Tissue Doppler echocardiographic and color M-mode estimation of

left atrial pressure in infants*

Fabrice Larrazet, MD; Kamel Bouabdallah, MD; Emmanuel Le Bret, MD; Pascal Vouhé, MD;
Colette Veyrat, MD; François Laborde, MD

Objectives: Using recorded flow and tissue Doppler, we eval-
uated the relation of peak velocity of early transmitral Doppler
filling (E)/early diastolic velocity of the lateral mitral annulus (Ea)
ratio and of E/flow propagation velocity (Vp) ratio to mean left
atrial pressure in infants after surgery for congenital heart dis-
ease.
Design: Experimental design.
Setting: Pediatric intensive care unit.
Patients: Thirty-seven infants aged 4 (3– 8) months.
Interventions: Patients underwent postoperative invasive he-
modynamic monitoring with simultaneously obtained Doppler
measurements.
Measurements and Main Results: Values are expressed as
median (25th–75th percentiles). Heart rate was 145 (135–157)
beats/min. Left atrial pressure was 10 (8 –12) mm Hg with E/Ea 16
(12–19) and E/Vp 1.9 (1.3–2.4). E/Ea and E/Vp ratios were higher
in patients with left atrial pressure >10 mm Hg (n ⴝ 18), than in
patients with left atrial pressure <10 mm Hg (n ⴝ 19) (E/Ea, 16
[15–25] vs. 12 [9 –17], p ⴝ .01; E/Vp, 2.3 [1.9 –2.8] vs. 1.4 [1–1.9].
respectively, p ⴝ .001). At a cutoff point of 15, E/Ea sensitivity for
left atrial pressure >10 mm Hg was 17 of 18 (94%) with speci-
ficity 13 of 18 (72%). At a cutoff point of 2, E/Vp sensitivity for left
atrial pressure >10 mm Hg was 15 of 18 (83%) with specificity 16
of 18 (89%). Areas under the receiver operating characteristic
curves were 0.76 (E/Ea) and 0.83 (E/Vp).
Conclusions: Doppler ratios might be considered as promising
noninvasive tools for left atrial pressure evaluation in infants after
cardiac surgery. (Pediatr Crit Care Med 2005; 6:448 –453)
KEY WORDS: echocardiography; diastole; pediatrics; surgery;
catheters; cardiac shunt

N oninvasive estimation of left
ventricular (LV) filling pres-
sure can be obtained by com-
binations of mitral flow ve-
locity curves with other Doppler
variables. These include color M-mode,
tissue Doppler imaging (TDI) of mitral
annular motion, quantitative analysis of
the pulmonary venous signal, and re-
sponse of the mitral inflow to altered
loading conditions (1–7). LV early dia-
stolic flow propagation velocity (Vp), de-
rived from color M-mode Doppler, which
represents the time difference between
maximal velocity at the apex level and at
the mitral leaflet tips, as well as peak
early diastolic myocardial velocity (Ea) of
the mitral annulus, measured by TDI,
correlate well with the relaxation time
constant (␶) and are independent of pre-
load (3, 8 –12). Peak velocity of early
transmitral Doppler filling (E)/Vp and
*See also p. 496.
From the Department of Cardiology, Institut Mutu-
aliste Montsouris, Paris, France.
Copyright © 2005 by the Society of Critical Care
Medicine and the World Federation of Pediatric Inten-
sive and Critical Care Societies
DOI: 10.1097/01.PCC.0000164345.86775.35
E/Ea ratios have been shown to correlate
with mean pulmonary artery occlusion
pressure (PAOP) in adults (2, 3). During
cardiac surgery performed in infants, the
filling pressure can be obtained from a
catheter inserted into the left atrium.
These catheters might be the source of
sepsis, embolic events, or local hemor-
rhagic complications (13, 14). Noninva-
sive evaluation of left atrial pressure
(LAP) would be highly relevant for the
management of children in the postoper-
ative period. Clinical efficiency of the
Doppler ratios was recently shown un-
changed in cases of tachycardia in adults
(15). The infant population is character-
ized by a short diastole related to usual
tachycardia, high range of flow and TDI
velocities, and a small LV configuration.
