB. These authors noted an impairment of RV reserve in patients with prolonged aortic cross-clamp time and native right coronary artery (RCA) disease. Such patients demonstrated volume repletion-associated decreases in RVEFrn concomitant with increases in end-diastolic volume index. Finally, Stein et al4 have reported that hemodynamically meaningful RV depression (RVEFrn measurements) can occur postoperatively despite intraoperative preservation
From the Departments of Anesthesiology, Surgery, and Radiology, Yale University School of Medicine, New Haven, CT. Address reprint request,c to Terence Raffegy, MD, Department of Anesthesiology, Yale University School of Medicine, 333 Cedar St, New Haven, CT 06510. Copyright o 1993 by W.B. Saunders Company IOS3-0770/9310702-0007$03.0010 160
MB, BS, FFARCS,
MD, Edward
Prokop,
Stephen
Harris,
MD, and Teresa
MD,
O’Connor,
MPH
ceived inotropic support (epinephrine, 0.2 to 0.3 Pg/ kg/min). CPB was associated with significant decreases in max major axisu and 2DW (P -z 0.05) as compared to measurements determined prior to CPB. Maximum major axisu values pre-CPB were 0.35 ? 0.06 and 0.33 f 0.08 versus post-CPB values of 0.24 f 0.08.20~ values were 0.50 f 0.16 and 0.47 ir 0.10 versus post-CPB values of 0.37 -C 0.11. 2Dsn was maintained (values 0.43 ? 0.17 and 0.44 + 0.18 versus postCPB 0.52 r 0.18). The pre-CPB maximum major axislA/ maximum minor axism relationship was significant (y = 0.71). There was no correlation between these variables post-CPB. Maximum major axislg, changes were unrelated to CPB duration, aortic cross-clamp time, cardioplegia dose, antecedent RV function, and the presence/absence of RCA disease. Significant changes were also unrelated to changes in heart rate. There was no significant change in EDVI, EDP, and F?f@ measurements. These findings may indicate regional inhomogeneity of RV function or a spectrum of differing sensitivities to changes in RV performance of the respective measurement indices, Copyright b; 1993 by W-8. Saunders Company KEY WORDS: right ventricle, dimensional echocardiography
systolic
performance,
two-
of function. ‘These studies are limited to estimates of RV global performance. It can be speculated that measurement of echocardiographic RV systolic performance indices in multiple planes might be helpful in delineating RV regional dysfunction. Echocardiographic measurements, by definition, assess individual ventricular cross-sections. irrespective of their correlation with independent estimates of global ejection fraction in given situations. A dynamic change in the inter-relationships of the echocardiographic measurements could be interpreted as indicating inhomogeneity of performance, given a constant geometry and image acquisition technique. This concept prompted the following combined hemodynamic and echocardiographic study. METHODS Sixteen consecutive patients undergoing uncomplicated CAUG surgery (Table l), fulfilling institutional Human Investigation Committee criteria, were studied. Anti-angina1 therapy (calcium channel and P-adrenergic blocking drugs) was continued to the time of surgery. Five of the patients presented with an ongoing nitroglycerin infusion, which was maintained throughout the operative period (Table 2). Patients satisfied the criteria of no prior cardiac or thoracic surgical procedure and no bundle-branch block that could interfere with assessment of regional wall motion. Anesthetic and surgical management of all cases were by the same attendinganesthesiologist and surgeon, respectively. Premedication was with morphine, 0.1 mgikg, and scopolamine, 0.4 mg, intramuscularly. Anesthesia was induced and maintained with intravenous fentanyl and pancuronium bromide (93 to 113 kg/kg and 0.1 to 0.15 mgikg, respectively). Controlled ventilation (zero end-expiratory pressure) was maintained with an inspired oxygen
Journalof Cardiorhoracic and VascularAnesthewa,
Vol7, No 2
(April),1993:
pp
160.166
RV SYSTOLIC PERFORMANCE INDICES
161
Table 1. Baseline Characteristics
Thermodilution
Age (mean IT SD; years)
59 ? 14
Sex (male:female)
12:4
Previous myocardial infarction
11
Angiographic disease* Right coronary artery
11
Left main coronary artery
4
Left anterior descending artery
15
Left circumflex artery
10
Pre-induction right ventricular indices RVEF,,
0.43 & 0.13
EDVI (mL/mz)
110 * 33
EDP (mmHg)
7.1 ? 4.2
Cl (Llmin/mz)
3.4 2 1.0
PAP (mmHg) 170%
??
fraction
21 ) 6
stenosis. (F102)
of 1.0. Ventilation
(Controller
2, Air-Shields
was delivered Corporation,
ejection
fraction
(RVEFT~),
output
pulmonary artery, and a VdR (-V,) surface electrode10 (SwanGanz catheter 93A-41H-7.5F, American Edwards Laboratories, Irvine, CA) interfacing with an RVEF computer (REF-1, Baxter Healthcare Corp, Santa Ana, CA) featuring automatic constant adjustment with injectate temperature. RVEFTD and CO were determined at end-expiration using lo-mL injections of iced ( <3”C) 5% dextrose in water. Color-flow Doppler evidence of significant tricuspid regurgitation, as defined by a maximal planimetered regurgitant jet area > 0.3 cm2, was not present during the
via a time-cycled
PA) at 10 to 15 breaths per minute and a tidal volume of 10 to 12 mL/kg to maintain normocarbia as defined by arterial blood gas analysis and end-tidal mass spectrometric capnography (Advantage System, Perkin-Elmer Corporation, Pomona, CA). Cardiopulmonary bypass (CPB) with systemic hypothermia (28°C) was performed with the use of a Cobe heart-lung machine, (Cobe Laboratory Inc, Lakewood, CO). A sump drain was inserted in the left atria1 cavity through the right superior pulmonary vein. Topical hypothermia was afforded by the introduction of iced saline into the pericardial well. The hypothermic (4°C) asanguinous cardioplegia solution was delivered via the aortic root. The cardioplegia solution consisted of 5.6% dextrose in 0.9% normal saline with the addition of mannitol, 12.5 g/L, and potassium chloride, 20 mEq/L. Administration was by roller infusion pump via 14.gauge cannula (Aortic Root Cannula, DLP Inc, Grand Rapids, MI) connected by means of 3132 inch polyvinylchloride tubing and delivered at driving pressures of 150 to 200 mmHg.
ventilator
right ventricular
(CO), and intravascular pressure measurements (RV end-diastolic pressure [EDP] and mean pulmonary artery pressure [PAP]) were obtained with a hexalumen pulmonary artery right ventricular ejection fraction catheter.5-9 The catheter was equipped with a 50-msec response thermistor and ventricular,
cardiac
Hatboro,
measurement periods. I1 All measurements were performed in triplicate during sinus rhythm, with the stated result representing an average with less than 10% variation. The washout curve was analyzed by an exponential algorithm.’ The injectate lumen was positioned in the right atrium just above the tricuspid valve in standard fashion by analysis of transduced pressure waveforms as follows: when the transduced injectate port was passed from the right atrium into the right ventricle, as noted by the characteristic change in pressure waveform, the balloon was deflated and the catheter was withdrawn 2 to 3 cm. Values for CO and end-diastolic volume (EDV) were expressed as indices (CI and EDVI, respectively). Two-dimensional echocardiographic images were obtained with a transesophageal 5.0-MHz short-focus phased-array transducer (Imaging System 77020, Hewlett Packard Corp, Andover, MA). Standardized right ventricular views were as follows: (1) The longitudinal-axis (LA) view was provided by a plane transecting the heart from base to apex below the level of the aortic valve with at least the medial portion of the anterior mitral leaflet visible in the far field (Fig 1). This imaging view provided the maximum amplitude of excursion of the lateral border of the tricuspid
Table 2. Pharmacologic Variables PatientNo.
Post-intubation
Pre-CPB
1
0
0
2
0
0
Post-CPB NTG 1 kglkglmin NTG 1 pg/kg/min Epi 0.2 pg/kg/min
3
NTG 1 kg/kg/min
NTG 1 pg/kg/min
NTG 1 pg/kg/min
4
NTG 3 pg/kg/min
NTG 3 kg/kg/min
NTG 3 pg/kg/min
5
NTG 1 pg/kg/min
NTG 1 pg/kg/min
NTG 1 rg/kg/min
6
0
0
NTG 1 pg/kg/min
7
0
0
NTG 1 pg/kg/min
6
0
0
NTG 0.75 kglkgimin
9
NTG 2 pglkglmin
NTG 2 kg/kg/min
NTG 2 kg/kg/min
10
0
0
NTG 1 wg/kg/min
11
0
0
NTG 1 kg/kg/min SNP 4.5 kg/kg/min
12
NTG 1 pglkglmin
NTG 1 pglkglmin
SNP 0.75 pg/kg/min
SNP 0.75 bg/kg/min
NTG 1 pglkgimin
13
0
0
NTG 1 kglkglmin
14
0
0
NTG 1 pg/kg/min
15
0
0
SNP 0.5 pg/kg/min NTG 1 pg/kg/min SNP 0.5 Bg/kg/min Epi 0.5 pg/kg/min 16
0
0
NTG 1 pglkglmin SNP 1 wg/kg/min
Abbreviations: NTG, nitroglycerin; SNP, nitroprusside; Epi, epinephrine.
162
Fig 1. Two-dimensional echocardiographic longitudinal (top) and short-axis (bottom) images.
annulus consistent with simultaneous visualization of the free wall-septal junction; (2) short-axis (SA) images were obtained at the previously described transesophageal left ventricular papillary muscle level view, with the transducer oriented to position the RV ventricle close to the center of the field to optimize lateral resolution12~‘” (Fig 1). Endocardial image clarity was optimized by gray scale adjustment. Videotapes, recorded in real-time simultaneously with calibration grids and an ECG, were reviewed on an off-line analysis system providing forward, reverse. slow motion, and freeze viewing formats with the areas of interest traced by a hand-held digitizer. Adequate image clarity was obtained in all data sets. Special attention was paid to image right ventricular dimensions at their maximal excursion (minimum three cardiac cycles) and at zero end-expiratory pressure to obviate the effects of positive-pressure ventilation and cardiac motion on measured values. End-diastolic and end-systolic measurements included LA and SA planimetered area excursion fractions (2DrA and 2DsA, respectively)‘4.‘5 (Fig 2). and LA maximum major and minor axes shortening fractions (max major axisrA and minor axist,A, respectively) (Fig 3).*
*Maximal major axis shortening fraction rationale: The longitudinal major axis of the ventricle bisects the mid-annular plane of the tricuspid valve.1h The maximal major axis connects the apex of the ventricle to the lateral border of the annulus. The amplitude of motion from end-diastole to end-systole represents the excursion of this latter annular border. This measurement embodies the concept of Kaul et al” (tricuspid annular plane systolic excursion [TAPSE]), modified to include specific localization of the free wall-septal junction during both systole and diastole to confer validity under conditions of post-CPB ventricular translocation,
Data wcrc obtained following endotracheal intubation, after sternotomyipericardiotomy. and at the conclusion of the procedure following removal of the arterial and venous drainage cannulae and resumption of sinus rhythm (prior to chest closure). Pre-CPB echocardiographic measurements were analyzed in terms of max major axist.A/2Dr ,\ and max minor axis&2DLA relationships, respectively, to determine whether sternotomy and pericardiotomy-induced lack of restraint might preclude valid data comparison between post-intubation (closed chest) and open chest measurements. There was no evidence to reject the null hypothesis that the respective regressions were parallel and coincident (P > 0.05). Further statistical analyses were as follows: data sets were compared by repeated measures analysis of variance: interrelationships were analyzed by correlation analysis. All patients completed the protocol. Echocardiographic measurement variability was assessed by selecting 257~ of profiles for re-analysis by means of a random
and with the expression of the systolic excursion as a ratio rather than in terms of absolute values to facilitate data intercomparison and interpretation. Maximal minor axis shortening fraction rationale: The longitudinal minor axis of the ventricle is the distance between the septal and free wall endocardium in a plane perpendicular to the major axis, with the precise supero-inferior position of the maximal minor axis varying according to the intravascular volume status and the presence or absence of ventricular hypertrophy.rh.rn The maximal minor axis shortening fraction has previously been proposed as a right ventricular contractility index,ry and determinations are comparable in accuracy/inaccuracy to planimetered area excursion fraction measurements.?”
RV SYSTOLIC
PERFORMANCE
163
INDICES
~DLA
Diastole
MAX MAJOR AXIQ
Systole
Diastole
2DSA
Diastole
Systole
MAX MINOR AXIS,,
Systole
Fig 2. Representation of longitudinal- and short-axis plenimetered area excursion fraction (2Du and 2DSA, respectively).
number table. Observer
variabilities were calculated as estimate 1 minus estimate 2 divided by mean estimate 1 and 2, expressed in percent. Mean intra- and inter-observer variabilities did not exceed 8% and lo%, respectively.
Diastole
Systole
Fig 3. Representation of longitudinal axis maximal major and minor axis shortening fractions (max major axisu and mex minor exisw, respectively).
RESULTS
Durations of CPB, aortic cross-clamp time, and cardioplegia dose were 83 2 32 minutes, 48 +- 19 minutes, and 1.6 ? 0.4 L, respectively. AI1 patients were maintained on a nitroglycerin infusion (range, 1 to 3 pg/kg/min) following CPB. Five patients received a nitroprusside infusion (range, 0.75 to 3 pg/kg/min). Two patients received inotropic support (epinephrine, 0.2 and 0.5 kg/kg/min, respectively) (Table 2). CPB was associated with significant decreases in maximum major axisLA and 2DLA (P < 0.05) as compared to post-intubation and pre-CPB measurement sets (Table 3). Pre- and post-CPB maximum major axisLA and 2DLA measurements were significantly related (y = 0.64 and 0.65, respectively) (Fig 4). Pre-CPB maximum major axisLA and maximum minor axisLA measurements were significantly related (y = 0.71). There was no significant relationship between these latter variables following CPB (Fig 5). Heart rate (HR) increased significantly (P < 0.05) following CPB. Echocardiographic changes were unrelated to changes in heart rate. There were no significant changes
in RVEFro, CI, EDVI, EDP, and PAP (Table 3). Significant changes were unrelated to CPB duration, aortic cross-clamp time, cardioplegia dose, antecedent RV function, and the presence or absence of RCA disease. DISCUSSION
To the authors’ knowledge, exclusive of interventricular septum data, this report represents the first echocardiographic study of RV systolic function during CABG. The echocardiographic measurements depict the performance of individual RV cross-sections. Significant decreases occurred following CPB. These decreases were restricted to longitudinal axis measurements; specifically, maximal major axis and planimetered area values. The subsequent course of these changes was not studied. Accordingly, it can only be speculated whether spontaneous recovery occurred or whether changes were premonitory to further deterioration. In this series only two patients received inotropic support
RAFFEHY
164
i i .i,
04c
05s No lr. pcoc1 T = 0 64
0 50 r
4 m
045
j
.
2 G '5 I
04oc
z I
035
z! 0
.
0 . ?? e
. .
.
.
0 30.-
.
.
0
. .
s! CL 025
.
--
. a
.
0.15
0
e
0
0201_
I
02
??
.
??
i___-.a
03
04
05
I..I___1____,_i 02 03
0 10
06
04
05
Post-CPB 2D,,
Pre-CPB 2D,,
absence of RCA. Finally, the data suggest that differences in antecedent RV function, HR, and loading conditions could not explain the results. Accordingly, a precise etiology for the findings could not be documented. However, the RV has been shown to be susceptible to ambient rewarming the myocardial protection afduring CPB,27 mitigating forded by hypothermia, and it seems reasonable to postulate such an etiology. The constancy of the association between the longitudinal axis maximal major axis area values suggests that decreases were predominantly due to changes along the maximal major axis. The short-axis measurement plane transects that of the maximal major axis. Short-axis planimetered area values were maintained following CPB. This could be interpreted to suggest that the decreases were
following CPB. The involved dosages were small and this variable was unlikely to have played a major role. Vasodilator therapy was a facet of management. This is reflected by the modest end-diastolic volume indices in this population. Such therapy was relatively uniform. It should be noted that this factor may have had a direct impact on image quality by maintaining the ventricular wall close to the high-frequency transesophageal transducer and, thus, optimizing lateral resolution. This study was unable to define a relationship between the echocardiographic changes and either aortic crossclamp time or cardioplegia dose, variables commonly associated with ventricular dysfunction.3J-24 Similarly, though preexisting RCA disease can limit adequate cardioplegia delivery,3J5,2h changes were unrelated to the presence or
N = 16 p< 0003 r:oi1
0504 .?? 2
-l J .z
045.
040.
E a
-I 5 'm' I
. .
. .
g 0
030
.
035
.
. .
01
??
’
02
. 0
.
s
0.15 .
.
’
. .
LL
.
iI-
0.3
0
04
Pre-CPB Max Minor AxisLA
05
.
025-
J .
. .
1 2 kt! 0.20"
030-
020’
I
.
. .
025.
??
I 035
s I
N=16 NS
.
z '5 zs
.
040
0 55
Fig 5. Pre-cardiopulmonary and post-cardiopulmonary bypass longitudinal axis maximal major axis shortening fraction (max major ax&J versus longitudinal axis maximal minor axis shortening fraction (max minor axis&
06
Fig 4. Pm-cardiopulmonary and post-cardiopulmonary bypass longitudinal axis maximal major axis shortening fraction (max major axisJ verws longitudinal axis planimetered area excursion fraction (P&J.
I
.
. . .
,oL_I-.-J 01
02
0.3
0.4
Post-CPB Max Minor AxisLA
05
RV SYSTOLIC PERFORMANCE INDICES
165
Table 3. Hemodynamic and Echocardiagraphic
Values
Postintubation
RVEF,, Cl (L/minim*) HR (beats/min)
PWCPB
Post-CPB
0.41 ‘- 0.14
0.38 k 0.12
0.37 k 0.12
3.0 k 0.9
2.6 2 0.5
2.8 2 0.6
67k
92 k 12*
742
17
13
Diastolic max major axisLA (cm) Max major axisLA
7.4 + 1.2
7.8 + 1.3
7.3 f 1.2
0.35 + 0.06
0.33 2 0.08
0.24 r 0.08’
Diastolic max minor axisLA(cm) Max minor axisLA
4.2 + 1.0
4.4 -c 0.7
4.1 f 1.1
0.34 + 0.13
0.32 f 0.11
0.27 + 0.10
Diastolic 2D,, (cm21
25.0 k 6.9
24.8 + 6.0
23.1 + 6.7
~DLA
0.50 + 0.16
0.47 + 0.10
0.37 + 0.11*
Diastolic 2Dsn (cm2)
19.8 ? 10.4
16.7 + 7.3
15.9 + 6.9
2Dsa
0.43 5 0.17
0.44 + 0.18
0.52 + 0.18
EDVI (mL/m*)
100 2 25
105 5 25
88 t 33
EDP (mmHg)
6.0 2 4.1
4.5 2 3.2
5.4 2 3.2
PAP (mmHg)
17.8 -t 4.8
16.5 + 6.0
16.8 2 5.1
*Significantly different (P < 0.05) than post-intubation and pre-CPB values.
restricted to the upper portion of the RV. Further interpretation is theoretically possible because of the orthogonal relationship of the maximal major and minor axes. The CPB-associated dissociation of shortening fraction measurements determined in these axes suggests that decreases were confined to the base of the ventricle. A more definitive localization might have been afforded by centerline motion analysis. However, inadequate epicardial definition precluded such an evaluation in this series. The above interpretation is, of course, indirect and dependent on the absence of widely separated areas of local dysfunction. In addition, it should be borne in mind that the maximal minor axis measurements would be expected to be affected by dyskinesia of the interventricular septum. Paradoxical septal motion, left ventricular systolic correction of diastolic right ventricular displacement of the septum into the left ventricular cavity, would be anticipated to result in exaggeration of systolic excursion of the maximal minor axis. This would be expected to generate a shortening value directly reflecting right and left ventricular performance.
There are no published data relating to this issue. There is a precedent for preoperative echocardiographic estimates of cardiac systolic function to delineate spurious findings. Lehmann et al** have demonstrated that the apparently high prevalence of dyskinetic interventricular septal motion following CABG determined by echocardiography may be artifactual and, at least in part, due to cardiac translocation. Accordingly, it is important to address measurement technique limitations specifically referable to this study. Technical considerations pertaining to the echocardiographic measurements include the possibility that postCPB foreshortening of right ventricular images may have been responsible for the results. However, pre-CPB and post-CPB measured loading conditions were comparable. Further, systolic shortening measurements were expressed as a function of corresponding end-diastolic dimensions, and foreshortening would be expected to minimize decreases in these measurements. The complex geometry of the RV must also be considered. Inconsistencies in shape could conceivably furnish differing regions for analysis despite meticulous standardization of image acquisition technique. Certainly, at least in terms of echocardiographic estimation of RV volumes, Levine et alz9 have noted that findings derived from even multiple planes of interrogation may be suspect in individual cases. Again, the loading conditions during each of the measurement periods were not disparate, suggesting constancy of shape. A further issue that must be addressed is that the echocardiographic .and RVEFro findings, superficially, appear to be contradictory. Decreases in longitudinal axis performance indices occurred in the face of maintenance of RVEFm values. Though short-axis indices increased, this increase was not statistically significant, invoking the question as to how overall RVEF was maintained. It can only be assumed that further compensation occurred in areas that were not interrogated, such as the outflow tract, or in regions that were not accessible to transverse plane analysis. Equally, it may well be that the thermodilution technique estimates were less sensitive indicators of RV global performance than the echocardiographic measurements.
REFERENCES
1. Hines R, Barash P: Intraoperative right ventricular dysfunction detected by a right ventricular ejection fraction catheter. J Clin Monitoring 2:206-208,1986 2. Mangano D: Biventricular function after myocardial revascularization in humans: Deterioration and recovery patterns during the first 24 hours. Anesthesiology 62:571-577,1985 3. Boldt J, Kling D, Moosdorf R, et al: Influence of acute volume loading on right ventricular function after cardiopulmonary bypass. Crit Care Med 17:518-525,1989 4. Stein K, Breisblatt W, Wolfe C, et al: Depression and recovery of right ventricular function after cardiopulmonary bypass. Crit Care Med 18:1197-1200, 1990 5. Kay H, Afshari M, Barash P, et al: Measurement of ejection fraction by thermal dilution techniques. J Surg Res 34:337-346, 1983 6. Biondi J, Hines R, Rafferty T, et al: The effect of high-
frequency positive-pressure ventilation on right and left ventricular function. Anesth Analg 65:679-682, 1986 7. Dhainaut F-F, Brunet F, Monsallier J, et al: Bedside evaluation of right ventricular performance using a rapid computerized thermodilution method. Crit Care Med 15:148-152,1987 8. Urban P, Scheidegger D, Gabathuler J, et al: Thermodilution determination of right ventricular volume and ejection fraction: A comparison with biplane angiography. Crit Care Med 15:652-655, 1987 9. Voelker W, Gruber H, Ickrath 0, et al: Determination of right ventricular ejection fraction by thermodilution technique-A comparison to biplane cineventriculography. Intensive Care Med 14:461-466,1988 10. Klein H, Tordjman T, Ninio R, et al: The early recognition of right ventricular infarction: Diagnostic accuracy of the electrocardiographic V4R lead. Circulation 67:558-565, 1983
RAFFEHiV
166
11. Rafferty T, Durkin M, Mathew regurgitation: Color-flow Dopplerversus ography. Anesth Analg 72:S219, 1991
J: Estimation of tricuspid saline contrast echocardi-
12. Schluter M, Hinrichs A, Thier W, et al: ‘Transesophageal two-dimensional echocardiography: Comparison of ultrasonic and anatomic sections. Am J Cardiol 53:1173-1178, 19X4 13. Seward J, Khanderia echocardiography: Technique, tion, and clinical applications.
B, Oh J, et al: Transesophageal anatomic correlations. implementaMayo Clin Proc 63649-680, 1988
14. Jardin F, Gueret P, Dubourg 0, et al: Right ventricular volumes by thermodilution in the adult respiratory distress syndrome: A comparative study using two-dimensional echocardiography as a reference. Chest 88:34-39, 1985 15. Jardin F, Brun-Ney D, Hardy A, et al: Combined thermodilution and two-dimensional echocardiographic evaluation of right ventricular function during respiratory support with PEEP. Chest 99:162-168, 1991 16. Weyman H: Cross-sectional Echocardiography. phia, PA, Lea and Febiger, 1982. pp 382-395 17. Kaul S, Tei C, Hopkins J. Shah ventricular function using two-dimensional Heart J 107:526-531, 1984
Philadel-
P: Assessment of right echocardiography. Am
18. Bommer W, Weinert L, Neumann A, et al: Determination of right atrial and right ventricular size by two-dimensional echocardiography. Circulation 60:91-100, 1979 19. Feigenbaum H: Echocardiography. and Febiger. 1986, pp 158-167 20. Durkin
M, Rafferty
Philadelphia,
T: Two-dimensional
PA, Lea
echocardiographic
ti Al
determination of rtght ventricular- maximum minor- axis duncrrsrori fractional shortening. Anesthesiology ?3:AS42. 1990 2 I. Braimhridge M: Myocardial protection. in Taylor I(. (c(i) C‘ardtopulmonary Bypass-Principle and Management. London: Chapman and Ilall. 19X6, pp 375-39X 22. Takach 1‘. Glassman I Milewicz A. et al: C’ontmuous measurement of intramyocardial pH: relative importance of hypothermia and cardioplegic perfusion pressure and temperature Ann Thorac Surg 42:365-371. I%h 2.1.Behrendt D. Jochim K: Effect of temperature of cardiople gic solution. J Thorac C‘ardiovasc Surg 76:3.53-356. 1978 24. Chiu R. Blundell P, Scott J. Cain S: The importance ot monitoring intramyocardial temperature during hypothermic myocardial protection. Ann Thorac Surg 28:317-322, 1978 25. Boldt J. Kling D. Dapper I;. Hempelmann G: Myocardial temperature during cardiac operations: Influence on right ventricular function. J Thorac Cardiovasc Surg 100:562-568, 1990 26. Becker H, Vinter-Johansen .I. Buckberg G: Critical importance of ensuring cardioplegic delivery with coronary stenoses. .I Thorac Cardiovasc SurgXl:507-515, 1981 27. Gonzalez A. Brandon T. Fortune R: Acute right ventricular failure is caused by inadequate right ventricular hypothermia. J Thorac Cardiovasc Surg X9:38&39X. 19X5 2X. Lehmann K. Lee F. McKenzie W. et al: Onset of altered interventricular septal motion during cardiac surgery. Circulation 82:1325-13.34. 1990 29. Levine R, Gibson T. Aretz T. et al: Echocardiographic measurement of right ventricular volume. Circulation 69:497-SOS. 1984