Anesth Analg 2001;92:1117-1122
© 2001 International Anesthesia Research Society
CARDIOVASCULAR ANESTHESIA
Do Pulmonary Artery Catheters Cause or Increase Tricuspid or Pulmonic Valvular Regurgitation?
Steven V. Sherman, MD,
Michael H. Wall, MD,
Daniel J. Kennedy, MD,
Robert F. Brooker, MD*, and
John Butterworth, MD
Department of Anesthesiology, Wake Forest University School of Medicine, Winston-Salem, North Carolina; and *Cardiovascular Anesthesia Consultants, Las Vegas, Nevada
Address correspondence and reprint requests to Dr. Wall, Department of Anesthesiology, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157-1009. Address e-mail to mhwall{at}wfubmc.edu
 |
Abstract
|
|---|
There are few quantitative data on the extent or mechanism of pulmonary artery catheter (PAC)-induced valvular dysfunction. We hypothesized that PACs cause or worsen tricuspid and pulmonic valvular regurgitation, and tested this hypothesis by using transesophageal echocardiography. In 54 anesthetized adult patients, we measured color Doppler jet areas of tricuspid regurgitation (TR) in two planes (midesophageal [ME] 4-chamber and right ventricular inflow-outflow views) and pulmonic insufficiency (PI) in one plane (ME aortic valve long-axis view), both before and after we advanced a PAC into the pulmonary artery. Regurgitant jet areas and hemodynamic measurements were compared by using paired t-test. There were no significant changes in blood pressure or heart rate after passage of the PAC. After PAC placement, the mean PI jet area was not significantly increased. The mean TR jet area increased significantly in the right ventricular inflow-outflow view (+0.37 ± 0.11 cm2) (P = 0.0014), but did not increase at the ME 4-chamber view. Seventeen percent of patients had an increase in TR jet area 
1 cm2; 8% of patients had an increase in PI jet area 1 cm2.
Implications: In patients without pulmonic or tricuspid valvular pathology, placement of a pulmonary artery catheter (PAC) worsened tricuspid regurgitation, which is consistently visualized in the right ventricular inflow-outflow view, and often not seen in the midesophageal 4-chamber view. This is consistent with malcoaptation of the anterior and posterior leaflets. PAC-induced pulmonic insufficiency was rarely detected in the midesophageal aortic valve long-axis view. We conclude that a PAC is very unlikely to be the sole cause of severe tricuspid regurgitation or pulmonic insufficiency.
|
Introduction
|
|---|
After its introduction in the early 1970s (1), the pulmonary artery catheter (PAC) became widely used in cardiac surgery (2). Nevertheless, some complications from PACs are common and can be life threatening. These complications include arrhythmias, pulmonary artery (PA) rupture or pseudoaneurysm, intracardiac injury, and possible increased tricuspid and pulmonic valvular regurgitation (3). Cardiac valvular function alters measurements using PACs (4); however, few data exist on the effect of a PAC on cardiac valvular function. Previous studies evaluating right heart catheter-induced regurgitation were limited by their inability to measure regurgitation without cardiac catheterization (5–7), or because they predated the widespread availability of color Doppler imaging (8,9). No study has proposed a specific mechanism of valvular dysfunction.
Our hypothesis was that PACs cause or increase tricuspid regurgitation (TR) and cause or increase pulmonic insufficiency (PI). To test our hypothesis, we used transesophageal echocardiography (TEE) with a multiplane transducer to measure the area of the color Doppler tricuspid regurgitant jet and PI jet before and after PAC placement.
|
Methods
|
|---|
After institutional approval and written informed consent, 65 adult patients were enrolled. All patients were undergoing elective cardiovascular surgery in which the use of both a PAC and TEE was planned. Patients known preoperatively to have tricuspid or pulmonic valvular pathology were excluded.
After induction of anesthesia, a PAC (7.5F Swan-Ganz catheter 931HF-75; Baxter Healthcare Corp., Irvine, CA) was placed into the superior vena cava via an introducer (8.5 or 9F Cordis Introducer; Baxter Healthcare). The right internal jugular vein was used in 53 patients, and the left internal jugular vein in one patient. A multiplane TEE transducer (Hewlett-Packard model 21364A; Hewlett-Packard Corp., Andover, MA, or Toshiba model PEK-510MB; Toshiba Corp., Tochigi-ken, Japan) was then advanced into the esophagus. The probes were used at 5 MHz with Hewlett-Packard Sonos 1500 or 2000 systems (models 77035A or M2406A; Hewlett-Packard) or a Toshiba Power Vision System (model SSA-380A; Toshiba). Two-dimensional gain was optimized and color gain set just below the level producing artifactual noise. Doppler sample size was minimized to increase frame rate.