Our purpose was to explore the validity of
both Doppler ratios in infants after car-
diac surgery by comparing them with si-
multaneously obtained invasive LAP.
METHODS
Invasive Pressure Measurements. Thirty-
seven infants who underwent invasive hemo-
dynamic monitoring in the early postoperative
cardiac surgery period made up the study
group. There were 13 ventricular and 17 atrio-
ventricular septal defects, two truncus arteri-
osus, two tetralogies of Fallot, two hypertro-
phic cardiomyopathies, and one mitral valve
regurgitation. We excluded mitral valve re-
placement, persistent moderate mitral regur-
gitation, and arrhythmia. Interventions were
performed using normothermic bypass with
myocardial protection achieved by repeat
warm-blood cardioplegia. Treatment of atrio-
ventricular septal defect included ventricular
septal defect closure and atrial septal defect
closure by the two-patch technique and com-
plete closure of the mitral cleft (n ⫽ 6) asso-
ciated with commissuroplasty (n ⫽ 9) or pos-
terior annuloplasty (n ⫽ 2). The study was
approved by the institutional review board,
and parents signed informed consent.
Fluid-filled catheters were placed into the
left atrium (centrocath 4 Fr), the radial artery,
and the pulmonary artery to monitor right
and left pressures in all patients. The trans-
ducers were positioned at the same level as
that of the heart. A zero pressure reference
was established before each measurement.
Children were ventilated with Siemens
Servo 300. The mode of ventilation was pres-
sure regulated plus volume control. Drugs
used for sedation and anesthesia were mida-
zolam intravenous infusion 0.05– 0.2 mg/
kg/hr and sufentanil 0.5–2 ␮g/kg/hr. Inotropic
drugs were infused in all but three infants

448 Pediatr Crit Care Med 2005 Vol. 6, No. 4


Figure 1. Doppler echocardiographic recordings in an infant with left atrial pressure ⬍10 mm Hg after
cardiac surgery. Top, transmitral Doppler flow; middle, pulsed Doppler wave at the lateral corner of the
mitral annulus; bottom, color M-mode transmitral flow propagation). E, early transmitral Doppler
filling velocity; HR, heart rate; Ea, early diastolic velocity at the mitral annulus; Dist; Vp, flow
propagation velocity.
during this period (epinephrine, n ⫽ 15; do-
butamine, n ⫽ 24; milrinone, n ⫽ 11 pa-
tients).
Conventional Echocardiographic and
Doppler Measurements. Doppler echocardiog-
raphy was performed (5-MHz probe; System
Five; General Electrics, WI) simultaneously
with the invasive recordings, in a blinded fash-
ion to LAP during the end phase of expiration.
One operator (FL) performed all Doppler re-
cordings. Another operator (KB) performed 15
Doppler recordings to measure the interindi-
Pediatr Crit Care Med 2005 Vol. 6, No. 4
vidual variability. Ejection fraction was calcu-
lated according to the Simpson modified
method. The left atrium appeared to be filled
when the atrial septum did not move toward
the left atrial cavity. Mitral Doppler flow re-
cordings were obtained from the apical win-
dow with the pulsed Doppler sample volume
placed at the mitral valve tips (Fig. 1, top, and
Fig. 2, top). On mitral inflow recordings, we
measured E (cm/sec) and the deceleration
time. Variables were averaged over five cardiac
cycles.
TDI Analysis. Annular velocities were ob-
tained from TDI. The TDI program was set to
the pulsed-wave TDI modality with a 5-MHz
probe. From the apical four-chamber view, a
2– 4 mm sample volume was placed at the
lateral corner of the mitral annulus (Fig. 1,
middle, and Fig. 2, middle). Five TDI record-
ings were performed at end-expiration at a
sweep speed of 100 mm/sec. E/Ea was calcu-
lated as the ratio of E divided by Ea.
Flow Propagation Analysis. We used the
color M-mode flow velocity displays from the
apical four-chamber view. The M-mode cursor
was placed through the center of the mitral
inflow region, parallel with the direction of
flow observed by color Doppler flow imaging
(3, 9). Vp was measured as the slope of the first
aliasing velocity (usually approximately 50 –
75% of E) during early filling, from the mitral
valve plane to ⱖ2 cm distally into the LV
cavity (Fig. 1, bottom, and Fig. 2, bottom) at a
sweep speed of 100 mm/sec. E/Vp was calcu-
lated as the ratio of E divided by Vp.
Statistical Analysis. Data are expressed as
median values (25th–75th percentiles). Linear
regression was used to assess the relation of
Doppler variables to LAP. A binary logistic
regression analysis tested the variables associ-
ated with LAP ⬎10 mm Hg. Receiver operat-
ing characteristic curves were constructed for
the individual Doppler variables for the pre-
diction of LAP ⬎10 mm Hg. Comparisons
between patient groups having LAP ⱕ10 mm
Hg (group A) or ⬎10 mm Hg (group B) were
assessed with the Mann-Whitney nonparamet-
ric test. Statistical significance was defined as
p ⬍ .05 (SPSS 10.1, Chicago, IL). We also
studied the diagnostic reliability of ratios ac-
cording to the presence or not of mitral valve
repair. Intra- and interobserver variabilities
were calculated at different time for Ea and Vp
in a subgroup of 15 infants with a stable he-
modynamic status. It was expressed as a per-
centage of variation for reproducibility and as
a percentage of error between readings: re-
cording 1 ⫺ recording 2/mean of both record-
ings. A Bland-Altman plot was provided for the
interobserver variability of both measure-
ments with the limits of agreement defined as
mean difference ⫾2 SD of the differences.
RESULTS
Infants (22 girls, 15 boys) aged 4 (3– 8)
months underwent echocardiographic
and Doppler examination 7 (5–24) hrs
following cardiac surgery. Two infants
were transfused after bleeding due to
atrial catheter removal. E and Ea were
analyzable in all patients (100% feasibil-
ity). Vp was measurable in all but one
patient (97% feasibility). Intraobserver
variabilities were 7.2 ⫾ 8.4% for Ea and
6.1 ⫾ 5.8% for Vp. Interobserver variabil-
ities were 8.3 ⫾ 10.5% for Ea and 7.9 ⫾
5.9% for Vp. A Bland-Altman plot repre-
449

Figure 2. Doppler echocardiographic recordings in an infant with left atrial pressure ⬎10 mm Hg after
cardiac surgery. Top, transmitral Doppler flow; middle, pulsed Doppler wave at the lateral corner of the
mitral annulus; bottom, color M-mode transmitral flow propagation. E, early transmitral Doppler
filling velocity; HR, heart rate; Ea, early diastolic velocity at the mitral annulus; Dist; Vp, flow
propagation velocity.
senting interobserver variability for both
ratios is shown on Figure 3.
Heart rate was 145 (135–157) beats/
min. Mean arterial blood pressure was 71
(62– 80) mm Hg. Systolic pulmonary arte-
rial blood pressure was 30 (25–34) mm Hg.
LAP was 10 (8 –12) mm Hg with extreme
values ranging from 5 to 16 mm Hg. E/Ea
was 16 (12–19) and E/Vp was 1.9 (1.3–2.4).
Table 1 lists median values of variables for
both groups A and B. The distribution of
450
the septal defects did not differ significantly
between groups. Mild mitral regurgitation
as well as small residual shunts tended to
be more frequently observed in group B.
Fusion of E and late transmitral Dopp-
ler filling velocity (A) was present in ten
cases (27%) and was more frequently ob-
served in patients with higher heart rates
(146 [145–170] compared with 141 (130 –
151), p ⬍ .05) and higher E/Vp (2.3 [1.9 –
3] compared with 1.6 [1.2–2.1], p ⬍ .05).
Patients with mitral valve repair (n ⫽
17) had higher E values than others (1
[0.9 –1.1] vs. 0.7 [0.6 –1.1] m/sec, respec-
tively, p ⬍ .05). Therefore, E/Ea and E/Vp
were significantly higher than in patients
without mitral valve surgery (E/Ea, 17 [15–
24] vs. 13 [9 –16], p ⬍ .05; and E/Vp, 2
[1.8 –2.4] vs. 1.4 [1.1–2.1], respectively, p ⬍
.05).
