(Circulation. 1997;96:3459-3465.)
© 1997 American Heart Association, Inc.
Articles |
From the Section of Cardiology, Department of Internal Medicine, Ochsner Medical Institutions, New Orleans, La.
| Abstract |
|---|
|
|
|---|
Methods and Results Thirty-nine patients with significant coronary artery disease and resting wall motion abnormalities underwent DE (2.5 to 20 µg · kg-1 · min-1) and wall motion analysis (16-segment model). MCE was performed with selective intracoronary injections of sonicated meglumine (2 cm3). Myocardial viability was defined as presence of contrast effect by MCE and contractile reserve or an ischemic response by DE. Functional recovery (improvement in wall motion) was assessed after revascularization (percutaneous transluminal coronary angioplasty, n=20; coronary artery bypass surgery, n=19). When the two groups of patients were analyzed, MCE was associated with excellent sensitivities (84%) yet poor specificities (19% to 26%); DE had lower sensitivities (79% to 80%) but also poor specificities (30% to 36%). The combination of both was associated with excellent sensitivities (90% to 93%) and modest specificities (48% to 50%) for predicting functional recovery. A biphasic response with DE was infrequent (14% to 42%) but highly specific of functional recovery (84% to 94%). MCE had an excellent negative predictive value for functional recovery (83%).
Conclusions The prediction of functional recovery post-revascularization can be enhanced by combining MCE and DE.
Key Words: perfusion contractility myocardium revascularization
| Introduction |
|---|
|
|
|---|
The purpose of this study was to evaluate the relative contribution of MCE and DE (alone or in combination) in the prediction of regional functional recovery in patients with resting wall motion abnormalities caused by chronic CAD. We hypothesized that knowing perfusion (MCE) and contractile reserve (DE) would be superior to knowing each individually in predicting myocardial functional recovery in chronic CAD. To test this hypothesis, we studied 39 patients using MCE and DE.
| Methods |
|---|
|
|
|---|
|
Dobutamine Echocardiography
A baseline two-dimensional echocardiogram was obtained with a
Sonos 1500 Hewlett-Packard ultrasound machine equipped with 2.5- and
3.5-MHz transducers. Standard parasternal long- and short-axis views
and apical four- and two-chamber views were obtained and recorded
on super VHS tape. Continuos ECG and hemodynamic
monitoring was done throughout the infusion of dobutamine
until the heart rate returned to baseline. Dobutamine was
given to 22 patients at doses of 5 and 10 µg ·
kg-1 · min-1
for 5 minutes. These doses are similar to the low doses of
dobutamine that have been found by various investigators to
be safe and effective in detecting contractile reserve in asynergic
regions.2 3 4 5 6 7 Dobutamine was administered to the
remaining 17 patients in doses ranging from 2.5 to 20 µg ·
kg-1 · min-1
because lower or higher doses than 5 to 10 µg ·
kg-1 · min-1
could potentially increase the sensitivity of DE.5 7
Infusion at each dose was continued for 5 minutes.
Visual analysis of wall motion and thickening was performed by an experienced investigator using the American Society of Echocardiography's 16-segment model,8 in a blinded fashion as described elsewhere,9 without any knowledge of patient identity and outcome and independent of the analysis of MCE. The images obtained at baseline and during dobutamine administration were digitized and placed in random order on quad-screen cine-loop formats (Horizon workstation, Freeland-Tomtec). The grading score used was as follows: 0=normal and/or hyperkinesis; 1=mild hypokinesis, 2=severe hypokinesis, 3=akinesis, and 4=dyskinesis. The DE was performed within a few hours of the MCE. All patients underwent a 3-month follow-up echocardiogram after revascularization.
For DE, potential for functional recovery was defined as either a decrease (ie, contractile reserve) or an increase (ie, ischemic response) in regional wall motion score during dobutamine administration. A change from dyskinesis to akinesis was not considered an improvement in function. A biphasic response was defined as an improvement in wall motion score at low-dose dobutamine infusion with subsequent worsening at a higher dose.5 Recovery of function was defined as a decrease in wall motion score of at least one degree at follow-up.
