(Chest. 2001;120:1322-1326.)
© 2001
American College of Chest Physicians
Invasive Arterial BP Monitoring in Trauma and Critical Care*
Effect of Variable Transducer Level, Catheter Access, and Patient Position
Ulysse G. McCann, II, MD;
Henry J. Schiller, MD;
David E. Carney, MD;
Judy Kilpatrick, RN;
Louis A. Gatto, PhD;
Andrew M. Paskanik and
Gary F. Nieman, BS
*
From the Cardiopulmonary and Critical Care Laboratory (Drs. McCann, Schiller, and Carney; Ms. Kilpatrick; Mr. Paskanik; and Mr. Nieman), Department of Surgery, SUNY Upstate Medical University, University Hospital, Syracuse, NY; and Department of Biological Sciences (Dr. Gatto), SUNY at Cortland, NY.
Correspondence to: Gary F. Nieman, BS, SUNY Health Science Center, Department of Surgery, 750 East Adams St, Syracuse, NY 13210
 |
Abstract
|
|---|
Objectives: (1) To determine the validity of current
recommendations for direct arterial BP measurement that suggest that
the transducer (zeroed to atmosphere) be placed level with the catheter
access regardless of subject positioning: and (2) to investigate the
effect of transducer level, catheter access site, and subject
positioning on direct arterial BP measurement.
Design:
Prospective, controlled laboratory study.
Setting:
Large animal laboratory.
Subjects: Five Yorkshire
pigs.
Interventions: Anesthetized animals had 16F
catheters placed at three access sites: aortic root, femoral artery,
and distal hind limb. Animals were placed in supine, reverse
Trendelenburg 35°, and Trendelenburg 25° positions with a
transducer placed level to each access site while in every
position.
Measurements and main results: For each
transducer level, five systolic and diastolic pressures were measured
and used to calculate five corresponding mean arterial pressures (MAPs)
at each access site. When transducers were at the aortic root, MAP
corresponding to aortic root pressure was obtained in all positions
regardless of catheter access site. When transducers were moved to the
level of catheter access, as current recommendations suggest,
significant errors in aortic MAP occurred in the reverse Trendelenburg
position. The same trend for error was noted in the Trendelenburg
position but did not reach statistical significance.
Conclusions: (1) Current recommendations that suggest
placing the transducer at the level of catheter access regardless of
patient position are invalid. Significant errors occur when subjects
are in nonsupine positions. (2) Valid determination of direct arterial
BP is dependent only on transducer placement at the level of the aortic
root, and independent of catheter access site and patient
position.
Key Words: arterial BP • catheter • critical care • invasive • monitoring • patient positioning • reverse Trendelenburg • supine • transducer level • trauma • Trendelenburg
|
Introduction
|
|---|
Direct
arterial BP monitoring is standard practice in a majority of trauma and
critical-care patients and the most common invasive monitoring device
used in the ICU.1
This routine procedure may become
problematic when patients are not in the supine position. In fact,
increasingly aggressive positioning of these patients may introduce
error into arterial pressures measured by this technique because of
altered transducer and catheter locations. A surprising degree of
controversy and confusion is evident in medical2
and
nursing journals3
4
and in critical-care
texts5
regarding arterial BP measurement in nonstandard
scenarios frequently encountered in trauma and critical care. Even
among specialists, there is a persistent misunderstanding regarding the
basics of how hydrostatic pressure influences measured pressures in a
fluid-filled catheter system.6
Current standards of hemodynamic monitoring recommend zero reference
point (a transducer zeroed to atmosphere) for all
fluid-filled monitoring systems be placed level with the catheter tip
to negate the effects of hydrostatic and atmospheric pressure. These
guidelines further emphasize the relationship between transducer and
catheter-tip location as the crucial factor in valid direct arterial BP
measurement.7
We questioned the validity of this approach
based on accepted physics principles and clinical experience with
patients. We hypothesized instead that the transducer (zero-reference
point) should always be placed level to the aortic root when measuring
arterial BP, regardless of patient position or catheter-tip location,
and that to do otherwise would introduce significant error. These
experiments were conducted to test this hypothesis by studying the
effect of three variables on direct arterial BP measurement: (1)
transducer level (zero reference point), (2) catheter access site, and
(3) patient position.
|
Materials and Methods
|
|---|
Yorkshire pigs (n = 5) weighing from 18 to 33 kg were
preanesthetized with atropine, 0.05 mg/kg; acepromazine, 1.1
mg/kg; and ketamine, 22 mg/kg IM. Peripheral IV access was established,
and sodium pentobarbital, 50 mg/mL, was administered to effect.
Continuous anesthesia with sodium pentobarbital, 6 mg/kg/h, was
delivered via an infusion pump (model 907; Harvard Apparatus;
Holliston, MA), and pancuronium bromide was administered in a
small bolus to maintain paralysis. Animals were then surgically
prepared using a clean technique for hemodynamic and ventilatory
monitoring. Surgical preparation included tracheal intubation via
tracheostomy, direct bladder catheterization, central venous
cannulation of the internal jugular vein (7F triple lumen; Arrow
International; Reading, PA) and 16F intra-arterial catheters (Cathlon;
Johnson & Johnson; Arlington, TX) placed by cutdown in the right
distal hind limb, left femoral artery, and the aortic root via the left
carotid artery. Catheters were connected via identical lengths of
tubing to three separate transducer/stopcock assemblies (Argon Model
049–992-000A; CB Sciences; Dover, TX) that were mounted on a movable
platform (Cobe Cardiovascular; Arvada, CO) attached to an IV pole (Fig 1
). Measurements were recorded on a recording and analysis system
(PowerLab/16s; ADInstruments; Mountain View, CA) and computer (Dell
Dimension XPS R400; Dell Computer; Round Rock, TX) [Fig 1
].
Animals received ventilation (Amadeus FT; Hamilton Medical;
Rhäzüns, Switzerland) at standard settings: tidal volume, 7
to 10 mL/kg, fraction of inspired oxygen, 0.50; and positive
end-expiratory pressure, 5 mm Hg. Respiratory rate was titrated to keep
minute volume near 4.0 L/min. Baseline arterial blood gases were
measured (model ABL5; Radiometer; West Sussex, UK) to ensure normoxia
at the beginning of each protocol.
View larger version (23K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1.. Experimental apparatus and monitoring:
simultaneous measurement of aortic, femoral, and distal lower-extremity
arteries; movable transducer platform; and positioning table.
|
|
Protocol
Prior to each experiment, the three transducers were separately
zero-calibrated to the recording and analysis system using a standard
mercury manometer (Baumanometer; W.A. Baum; Copiague, NY). Animals were
placed in supine position on a positional operating table (model
25000115; Shor-line; Cowbridge, UK) with the transducer platform
leveled to the aortic root with a 24-inch carpenter’s level (Fig 2 , left, A). The three catheter tips—aortic,
femoral, and distal—were all within 2 to 3 cm of the same horizontal
plane in this position. Each transducer was zeroed, and then arterial
BP tracings for the three catheter locations were recorded
simultaneously for 5 min. Animals were then placed in a 35° reverse
Trendelenburg position (Fig 2
, middle, B). Five
minutes were allowed to elapse before any further measurements were
taken. The transducer platform was then leveled with the aortic root,
zeroed, and arterial BP tracings for the three catheter locations were
again recorded simultaneously for 5 min. With the position of the
animal unchanged, the transducer platform was moved down the IV pole
and leveled with the femoral catheter. Again the transducers were
zeroed and arterial tracings for the three catheter locations were
simultaneously recorded. The transducer platform was then moved to the
level of the distal (hind limb) catheter, zeroed, and arterial tracings
recorded as before. Measurements were also taken of the vertical
distance along the IV pole between aortic and femoral transducer
levels, and between femoral and distal transducer levels so that these
distances could be correlated with measured pressure differences.
Animals were returned to the supine position for 5 min, and then the
same protocol was repeated in a 25° Trendelenburg position (Fig 2 ,
right, C). Following the experiment, animals were
killed with an overdose of pentobarbital (90 mg/kg IV). The experiments
described in this study were performed in compliance with standard
practices of care for laboratory animals. The protocol was
approved by the Committee for the Humane Use of Animals at the SUNY
Health Science Center, Syracuse, NY.
View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2.. Experimental protocol and movements.
Left, A: supine position with all
catheters in same plane and a single transducer level.
Middle, B: 35° reverse Trendelenburg
position with catheters in different planes and variable transducer
levels moved downward in numbered order at each catheter level.
Right, C: 25° Trendelenburg position
with variable catheter planes and corresponding transducer levels
moving upward in numbered order.
|
|
Data Analysis
Seven data sets, each composed of three arterial tracings
(aortic, femoral, and distal) were obtained as described above: one set
from the supine position and three sets each from the Trendelenburg and
reverse Trendelenburg positions. Each data set of tracings was
converted into numeric systolic and diastolic means by the recording
and analysis system by taking 20-s time intervals each minute during
the 5-min tracing. Mean arterial pressure (MAP) was then calculated for
all measurements using the standard formula,
Thus, each numeric data set consists of 45 measurements: 15
measurements for each catheter location (aortic, femoral, distal) made
up of 5 systolic measurements, 5 diastolic measurements, and 5 MAP
measurements. We generated seven numeric data sets for each of the five
animals, accounting for > 1,575 measurements. Only the MAPs were used
to generate the final mean values for each catheter location.
Statistical analysis was performed within positions and between
transducer levels using Instat 2.0. All reported values are
means ± SD and were compared using analysis of variance. Differences
were considered significant at a probability level of 95%
(p < 0.05).
|
Results
|
|---|
As demonstrated in Table 1
, no significant differences in arterial pressures can be attributed to
variable catheter site in the supine position. Animals placed in the
reverse Trendelenburg position, with transducers at the aortic root,
consistently showed a decrease in MAP due to orthostasis that was also
independent of catheter site. However, as the transducer was lowered to
the femoral and then distal catheter access sites, as current
guidelines recommend, significant errors in measured arterial pressures
occurred when compared to actual aortic root pressure being monitored
almost simultaneously. The same trends were noted in the Trendelenburg
position but did not reach statistical significance.
Measurements between each transducer level in centimeters showed that
on average, every centimeter the transducer moved down resulted in a
0.63-mm Hg pressure decrease (Table 2
). This corresponds closely with the known value of 0.73 mm Hg change
for each 1 cm H2O in either direction.
|
Discussion
|
|---|
Invasive monitoring in trauma and critical care presents unique
challenges due to the variable nature of injuries. It is often
impossible to follow routine protocol with respect to patient
positioning and monitoring devices such as catheters and transducers.
Head injuries,8
congestive heart failure, respiratory
insufficiency,9
and hemorrhagic shock are
situations that may call for aggressive nonsupine positioning. Burn
patients also present numerous problems for catheter access sites and
positioning due to the location and percentage of total body surface
area involved. Although these circumstances may be transient, many will
be prolonged. Even so, each of these situations involves a decision on
where to place the transducer for valid direct arterial BP
measurements.
The purpose of these experiments was to test the effects and
relationships of three variables—transducer level (zero reference
point), catheter access site, and patient position—on direct arterial
BP measurement. In doing so, we specifically sought to determine the
validity of current guidelines that recommend that all fluid-filled
monitoring systems be placed level with the catheter
tip.3
4
7
10
While such recommendations are valid for
patients in the supine position, they are prone to significant error in
other circumstances. We found that in the reverse Trendelenburg
position of 35°, errors in MAP of almost 30 mm Hg occurred in distal
arteries when the transducer was placed at the level of the catheter
tip. The same trend for error was noted in the Trendelenburg position
but did not reach statistical significance.
The clinical implications of such errors are important to recognize,
and a hypothetical example may be illustrative. At our institution, it
is not uncommon to find a burn patient with only dorsalis pedis
arteries available for arterial monitoring, who is in the
reverse Trendelenburg position or sitting up in bed with a normal BP
(120/80 mm Hg). On further inspection, one may note that the transducer
is at the level of the catheter tip, down from the dorsalis pedis
artery, from 20 to 30 cm below the correct level to transduce. When the
transducer is moved to the aortic root, the patient has a clearly
hypotensive pressure of 90/65 mm Hg. Thus, using current guidelines in
this scenario represents a serious overestimation of tissue-perfusion
pressure of the patient and a potential for further errors in clinical
judgment.
The important findings to be emphasized from this study are: (1) that
valid arterial BP measurements are obtained only when the transducer is
placed at the level of the aortic root, and (2) that direct arterial BP
measurement is independent of catheter access site and patient position
if the transducer is at the proper level. In other words, a patient may
be in any position, with a catheter in any artery, and the clinician
will be able to obtain a valid arterial BP as long as the transducer is
level with the aortic root. When this condition is met, it makes no
difference where the catheter is inside the system, or what position
the system happens to be in. Although these concepts may seem
counterintuitive, they can be confirmed by simple manipulation of a
fluid-filled catheter transducer system and graduated cylinder.
The principle behind these results, established by Courtois et
al,6
is that the proper level for a transducer to negate
the effects of hydrostatic pressure is always at the top of the fluid
column in the system being analyzed (in this case, the aortic root).
The reason the aortic root is the only reference position that will
accurately reflect arterial BP is twofold. First, the central MAP and
particularly the aortic mean is the key component in coronary and
cerebral perfusion. This is the pressure that is sensed by baroreceptor
mechanisms. Furthermore, this is the pressure that is indirectly
measured using standard sphygmomanometer techniques. Second, clinicians
are not interested and should not care what the pressure is in a distal
peripheral artery. Not only is the value obtained inaccurate due to the
effects of hydrostatic pressure, it is altogether the wrong pressure
being obtained.
Although this study makes an important clarification regarding direct
arterial BP measurement applicable to clinical practice, it raises
other questions. For instance, where is the critical point, in degrees,
at which the reverse Trendelenburg position begins to cause
statistically significant errors in arterial BP? Also, why did
measurements in the Trendelenburg position not reach statistical
significance? At what point, if any, would they reach significance?
Although it may have been due to the fact that there was less
inclination (25° in the Trendelenburg position compared to 35° in
the reverse Trendelenburg position) making the distance from aortic
root to catheter access site too short to provide an adequate fluid
column, other factors may be influential. It may be a function of the
differing capacitance in the arterial system above and below the aortic
root. Obviously there are differences in hydrostatic gradients that
exist from head to heart compared to foot to heart when in nonsupine
positions. Yet another consideration is the effect lateral rotational
therapy11
in the ICU may have on direct arterial pressure
measurements. These will be topics of future study.
|
Conclusion
|
|---|
In conclusion, we demonstrated in a clinically relevant, large
animal model the effect of position, catheter access site, and
transducer level on arterial BP measurement. Although these concepts
may be fairly well established in the minds of critical-care
physicians, there is considerable confusion evident in others:
particularly the practices and guidelines of ICU nurses. We found that
current recommendations that suggest the transducer be placed level
with the catheter tip are invalid in the nonsupine positions frequently
encountered in trauma and critical care. Valid direct arterial BP
measurements depend only on having the transducer level with the aortic
root. We believe these findings are an important clarification
deserving emphasis and application in the clinical care of trauma and
critical-care patients.
|
Footnotes
|
|---|
Abbreviation: MAP = mean arterial pressure
Presented in part at the 65th Annual International Scientific Assembly
of the American College of Chest Physicians, Chicago, IL, November
1–4, 1999.
|
References
|
|---|
-
Muakkassa, FF, Rutledge, R, Fakhry, SM, et al (1990) ABGs and arterial lines: the relationship to unnecessarily drawn arterial blood gases. J Trauma 30,1087-1093[Medline]
-
Parry, T, Hirsch, N, Fauvel, N (1995) Comparison of direct blood pressure measurement at the radial and dorsalis pedis arteries during surgery in the horizontal and reverse Trendelenburg positions. Anaesthesia 50,553-555[Medline]
-
Chulay, M, Holland, S (1996) Where should the transducer be leveled for radial or femoral arterial pressure monitoring? Crit Care Nurs 16,103-107
-
Berro, EA (1997) Further issues in transducer placement [letter]. Crit Care Nurs 17,14-16
-
Ahrens, TS, Taylor, LA (1992) Technical considerations in obtaining hemodynamic waveform values. Hemodynamic waveform analysis ,222 W.B. Saunders Philadelphia, PA.
-
Courtois, M, Fattal, PG, Kovacs, SJ, et al (1995) Anatomically and physiologically based reference level for measurement of intracardiac pressures. Circulation 92,1994-2000[Abstract/Free Full Text]
-
Darovic, GO, Vanriper, S, Vanriper, J (1995) Fluid-filled monitoring systems. Darovic, GO eds. Hemodynamic monitoring: invasive and noninvasive clinical application 2nd ed. ,164-170 W.B. Saunders Philadelphia, PA.
-
Emrich, JF, Lumb, PD (1994) Neurologic trauma. Moylan, JA eds. Surgical critical care ,568-579 Mosby St. Louis, MO.
-
Mure, M, Martling, CR, Lindahl, SGE (1997) Dramatic effect on oxygenation in patients with severe acute lung insufficiency treated in the prone position. Crit Care Med 25,1539-1544[CrossRef][ISI][Medline]
-
Ahrens, T, Chulay, M (1997) Hemodynamic monitoring. Chulay, M Guzzetta, C Dossey, B eds. AACN handbook of critical care nursing ,83-118 Appleton and Lange Stamford, CT.
-
Anzueto, A, Peters, JI, Seidner, SR, et al (1997) Effects of continuous bed rotation and prolonged mechanical ventilation on healthy, adult baboons. Crit Care Med 25,1560-1564[CrossRef][ISI][Medline]
TOP
Abstract
Introduction