Blood pressure (BP) and heart rate (HR) were recorded at the start of the first TEE measurement, and monitored throughout the study. Measurements were made only when BP and HR were within 10% of the baseline value. TR and PI jets were measured on-line by a nonblinded echocardiographer (SVS performed 48 of the 54 studies). The area of the color Doppler TR jet was measured in two planes (midesophageal [ME] 4-chamber view and ME right ventricular [RV] inflow-outflow view). The PI jet was measured in the ME long-axis plane at the level of the longitudinal view of the aortic valve. These imaging planes are shown in Figure 1 and described in the American Society of Echocardiography/Society of Cardiovascular Anesthesiologists guidelines for performing a comprehensive multiplane intraoperative TEE examination (10). The largest-appearing jet during a period of at least 10 cardiac cycles and 2 respiratory cycles was captured in the ultrasonograph’s memory loop and only the turbulent portion of the regurgitant jet was traced and measured. Cardiac cycles after premature ventricular contractions were excluded. TR jets were measured only in midsystole.
View larger version (67K):
[in this window]
[in a new window]
|
Figure 1. A, Color Doppler image of tricuspid regurgitation, midesophageal 4-chamber view (0°). B, Color Doppler image of tricuspid regurgitation, right ventricular inflow-outflow view (43°). C, Color Doppler image of pulmonic insufficiency, midesophageal long-axis view (109°).
|
|
The PAC was then advanced into the PA, and the PA pressure, central venous pressure, and cardiac output were recorded. The regurgitant jets were mea-sured again in the same echocardiographic planes. The machine settings were not changed between TEE examinations. The entire examination was recorded on S-VHS videocassettes. All measurements were made before sternotomy.
Interrater variability of the Doppler regurgitant jet areas was tested by two echocardiographers (DJK and MHW), blinded to the patient, in 20 of the 54 patients (10 patients each) using intraclass correlations (r2) (11,12). It was not possible to blind the echocardiographers to the presence or absence of a PAC because it was easily seen in most views.
Two echocardiographers (SVS, MHW) who were blinded to the patient and each other, also reviewed off-line the number and direction of TR jet after the passage of the PAC at the ME RV inflow-outflow view. The jets were described as centrally, anteriorly, or posteriorly directed and assigned a clinical grade (0, 1+, 2+, 3+, and 4+) to the severity of the TR at the RV inflow-outflow view before and after passage of the PAC. The TR grade was based on the apparent size ratio between the jet area and the right atrial area (13). Any disagreements in these measurements were resolved by consensus using a third blinded echocardiographer (DJK).
All three echocardiographers (SVS, DJK, MHW) who acquired or evaluated on-line or off-line images are certified in Perioperative TEE by the National Board of Echocardiography.
The mean regurgitant jet areas before and after PA catheterization were compared by paired t-test. BP and HR were compared before and after passage of PAC by using paired t-test. The effects of covariates on valvular regurgitation were tested by using Spearman’s rank test. Intraclass correlation was calculated from variance components by using the SAS Subroutine Proc Varicomp (SAS Institute, Cary, NC). All statistical calculations were accomplished by using SAS (SAS version 8). P < 0.05 was considered significant.
|
Results
|
|---|
Eleven patients gave consent but were not included in the study for the following reasons: two patients demonstrated esophageal abnormalities; the attending anesthesiologist decided that TEE was not necessary for three patients; the PAC could not be inserted in a timely manner in three patients; and in three patients, the investigators were not available at the time of surgery. The remaining 54 patients were included. Two patients without known tricuspid valve disease were discovered to have severe TR on the comprehensive intraoperative TEE examination. These patients were included in the analysis. The mean age of the patients was 67 yr (SD = 10). Most of the patients (67%) underwent aortocoronary bypass alone, 16% underwent aortocoronary bypass in combination with aortic valve replacement, and 6% underwent aortic valve replacement alone. The remainder (11%) underwent other procedures, all using cardiopulmonary bypass. Significant (3+ or 4+) mitral regurgitation was present in 12% of the patients, and 21% of the patients had a left ventricular ejection fraction 
40% (mea-sured during the preoperative TEE examination).