Coronary Angiography and MCE
Coronary angiography was performed in all patients using
the Judkins technique. MCE was performed after angiography with
intracoronary injections of sonicated meglumine diatriazoate (2
cm3). A Heat Systems sonicator model XL-2020 was used to
prepare the ultrasound contrast agent before each injection and as
previously described.10 Attempts were made in every
patient to acquire all views described above. Acoustic power, reject,
and time-gain compensation settings were adjusted at baseline to
optimize image quality and were left unaltered for the remainder of the
study.10 Images were obtained before, during, and after
the injection of contrast into the left main and right coronary
arteries. In all instances, imaging was continued until complete
disappearance of the contrast effect.
Digitized, sequential end-diastolic frames were placed on cine-loop format and used to analyze the presence or absence of echo-contrast effect. Contrast effect (ie, presence of myocardial perfusion) was assessed for each segment of the same 16-segment model used for DE. The analysis of contrast effect (perfusion) was performed by an investigator blinded to the wall motion scores. Analysis of the degree of contrast effect was visually performed using the following scores: 0=no contrast effect, 1=patchy and barely visible contrast effect, 2=moderate contrast effect, and 3=intense contrast effect.
For MCE, presence of viability was defined as the presence of contrast
effect (score of
1). Absence of contrast effect indicated absence of
myocardial viability.
Treatment Groups
The decision to revascularize by either CABG or PTCA was left to
the referring physician.
Statistical Analysis
Sensitivities and specificities in detecting myocardial
viability, and thus potential recovery of function, were calculated for
DE and MCE independently and in combination. Criteria used to define
viability for each technique were as follows.
For DE, true positives were the regions that demonstrated either the presence of contractile reserve, an ischemic response, or a biphasic response during dobutamine administration at any dose and that also demonstrated improvement in function at follow-up. False positives were the regions that met these criteria but demonstrated no improvement at follow-up. True negatives were the regions that demonstrated no change in wall motion score at any dose of dobutamine and that demonstrated no improvement in function at follow-up. False negatives were the regions that met these criteria yet demonstrated an improvement at follow-up.
For MCE, true positives were the regions that demonstrated presence of
contrast effect (score
1) and an improvement in wall motion score at
follow-up (Figs 1
and 2
). If no improvement was observed in
these regions, they were considered false positives. True negatives
were the regions that showed no contrast effect and no improvement in
wall motion score at follow-up. False negatives were regions that
showed no contrast effect yet had an improvement at follow-up.
|
|
When DE and MCE (Figs 1
and 2
) were combined, the criteria described
above had to be simultaneously present. Positive and
negative predictive values were also derived.
Interobserver and Intraobserver Variabilities
Concordance between and within observers to the presence or
absence of perfusion defects in our laboratory is 95% and 100%,
respectively.11 In that study, the size of the perfusion
defect showed interobserver and intraobserver variabilities of
7.2±8.1% and 6.3±7.4%, respectively. Complete concordance for
grading contrast effect in our laboratory is 66%, and concordance
within one contrast grade is 93%, as previously
reported.12
Interobserver and intraobserver variabilities in grading wall motion
abnormalities were assessed in 666 and 734 segments, respectively, as
mean±SD and as the degree of concordance by
statistics. The
mean±SD of observers 1 and 2 was 1.98±1.15 and 2.01±1.07,
respectively. No significant difference by paired t test was
observed. Complete concordance in the degree of wall motion abnormality
was seen in 50% of segments; concordance within one degree of
difference was seen in 86%.
For intraobserver variability, the original mean value of observer 1 was 0.81±1.01; several months later, the same reviewer obtained a value of 0.93±1. No significant difference by paired t test was observed. Intraobserver variability in wall motion abnormality revealed complete concordance in 61% and within one degree of difference in 89%.
| Results |
|---|
|
|
|---|
Dobutamine Echocardiography
Of the 624 potential regions for analysis (maximum, 16 per
patient), 550 (88%) could be adequately visualized and thus were
included in this analysis. Eighty-one regions (14.7%) had
normal contractility. The remaining 469 regions
(85.2%) were found to have abnormal resting function and form the
basis of this report. In all cases, the windows that provided the best
endocardial border definition were used for interpretation. The
sensitivity and specificity of DE in detecting functional recovery are
depicted in Table 2
.
|
Myocardial Contrast Echocardiography
A total of 425 regions were adequate for analysis by MCE.
