Download Reliability of Central Venous Pressure as an

Reliability of Central Venous Pressure as an
Indicator of Left Atrial Pressure*
A Study in Patients with Mitral Valve Disease
Hubert Bell, M.D.," David Stubbs, M.D., and David Pugh, M.D.OQ
Central venous pressure (CVP) measurements have been used extensively in the
management of critically ill patients. The value of CVP determinations was
studied by determining the relationship of left atrial and right atrial pressares in
200 patients undergoing transsepta1 heart catheterization. All patients had mitral
valve disease. The correlation coefficient of left atrial and right atrial pressure was
0.48. Low right atrial pressure (below 3 mm Hg) wss seen &I 18 of 91 patients
with left atrial pressure 20 mm H g or greater. In 25 patients, acute pressure
changes were induced by volume expansion, pachng, or drug infusion ( h o p
terenol or phenylephrine). Increases in left atrial pressure hquently failed to be
reflected in right atrial pressure changes. C e n W venous pressare may be
misleading as a guide to left atrial pressure.
entral venous pressure (OVP) measurements
have been used extensively to provide guidelines in the treatment of critically ill patients. Since
CVC and right atrial pressure are identical,' the
clinical usefulness of CVP monitoring depends upon
the relationship between left atrial and right atrial
pressures. The superiority of CW over other methods of determining venous pressure has been established.2 Important clinical decisions have been
based on CVP measurements and on the changes
which occur during the course of observation and
treatment.3-6 The following study was performed to
determine the relationship of CVP to left atrial
pressure.
Patient material consisted of 200 consecutive ~ a t i e n t swith
mitral valve disease undergoing cardiac catheterization. Patients with relatively pure mitral stenosis or mitral insdciency were included as well as patients with "mixed" mitral
valve disease. Patients with associated aortic or tricuspid
valve disease were excluded from the study. Transseptal
technique was used to introduce a left atrial catheter ( Brockenbrough 8#F) in all patients. Pressure recordings were
'From the De artment of Medicine, Cardiovascular Section,
University o h a n s a s Medical Center, Kansas City, Kansas.
Supported in part by NIH Grant 5 TO1 HE05670-05.
"Assistant Professor of Medicine.
obtained with use of Statham pressure transducer (model
P23D) and Electronics for Medicine recorder.
Baseline right atrial pressure recordings were made
through the transseptal catheter immediately before it was
advanced into the left atrium. Pulmonary artery pressure
measurements were obtained using a NIH 6 or 7F Dacron
thin wall catheter and were recorded simultaneously with
right and left atrial pressures. The Fick method was used to
obtain cardiac output.
In 25 patients, a significant rise in atrial pressure was
induced by transvenous pacing, infusion of isoproterenol or
phenylephrine, or volume expansion. The transsepta1 catheter
was then rapidly withdrawn back into the right atrium while
pressures were continuously recorded. Changes in pulmonary
artery pressure were also continuously recorded. The rise in
left atrial pressure was sustained for at least 15 minutes and
in some patients for one hour. Volume expansion was accomplished by rapid infusion of 200 to 1000 ml of 5 percent
dextrose and water into the venous circulation. Pacing was
performed at a rate of 100-150/minute from the right atrium
or right ventricle. Isoproterenol was infused at a rate of 5-10
pg/min. Phenylephrine was infused slowly over five to ten
minutes to a total dose of 2-3 mg.
The relationship between left atrial and right
atrial pressures is summarized in Table 1 and Figure
1. The correlation coefficient of left atrial and right
h t pressure
atrial pressure was 0.48. ~ i ~ atrial
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21508/ on 05/03/2017
BELL, STUBBS AND PUGH
FIGURE1 . Relationship between left atrial and right atrial pressures with standard error.
equaled 0.228 of left atrial pressure. Four patients
( 4/31 ) with markedly elevated left atrial pressure
(26 mm Hg or greater) had low right atrial
pressures ( 3 mm Hg or less). Eighteen patients
(18191) with left atrial pressure 20 m m Hg or
greater had low right atrial pressure ( 3 mm Hg or
less). Three patients (3139) with normal left atrial
pressure ( 12 mm Hg or less) had elevated right
atrial pressure ( 8 mm Hg or greater).