Relation of LAP to E/Vp and E/Ea. The
transmitral flow velocity, TDI of mitral
annular motion, and color M-mode mi-
tral flow are represented in Figure 1 (in-
fant with LAP ⬍10 mm Hg) and in Figure
2 (infant with LAP ⬎10 mm Hg). E/Vp
correlated with E/Ea (r ⫽ .6, p ⬍ .01).
E/Ea and E/Vp ratios were higher in
group B than in group A (E/Ea, 16 [15–
25] vs. 12 [9 –17] p ⫽ .01; and E/Vp, 2.3
[1.9 –2.8] vs. 1.4 [1–1.9], respectively, p ⫽
.001, Fig. 4). Age, weight, height, and E
appeared as confounding factors for the
comparison of Doppler ratios between
groups (Table 1). Left atrial diameter
tended to be higher in group B (p ⫽ .07).
In both groups, the left atrium appeared
to be filled in the same proportion (8 of
19 vs. 8 of 18). There was no relation
between the left atrial aspect and Doppler
ratios. We observed a weak but significant
relation between LAP and both ratios (r
⫽ .41, p ⬍ .05 for E/Ea, and r ⫽ .41, p ⬍
.05 for E/Vp). The multiple linear regres-
sion analysis revealed that E/Vp was in-
dependently associated with LAP values
(p ⬍ .05). The binary logistic regression
analysis demonstrated that both ratios
E/Ea and E/Vp were significantly associ-
ated with high LAP values (Table 2). Ta-
ble 3 lists sensitivity, specificity, predic-
tive values, and receiver operating
characteristic curve areas of both ratios
to distinguish LAP values ⬎10 mm Hg in
all infants but also in the group of pa-
tients with (n ⫽ 17) or without (n ⫽ 20)
mitral valve surgery. Doppler ratios were
more efficient in the group of patients
without mitral valve surgery than in the
other group. E/Ea was more sensitive but
less specific than E/Vp for detecting LAP
values ⬎10 mm Hg (Table 3).
DISCUSSION
Our results indicate that the combina-
tion of Doppler mitral peak E velocity
with color-M mode propagation velocity
Vp and TDI annular velocities Ea may be
helpful in screening infants with LAP
⬎10 mm Hg after cardiac surgery, espe-
cially in infants without mitral valve sur-
gery. Our study demonstrated that the
Pediatr Crit Care Med 2005 Vol. 6, No. 4
 Main Doppler Methods for Left Ven-
tricular Filling Pressure Assessment.
Doppler transmitral early diastolic inflow
velocities provide useful information on
LV filling hemodynamics (16). This is due
to the sensitiveness of mitral inflow ve-
locity profile to changes in preload. In
counterpart, mitral inflow velocity profile
is also sensitive to changes in afterload,
rate of relaxation, and heart rate. Several
methods have been proposed to overcome
these confounding effects. A first trend of
research relied on the combination of
TDI to flow traces (11). The relative pre-
load independency of the mitral annulus
prompted researchers to devise a Doppler
Figure 3. Bland-Altman plot showing the interobserver variability of early transmitral Doppler filling ratio combining E with velocities of the
velocity (E)/early diastolic velocity at the mitral annulus (Ea) and E/flow propagation velocity (Vp) annular recording, Ea. The ratio E/Ea was
measurements with the limits of agreement in 15 infants. well correlated with PAOP in adults.
However, the relation between E/Ea and
PAOP varies between studies but also ac-
Table 1. Comparison between infants with left atrial pressure (LAP) ⱕ10 mm Hg (group A) and LAP ⬎
cording to the LV ejection fraction (r
10 mm Hg (group B)
values range from .47 to .87) (2, 17).
Group A Group B In our population, the relation be-
(LAP ⱕ 10 mm Hg) (LAP ⬎ 10 mm Hg) tween E/Ea and LAP was weaker than the
relation observed in adult studies con-
No. of patients 19 18 cerning the Doppler ratio and PAOP. This
Age, mos 4 (3–4) 6 (3–15)a might be due to the extreme condition
Weight, kg 4.7 (3.8–5.2) 5.2 (4.4–8.5)a
Height, cm 56 (52–60) 65 (57–76)a
represented by our study group (recent
Heart rate, beats/min 145 (140–170) 143 (131–145) bypass with cardioplegia) but also to the
Time from surgery—Doppler evaluation 6 (5–24) 7 (5–24) low range of LAP value (5–16 mm Hg)
Hemoglobin, g/dL 12.6 (12–13) 12.7 (11.7–13) compared with that of other adult pa-
Ventricular septal defects, n (%) 8 (42) 5 (28) tients in sinus tachycardia with high
Atrioventricular septal defects, n (%) 8 (42) 9 (50)
LVEDD, mm 20 (17–25) 22 (18–26) range value of PAOP (6 – 43 mm Hg) (15).
LVEF, % 50 (40–55) 40 (35–50) Another trend of research was focused
LA diameter, mm 17 (15–19) 18 (16–21) on Vp (8, 9). The apex is a prominent
E, m/sec 0.7 (0.6–1) 1 (0.8–1.1)a source of recoil during early diastole and
E deceleration time, msec 90 (70–115) 80 (74–123)
Ea, cm/sec 6 (4–7) 5 (4–7)
contributes to LV filling by a series of
Vp, cm/sec 50 (41–60) 40 (35–49) intraventricular gradients drawing blood
E and A fusion, n (%) 4 (21) 6 (35) from base to apex. In coronary artery
Mild mitral regurgitation, n (%) 3 (16) 5 (28) disease and cardiomyopathies, these gra-
Small residual shunts 0 3 (17) dients are markedly reduced and flow
LAP, mm Hg 8 (7–9) 12 (11–15)a
propagation is slowed (10). Experiments
LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; LA, left showed that Vp was inversely correlated
atrial; E, early transmitral Doppler filling velocity; Ea, early diastolic velocity at the mitral annulus; Vp, with ␶ (r ⫽ ⫺.56 in children and r ⫽
flow propagation velocity; A, late transmitral Doppler filling velocity. ⫺.86 in adults but without a linear rela-
a
p ⬍ .05. Values indicates median (25th–75th percentiles). tion) (7, 18). Vp is also insensitive to
variations in preload (18). This was the
rationale for the calculation of E/Vp.
Doppler ratio calculation remained feasi- and sensitivity. E/Ea proved to have a Studies performed in catheterization lab-
ble in the majority of infants. These vari- high sensitivity in all infants and a high oratories have shown r values ranging
ables have not been previously compared specificity in the subgroup of infants from .71 (children) to .87 (adults) (3, 7).
with direct measurement of LAP in a clin- without mitral valve surgery but a poor Although the relation between E/Vp and
ical situation, which remains a challenge specificity in infants with mitral valve LAP was weaker in our population than
for the noninvasive estimation of the left plasty. E/Vp had a higher specificity but a that observed in previous studies, the test
filling pressure and which concerns our weaker sensitivity than E/Ea to detect allowed the detection of high LAP in 84%
population of infants. Despite the ex- high LAP values in all groups. Further- of the cases. Border et al. (7) found a
treme situation represented by the post- more, in contrast with a high specificity significant correlation between LV end-
operative status of infants with cardiac (100%) of E/Vp for subjects without mi- diastolic pressure and E/Vp but not with
surgical intervention, ratios E/Ea ⬎15 tral valve repair, our results pointed out a E/Ea in a group of 20 patients. However,
and E/Vp ⬎2 could screen patients with lessened specificity (63%) for infants with these patients were older than the infants
LAP ⬎10 mm Hg with good specificity mitral valve surgery. of the present study and had a wide range

Pediatr Crit Care Med 2005 Vol. 6, No. 4 451

 Doppler ratios. It entails a frequent merg-
ing of E and A waves (27% of infants in
our study group) on transmitral flow ve-
locity recordings, due to the long dura-
tion of E and A flow waves, whereas Ea
and late diastolic velocity at the mitral
annulus remain often distinguishable on
TDI recordings. Applicability of E/Ea to
adults with tachycardia remained effi-
cient within the limit of 110 (range 100 –
130) beats/min, still lower than in the
present study (15).
The second factor consists of the mag-
nitude of early diastolic flow and TDI
velocities in children and young adults.
Previous studies have shown a negative
Figure 4. Box plot graphic comparison of peak velocity of early transmitral Doppler filling (E)/early correlation between age and early dia-
diastolic velocity of the lateral mitral annulus (Ea) ratio (left) and of E/flow propagation velocity (Vp) stolic velocities, which applied to flow
ratio (right) between infants with LAP ⱕ10 mm Hg (group A, n ⫽ 19) and patients with LAP ⬎10 mm and tissue (19, 20). Investigations of mi-
Hg (group B, n ⫽ 18). tral annular motion have been performed
in 30 normal children (21). Recently, the
Table 2. Factors associated with high left atrial pressure (⬎10 mm Hg) values (binary logistic age effect on E and Ea was studied in 151
regression analysis) healthy consecutive children aged 1–18
yrs: Although there were statistically sig-
CI for Exp(B) 95% nificant differences in the velocity of
some of the TDI indexes among pediatric
Coefficient B Wald p Exp(B) Inferior Superior age groups, there was no significant lin-
ear correlation with age (22). Infants
E/Ea 2.7 5 0.03 14.8 1.36 162.01
E/Vp 1.9 4 0.04 6.4 1.18 42.37 have the lowest Ea velocities, which in-
crease with age (particularly during the
Exp, exponential; CI, confidence interval; E, early transmitral Doppler filling velocity; Ea, early first year of life). Infants have the highest
diastolic velocity at the mitral annulus; Vp, flow propagation velocity. E/Ea ratio, which decreases with age, car-
diac growth, and decreased heart rate
Table 3. Sensitivity, specificity, predictive value, and area under the receiver operating characteristic (22).
(ROC) curve of Doppler ratios in detection of left atrial pressure ⬎ 10 mm Hg Low Ea values observed in our study
population might be the consequence of
E/Ea E/Vp impaired ventricular systolic and dia-
stolic function related to the underlying
No mitral valve repair n ⫽ 20 n ⫽ 19
Sensitivity, n (%) 8/9 (89) 6/8 (75)
disease and the period of examination
Specificity, n (%) 10/11 (91) 11/11 (100) (very early postoperative phase).
Positive predictive value, n (%) 8/9 (89) 6/6 (100) Decreased LV long axis in infants is
Negative predictive value, n (%) 10/11 (91) 11/13 (85) the third factor of possible Doppler ratio
Area under the ROC curve 0.81a 0.91b shortcomings to be addressed. A higher
Mitral valve repair n ⫽ 17 n ⫽ 17
Sensitivity, n (%) 9/9 (100) 7/9 (78) peak early diastolic LV filling rate has
Specificity, n (%) 3/8 (38) 5/8 (63) been reported in adults than in children
Positive predictive value, n (%) 9/15 (64) 7/10 (70) (23). The mechanism was attributed to
Negative predictive value, n (%) 3/3 (100) 5/7 (71) the small mitral ring rather than to the
Area under the ROC curve 0.63 0.72
All infants n ⫽ 37 n ⫽ 36
small LV size or to E magnitude. The
Sensitivity, n (%) 17/18 (94) 13/17 (76) spreading of the E wave beyond the an-
Specificity, n (%) 13/19 (68) 16/19 (84) nulus toward the apex raises a method-
Positive predictive value, n (%) 17/23 (74) 13/16 (81) ological issue. A small LV long axis con-
Negative predictive value, n (%) 13/14 (93) 16/20 (80) figuration representing about half of the
Area under the ROC curve 0.75a 0.83b
adult size does not facilitate LV color
E, early transmitral Doppler filling velocity; Ea, early diastolic velocity at the mitral annulus; Vp, spreading. Therefore, it might bias the
flow propagation velocity. calculation of Vp. To overcome this prob-
a
p ⬍ .05; b p ⬍ .005. lem, we modified the method previously
proposed by shortening the distance from
the mitral valve plane to the LV cavity (3,
of ages, a broad spectrum of different lenging for an efficient noninvasive eval- 8, 10). The method we applied will need
diagnoses, and different heart rates. uation of LAP in infants. First, the further validation in other settings.
Specific Conditions Related to an In- occurrence of tachycardia shortens dias- Potential Implications. Capability of a
fant Population. Three factors are chal- tole and is particularly concerning for the noninvasive technique to assess LAP val-

452 Pediatr Crit Care Med 2005 Vol. 6, No. 4

ues in this population might have a major
practical impact. The measurements of
E/Ea and E/Vp were reproducible and
easy to obtain. Infants are highly fragile
in the postoperative period. Dobutamine
and catecholamines are frequently
needed to maintain a convenient hemo-
dynamic status. Accordingly, invasive
monitoring is frequently the rule. Moni-
toring requires the insertion of a small
catheter in the left atrium to detect early
LV dysfunction and increased right ven-
tricular afterload. Unfortunately, the
placement of a left atrial catheter may be
complicated by systemic gaseous emboli
and sepsis. Hemorrhage may also follow
catheter removal (13, 14). We progres-
sively abandoned invasive LAP measure-
ments. This attitude needs frequent
Doppler examinations. The impact of
such a change in infant monitoring has
to be evaluated.
Limitations. The LAP cutoff value of
10 mm Hg was selected because values
⬎10 mm Hg are considered as abnormal
after interventions performed using nor-
mothermic bypass with myocardial pro-
tection achieved by warm-blood cardio-
plegia. However, we should keep in mind
that as left ventricular compliance wors-
ens, a higher left atrial pressure is fre-
quently required in such a way that LAP
⬎10 mm Hg is not necessarily a critical
value. In other cases, this value and even
lower filling pressures are associated with
left ventricular distension and worsening
hemodynamics. The second reason to se-
lect LAP cutoff value of 10 mm Hg was
that it represented the median value and
therefore allowed us to distinguish two
groups with almost identical numbers of
patients.
The Doppler ratios were thought to be
affected by the mitral valve surgery.
Higher peak E velocities were due to
some restriction of the atrioventricular
apparatus resulting from the interven-
tion. This was probably the explanation
for the poor specificity observed in this
subgroup of patients. On the other hand,
this surgery is not exceptional in this
infant population and it was worthwhile
to define the reliability of Doppler ratios.
We used one single recording for the
annular trace. We assumed that the lat-
eral annulus was easier to interpret than
the septal annular area in these infants
having a majority septal defect repara-
tion. Decreased applicability led us to dis-
card pulmonary venous trace recordings
in our infant population. Although our
results are encouraging, there is still an
Pediatr Crit Care Med 2005 Vol. 6, No. 4
important role of left atrial catheters in
the early postoperative management of
infants with an unstable hemodynamic
state in order to obtain a continuous
monitoring that is less time-consuming
and operator dependent than the echo-
cardiographic examination.
CONCLUSIONS
In a population consisting of infants
with a specific clinical presentation,
Doppler ratios appeared as promising
tools for the noninvasive assessment of
LAP.
REFERENCES
1. Appleton CP, Galloway JM, Gonzalez MS, et al:
Estimation of left ventricular filling pressures
using two-dimensional and Doppler echocardi-
ography in adult patients with cardiac disease.
Additional value of analyzing left atrial size, left
atrial ejection fraction and the difference in
duration of pulmonary venous and mitral flow
velocity at atrial contraction. J Am Coll Cardiol
1993; 22:1972–1982
2. Nagueh SH, Katherine J, Middleton J, et al:
Doppler tissue imaging: A noninvasive tech-
nique for evaluation of left ventricular relax-
ation and estimation of filling pressures.
J Am Coll Cardiol 1997; 30:1527–1533
3. Garcia MJ, Ares MA, Asher C, et al: An index
of early left ventricular filling that combined
with pulsed Doppler peak E velocity may
estimate capillary wedge pressure. J Am Coll
Cardiol 1997; 29:448 – 454
4. Kuecherer HF, Muhiudeen IA, Kusumoto
FM, et al: Estimation of mean left atrial pres-
sure from transesophageal pulsed Doppler
echocardiography of pulmonary venous flow.
Circulation 1990; 82:1127–1139
5. Dumesnil JG, Gaudreault G, Honos GN, et al:
Use of Valsalva maneuver to unmask left ven-
tricular diastolic function abnormalities by
Doppler echocardiography in patients with
coronary artery disease or systemic hyper-
tension. Am J Cardiol 1991; 68:515–519
6. Hurrel DG, Nishimura RA, Ilstrup DM, et al:
Utility of preload alteration in assessment of
left ventricular filling pressure by Doppler
echocardiography: A simultaneous catheter-
ization and Doppler echocardiographic
study. J am Coll Cardiol 1997; 30:459 – 467
7. Border WL, Michelfelder EC, Glascock BJ, et
al: Color M-Mode and Doppler tissue evalua-
tion of diastolic function in children: Simul-
taneous correlation with invasive indices.
J Am Soc Echocardiogr 2003; 16:988 –994
8. Brun P, Tribouilloy C, Duval AM, et al: Left
ventricular flow propagation during early
filling is related to wall relaxation: A color
M-mode Doppler analysis. J Am Coll Cardiol
1992; 20:420 – 432
9. Takatsuji H, Mikami T, Urasawa K, et al: A
new approach for evaluation of left ventricu-
lar diastolic function: Spatial and temporal
analysis of left ventricular filling flow propa-
gation by color M-mode Doppler echocardi-
ography. J Am Coll Cardiol 1996; 27:365–371
10. Stugaard M, Smiseth OA, Risoe C, et al: In-
traventricular early diastolic filling during
acute myocardial ischemia: Assessment by
multigated color M-mode Doppler. Circula-
tion 1993; 88:2705–2713
11. Sohn DW, Chai IH, Lee DJ, et al: Assessment
of mitral annulus velocity by Doppler tissue
imaging in the evaluation of left ventricular
diastolic function. J Am Coll Cardiol 1997;
30:474 – 480
12. Nagueh SF, Sun H, Kopelen HA, et al: He-
modynamic determinants of the mitral an-
nulus diastolic velocities by tissue Doppler.
J Am Coll Cardiol 2001; 37:278 –285
13. Feerick AE, Church JA, Zwischenberger J, et
al: Systemic gaseous microemboli during left
atrial catheterization: A common occur-
rence? J Cardiothorac Vasc Anesth 1995;
9:395–398
14. Yokoyama M, Wada J: Placement of left atrial
pressure monitoring catheter: Use of an elastic
surgical suture. Ann Thorac Surg 1987; 43:226
15. Nagueh SF, Mitaki I, Kopelen HA, et al:
Doppler estimation of left ventricular filling
pressure in sinus tachycardia. Circulation
1998; 98:1644 –1650
16. Okamoto M, Sakura E, Shimamoto H, et al:
Analysis of mitral inflow velocity pattern in
relation to left ventricular end-diastolic pres-
sure. J Cardiogr 1986; 16:941–948
17. Ommen SR, Nishimura RA, Appleton P, et al:
Clinical utility of Doppler echocardiography
and tissue Doppler imaging in the estimation
of left ventricular filling pressures. Circula-
tion 2000; 102:1788 –1794
18. Garcia MJ, Smedira NG, Greenberg NL, et al:
Color M-Mode Doppler flow propagation veloc-
ity is a preload insensitive index of left ventric-
ular relaxation: Animal and human validation.
J Am Coll Cardiol 2000; 35:201–208
19. Schmitz L, Koch H, Bein G, et al: Left ven-
tricular diastolic function in infants, chil-
dren, and adolescents. Reference values and
analysis of morphologic and physiologic de-
terminants of echocardiographic Doppler
flow signals during growth and maturation.
J Am Coll Cardiol 1998; 32:1441–1448
20. Palka P, Lange A, Fleming AD, et al: Age-
related transmural peak mean velocities and
peak velocity gradients by Doppler myocar-
dial imaging in normal subjects. Eur Heart J
1996; 17:940 –950
21. Rychik J, Tian ZY: Quantitative assessment of
myocardial tissue velocities in normal chil-
dren with Doppler tissue imaging. Am J Car-
diol 1996; 77:1254 –1257
22. Swaminathan S, Ferrer PL, Wolff GS, et al:
Usefulness of tissue Doppler echocardiogra-
phy for evaluating ventricular function in
children without heart disease. Am J Cardiol
2003; 91:570 –574
23. Takenaka K, Dabestani A, Waffarn F, et al:
Effect of left ventricular size on early dia-
stolic left ventricular filling in neonates and
in adults. Am J Cardiol 1987;59:138 –141.
453