|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Divisions of Cardiovascular Diseases and Internal Medicine (Drs. Tsang, Hayes, Freeman, and Seward, and Mss. Barnes and Osborn Butler) and Thoracic and Cardiovascular Surgery (Dr. Dearani), Mayo Clinic and Mayo Foundation, Rochester, MN.
Correspondence to: Teresa S. M. Tsang, MD, Mayo Clinic, 200 First St, SW, Rochester, MN 55905
| Abstract |
|---|
|
|
|---|
Design: From the prospective Mayo Clinic Registry of Echo-guided Pericardiocentesis (February 1979 to June 1998), 245 procedures performed for clinically significant postoperative effusions were identified. Clinical features, effusion causes, echocardiographic findings, and management outcomes were studied and analyzed. Cross-referencing the registry with the Mayo Clinic surgical database provided an estimate of the incidence of significant postoperative effusions and the number of cases in which primary surgical management was chosen instead of pericardiocentesis.
Results: Use of anticoagulant
therapy was considered a significant contributing factor in 86% and
65% of early effusions (
7 days after surgery) and late effusions
(> 7 days after surgery), respectively. Postpericardiotomy syndrome
was an important factor in the development of late effusions (34%).
Common presenting symptoms included malaise (90%), dyspnea (65%), and
chest pain (33%). Tachycardia, fever, elevated jugular venous
pressure, hypotension, and pulsus paradoxus were found in 53%, 40%,
39%, 27%, and 17% of cases, respectively. Transthoracic
echocardiography permitted rapid diagnosis and hemodynamic assessment
of all effusions except for three cases that required transesophageal
echocardiography for confirmation. Echo-guided pericardiocentesis was
successful in 97% of all cases and in 96% of all loculated effusions.
Major complications (2%), including chamber lacerations (n = 2) and
pneumothoraces (n = 3), were successfully treated by surgical repair
and chest tube reexpansion, respectively. Median follow-up duration for
the study population was 3.8 years (range, 190 days to 16.4 years). The
use of extended catheter drainage was associated with reduction in
recurrence for early and late postoperative effusions by 46% and 50%,
respectively.
Conclusions: The symptoms and physical findings of clinically significant postoperative pericardial effusions are frequently nonspecific and may be inadequate for a decision regarding intervention. Echocardiography can quickly confirm the presence of an effusion, and pericardiocentesis under echocardiographic guidance is safe and effective. The use of a pericardial catheter for extended drainage is associated with lower recurrence rates, and the majority of patients so treated do not require further intervention.
Key Words: cardiac tamponade echo-guided pericardiocentesis postoperative pericardial effusions two-dimensional echocardiography
| Introduction |
|---|
|
|
|---|
We reviewed a 19-year experience with clinically significant pericardial effusions after chest or open heart surgery and evaluated the clinical presentation, echocardiographic features, and short-term and longer-term patient outcomes associated with echocardiographically (echo-) guided pericardiocentesis.
| Materials and Methods |
|---|
|
|
|---|
Detailed records of pericardiocentesis procedures from the Echo-guided Pericardiocentesis Registry were reviewed to determine the clinical characteristics, timing of presentation, and echocardiographic diagnostic features of postoperative pericardial effusions. In addition, all available Mayo Clinic charts and external records on these patients were studied. Data collected included demographics, type of operation performed, timing of effusion diagnosis relative to surgery, clinical presentation, use of anticoagulants, echocardiographic findings, characteristics of pericardial effusion, success and complications associated with the pericardiocentesis procedure, and patient outcomes (need for other interventions, recurrence of effusion, and survival). Telephone interviews with patients, their families, or their primary physicians were conducted when necessary. All subjects were followed up for at least 6 months, except for two patients whose postoperative effusions developed in 1998. Therefore, follow-up data were complete for 206 patients (99%).