In contrast, mean pulmonary artery pressure was
elevated (25 mm Hg or greater) in 90 of 91 patients
with left atrial pressure 20 mm Hg or greater. The
correlation coefficient of left atrial and mean pulmonary artery pressure was 0.70 (Fig 2 ) . Mean
pulmonary artery pressure equaled 1.23 of left atrial
pressure.
The effect of acute elevation of left atrial pressure
on right atrial and pulmonary artery pressure is
summarized in Table 2 and Figures 3 and 4. No
consistent pattern of right atrial or pulmonary
artery pressure change was observed in response to
the left atrial pressure rise. Although in 16 patients a
Table
1-7
3 or less
MPA
......
..
I-Division
of 200 patients according t o
pressure relationships
RA Pressurr
(mm HR)
8 o r greater
rise in left atrial pressure was reflected by a rise in
right atrial pressure, in four of these a left atrial
pressure rise of 10 mm Hg or greater was associated
with a 1 mm Hg or less rise in right atrial pressure.
In two patients, right atrial pressure increased
despite no increase in left atrial pressure. In seven
patients, a rise in left atrial pressure was not
associated with a rise in right atrial pressure.
L A Pressure (mm Hg)
11 or
less
12-19
20-25
3
14
8
27
22
-
35
-
39
70
10
36
14
-
60
26 or
Total
greater Patients
17
10
4
31
38
87
75
200
FIGURE2. Relationship between left atrial and mean prllmonary artery pressures with standard error.
CHEST, VOL. 59, NO. 2, FEBRUARY 1971
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21508/ on 05/03/2017
CENTRAL VENOUS PRESSURE AS INDICATOR OF LEFT ATRIAL PRESSURE
16 ' ARA
mm Hg
8
0
x
VOLUME EXPANSION
WING
NEOSYNEPHRINE
-
0
.
x
FIGURE
3. Changes in left atrial
and right atrial pressures with
volume expansion, pacing, and
drug infusion.
A LA
mn Hg
-4
0
4
8
12
I
//
1
I
1
16
"
28
32
36
D~scussro~
Important clinical decisions on critically ill patients have been based on right atrial (CVP)
measurements. A low initial central venous pressure
associated with low cardiac output has indicated
that volume expansion therapy should be initiated.3
A rapid rise in central venous pressure, or a high
initial CVP is usually an indication that volume
expansion therapy should not continue.* Digitalis
has been used when low cardiac output has been
associated with a high CVP.5 Decisions regarding
therapy for arrhythmias, eg digitalis, cardioversion,
are based on the effect the arrhythmia has had on
28
o
24
x
-
I
CVP. CVP monitoring
- has also been used to determine the safety of vasoactive drugs, eg phenylephrine and isoproterenol.6 In our study, volume
expansion, rapid pacing, phenylephrine and isoproterenol were selected to induce left atrial pressure
changes because of their practical clinical implications.
It would be ideal if left atrial pressure could be
monitored in critically ill patients, but technical
difficulties make this impractical for most patients.
CVP is readily available and is assumed to reflect
left atrial Pressure. Left atrial and right atrial
pressures are similar under certain circumstances ie,
VOLUME EXPANSION
PACING
NEOSYNEPHRINE
ISOPROTERENOL
X
20 A MPA
16- m m ~ q
12
-
0
FIGURE
4. Changes in left atrial
and pulmonary artery pressures
with volume expansion, pacing,
and d ~ infusion.
g
-4
0
4
8
.
.
12
16
20
CHEST, VOL. 59, NO. 2, FEBRUARY 1971
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21508/ on 05/03/2017
24
I
I
28
32
I
36
BELL, STUBBS AND PUGH
Table
2--Hemodynaniicr before and during volume
exponrion
Patient Lesion
MS
MS
MI
MS
MS
MS/MI
MS
MS
MS/MI
MS
MS/MI
MS
MI
MI
MI
MI
MI
MI
MI
MS
MS/MI
MS
MS
MS
MS
Time
Pressure
Cardiac
(mm Hg)
Output
LA
RA MPA (L/min)
Baseline
Volume
Baseline
Volume
Baseline
Volume
Baseline
Volume
Baseline
Volume
Baseline
Pacing
Baseline
Pacing
Baseline
Pacing
Baseline
Pacing
Baseline
Pacing
Baseline
Pacing
Baseline
Pacing
Baseline
Neosynephrine
Baseline
Neosynephrine
Baseline
Neosynephrine
Baseline
Neosynephrine
Baseline
Neosynephrine
Baseline
Neosynephrine
Baseline
Neosynephrine
Baseline
Isoproterenol
Baseline
Isoproterenol
Baseline
Isoproterenol
Baseline
Isoproterenol
Baseline
Isoproterenol
Baseline
Isoproterenol
constrictive pericarditis and large atrial septa1 defects.' However, since elevated right atrial pressures
usually occur only with right heart failure and
elevated right ventricular end-diastolic pressures,
elevated left atrial pressure may occur with normal
CVP. Elevated CVP may occur with normal left
atrial pressures in cor pulmonale. The information in
our study emphasizes this lack of relationship since
20 percent of patients with left atrial pressure
greater than 20 mm Hg have low right atrial
pressure below 3 mm Hg. Similarly, 8 percent of
patients with right atrial pressure greater than 8 mm
Hg have normal left atrial pressures (below 12 mm
Hg). Consequently, CVP frequently cannot be used
as a measure of left atrial pressure.
Volume expansion therapy has gained widespread
popularity. By augmenting atrial pressure, the rapid
infusion of intravenous fluids may augment stroke
volume in accordance with the Frank-Starling principle. A plateau is eventually reached so that
additional increments of fluid may result in increased filling pressure with very little augmentation
of stroke volume.8 If volume expansion is continued
past this point, pulmonary edema will eventually
result. Recently, volume expansion therapy has been
advocated even in the patient in cardiogenic shock
following myocardial infarction.9 Cohn and Luria3
have stated that right and left atrial pressures rise
symmetrically during volume expansion in normal
subjects and that consistent rises in CVP are seen
before left atrial pressure is increased to a dangerous level. A CVP rise of 3 cm water or more with
each 100 ml of volume infused has been used as a
guideline indicating abnormal rises in left atrial
pressure.3 In our study, right atrial pressure did
increase in four of five patients during volume
infusion. However, in patient 5 this rise was associated with a fall in left atrial pressure. In patient 3,
right atrial pressure fell while left atrial pressure
rose to 36 mm Hg. In patient 2, a 1 mm rise in right
atrial pressure was associated with a 6 mm rise in
left atrial pressure.
Baseline mean pulmonary artery pressure showed
good correlation with left atrial pressure. Mean
pulmonary artery pressure was used in this study
because of the technical difficulties of obtaining
reliable pulmonary artery end-diastolic pressure.
While marked elevations in mean pulmonary artery
pressure occasionally occurred with only modestly
elevated left atrial pressures, the absence of pulmonary hypertension reliably indicated that left atrial
pressure was not markedly elevated. This lends
support to the concept that mean pulmonary artery
pressure rather than CVP should be used to make
decisions regarding therapy. While the changes in
mean pulmonary artery pressure with rises in left
atrial pressure were not consistent, a normal mean
pulmonary pressure excluded a marked elevation of
left atrial pressure.
Patients with mitral stenosis and insufficiency
were used in this study because they provided a
readily accessible group with chronic elevations of
CHEST, VOL. 59, NO. 2, FEBRUARY 1971
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21508/ on 05/03/2017
CENTRAL VENOUS PRESSURE AS INDICATOR OF LEFT ATRIAL PRESSURE
left atrial pressure. Mitral stenosis is uncommon
among critically ill patients. The mechanism of left
atrial pressure elevation is not the same in rnitral
stenosis as in other forms of heart disease. In
patients with mitral stenosis left atrial pressure
cannot be used as an indicator of left ventricular
end-diastolic pressure. However, both may have low
cardiac outputs and chronically elevated left atrial
pressure. Elevation of right heart pressure is presumably due to elevated left atrial pressures.10
Furthermore, mitral insuiliciency occurs with surprising frequency in critically ill patients. This may
be due t o papillary muscle dysfunction or rupture or
merely due to a markedly dilated left ventricle.11
Poor correlation exists between baseline left atrial
and right atrial pressures. Significant differences
between these pressures may occur with a frequency of 20 percent. Rapid changes in left atrial
pressure induced by volume expansion, tachycardia,
or vasoactive drugs may not be reflected in right
atrial or pulmonary artery pressure changes. Baseline mean pulmonary artery pressure may be helpful in predicting left atrial pressure. If the mean
pulmonary artery pressure is not elevated, it is likely
left atrial pressure is normal or only slightly elevated.
173
REFERENCES
1 Zimmerman HA: Intravascular Catheterization, Springfield, Charles C Thomas, 1966
2 Weil MH, Shubin H, Rosoff L: Fluid repletion in circulatory shock. JAMA 192:668,1965
3 Cohn JN, Luria MH: Studies in clinical shock and hypotension. JAMA 193:891, 1964
4 Tristani FE, Cohn JN: Masked heart failure in shock.
Circulation, Supplement 3,33 and 34, 1966
5 Siege1 JH, Creenspan M, Del Guercio LRM: Abnormal
vascular tone, defective oxygen transport and myocardial
failure in human septic shock. Ann Surgery 165:504,
1967
6 Hardaway RM, James PM, Anderson RW, et al: Intensive
study and treatment of shock in man. JAMA 199:115,
1967
7 Albers WH, Hugenholb PC, Nadas AS: Constrictive
pericarditis and atrial septal defect, secundum type. Amer
J Cardiol23:850, 1969
8 Burton AC: Physiology and Biophysics of the Circulation,
Chicago, Year Book, 1965
9 Nixon PC, Ikram H, Morton S: Cardiogenic shock treated
with infusion of dextrose solution. Amer Heart J 73:843,
1967
10 Linhart JW, Hildner FJ, Barold SJ, et al: Myocardial
function in patients with coronary artery disease. Amer J
Cardiol23:379, 1969
11 Friedberg CF: Diseases of the Heart, Philadelphia, W. B.
Saunders Co., 1966
Reprint requests: Dr. Bell, University of Kansas Medical
Center, Kansas City, Kansas 66103
RESIDENT FELLOWSHIP PROGRAM
American College of Chest Physicians
In an effort to advance knowledge of chest diseases
and their treatment, the American College of Chest
Physicians has established a Resident Fellowship program
through which medical graduates from other countries
may receive assistance in postgraduate medical training
in chest diseases in the United States. Under this program, the College aids qualified physicians in obtaining
suitable postgraduate residencies in cardiopulmonary
diseases at accredited hospitals or institutions in the
United States. At this time, the College is offering the
following fellowship for 1972:
Eudowood Fellowship for Tuberculosis with a grant
of $2,500.00 per year for postgraduate training in the
United States in tuberculosis.
All applicants must be ECFMG certified and the
Resident Fellowship program of the College requires
that the candidate be assured of a responsible position in
the field of chest diseases when he returns to his country.
Physicians who are interested should write to Dr.
Arthur M. Olsen, Chairman of the Council on International Mairs, in care of the ACCP, 112 East,Chestnut
Street, Chicago 6061 1.
CHEST, VOL. 59, NO. 2, FEBRUARY 1971
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21508/ on 05/03/2017