There were no significant changes in BP or HR comparing the pre-PAC measurements to those measured after advancing the PAC. All patients were in sinus rhythm for all measurements. Before PAC insertion, all tricuspid valves (n = 54), and 41 of 52 pulmonic valves, showed at least trace (1+) regurgitation. Post-PAC, the mean regurgitant jet areas were not significantly larger at the pulmonary valve or the tricuspid valve at the ME 4-chamber view. The mean TR jet area increased by 0.37 ± 0.11 cm2 in the ME RV inflow-outflow view (P = 0.0014). We were not able to obtain satisfactory ME 4-chamber images in three patients or ME aortic valve short-axis images in two patients (see Table 1). Four of 51 patients (8%) had an increase in TR jet area 1 cm2 when measured at the ME 4-chamber view, whereas 9 of 54 patients (17%) had an increase in TR jet area 1 cm2 when measured at the ME RV inflow-outflow view. Four of 52 patients (8%) had an increase in PI jet area 1 cm2.
When the Spearman rank correlation analysis was applied to change in regurgitant jet area and covariate characteristics (age, ejection fraction, central venous pressure, and mean PA pressure), no significant correlation was found with any of the variables. When patients with moderate to severe TR (or area 2 cm2) pre-PAC were excluded, or when patients with an ejection fraction >40 or 40 were analyzed separately, passage of the PAC increased TR regurgitant area significantly (P = 0.007, 0.011, 0.028) at 40°, but there was no significant increase in TR at 0° or PI. When patients with moderate to severe TR (or area 2 cm2; n = 8) were analyzed separately by Wilcoxon’s signed rank test, there was no significant change in TR or PI at any view; however, with only eight patients in this group, this analysis is poorly powered.
Placement of a PAC did not result in severe (4+) TR in any patient, and the probability that placement of a PAC results in severe TR was very small (95% confidence intervals [CI]; 0%–8%); however, placement of a PAC worsens TR in 44% of patients (95% CI; 24%–65%). Moreover, placement of a PAC can result in moderate (3+) TR in 11% of patients (95% CI; 3%–25%).
Two patients’ jets changed from central to anterior, two patients’ jets changed from central to posterior, two changed from anterior to central, and one changed from posterior to central after passage of the PAC. One subject had a decrease in the number of TR jets and four patients had an increase in the number of TR jets. Fourteen patients had an increase in TR by at least one clinical grade ( Fig. 2).
View larger version (30K):
[in this window]
[in a new window]
|
Figure 2. Clinical grade (1+, 2+, 3+, 4+) of tricuspid regurgitation at the right ventricular inflow-outflow (40°) view—before pulmonary artery catheter (PAC) placement, after PAC placement, and change in grade.
|
|
Intraclass correlations are shown in Table 2. The intraclass correlations are the proportion of total variability attributed to between-patient variability. Thus, if the echocardiographers were in complete agreement for every measure, all variability would be attributed to the patients, and then the intraclass correlation would equal one (11,12). We also have reported the square root of the intraclass correlations as is often done when referring to correlation coefficients (r). These show excellent agreement between measurements obtained by all three echocardiographers, except for the post-PAC placement pulmonary insufficiency jet area measurement.
|
Discussion
|
|---|
Our findings were unexpected. We predicted that we would find a small increase in mean TR jet size at both image planes; however, we found that the mean TR jet size increased significantly in the ME RV inflow-outflow view but not in the 4-chamber view. This asymmetric increase in regurgitant jet size suggests PAC-induced malcoaptation of the anterior and posterior leaflets, which are the leaflets best visualized at the ME RV inflow-outflow plane (14). It is unlikely that a single leaflet is affected by a PAC, because few patients had a change in TR jet direction after catheterization, and single leaflet dysfunction typically produces eccentric regurgitant jets. A PAC causing malcoaptation of the anterior and posterior leaflets is logical when one considers the orientation of the valve in a supine patient and the orientation of a PAC as it enters the right atrium from the superior vena cava. The curvature of the PAC as it enters the RV outflow tract could course between the anterior and posterior leaflets. The increase in regurgitation in most patients was small, but 17% of patients had an increase in TR jet area 1 cm2, and 30% had an increase in TR severity by at least one clinical grade. The mean increase in PI was small, and few (8%) patients had an increase in PI 1 cm2.
Initial attempts to measure the effects that a catheter might have on the function of right heart valves used chemical indicators, special catheters, and radiographic contrast (5–7). These techniques were limited by three problems. First, it was not possible to mea-sure regurgitation in the absence of a catheter. Second, these studies were performed before it was known that trace TR and PI are very common in normal patients. Third, it is difficult to quantify valvular regurgitation by using these methods. The conclusion from most of these studies was that right heart catheterization causes some degree of regurgitation in most patients.
When Doppler echocardiography was introduced, it became possible to measure regurgitation without right heart catheterization. In 1989, Shandling et al. (8) used pulse wave (PW) and continuous wave (CW) Doppler to detect improvement in regurgitation while removing catheters and pacemakers from the right heart. They measured very little change in TR or PI, likely reflecting the inherent difficulties in quantifying regurgitation by using PW and CW Doppler. Ultimately, color Doppler imaging was introduced, potentially allowing a comprehensive evaluation of regurgitation before right heart catheterization. Stewart et al. (9) used transthoracic echocardiography to grade TR in 25 intensive care unit patients with PACs in place and compared these TR grades with echocardiograms performed at some point before catheterization. Unfortunately, they used a semiquantitative grading system based on the length of the PW TR jet, which ignores the presence of multiple jets, and inaccurately measures eccentric regurgitant jets. Additionally, a mean of 1.5 days elapsed between their echocardiographic examinations, during which time many uncontrolled hemodynamic changes could have occurred, such as diuresis, myocardial infarction, sepsis, and mechanical ventilation. Also, transthoracic echocardiography was used, which allowed only a single view of the tricuspid valve in some of their patients, likely contributing to failure to detect any TR before catheterization in 16 of their 25 patients. They found a 48% incidence of "catheter induced" TR by using PW Doppler, and a 28% incidence by using color Doppler. In addition, they did not report the clinical grades of TR seen with color Doppler. Using TEE, our study detected at least trace TR before catheterization in all 54 patients, and the measurable hemodynamics were not significantly different among echocardiographic measurements.
Our study has several limitations. We chose to measure the largest-appearing regurgitant jet during a period of 10 cardiac cycles and 2 respiratory cycles, rather than to measure several consecutive jets and use the mean area of those jets. This method was chosen for its speed of performance and ability to measure the jets on-line. Because all of our patients had sinus rhythm, we cannot comment on the effect of a PAC in patients with atrial fibrillation. Patients in atrial fibrillation may have variable RV filling time and stroke volumes, causing larger regurgitant jets after longer cardiac cycles. However, this variability would also exist after passage of the PAC and during the second set of measurements. Our study was conducted on supine, hemodynamically stable, elective cardiac surgery patients who were receiving general anesthesia and mechanical ventilation. Although these conditions did not change during the course of our study, we do not know whether mechanical ventilation might alter our measurements when compared with spontaneous respiration attributed to changes in venous return and intrathoracic pressure. Moreover, the vasodilating effects of general anesthesia may tend to reduce any valvular regurgitation relative to what might be observed in awake patients undergoing PA catheterization.
In addition, all PACs (except one) were passed from the right internal jugular vein, and the measurements were made immediately after the PAC was advanced into the PA. With time, the PAC may soften, warmed by blood, resulting in less of an effect on the valves. We also cannot comment on what effect changes in pulmonary vascular resistance may have had in the amount of pulmonic regurgitation or TR, and unfortunately, we did not perform a PW or CW Doppler evaluation on the TR or PI jets so we cannot estimate PA pressures before or after the PAC was placed. However, with stable hemodynamics and mechanical ventilation and the short time interval between the two sets of measurements, it is unlikely that changes in pulmonary vascular resistance caused the isolated increase in TR as described. Another limitation is that the PI jet was measured in only one TEE view. This method of studying the pulmonic valve was chosen because finding and measuring PI from alternate views is frequently challenging and time consuming.
Our intraclass correlations were excellent except for the post-PAC PI jet. This poor correlation was attributed to one measurement that was dramatically different between the two observers of a poorly imaged PI jet. Without this one jet, the intraclass correlation would have been 0.94 (r = 0.971), and the post-pre intraclass correlation would have been 0.882 (r = 0.939).
We recognize that it may not be possible to extrapolate the results of this study to patients under different conditions, such as different patient positions, different catheterization sites, awake patients, or patients with a PAC that has been in place longer than a few minutes.
Despite these limitations, using TEE with color Doppler allowed us to examine catheter-induced regurgitation in a more quantitative manner than previous studies, and suggested a mechanism to explain our findings. To test our explanation that catheter-induced TR is caused by malcoaptation of the anterior and posterior leaflets, further investigation with magnetic resonance imaging or three-dimensional echocardiography may be useful, although these methods also have limitations.
In conclusion, the clinical implications of this study are that a PAC has minimal to no effect on PI, but can worsen TR in the ME RV inflow-outflow view. Also, the significant increases in TR were only seen in the ME RV inflow-outflow view, but not in the ME 4-chamber view after placement of a PAC. This suggests that the mechanism of TR is catheter-induced malcoaptation of the anterior and posterior tricuspid valve leaflets. Finally, if severe TR is seen, especially in multiple planes, it is very unlikely to be solely caused by the PAC, and other etiologies of TR need to be evaluated.
|
Acknowledgments
|
|---|
Supported, in part, by the Department of Anesthesiology, Wake Forest University School of Medicine, Winston-Salem, North Carolina.
|
Footnotes
|
|---|
Current address for RFB is Cardiovascular Anesthesia Consultants, 9309 Provence Garden Lane, Las Vegas, NV 89145.
|
References
|
|---|
-
Swan HJ, Ganz W, Forrester J, et al. Catheterization of the heart in man with use of a flow-directed balloon-tipped catheter. N Engl J Med 1970; 283: 447–51.
-
Wray DL, Hughes CW, Fine RH, Thomas SJ. Anesthesia for cardiac surgery. In: Barash PG, Cullen J, Stoelting BF, eds. Clinical anesthesia. 2nd ed. Philadelphia: Lippincott-Raven, 1996: 1021–57.
-
Elliott CG, Zimmerman GA, Clemmer TP. Complications of pulmonary artery catheterization in the care of critically ill patients: a prospective study. Chest 1979; 76: 647–52.[Abstract/Free Full Text]
-
Rafferty T, Durkin M, Hines RL, et al. The relationship between "normal" transesophageal color-flow Doppler-defined tricuspid regurgitation and thermodilution right ventricular ejection fraction measurements. J Cardiothorac Vasc Anesth 1993; 7: 167–74.[Medline]
-
Skelton R, Corday E. Detection of valvular insufficiency by a platinum electrode technic: a preliminary study. Am J Cardiol 1963; 11: 373.[Medline]
-
Sobol B, Bottex G, Emirgil C, Gissen H. Valvular insufficiency occurring during cardiac catheterization. Am J Cardiol 1964; 14: 533–6.[Medline]
-
Ubago JL, Figueroa A, Colman T, et al. Right ventriculography as a valid method for the diagnosis of tricuspid insufficiency. Cathet Cardiovasc Diagn 1981; 7: 433–41.[Medline]
-
Shandling AH, Lehmann KG, Atwood JE, et al. Prevalence of catheter-induced valvular regurgitation as determined by Doppler echocardiography. Am J Cardiol 1989; 63: 1369–74.[Medline]
-
Stewart D, Leman RB, Kaiser J, Mann DL. Catheter-induced tricuspid regurgitation: incidence and clinical significance. Chest 1991; 99: 651–5.[Abstract/Free Full Text]
-
Shanewise JS, Cheung AT, Aronson S, et al. ASE/SCA guidelines for performing a comprehensive intraoperative multiplane transesophageal echocardiographical examination: recommendations of the American Society of Echocardiography Council for Intraoperative Echocardiography and the Society of Cardiovascular Anesthesiologists Task Force for Certification in Perioperative Transesophageal Echocardiography. Anesth Analg 1999; 89: 870–84.[Free Full Text]
-
Armitage P, Berry G. Statistical methods in medical records. 3rd ed. Section 8.7, Intraclass correlation. Oxford: Blackwell Science, 1994:272–6.
-
Rosner BA. Fundamentals of biostatistics. 5th ed. Section 12.9, The intraclass correlation coefficient. Pacific Grove, CA: Duxbury, 2000:562–6.
-
Fisher EA, Goldman ME. Simple, rapid method for quantification of tricuspid regurgitation by two-dimensional echocardiography. Am J Cardiol 1989; 63: 1375–8.[ISI][Medline]
-
Maxted W, Nanda NC, Kim K-S, et al. Transesophageal echocardiographic identification and validation of individual tricuspid valve leaflets. Echocardiography 1994; 11: 585–96.
Accepted for publication December 28, 2000.
Top
Abstract
Introduction