Procedures and Techniques
The open heart procedures in this study included isolated
coronary artery bypass graft (CABG) surgery or valvular surgery,
combined CABG and valvular surgery, heart transplantation, operations
for congenital heart disease, and ascending aortic surgery. There were
some noncardiothoracic procedures, and there was one abdominal aortic
surgery. All surgical procedures were performed in accordance with the
standard techniques of the period.
All patients in the study initially underwent echo-guided pericardiocentesis as per Mayo Clinic technique.4 5 Standard two-dimensional (2D) images were obtained using commercially available equipment. Doppler echocardiography, in conjunction with the use of a respirometer, for assessment of hemodynamic significance of an effusion was initiated in 1987. The location and distribution of the pericardial effusion were confirmed by 2D echocardiography, and an ideal entry site was defined.
On entry into the pericardial fluid with the polytef-sheathed needle (Deseret; Becton Dickenson; Franklin Lakes, NJ), the steel needle core was withdrawn, and only the sheath was advanced. At the discretion of the echocardiologist, and invariably when bloody pericardial fluid was removed, agitated saline contrast material was injected to confirm the position of the sheath. When extended pericardial catheter drainage was to be used, a standard dilator and an introducer sheath (5 to 8F) were advanced over the wire. The guidewire would be withdrawn, and a pigtail angiographic catheter (60 to 65 cm) would be inserted through the sheath. The effusion was first evacuated as completely as possible, and subsequent intermittent aspirations were performed every 4 to 6 h until the drainage had decreased to < 25 to 30 mL during 24 h. A few patients required transesophageal echocardiography (TEE) for the diagnosis of pericardial effusion because of inadequate transthoracic images.
Definitions
Early and late postoperative pericardial effusions were defined
as those occurring within 7 days after surgery and > 7 days after
surgery, respectively. Postpericardiotomy syndrome was considered a
likely cause if the development of postoperative pericardial effusion
occurred in the context of fever (> 38°C) and pericarditic chest
pain, and in the presence of a rub. Hypotension referred to systolic BP
< 90 mm Hg, and tachycardia, heart rate > 100 beats/min. The JVP
was considered elevated if the height of the jugular venous pulse was
> 8 cm above the right atrium by physical examination.
2D-echocardiographic features of tamponade included right atrial compression during late diastole,6 right ventricular collapse during early diastole,7 abnormalities of mitral valve motion,8 dilated inferior vena cava with lack of inspiratory collapse,9 and swinging heart.10
Doppler hemodynamic findings of tamponade included decreased left ventricular filling with inspiration leading to delay of mitral valve opening, lengthened isovolumic relaxation time, and decreased mitral E velocity.11 Opposite changes occurred on expiration, and reciprocal changes occurred on the right side. Inspiratory decrease and expiratory increase in pulmonary venous diastolic forward flow as well as expiratory increase in hepatic venous diastolic flow reversal were characteristic findings.11 12 13 Findings of predominantly systolic forward flow in the superior vena cava, with a decrease or loss of diastolic component and increased flow reversals on expiration, were accepted as evidence of tamponade.13 Hemodynamic significance of an effusion during TEE was determined by transmitral and pulmonary venous flow patterns with simultaneous respirometry.14 15
Pericardial effusions were categorized as large in adults (> 16 years old) if drainage was > 400 mL. In children, the size of an effusion was based on visual qualitative evaluation of echocardiographic findings. The initial pericardiocentesis was considered successful if pericardial fluid was drained without incident, achieving the therapeutic indication for the procedure. Recurrence was defined as any reaccumulation of fluid requiring intervention. Early or late recurrence referred to whether reaccumulation occurred within 30 days of pericardiocentesis or thereafter. Major complications of the pericardiocentesis procedure included any adverse events that required intervention. Minor complications did not require any treatment except for appropriate monitoring and follow-up.
A low platelet count was defined as < 80 x 109/L. Patients were considered to have achieved therapeutic anticoagulation when their partial thromboplastin time was 50 to 75 s, prothrombin time was 1.7 to 2.4 s, or prothrombin time international ratio was 2.0 to 3.5. Excessive anticoagulation referred to the presence of values exceeding the upper limits of these ranges.