Of these, 337 had abnormal contractility at baseline
and 274 (81.3%) demonstrated presence of myocardial perfusion, yet
only 147 (53.6%) showed improvement in function at follow-up. There
was no evidence of myocardial perfusion in 63 regions (18.7%). Of
these, 47 (75.8%) demonstrated no improvement in wall motion score at
follow-up. The sensitivities and specificities by MCE in detecting
functional recovery are depicted in Table 2
.
Degree of Perfusion and Improvement at Follow-Up
The contrast score was evaluated in the 337 regions. A score of 0
was observed in 63 regions (18.7%), as was a score of 1 in 61 regions
(18.1%), 2 in 116 regions (34.4%), and 3 in 97 (28.7%) regions. When
all segments were analyzed together (regardless of which group
the patient belonged), 18 of 63 (28.5%) with a score of 0, 32 of 61
(52.4%) with a score of 1, 51 of 116 (43.9%) with a score of 2, and
42 of 97 (43.3%) with a score of 3 showed an improvement in
contractility at follow-up. When this analysis
was performed by group (Table 3
), those
undergoing CABG had the highest improvement in wall motion
(P<.03 from PTCA), particularly in regions where the
perfusion was poor. Interestingly, a significant number of regions with
no perfusion showed improvement in both groups.
|
Combination of DE and MCE
Two hundred ninety-one regions were available for comparative
analysis by DE and MCE. Two hundred thirty-two regions (79%)
were considered viable by MCE, whereas 199 regions (68%,
P<.01) were considered viable by DE criteria. Concordance
between MCE and DE for the presence or absence of viability was 72.9%
and 43%, respectively. At follow-up, 113 of 201 regions meeting
viability criteria by the combination of these techniques improved at
follow-up (positive predictive value, 56.2%), whereas the remaining
134 (43.8%) did not improve. The sensitivities and specificities of
the MCE-DE combination are shown in Table 2
. Only 24 of the 106 regions
showing perfusion yet no viability by DE criteria (22.6%) improved at
follow-up.
On the other hand, of the 66 regions that demonstrated no perfusion by MCE, 13 (19.7%) met viability criteria by DE. Of these 66 regions, only 17 improved at follow-up (25.7%). Thirty-two regions were considered not viable by both techniques, and of these, only 6 improved (18.7%).
A
2 grade improvement in wall motion in perfused segments resulted in
the best relation of sensitivity and specificity (Table 4
). The presence of a biphasic response
in perfused segments resulted in a high specificity and was observed in
10 patients (Table 4
).
|
Treatment Modality and Recovery of Function
A total of 214 of the 469 (45.6%) regions with asynergy improved
at follow-up. Ninety-four (43.8%) regions improved in the PTCA group,
and 120 (56.1%) improved in the CABG group. Regardless of the group
examined, most of the segments showing improvement in function were
akinetic or severely hypokinetic at baseline (Table 5
). Improvement in function in two or
more segments was observed in 15 patients in the PTCA group and in 17
patients in the CABG group. In the PTCA group, the
echocardiographic EF at baseline was 23.9±4.5 and
improved to 28.2±7.8 (P<.04). In the CABG group, the EF
improved from 25.7±9.7 at baseline to 34.4±12.3 (P<.001)
at follow-up.
|
| Discussion |
|---|
|
|
|---|
Present Study
In this study, we examined the ability of DE and MCE to predict
the recovery of regional function in patients with chronic CAD and
resting asynergy. The principal findings of this study follow.
First, DE and MCE exhibited good sensitivities but poor specificities in the two groups of patients studied. The combination of both techniques proved better than either one alone in predicting potential functional recovery.
Second, the absence of regional myocardial perfusion as assessed by MCE identifies regions that are unlikely to improve function at follow-up; therefore, its absence is a strong negative predictor of functional recovery despite revascularization.
Third, perfused segments (as assessed by MCE) are very likely to improve if a biphasic response and/or more than two grades of improvement in wall motion score is noted during DE.
DE in Myocardial Viability
Pierard et al2 studied 17 patients treated with
thrombolysis within 3 hours of an acute MI. Patients
found to have normal perfusion and glucose uptake on positron emission
tomography scan showed an improvement in contractility
with dobutamine and at follow-up. Smart et al4
studied patients 2 days after MI with DE. After
revascularization, low-dose DE correctly predicted
improvement in regional function in 19 of 22 patients. Barrilla et
al3 studied 21 patients after MI. All but 1 patient who
improved with DE before surgery improved at follow-up. Finally, in an
experimental model of acute MI, Sklenar et al16 observed a
close inverse correlation between infarct size and wall thickening
during dobutamine administration.
In our study, we observed a high sensitivity but a poor specificity of DE. The lower specificities observed could potentially be explained by the fact that resting wall motion abnormalities could have been secondary to infarction, stunning, hibernation, and/or combinations of these three mechanisms. The higher specificities observed in the above-mentioned studies could relate to a high incidence of myocardial stunning after MI; this entity is well known to respond to the administration of dobutamine. We observed that a positive response during DE overestimated the potential recovery of function. Similar findings were observed by Afridi et al.5 In their study, segments that exhibited either sustained improvement (85%) or worsening (65%) of function during DE failed to improve at follow-up. On the other hand, a high specificity of DE was observed with the biphasic response in both the Afridi et al5 study and ours. However, this response is seen relatively infrequently in asynergic regions (21% in the study of Afridi et al; 25% in ours) and thus is of limited clinical value. In our study, a two-grade improvement in function resulted in specificities ranging from 82% to 85%; however, these were only modestly sensitive (53% to 62%).
In our study, 24% of the regions showed a negative DE response yet improved at follow-up. Similar results were observed by Cigarroa et al.7 These authors studied 29 patients with resting wall motion abnormalities, coronary stenoses >70%, and EFs <0.45. Regional function improved after revascularization (PTCA in 4 and CABG in 25) in 9 of the 11 patients found to have contractile reserve with DE; however, it also improved in 12 of 14 patients without it.
Panza et al17 in 30 patients with CAD and EF <0.45. A positive response with dobutamine was observed in 72% of hypokinetic segments, 45% of akinetic segments, and 0 of the 7% dyskinetic segments. On the other hand, 84% of all asynergic segments were considered viable by thallium. Their study, like ours, suggests that DE underestimates the number of viable segments. Likewise, Takeuchi et al18 studied 40 patients with DE and thallium-SPECT. Thallium redistribution was seen in up to 50% of patients in whom no improvement in wall motion was observed with DE, suggesting that unchanged wall motion abnormalities during DE do not necessarily represent nonviable myocardium.
LaCanna et al6 reported similar sensitivity but a much higher specificity than our study for predicting reversible wall motion after CABG. Patients with low EFs, wall thickness <5 mm, ventricular aneurysms, and/or revascularization were excluded. Furthermore, the decision to perform CABG was supported by the demonstration of viability by rest-redistribution thallium. Thus, in all likelihood, the exclusion criteria used biased the study toward a positive DE result.
MCE in Viability
Ito et al1 studied 39 patients during anterior MI.
Despite angiographic patency in all patients, they observed a lack of
perfusion in the infarcted territory in 9 patients. At follow-up, only
patients showing reflow after reperfusion showed large improvements in
function. Sabia et al19 observed an improvement in
function after MI only in patients with collateral flow by MCE. In a
related study, Sabia et al20 observed that despite
prolonged coronary occlusions, patients can have marked
improvements in function after PTCA when collateral flow by MCE is
present.
In our study, MCE overestimated the potential recovery of function; only 39% of the regions in the PTCA group and 54% in the CABG group improved at follow-up, despite the presence of perfusion at baseline. Potential explanations for these findings are as follows: First, abnormal function can occur in normally perfused segments adjacent to asynergic segments because of the tethering phenomenon.21 Second, dobutamine could result in improvement in contractility in segments with subendocardial necrosis yet intact subepicardium. However, if enough subendocardium is necrosed, those segments may not result in any improvement in resting function at follow-up.22
In our study, the degree of contrast enhancement did not bear a clear
relation to eventual functional improvement (Table 3
). Nearly twice as
many regions with suboptimal perfusion (grades 1 or 2+) improved with
PTCA and/or bypass surgery, suggesting that a significant number of
patients might have had a substrate of myocardial hibernation. An
explanation for why this improvement occurred more often in CABG
patients with grade 1+ perfusion than in the PTCA patients could relate
to the fact that CABG, through multivessel
revascularization, improved flow to areas supplied
by more than one vessel.
It has been previously observed that not all perfused regions improve in function.1 Furthermore, adequate perfusion by MCE might occur despite the presence of infarction,1 particularly in the presence of islands of viable myocytes.23 Thus, its presence might not necessarily result in improvement of function. Vanoverschelde et al24 noted that in patients with CAD and resting asynergy in regions perfused by underdeveloped or immature collaterals, regional flow can be insufficient in preventing the development of intermittent episodes of ischemia. Thus, recovery from repetitive episodes of such ischemia could be hampered by incomplete reperfusion.
Lack of perfusion, as assessed by MCE in our study, was associated with lack of recovery of function in the majority of such regions. Multiple studies with MCE have shown a similar negative predictive value for recovery of function in the absence of perfusion.1 19 20
Combination MCE and DE
In our study, potential improvement after
revascularization was identified with a sensitivity
of 74% to 75% and a specificity of 67% to 68% when we combined the
use of presence of perfusion (by MCE) with any degree of improvement of
function during DE (Table 4
). These outcomes could relate to the
success of revascularization and to the presence of
islands of infarcted tissue with viable yet asynergic
myocardium.
Using MCE and DE, deFilippi et al25 studied 35 patients with CAD and a mean EF of 0.36. Twenty-three patients underwent revascularization. At follow-up, 81% of the hypokinetic segments that were revascularized improved compared with only 32% of the akinetic segments. In contrast, only 7% of the hypokinetic and 9% of the akinetic and nonrevascularized segments improved. Like us, these authors found that perfusion (by MCE) was present more often than contractile reserve (by DE) in hypokinetic (97% versus 91%) and in akinetic segments (56% versus 33%). The positive predictive value of MCE and DE in predicting functional recovery after revascularization was 76% and 85%, and 55% and 85% for akinetic segments, respectively. The negative predictive value for both techniques (including akinetic segments) was excellent (93% to 97%). As explained by these authors and by Bodenheimer et al23 and Vanoverschelde et al,24 the presence of perfusion in regions that fail to improve after revascularization is not unexpected because fibrosis may preclude contractile reserve, despite preserved perfusion. Contractile reserve requires a critical mass of functional myocytes within a given segment, whereas perfusion can be detected in regions in which functional integrity is precluded by an insufficient number of myocytes.
In patients with acute MI who are undergoing MCE before and after PTCA and DE 3 days from the MI, Bolognese et al26 found a similar sensitivity for both techniques for predicting late functional recovery but a lower specificity, positive predictive value, and accuracy for MCE.
We found that when perfusion was present by MCE and no response to dobutamine administration was observed, only one fifth of the regions improved at follow-up. This finding is of clinical significance because the addition of DE to perfusion data was helpful in identifying regions that have minimal potential for recovery. On the other hand, regions of asynergic myocardium with evidence of myocardial perfusion (MCE) and a positive dual response (DE) or more than a two-grade improvement in wall motion during dobutamine administration are very likely to show an improvement in function at follow-up.
As expected, the vast majority of nonperfused regions lacking any response to dobutamine did not improve at follow-up. Furthermore, in the absence of perfusion, and as assessed by MCE, only 25.7% of regions with a positive response to dobutamine showed improvement at follow-up. Therefore, in our study, the addition of DE was of limited value once the absence of myocardial perfusion was demonstrated by MCE.
Study Limitations
In the present study, dobutamine was infused at 5
and 10 µg · kg-1 ·
min-1, doses found useful in identifying
viable myocardium2 3 and the extent of
infarction size.16 Doses of 2.5 to 20 µg ·
kg-1 · min-1
were administered to only 17 patients, and although these extra doses
did not have an effect on the sensitivity and specificity of DE,
because of the limited numbers, the analysis of the biphasic
response is limited. Two significant limitations of our study relate to
the small number of patients studied and to the fact that no patient
was restudied angiographically after PTCA or CABG; thus, had the
revascularization failed, it would have resulted in
false-positive DE or MCE results. Future studies will need to include
angiographic follow-up to determine the true sensitivity and
specificity MCE and DE.
