Background
Implantable device leads can cause tricuspid regurgitation (TR) when they interfere with leaflet motion. The aim of this study was to determine whether lead-leaflet interference is associated with TR severity, independent of other causative factors of functional TR.
Methods
A total of 100 patients who underwent transthoracic two-dimensional and three-dimensional (3D) echocardiography of the tricuspid valve before and after lead placement were studied. Lead position was classified on 3D echocardiography as leaflet-interfering or noninterfering. TR severity was estimated by vena contracta (VC) width. Logistic regression analysis was used to identify factors associated with postdevice TR, including predevice VC width, right ventricular end-diastolic and end-systolic areas, fractional area change, right atrial size, tricuspid annular diameter, TR gradient, device lead age, and presence or absence of lead interference. Odds ratios were used to describe the association with moderate (VC width ≥ 0.5 cm) or severe (VC width ≥ 0.7 cm) TR, separately, using bivariate and stepwise multivariate logistic regression analysis.
Results
Forty-five of 100 patients showed device lead tricuspid valve leaflet interference. The septal leaflet was the most commonly affected (23 patients). On bivariate analysis, preimplantation VC width, right atrial size, tricuspid annular diameter, and lead-leaflet interference were significantly associated with postdevice TR. On multivariate analysis, preimplantation VC width and the presence of an interfering lead were independently associated with postdevice TR. Furthermore, the presence of an interfering lead was the only factor associated with TR worsening, increasing the likelihood of developing moderate or severe TR by 15- and 11-fold, respectively.
Conclusion
Lead-leaflet interference as seen on 3D echocardiography is associated with TR after device lead placement, suggesting that 3D echocardiography should be used to assess for lead interference in patients with significant TR.
A growing number of permanent pacemakers (PPMs), implantable cardioverter-defibrillators (ICDs), and cardiac resynchronization therapy (CRT) devices are being inserted each year in the United States, as the indications for device therapy expand. Despite the increasing number of implanted devices, relatively little information is available on how they interact with the tricuspid valve (TV) apparatus. Some studies suggest that they induce tricuspid regurgitation (TR), a valvular lesion that is increasingly appreciated to be associated with high morbidity and mortality, whereas other studies suggest that they do not.
Functional TR, commonly encountered in clinical practice, usually occurs in the presence of an anatomically preserved valve. It is frequently directly associated with an enlarged tricuspid annulus and/or a dilated right ventricle but can also be related to leaflet tethering and right ventricular (RV) remodeling as a consequence of increased right-sided pressures often secondary to left-sided heart disease.
We recently reported on an observational study that described the use of three-dimensional (3D) echocardiography for the visualization of device leads as they traverse the TV. We found that TR was present when implanted device leads were interfering with the normal TV leaflet motion. Although that study showed an association between the presence of a device lead and TR, it did not prove that the lead was a major cause of TR, because preimplantation studies were not available in the majority of patients. In addition, other factors were not considered as potential causes of TR, and the strengths of their associations with the severity of TR were not studied.
Accordingly, in the present study, we sought to determine whether device lead interference with the TV leaflets, as depicted by 3D echocardiography, is an important contributor to TR severity by comparing pre- and postimplantation studies or whether other factors known to increase TR severity, such as RV size and systolic function, tricuspid annular (TA) diameter, systolic pulmonary artery pressure (sPAP), and right atrial (RA) size, are more important predictors of TR in these patients.
Methods
Patient Population
We retrospectively studied 100 consecutive patients (49 men, 51 women; mean age, 67 ± 16 years; range, 26–93 years) with morphologically normal TVs who had device lead implantation in the right ventricle (PPM, ICD, or CRT with defibrillator or pacemaker) and had undergone complete two-dimensional (2D) and 3D transthoracic echocardiography before and after device implantation. Fifty-three of these patients were included in our previous publication. Demographic information, date and site of device generator implantation (left vs right chest pocket), and type of device were obtained by chart review. This study was approved by the institutional review board.
2D Transthoracic Echocardiography
Comprehensive 2D and color Doppler evaluation was performed by an experienced sonographer using an iE33 imaging system equipped with an S5 transducer (Philips Medical Systems, Andover, MA). Digital loops were stored and analyzed ( Figure 1 ) offline (Xcelera Workstation; Philips Medical Systems). TR was quantified using vena contracta (VC) width according to published guidelines. It represented the narrowest portion of the TR jet at or downstream from the orifice in mid-systole. The largest VC width obtained from either the RV inflow or apical four-chamber view was reported on both the pre- and postimplantation echocardiograms ( Figure 1 C). In addition, RA dimensions and area, RV basal and mid-end-diastolic dimensions and fractional area change, sPAP, and RA pressure were recorded according to published guidelines.
Maximal RA long-axis and short-axis dimensions were measured at ventricular end-systole, defined as the frame immediately before TV opening, when the RA chamber was at its greatest size in the apical four-chamber view. A straight line was traced connecting both sides of the tricuspid leaflet attachment points. The RA long-axis dimension was measured from the midpoint of this line to the superior border of the right atrium parallel to the interatrial septum. The short-axis dimension was then measured from the lateral wall of the RA to the interatrial septum, perpendicular to the RA long-axis dimension ( Figure 1 D). RA area was obtained from the apical four-chamber view by tracing the endocardial border at end-systole excluding the RA appendage ( Figure 1 D). TA diameter was also obtained from the apical four-chamber view at end-diastole starting at the lateral hinge point of the nonseptal TV leaflet and ending at the hinge point of the septal TV leaflet ( Figure 1 B). RV basal (maximal dimension in the lower third) and middle (maximal dimension in the middle third) cavity measurements were made in the RV-focused apical four-chamber view in end-diastole ( Figure 1 A). Fractional area change, defined as (end-diastolic area − end-systolic area)/end-diastolic area × 100, was also measured using the RV-focused apical four-chamber view. End-diastole and end-systole were defined as the frames depicting the largest and smallest RV cavity sizes, respectively ( Figure 1 E). Peak TR gradient was measured using the modified Bernoulli equation, from the maximal TR jet velocity ( Figure 1 F). sPAP was estimated using the TR gradient and adding an approximation of RA pressure based on inferior vena cava size and collapsibility ( Figure 1 G).
3D Transthoracic Echocardiography
Three-dimensional transthoracic echocardiographic studies were performed by using a Phillips iE33 ultrasound system (Philips Medical Systems) with a fully sampled matrix-array transducer (model X7-2t). The TV was imaged from the apical four-chamber view using full-volume and 3D zoom modes, as previously described. Three-dimensional acquisitions were performed using electrocardiographic gating over four consecutive cardiac cycles with a single breath-hold. Once acquired, the image was cropped and oriented to visualize the TV leaflets in the en face view (RV or RA perspective), depending on which orientation best depicted the device lead. For display, the TV was oriented with the septum in the 6 o’clock position, in accordance with American Society of Echocardiography guidelines.
Device Lead Location Relative to the TV Leaflets
Three-dimensional data sets were cropped (QLAB version 9.0; Phillips Medical Systems) and displayed to enable identification of the device lead position at the level of the tricuspid annulus. The device lead was described as interfering with leaflet motion (anterior, posterior, or septal), if it was noted to be impinging or adhering to a leaflet ( Videos 1–3 ; available at www.onlinejase.com ). If the device lead was in the commissure, the lead position was named according to the two leaflets between which it was located (anteroseptal, anteroposterior, or posteroseptal) ( Videos 4 and 5 ; available at www.onlinejase.com ). A lead was described as mobile or centrally located when it was neither in a commissural position nor interfering with leaflet function. In addition, device pulse generator position was determined as “left” or “right” using the chest x-ray taken closest to the time of 3D transthoracic echocardiography.
Statistical Analysis
Continuous variables are summarized as mean ± SD, and categorical variables are presented as absolute numbers and percentages, unless otherwise stated. Pre- and postdevice measurements were compared using paired t tests. Differences between leads resulting in “leaflet-interfering” and “noninterfering” groups were tested using the unpaired t test. Independent factors associated with postdevice VC width increase were established using multivariate linear regression analysis, including individual factors with significant associations in bivariate analysis. The association between each independent factor and VC width is reported as β, representing the slope of the regression line and describing the expected change of VC width as a consequence of a unitary increase of the relevant predictor.
Postdevice TR VC width was dichotomized to separate moderate (VC width ≥ 0.5 cm) from less than moderate (VC width < 0.5 cm) TR and severe (VC width ≥ 0.7 cm) from less than severe (VC width < 0.7 cm) TR. Associations between predevice echocardiographic parameters and dichotomized TR severity were investigated using stepwise, backward and forward, multivariate logistic regression. Odds ratios (OR) were used to describe the associations of factors with moderate or severe TR, separately, using both the bivariate and multivariate logistic regression analyses.
Finally, the mitral regurgitation scale for VC severity was used to divide TR VC into mild (grade 1, 0–0.3 cm), mild to moderate (grade 2, 0.3–0.5 cm), moderate (grade 3, 0.5–0.7), and severe (grade 4, ≥0.7 cm) categories to determine in which cases (interfering vs noninterfering leads) TR improved significantly (decreased by ≥1 grade) or worsened significantly (increased from grade 1 or 2 to grade 3 or 4 or from grade 3 to 4).
Statistical significance was defined as P ≤ .05. Statistical analysis was performed using SPSS Statistics 20 (IBM, Armonk, NY).
Results
Table 1 shows the summary of baseline characteristics. Of the 100 patients studied, 53% had ICDs, 20% had PPMs, and 27% had CRT devices. The majority of device generators (88%) were implanted in a left-sided chest pocket. The mean time between device implantation date and the postimplantation 2D or 3D echocardiographic study was 3.8 ± 3 years. The mean time between the preimplantation echocardiographic examination and device lead placement was 0.9 ± 1.6 years. The majority of patients (79%) underwent preimplantation echocardiography <1 year before device lead implantation.
| Characteristic | n ∗ | Before device placement ( n = 100) | n ∗ | After device placement ( n = 100) | P |
|---|---|---|---|---|---|
| Demographics | |||||
| Age (y) | 100 | — | — | 67 ± 16 (26–93) | — |
| Men | 100 | — | — | 49% | — |
| Device lead age † (years) | 100 | — | — | 3.8 ± 3 | — |
| Device information | |||||
| Pulse generator position on left | — | — | 97 | 84 (85%) | — |
| ICD | — | — | 100 | 53 (53%) | — |
| PM | — | — | 20 (20%) | — | |
| CRT defibrillator | — | — | 27 (27%) | — | |
| Echocardiographic parameters | |||||
| LVEF (%) | 98 | 35 ± 16 | 97 | 36 ± 16 | .79 |
| RA-T (mm) | 96 | 44 ± 9 | 100 | 46 ± 10 | .32 |
| RA-L (mm) | 95 | 53 ± 9 | 100 | 53 ± 11 | .78 |
| RA area (cm 2 ) | 95 | 22 ± 7 | 100 | 23 ± 9 | .02 |
| TA diameter (mm) | 98 | 37 ± 6 | 99 | 37 ± 7 | .76 |
| RVEDDb (mm) | 97 | 46 ± 7 | 100 | 50 ± 9 | <.01 |
| RVEDDm (mm) | 96 | 33 ± 7 | 99 | 39 ± 8 | <.01 |
| RVEDA (cm 2 ) | 98 | 27 ± 8 | 100 | 30 ± 9 | <.01 |
| RVESA (cm 2 ) | 98 | 17 ± 8 | 100 | 20 ± 9 | <.01 |
| RV FAC (%) | 98 | 38 ± 14 | 100 | 36 ± 13 | .13 |
| TR gradient (mm Hg) | 86 | 39 ± 16 | 93 | 40 ± 16 | .51 |
| RA pressure (mm Hg) | 96 | 7 ± 5 | 94 | 7 ± 5 | .90 |
| sPAP (mm Hg) | 86 | 46 ± 18 | 93 | 46 ± 18 | .79 |
| TR VC width (cm) | 100 | 0.31 ± 0.36 | 100 | 0.57 ± 0.49 | <.01 |
| TR severity | |||||
| Mild (0–0.3 cm) | 60 | 30 | |||
| Mild to moderate (0.3–0.5 cm) | 15 | 19 | |||
| Moderate (0.5–0.7 cm) | 8 | 22 | |||
| Severe (≥0.7 cm) | 17 | 29 |
∗ Number of patients for whom data were available.
† Obtained from determining the time between 3D echocardiography and device implantation date.
When comparing pre- and postimplantation echocardiograms, parameters that were significantly increased included RV end-diastolic and end-systolic areas and dimensions, RA area, and TR VC width. None of the other parameters, including left ventricular ejection fraction and TA diameter, significantly changed between pre- and postimplantation ( Table 1 ).
In accordance with clinical practice, 2D images were reviewed before 3D images. In cases with little or no TR device lead present, noninterference was assumed. In cases with moderate or more TR, often with the TR jet directed along the wire, device lead interference with normal leaflet motion was suggested. An attempt was not made to specify with which of the three TV leaflets the lead was interfering, as this could not be diagnosed with certainty. Leaflet interference was easier to detect on dynamic 3D images of the TV when visualized from the RV perspective ( Videos 1–4 ; available at www.onlinejase.com ). Forty-five patients (45%) showed device lead interference with the TV leaflet. The septal leaflet was the most commonly affected (23 cases, 51% of the total interferences) followed by the posterior leaflet (19 cases, 42% of the interferences) and anterior leaflet (three cases) ( Figure 2 ). Leaflet interference was most commonly due to device lead impingement ( Figure 2 ) or adherence to the leaflet ( Figure 3 ). Fifty-five patients did not have device lead–associated leaflet interference. In these patients, device leads were seen in either the posteroseptal commissure (26 cases) ( Figure 4 ) or the middle of the valve (26 cases) ( Figure 5 ). Videos 3 and 4 (available at www.onlinejase.com ) illustrate a device lead in the posteroseptal ( Video 3 ) and middle ( Video 4 ) locations. Figure 6 details the number of device leads in interfering and noninterfering locations.
Differences between patients with and without device lead–associated leaflet interference are summarized in Table 2 . There were no significant intergroup differences on predevice echocardiograms. On the postdevice echocardiograms, in the presence of an interfering lead, the TR VC width was larger, as were the TR gradient, sPAP, and RA areas. Of note, the time between implantation and echocardiographic recording was similar between the interfering and noninterfering groups ( Table 2 ).
| Parameter | Before device implantation | After device implantation | ||||
|---|---|---|---|---|---|---|
| Noninterfering leads ( n = 55) | Interfering leads ( n = 45) | P ∗ (predevice) | Noninterfering leads ( n = 55) | Interfering leads ( n = 45) | P † (postdevice) | |
| Age (y) | — | — | — | 66 ± 14 | 67 ± 19 | .94 |
| Men | — | — | — | 60% | 36% | |
| Device age (y) | — | — | — | 3.3 ± 2.7 | 4.5 ± 3.4 | .07 |
| LVEF (%) | 35 ± 16 | 35 ± 17 | .97 | 36 ± 16 | 36 ± 16 | .92 |
| RA-T (mm) | 43 ± 8 | 45 ± 10 | .24 | 44 ± 8 | 47 ± 13 | .30 |
| RA-L (mm) | 52 ± 9 | 54 ± 10 | .50 | 52 ± 10 | 55 ± 13 | .16 |
| RA area (cm 2 ) | 21 ± 6 | 23 ± 9 | .29 | 22 ± 7 | 25 ± 10 | .03 |
| TA diameter (mm) | 37 ± 6 | 37 ± 7 | .72 | 37 ± 8 | 38 ± 6 | .46 |
| RVEDDb (mm) | 45 ± 7 | 46 ± 8 | .79 | 49 ± 10 | 51 ± 8 | .33 |
| RVEDDm (mm) | 33 ± 7 | 33 ± 8 | .90 | 38 ± 8 | 40 ± 8 | .30 |
| RVEDA (cm 2 ) | 27 ± 8 | 26 ± 8 | .40 | 28 ± 9 | 31 ± 9 | .18 |
| RVESA (cm 2 ) | 17 ± 7 | 17 ± 8 | .97 | 19 ± 9 | 21 ± 9 | .32 |
| RV FAC (%) | 39 ± 14 | 37 ± 14 | .41 | 36 ± 13 | 35 ± 14 | .80 |
| TR gradient (mm Hg) | 36 ± 14 | 42 ± 18 | .13 | 36 ± 15 | 45 ± 16 | <.01 |
| RA pressure (mm Hg) | 6 ± 5 | 7 ± 5 | .21 | 7 ± 5 | 6.8 ± 5.0 | .99 |
| sPAP (mm Hg) | 43 ± 15 | 50 ± 21 | .08 | 42 ± 18 | 51 ± 18 | .01 |
| TR VC width (cm) | 0.26 ± 0.38 | 0.35 ± 0.32 | .25 | 0.38 ± 0.45 | 0.79 ± 0.44 | <.01 |
| TR severity | ||||||
| Mild (0–0.3 cm) | 38 | 22 | 26 | 4 | ||
| Mild to moderate (0.3–0.5 cm) | 7 | 8 | 14 | 5 | ||
| Moderate (0.5–0.7 cm) | 1 | 7 | 9 | 13 | ||
| Severe (≥0.7 cm) | 9 | 8 | 6 | 23 | ||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