Statistical Analysis
Data were reported as absolute numbers and frequency
percentages, mean ± SD, and median with range. Continuous and
categorical variables were compared using the Student's t
test and
2, respectively. All p values were
two-tailed, with p < 0.05 considered significant.
| Results |
|---|
|
|
|---|
|
Echocardiographically, large effusions (> 400 mL) without classic criteria for tamponade accounted for 29% of the pericardiocenteses performed postoperatively. Pericardial tamponade was confirmed in 171 cases (70%) by 2D or Doppler echocardiography, or both (Table 2 ). TEE was necessary for identifying the effusion and confirming tamponade in three cases (1%).
|
Postpericardiotomy syndrome was the attributing cause for 75 postoperative effusions (31%) but was an important factor in the development of late effusion, accounting for 34% of the late postoperative effusions. It was a less common cause for early effusion.
Estimated Incidence of Tamponade After Open Heart Surgery
From the total number of open heart procedures performed between
February 1979 and June 1998 and the number of pericardiocenteses
performed during the same period, the estimated incidence of
postoperative clinically significant effusions requiring
pericardiocentesis was 0.8% (Table 3
). Pericardiocentesis for clinically significant effusion was most often
performed after heart transplant (8.4%) and least often with isolated
CABG surgery (0.2%; Table 3
). When the Mayo Surgical Index was
cross-referenced with the Echocardiography Database and Echo-guided
Pericardiocentesis Registry, 48 cases were identified during this
period in which emergency surgical drainage was performed for suspected
pericardial tamponade or intrapericardial bleeding that developed after
open heart procedures. In these cases, patients did not undergo
echocardiography or echo-guided pericardiocentesis before emergency
surgical decompression. Data on these patients and outcomes have not
been included in this paper. However, when these cases were taken into
consideration, the estimated incidence of significant postoperative
effusion requiring intervention increased to approximately 1%.
|
The rate of major complications was 2%. These included two chamber lacerations requiring surgical repair and three pneumothoraces managed with chest tube reexpansion. Minor complications (1.2%) included two small pneumothoraces and one pleural-pericardial fistula. Pericardiocentesis was unsuccessful in fulfilling the therapeutic indications in eight cases (3%) that were surgically managed. When examined in relation to the distribution of the effusion, five of the eight failed procedures were attempts to drain circumferential effusions, and three were for posterior loculated effusions. Thus, on the basis of location alone, failure rate was 3% for the 174 circumferential effusions and 8% for the 37 posterior loculated effusions (p < 0.001; Table 2 ). Although the three loculated effusions that were not amenable to treatment by pericardiocentesis were posterior in location, 34 other posterior effusions were drained successfully by echo-guided pericardiocentesis, giving a 92% success rate (Table 2) . Considering all loculated effusions (anterior, posterior, and others) in general, 68 of 71 (96%) were successfully decompressed by echo-guided pericardiocentesis. Thus, the success rate was comparable to that for circumferential effusions (97%).
Use of Concomitant Medications
Prednisone was used concomitantly in 12 (5%) and nonsteroidal
anti-inflammatory drugs or aspirin in 189 (77%) of the cases. The use
of prednisone had decreased substantially through the 1980s, and it is
no longer used for treatment of postoperative effusion. Nonsteroidal
agents were commonly prescribed for postpericardiotomy syndrome, and
they continue to be used frequently as adjunctive therapy in the
treatment of postoperative effusions. The choice of nonsteroidal agent
was based on individual physician preference. Aspirin was prescribed
mainly as part of the regimen for treatment of coronary artery disease.
Recurrence
Recurrence of an effusion was more frequent when extended catheter
drainage was not used with the initial pericardiocentesis procedure. Of
the 208 patients who underwent echo-guided pericardiocentesis as
initial management strategy, 102 patients (49%) had extended catheter
drainage incorporated as primary treatment and 106 patients (51%) did
not. Eleven patients (11%) and 22 patients (21%) from the two groups,
respectively, required repeat pericardiocenteses because of recurrence.