Clinical Implications
The prediction of recovery of function after
revascularization can represent a difficult
problem because viable and nonviable tissues can coexist. By combining
DE and MCE, the overestimation and underestimation of myocardial
viability can be reduced. DE and MCE at the time of
catheterization can offer a timely alternative to other
more expensive methods.
Conclusions
The combination of perfusion data with contractile reserve data
enhances the prediction of functional recovery after
revascularization.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
| Footnotes |
|---|
Received April 7, 1997; revision received June 20, 1997; accepted June 26, 1997.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. G. Camici, S. K. Prasad, and O. E. Rimoldi Stunning, Hibernation, and Assessment of Myocardial Viability Circulation, January 1, 2008; 117(1): 103 - 114. [Full Text] [PDF] |
||||
![]() |
E. Toledo, R. M. Lang, K. A. Collins, G. Lammertin, U. Williams, L. Weinert, G. Bolotin, P. D. Coon, J. Raman, L. D. Jacobs, et al. Imaging and Quantification of Myocardial Perfusion Using Real-Time Three-Dimensional Echocardiography J. Am. Coll. Cardiol., January 3, 2006; 47(1): 146 - 154. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. N. Mazzadi, M. F. Janier, B. Brossier, X. Andre-Fouet, E. McFadden, D. Revel, and P. Croisille Dobutamine-tagged MRI for inotropic reserve assessment in severe CAD: relationship with PET findings Am J Physiol Heart Circ Physiol, May 1, 2004; 286(5): H1946 - H1953. [Abstract] [Full Text] [PDF] |
||||
![]() |
The Task Force on the Management of Acute Myocardi, F. Van de Werf, D. Ardissino, A. Betriu, D. V. Cokkinos, E. Falk, K. A.A. Fox, D. Julian, M. Lengyel, F.-J. Neumann, et al. Management of acute myocardial infarction in patients presenting with ST-segment elevation Eur. Heart J., January 1, 2003; 24(1): 28 - 66. [Full Text] [PDF] |
||||
![]() |
B. B. Chin, G. Esposito, and D. L. Kraitchman Myocardial Contractile Reserve and Perfusion Defect Severity with Rest and Stress Dobutamine 99mTc-Sestamibi SPECT in Canine Stunning and Subendocardial Infarction J. Nucl. Med., April 1, 2002; 43(4): 540 - 550. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Distante, R. Dankowski, P. Mincarone, C.G. Leo, E. Gianicolo, P. Voci, M.A. Morales, and D. Rovai Contrast echocardiography and medical economics: looking into the crystal ball Eur. Heart J. Suppl., March 1, 2002; 4(suppl_C): C39 - C47. [Abstract] [PDF] |
||||
![]() |
V. Mor-Avi, E. G. Caiani, K. A. Collins, C. E. Korcarz, J. E. Bednarz, and R. M. Lang Combined Assessment of Myocardial Perfusion and Regional Left Ventricular Function by Analysis of Contrast-Enhanced Power Modulation Images Circulation, July 17, 2001; 104(3): 352 - 357. [Abstract] [Full Text] [PDF] |
||||
![]() |
G LaCanna, S.H Rahimtoola, O Visioli, R Giubbini, O Alfieri, M Zognio, E Milan, C Ceconi, M Gargano, R LoRusso, et al. Sensitivity, specificity, and predictive accuracies of non-invasive tests, singly and in combination, for diagnosis of hibernating myocardium Eur. Heart J., August 2, 2000; 21(16): 1358 - 1367. [Abstract] [PDF] |
||||
![]() |
J. J. Bax, D. Poldermans, A. Elhendy, J. H. Cornel, E. Boersma, R. Rambaldi, J. R. T. C. Roelandt, and P. M. Fioretti Improvement of left ventricular ejection fraction, heart failure symptoms and prognosis after revascularization in patients with chronic coronary artery disease and viable myocardium detected by dobutamine stress echocardiography J. Am. Coll. Cardiol., July 1, 1999; 34(1): 163 - 169. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1997 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |