Prospective study comparing different echocardiographic measurements of pulmonary capillary wedge pressure in patients with organic heart disease other than mitral stenosis

Compromised left atrial function and increased size predict raised cavity pressure: a systematic review and meta-analysis

AIM

This meta-analysis assesses left atrial (LA) cavity and myocardial function measurements that predict pulmonary capillary wedge pressure (PCWP).

METHODS

PubMed-MEDLINE, EMBASE, Scopus, Google Scholar and the Cochrane Central Registry were searched up to December 2018 for studies on the relationship of LA diameter, LA indexed volume (LAVI max, LAVI min), peak atrial longitudinal (PALS), peak atrial contraction (PACS) strain and total emptying fraction (LAEF) with PCWP. Eighteen studies with 1343 patients were included. Summary sensitivity and specificity (with 95% CI) for evaluation of diagnostic accuracy and the best cut-off values for different LA indices in predicting raised PCWP were estimated using summary receiver operating characteristic analysis.

RESULTS

The pooled analysis showed association between PCWP and LA diameter: Cohen's d = 0·87, LAVI max: d = 0·92 and LAVI min: d = 1·0 (P<0·001 for all). A stronger correlation was found between PCWP and PALS: d = 1·26, and PACS: d = 1·62, total EF d = 1·22 (P<0·0001 for all). PALS ≤19% had a summary sensitivity of 80% (65-90) and summary specificity of 77% (52-92), positive likelihood ratio (LR+) 3·74, negative likelihood ratio (LR-) <0·25 and DOR > 15·1 whereas LAVI ≥34 ml m^-2 had summary sensitivity of 75% (55-89) and summary specificity 77% (57-90), with LR+ >3, LR- 0·32 and DOR >10·1.

CONCLUSIONS

Compromised LA myocardial function and increased size predict raised cavity pressure. These results should assist in optimum follow-up of patients with fluctuating LA pressure.

Non-invasive estimation of pulmonary capillary wedge pressure at rest and during exercise by electrocardiography, phonocardiography and Doppler echocardiography

Non-invasive estimation of the mean pulmonary capillary wedge pressure was accomplished by simultaneous electrocardiographic, phonocardiographic and continuous wave Doppler echocardiographic recordings. The interval from the onset of the QRS complex to the Doppler determined mitral valve closure (Q-MC) and the interval from the phonocardiographic aortic component of the second heart sound to the Doppler determined mitral valve opening (A2-MO) were measured. The non-invasive registrations were carried out simultaneously with direct measurements of the wedge pressure. In an initial group of 22 patients, a significant correlation was observed between the intervals alone and the wedge pressure, r = 0.60, SEE = +/- 6.9 mmHg, p less than 0.01, for the Q-MC interval and r = -0.70, SEE = 6.2 mmHg, p less than 0.001 for the A2-MO interval. A closer correlation was observed between the ratio Q-MC/A2-MO and the measured wedge pressure, r = 0.93, SEE = +/- 3.1 mmHg, p less than 0.001. The linear regression equation, PCW = 19.5 (Q-MC/A2-MO) + 3.0 (mmHg), was applied prospectively to a second group of 23 patients. Again the relationship between estimated and measured wedge pressure was highly significant, r = 0.90, SEE = +/- 3.1 mmHg, p less than 0.001. Twenty-two patients were also studied during an exercise test, and acceptable non-invasive recordings were obtained in 19 of them. The change in estimated wedge pressure during activity related closely to the change in actual wedge pressure, r = 0.80, SEE = +/- 5.7 mmHg, p less than 0.001. A simplified equation suitable for routine clinical practice, PCW = 24 (Q-MC/A2-MO) (mmHg), yielded almost equally accurate estimates of the wedge pressure over a wide range of pressures. The simplicity and reasonable accuracy of Doppler-assisted estimation of the wedge pressure makes it useful in the evaluation and follow-up of patients with suspected cardiac disorders. The method may assist in evaluating the effects of diagnostic or therapeutic procedures, since it is sufficiently sensitive to detect acute directional changes in wedge pressure.

Estimation of mean left atrial pressure from transesophageal pulsed Doppler echocardiography of pulmonary venous flow

To determine whether pulmonary venous flow and mitral inflow measured by transesophageal pulsed Doppler echocardiography can be used to estimate mean left atrial pressure (LAP), we prospectively studied 47 consecutive patients undergoing cardiovascular surgery. We correlated Doppler variables of pulmonary venous flow and mitral inflow with simultaneously obtained mean LAP and changes in pressure measured by left atrial or pulmonary artery catheters. Among the pulmonary venous flow variables, the systolic fraction (i.e., the systolic velocity-time integral expressed as a fraction of the sum of systolic and early diastolic velocity-time integrals) correlated most strongly with mean LAP (r = -0.88). Of the mitral inflow variables, the ratio of peak early diastolic to peak late diastolic mitral flow velocity correlated most strongly with mean LAP (r = 0.43), but this correlation was not as strong as that with the systolic fraction of pulmonary venous flow. Similarly, changes in the systolic fraction correlated more strongly with changes in mean LAP (r = -0.78) than did changes in the ratio of peak early diastolic to peak late diastolic mitral inflow velocity (r = 0.68). We conclude that in the surgical setting observed, pulmonary venous flow from transesophageal pulsed Doppler echocardiography can be used to estimate mean LAP. This technique may provide a rapid, simple, and relatively noninvasive means of gauging this variable in patients undergoing intraoperative transesophageal echocardiography.

Dyspnoea of cardiac origin in 67 year old men: (2). Relation to diastolic left ventricular function and mass. The study of men born in 1913

The relation of cardiac dyspnoea to diastolic left ventricular dysfunction was examined in a sample of 67 year old men from the general population of Gothenburg, Sweden. Forty two men with cardiac dyspnoea and 45 controls were selected from the screened cohort of 644 men. M mode echocardiography, apexcardiography, and phonocardiography were used to evaluate heart sounds, diastolic time intervals, aortic root motion (atrial emptying index); peak rate of change in left ventricular dimension, left atrial and ventricular size; and left ventricular mass. There was a significant relation between dyspnoea grade and left ventricular mass and posterior wall thickness. Dyspnoea grade also correlated significantly with the amplitude of the rapid filling wave and the third heart sound, atrial emptying index and left atrial size, the pulmonary component of the second heart sound, and the dimension of the right ventricle. In mild to moderate dyspnoea fractional shortening was normal, but posterior wall thickness and left atrial dimension were increased. The time from the second heart sound to the O point of the apexcardiogram, adjusted for heart rate, was significantly prolonged in mild to moderate dyspnoea, but not in severe dyspnoea. There was a significant decrease of rate adjusted isovolumic relaxation time, probably secondary to altered loading conditions, in severe dyspnoea, but not in mild to moderate dyspnoea. When the effect of systolic function was excluded multivariate analyses showed that the relation between dyspnoea grade and left atrial dimension persisted. The finding that diastolic abnormalities of the heart contributed to the generation of cardiac dyspnoea may have implications for treatment.

Echocardiographic estimate of left atrial pressure in children

Echocardiographic estimates of left atrial pressure using the Q-MVC/AVC-E ratio were obtained from 21 patients in the early postoperative period and compared to simultaneously recorded mean left atrial pressure. A good correlation was obtained between mean left atrial pressure and the Q-MVC/AVC-E ratio (r = 0.81, p less than 0.001) with a regression equation of: left atrial pressure = 11.09 (Q-MVC/AVC-E) + 0.84. The present data compared to those of previous investigators reveal that, although there are quantitative differences between each of the studies, the correlation coefficient of each study is good. Echocardiography provides a useful noninvasive estimate the left atrial pressure; however, each laboratory must establish its own normals.

Noninvasive evaluation of systolic and diastolic function in severe congestive heart failure secondary to coronary artery disease or idiopathic dilated cardiomyopathy

The usefulness of systolic time intervals, diastolic time intervals and echocardiography in evaluating left ventricular (LV) function was determined in 69 patients with severe congestive heart failure. All systolic time intervals were markedly abnormal (preejection period/LV ejection time 0.59 +/- 0.18 vs 0.30 +/- 0.04, preejection period index 170 +/- 37 vs 117 +/- 11, LV ejection time index 372 +/- 26 vs 410 +/- 17; patients vs control subjects, p less than 0.05). Diastolic time intervals in patients were not different from those in control subjects. Echocardiographic measurements were all markedly abnormal (LV end-diastolic dimension 6.9 +/- 1.0 vs 4.8 +/- 0.4 cm, patients vs control subjects, p less than 0.05). No pattern of abnormalities distinguished ischemic cardiomyopathies from idiopathic dilated cardiomyopathies. The presence of LV conduction delay did not substantially alter results, except that exclusion of patients with LV conduction delay normalized the total time of systole (QA2) index (from 542 +/- 40 to 531 +/- 31 ms) and reduced but did not normalize prolongation in the preejection period index (from 170 +/- 37 to 162 +/- 29 ms). No systolic or diastolic interval strongly correlated with any hemodynamic or other independent measure of LV performance. Twenty-four patients were given inotropic or unloading agents, which significantly improved hemodynamic values. Systolic and diastolic intervals were measured at baseline and at maximal hemodynamic effect. The correlation of changes in hemodynamics with changes in systolic and diastolic intervals was only modest. Thus, although systolic time intervals and associated echocardiographic measurements can detect abnormal LV function, they cannot reliably detect a change in LV function or distinguish gradations of abnormality.(ABSTRACT TRUNCATED AT 250 WORDS)

Relation between mitral valve closure and early systolic function of the left ventricle

In order to investigate relations between mitral valve closure and mechanical events at the onset of left ventricular systole, simultaneous M mode echocardiograms and phonocardiograms were recorded with the apexcardiogram and its first differential (dA/dt) in 25 normal subjects and 88 patients with heart disease. The timing of mitral and aortic valve closure and the onset and peak rate of rise of the apexcardiogram with respect to the Q wave of the electrocardiogram were measured. There was considerable variation in the intervals from Q to mitral valve closure (Q-MVC) and from Q to peak dA/dt and in the isovolumic contraction time between normal subjects. There was no consistent abnormality of these intervals in patients with coronary artery or valvar disease, and no relation between the interval from Q to mitral valve closure and end diastolic pressure. When the timing of the first heart sound and peak dA/dt were considered together, however, clear abnormalities became apparent. In normal subjects, the intervals Q-MVC and Q to peak dA/dt were significantly correlated. In coronary artery disease, the expected relation between Q-MVC and Q to peak dA/dt was found only when end diastolic pressure was normal and was lost when end diastolic pressure was raised. Mitral stenosis was associated with delayed mitral closure in a few cases only, but in chronic aortic regurgitation closure was consistently early with respect to the apexcardiogram. In patients with atrial fibrillation and a normal mitral valve the timing of mitral valve closure with respect to the apexcardiogram was normal, which is inconsistent with an atrial contribution to the timing of mitral valve closure. Thus when considered in isolation the timing of mitral valve closure and the duration of isovolumic contraction time gave little information about cardiac function. Nevertheless, a predictable relation exists between mitral valve closure and the onset of left ventricular mechanical systole in normal subjects, which can be used to identify characteristic alterations in patients with heart disease.

Echophonocardiographic estimates of pulmonary artery wedge pressure in mitral stenosis

The efficacy of noninvasive indexes for predicting pulmonary artery wedge (PAW) pressure was reviewed in 77 patients with mitral stenosis. M-mode echocardiography and phonocardiography were used to measure the aortic valve closure-mitral valve E-point interval (A2-E) and the electrocardiographic Q wave-mitral valve closure interval (Q-C) close to the time of diagnostic cardiac catheterization. During catheterization, in 65 patients PAW pressure was measured and in 12 left atrial (LA) pressure was measured. The A2-E and Q-C intervals taken alone had only modest correlation with PAW pressure (r = -0.54 and r = 0.46, respectively). The correlation was weakest in patients with atrial fibrillation and best in sinus rhythm when heart rate variation between invasive and noninvasive studies was within +/- 5 beats. Substitution of V-wave pressure for mean PAW pressure and correction for variation in blood pressure improved the A2-E correlation (r = -0.64), as did combining the A2-E and Q-C intervals into a ratio [(Q-C)/(A2-E)] (r = 0.62). However, the best results were obtained in patients where LA pressure was measured directly (r = -0.91 for A2-E), suggesting the PAW pressure is not always an accurate reflection of LA pressure. In conclusion, many factors in addition to LA pressure affect the Q-C and A2-E intervals which, in many situations, decrease their predictive value. However, if used appropriately, these intervals may allow an estimation of PAW pressure.