The mean duration of extended catheter drainage used with initial
pericardiocentesis was 3.2 days (median, 3 days; range, 1 to 18 days)
and was associated with a 48% decrease in recurrence (p < 0.001).
The mean volume of catheter drainage subsequent to initial aspiration
was 360 mL (range, 0 to 2,627 mL). The primary use of extended catheter
drainage in early and late postoperative effusions provided reductions
in recurrence of 44% and 50%,
respectively
(Table 4
).
|
For the remaining 18 patients, surgery was undertaken because pericardiocentesis was unsuccessful or failed to achieve a lasting desirable outcome, with persistence of drainage or subsequent recurrence of effusion. Nine of these patients had intrapericardial hematoma, all cases developed within 15 days of the cardiothoracic surgery, and surgical evacuation of hematoma was necessary after pericardiocentesis.
Follow-up and Survival
Except for two patients who underwent pericardiocentesis in 1998,
all patients had at least 6 months of follow-up. Median follow-up time
was 3.8 years (range, 190 days to 16.4 years). Pericardial constriction
occurred in two patients (1%) at an average of 2 years after
operation.
Seven patients died within 30 days after operation. The cause of these deaths was attributed to postoperative tamponade that was managed unsuccessfully by surgical decompression after failed pericardiocentesis in two patients (both with intrapericardial hematoma), perioperative complications in two patients, and the underlying disease for which the surgery was performed in three patients. Four additional patients survived the first 30 days but died within 6 months of primary surgery. These deaths were caused by underlying cardiac disease in one patient, perioperative complications in two patients, and unrelated cause in one patient.
Pattern of Practice in the Management of Postoperative Effusion
During the period from 1980 to 1997, the number of surgical
procedures performed for treatment of pericardial effusions decreased
steadily (Fig 1
). During the same time, use of a pericardial catheter with echo-guided
pericardiocentesis for extended drainage increased progressively (Fig 2
). There was evidence for decreased risk of recurrence of effusion and
use of surgery for management with increased use of a pericardial
catheter.5
16
|
|
| Discussion |
|---|
|
|
|---|
Proposed predisposing factors for development of postoperative effusions included valve surgery, use of anticoagulants,18 19 20 21 coagulation disorders, excessive mediastinal drainage,17 postpericardiotomy syndrome,22 and autoimmune reactions.23 In this study, the use of anticoagulant therapy was thought to be an important factor associated with development of a significant effusion in 68% of the cases and postpericardiotomy syndrome in 31% of the cases.
The incidence of early, hemodynamically significant effusion was low (approximately 0.1%) in our series, as in other studies.24 25 Most commonly, it occurred as a result of postoperative bleeding. Anticoagulant therapy was thought to be a contributing factor in 75% of these early effusions. The incidence of late, hemodynamically significant effusion in this study was approximately 0.8%. This falls within the range of 0.1 to 6% described in the literature.26 Anticoagulant use appeared to be a contributing factor in the majority of these late cases (65%). Postpericardiotomy syndrome was the attributing cause for a large number of the late effusions, accounting for 34% of effusions that presented beyond the first week postoperatively.
Clinical Features of Postoperative Effusions
The presenting clinical features differed according to time after
operation (early vs late). Early effusions were most frequently related
to postoperative bleeding, and 50% of these patients were hypotensive
and 58% were tachycardic. General malaise, dyspnea, and chest pain
were common symptoms (Table 1)
.
With late effusions, general malaise was most common (95%), followed by dyspnea (69%), fever (42%), and chest pain (34%). Hypotension was relatively uncommon (22%), although tachycardia (52%) and increased JVP (40%) were equally frequent findings as in patients with early effusions.
Diagnosis of hemodynamically significant effusion by physical findings alone is often a challenge. In one series,25 hypotension, pulsus paradoxus, and increased JVP were absent in 30%, 40%, and 50%, respectively, of the patients with echocardiographic evidence of tamponade. In another series,27 only 14% of the patients presenting with late posterior tamponade had pulsus paradoxus and increased JVP. In this current series, pulsus paradoxus was documented in only 17%, hypotension in 27%, and increased JVP in 39% of the cases. Therefore, clinical findings alone did not provide an adequate basis for decision regarding intervention.
Diagnosis of Hemodynamically Significant Pericardial Effusions by
Echocardiography
2D and Doppler echocardiography readily confirmed the presence of
an effusion and provided accurate assessment of its hemodynamic
significance. In only three cases in this series was TEE necessary for
confirming the diagnosis. The utility of TEE in postoperative effusions
was described previously, particularly in the context of loculated
effusion or intrapericardial hematoma, when transthoracic
echocardiography was not adequate.1
14
28
29
30
Several
factors in the postoperative patient may contribute to the need for
TEE: the surgical site may preclude use of the optimal transthoracic
window, chest tubes may prevent proper positioning of the patient, and
some loculated effusions may not be amenable to transthoracic imaging.
As in any transthoracic imaging, characteristics of the patient,
including certain body habitus and lung conditions, may contribute to
technical difficulties.
Echo-Guided Pericardiocentesis, Catheter Drainage, and Outcomes
Echo-guided pericardiocentesis as the initial management strategy
was successful in 97% of the cases in this series, and it was the only
form of therapy necessary for 171 patients (82%). These results are
similar to those found in a smaller series, in which echo-guided
pericardiocentesis was the only treatment in 79% of the patients who
experienced postoperative effusions.27
Of the eight cases
in the current study that could not be drained successfully by
echo-guided pericardiocentesis, three were loculated in a posterior
location.
Posterior loculated effusions were not uncommon in postoperative patients and posed greater difficulty in terms of diagnosis and management. In one series of 42 cases, 6 (14.3%) were loculated effusions, all in a posterior location.31 In another series of 11 cases, 7 (63.6%) were loculated, and again all in a posterior location.32 Although loculated effusions appeared to be least amenable to echo-guided pericardiocentesis, the success rate was nevertheless 92% in this study. The usefulness of this drainage technique for posterior effusion was also demonstrated in a smaller study in which 10 of 14 cases of tamponade were successfully relieved.27 Thus, echo-guided pericardiocentesis should be considered as primary management strategy for posterior effusions, as for effusions elsewhere.
Safety of echo-guided pericardiocentesis was confirmed by earlier series.5 27 33 34 Compared with blind pericardiocentesis, routinely performed via a subxiphoid approach, with morbidity and mortality rates of 50% and 6%, respectively,35 36 37 echo-guided pericardiocentesis was associated with much lower complication rates.16 33 34 38 With the optimal entry site determined by echocardiography,5 the most commonly chosen point of entry was located on the chest wall (86% in this series). Subcostal entries accounted for only 12% in this study. No death occurred as a result of the pericardiocentesis, and the major complication rate was low at 2%. No long-term sequelae were associated with any of the complications encountered.
The use of a pericardial catheter in conjunction with echo-guided pericardiocentesis has been shown to decrease risk of recurrence of an effusion.16 34 39 In this study, such adaptation to the primary treatment was associated with a reduction in recurrence rate of approximately 50%.
Mortality Associated With Pericardial Effusions After Open Heart
Surgery
Late tamponade has been considered a more dangerous complication
of open heart surgery, associated with higher mortality than early
tamponade.20
26
40
In the current series, the 30-day
mortality in postoperative patients with significant pericardial
effusion was 3%. No patient died as a result of early tamponade. Two
deaths were attributed to late tamponade that was treated
unsuccessfully by pericardiocentesis and attempted surgical
decompression. The underlying conditions and other perioperative
complications were considered to be more directly responsible for the
deaths of five other patients.
Limitations of Study
Although the prospective registry encompassed detailed records of
echo-guided pericardiocentesis procedures, data requiring retrospective
review and verification were subject to bias. As there was no
comparison group for this patient series, the clinical usefulness of
symptoms and signs for identifying postoperative patients who
experienced hemodynamically significant pericardial effusion cannot be
addressed. Because catheter drainage was used with increasing frequency
during the course of the study period, a greater proportion of patients
who had extended catheter drainage underwent pericardiocentesis in the
later years. Thus, a positive impact on outcomes of these patients
caused by concomitant advances in medicine cannot be excluded. The
strength of this study originated from the large number of consecutive
pericardiocentesis procedures performed during almost 20 years and
complete follow-up data in 99% of the patients.
| Conclusion |
|---|
|
|
|---|
| Footnotes |
|---|
Abbreviations: CABG = coronary artery bypass graft; 2D = two-dimensional; JVP = jugular venous pressure; TEE = transesophageal echocardiography
Received for publication October 14, 1998. Accepted for publication March 15, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. M. Tuladhar, M. Noursadeghi, J. J. Boyle, J. S. Friedland, and P. Hornick Tuberculous pericardial effusion after coronary artery bypass graft. Ann. Thorac. Surg., October 1, 2006; 82(4): 1519 - 1521. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Meurin, H. Weber, N. Renaud, F. Larrazet, J. Y. Tabet, P. Demolis, and A. Ben Driss Evolution of the Postoperative Pericardial Effusion After Day 15: The Problem of the Late Tamponade Chest, June 1, 2004; 125(6): 2182 - 2187. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Geffroy, H. Beloeil, E. Bouvier, A. Chaumeil, P. Albaladejo, and J. Marty Prolonged right ventricular failure after relief of cardiac tamponade: [Insuffisance ventriculaire droite prolongee apres evacuation d'une tamponnade pericardique] Can J Anesth, May 1, 2004; 51(5): 482 - 485. [Abstract] [Full Text] [PDF] |
||||
![]() |
H Luckraz, S Kitchlu, and A Youhana Haemorrhagic peritonitis as a late complication of echocardiography guided pericardiocentesis Heart, March 1, 2004; 90(3): e16 - 16. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. D. Cheitlin, W. F. Armstrong, G. P. Aurigemma, G. A. Beller, F. Z. Bierman, J. L. Davis, P. S. Douglas, D. P. Faxon, L. D. Gillam, T. R. Kimball, et al. ACC/AHA/ASE 2003 guideline update for the clinical application of echocardiography: summary article: a report of the American college of cardiology/American heart association task force on practice guidelines (ACC/AHA/ASE committee to update the 1997 guidelines for the clinical application of echocardiography) J. Am. Coll. Cardiol., September 3, 2003; 42(5): 954 - 970. [Full Text] [PDF] |
||||
![]() |
M. D. Cheitlin, W. F. Armstrong, G. P. Aurigemma, G. A. Beller, F. Z. Bierman, J. L. Davis, P. S. Douglas, D. P. Faxon, L. D. Gillam, T. R. Kimball, et al. ACC/AHA/ASE 2003 Guideline Update for the Clinical Application of Echocardiography: Summary Article: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASE Committee to Update the 1997 Guidelines for the Clinical Application of Echocardiography) Circulation, September 2, 2003; 108(9): 1146 - 1162. [Full Text] [PDF] |
||||
![]() |
E W Y Cheung, S A Ho, K K Y Tang, A K T Chau, C S W Chiu, and Y F Cheung Pericardial effusion after open heart surgery for congenital heart disease Heart, July 1, 2003; 89(7): 780 - 783. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. M. Seferovic, A. D. Ristic, R. Maksimovic, V. Tatic, M. Ostojic, and V. Kanjuh Diagnostic Value of Pericardial Biopsy: Improvement With Extensive Sampling Enabled by Pericardioscopy Circulation, February 25, 2003; 107(7): 978 - 983. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. Vijay and J. P. Gold Late Complications of Cardiac Surgery Card. Surg. Adult, January 1, 2003; 2(2003): 521 - 537. [Full Text] |
||||
![]() |
T. O. Cheng What To Do When Bloody Fluid Is Obtained on Pericardiocentesis? Chest, May 1, 2000; 117(5): 1525 - 1526. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |