Background
Attention to resource utilization has led to increased scrutiny of the appropriateness of initial diagnostic imaging studies on the basis of current guidelines. Far less attention has been paid to examining the lack of appropriate follow-up studies.
Methods
A retrospective cross-sectional analysis was performed of 3,781 consecutive outpatients referred for transthoracic echocardiography (TTE) from July to December 2008. Data from the electronic medical records were extracted to see if patients with at least moderate left-sided valvular stenosis or regurgitation underwent subsequent echocardiographic studies within 60 days of the period recommended by the 2006 American College of Cardiology and American Heart Association valve guidelines document.
Results
Of 342 outpatients with at least moderate valve dysfunction, 38 (11%) were excluded for reasons that precluded the need for a follow-up study (e.g. death, surgery). Of the remaining 304 patients, only 179 (59%) underwent follow-up echocardiography within the recommended period. Rates of timely follow-up TTE were higher when ordering physicians were cardiologists or cardiovascular surgeons (65%) compared with primary care physicians or internal medicine specialists (45%) ( P < .01). Follow-up rates were significantly different for aortic stenosis (77%), mitral stenosis (67%), aortic regurgitation (49%), and mitral regurgitation (49%) ( P < .01). Patients receiving timely follow-up TTE were younger (66 ± 15 vs 71 ± 15 years, P = .002) and more likely to be male (odds ratio, 1.79; 95% CI, 1.12–2.85; P = .01).
Conclusions
To the authors’ knowledge, this is the first study demonstrating low rates of compliance with guideline-recommended monitoring TTE in patients with at least moderate valve dysfunction. Cardiac practitioners have significantly better compliance. Strategies are needed to improve timely follow-up care in this population.
There is increasing scrutiny on potential overutilization of diagnostic medical imaging because of both costs and potential morbidity. In particular, noninvasive diagnostic cardiovascular imaging has received more attention as its growth rate exceeds those of other physician services without a similar increase in disease prevalence. Several studies have examined appropriateness for inpatient, outpatient, academic, and community referrals for echocardiography, with most documenting 80% to 85% appropriateness. These data show variations in guideline adherence across specialties and are focused on identifying inappropriate referrals in order to contain costs. However, there is a paucity of studies examining rates of timely appropriate follow-up studies (or otherwise inadequate monitoring) when patients have pathologies for which noninvasive monitoring is recommended.
Patients with valvular heart disease require such regular monitoring. These patients have an increased risk for adverse events, with a mortality risk ratio of 1.75 compared with those without valvular disease. Serial transthoracic echocardiography (TTE) is an integral component of monitoring valve disease progression, in addition to identifying changes in left ventricular function and size that may indicate the need for interventions even in the absence of symptoms, before irreversible remodeling or death occurs. Thus, inadequate monitoring can potentially lead to excess morbidity and mortality. Valve guidelines have been established as to when repeat TTE is recommended and remain similar in the 2014 update.
To determine the impact of increased adherence to the appropriateness criteria on echocardiography volume, one needs to consider the net effect of decreased inappropriate referrals for TTE in combination with increased appropriate referrals. The prevalence of at least moderate valvular heart disease has been estimated to be 0.7% in those <40 years of age, increasing to >13% in those >70 years of age. Therefore, the overall impact to TTE volumes is expected to be significant.
To address this issue, we sought to determine the rates of timely referral for guideline-recommended surveillance TTE in this patient group.
Methods
The study was reviewed and approved by the hospital institutional review board. Informed consent was waived.
This was a retrospective, cross-sectional analysis of consecutive patients who underwent outpatient TTE during a 6-month period of enrollment (July 1 to December 31, 2008) at the Beth Israel Deaconess Medical Center, a large academic medical center in Boston. The electronic echocardiography report database was screened to identify patients with at least moderate left-sided valve disease (i.e., aortic regurgitation or stenosis, mitral regurgitation or stenosis ) ( Table 1 ). We excluded patients whose valvular disease severity was deemed to be less than moderate by the either the ordering physician or the patient’s attending physician ( n = 5) after transthoracic echocardiographic review.
| Valve lesion | Criteria |
|---|---|
| >Moderate aortic stenosis | Jet velocity > 3 m/sec Mean gradient > 25 mm Hg Valve area < 1.5 cm 2 |
| >Moderate aortic regurgitation | Color Doppler jet width > 25% of LVOT or >0.3 cm Doppler vena contracta width > 0.3 cm Regurgitant volume > 30 mL/beat Regurgitant fraction > 30% Regurgitant orifice area > 0.10 cm 2 |
| >Moderate mitral stenosis | Mean gradient > 5 mm Hg Pulmonary artery systolic pressure > 30 mm Hg Valve area < 1.5 cm 2 |
| >Moderate mitral regurgitation | Color Doppler jet area > 4 cm 2 or >20% left atrial area Doppler vena contracta width > 0.3 cm Regurgitant volume > 30 mL/beat Regurgitant fraction > 30% Regurgitant orifice area > 0.2 cm 2 |
The patients’ electronic medical records and TTE reports were used to abstract sociodemographic, clinical, and echocardiographic data. The primary focus was on whether a patient underwent follow-up TTE within the guideline-recommended time period.
A patient was deemed to have appropriate follow-up if he or she underwent subsequent TTE within 60 days of the period recommended by the 2006 American College of Cardiology and American Heart Association valve guidelines document ( Table 2 ). The 2014 valve guidelines remain similar and included here for reference ( Table 2 ). For patients with multiple valve lesions, we selected the lesion for which follow-up TTE within the shortest period recommended by the guidelines. We also attempted to identify mitigating circumstances by which follow-up TTE was deemed unnecessary.
| Valve lesion | Moderate | Severe |
|---|---|---|
| 2006 guidelines | ||
| Aortic stenosis | 1–2 y | 1 y |
| Aortic regurgitation | 1 y | 6–12 mo ∗ |
| Mitral stenosis | 1–2 y | 1 y |
| Mitral regurgitation | 12 mo | 6–12 mo |
| 2014 guidelines | ||
| Aortic stenosis | 1–2 y | 6–12 mo |
| Aortic regurgitation | 1–2 y | 6–12 mo † |
| Mitral stenosis | 1–2 y | 1 y |
| Mitral regurgitation | 1–2 y | 6–12 mo † |
∗ If left ventricular end-diastolic diameter > 60 mm, every 6 to 12 months; if stable every 12 months; if left ventricular end-systolic diameter > 50 mm and left ventricular end-diastolic diameter > 70 mm, every 4 to 6 months.
Our electronic medical system has the ability to document, in a systematic fashion, all transfers of care out of the system. Attempts to contact patients for follow-up, whether successful or not, are documented in the electronic medical record. If TTE was performed outside our system within the recommended period, we classified this as “appropriate” follow-up.
Statistical Analysis
Continuous variables are expressed as mean ± SD. Categorical variables are expressed as counts and percentages. Potential predictors of appropriate follow-up TTE were evaluated with unpaired t tests for continuous variables and χ 2 or Fisher exact tests for categorical variables as appropriate. Univariate and multivariable logistic regression was used to further analyze the results to determine the adjusted independent relationship of each potential predictor with appropriate follow-up. The first model evaluated the difference in follow-up across the different valve lesions. Then individual models for each valve lesion were constructed. All potential predictive variables were initially included in the model and removed by backward selection if the P value was >.10. Age and gender were required to remain in all models as likely confounding factors. A two-tailed P value < .05 was considered to be indicate statistical significance. Statistical analysis was performed using SAS for Windows version 9.3 (Cary, NC).
Results
During the 6-month period, a total of 3,781 outpatient transthoracic echocardiographic studies were performed, including 342 in patients with at least moderate left-sided valve disease. Of these, 38 patients (11%) were excluded for reasons that precluded the need for follow-up TTE, including death ( n = 22), valve surgery ( n = 5), follow-up cardiac magnetic resonance imaging instead of TTE ( n = 4), palliative care ( n = 4), and transfer of care outside our system ( n = 3).
Lack of Referral
Of the remaining 304 patients included in this analysis, 41% ( n = 126) failed to undergo follow-up TTE within the guideline-recommended period ( Table 3 ). Twenty-two percent of patients had no follow-up TTE during a follow-up period of 2.61 ± 0.14 years. The actual times elapsed since the initial transthoracic echocardiographic study are illustrated in Figures 1 and 2 .
| Characteristic | All | Appropriate follow-up TTE | No appropriate follow-up TTE | P |
|---|---|---|---|---|
| n (%) | 304 | 178 (59%) | 126 (41%) | |
| Age (y) | 68 ± 15 | 66 ± 15 | 71 ± 15 | .002 |
| Men | 168 (55%) | 109 (61%) | 59 (47%) | .03 |
| BMI (kg/m 2 ) | 27 ± 6 | 27 ± 6 | 26 ± 6 | .07 |
| BP (mm Hg) | ||||
| Systolic | 126 ± 19 | 124 ± 18 | 129 ± 16 | .04 |
| Diastolic | 71 ± 11 | 71 ± 10 | 70 ± 11 | .52 |
| Heart rate (beats/min) | 69 ± 15 | 70 ± 15 | 68 ± 15 | .27 |
| Atrial fibrillation | 9 (2.8%) | 2 (1.1%) | 5 (4.0%) | .13 |
| LVEF (%) | 57 ± 13 | 57 ± 13 | 56 ± 14 | .43 |
The appropriate follow-up rates were significantly different for stenotic (76%) versus regurgitant (49%) lesions (aortic stenosis [77%], mitral stenosis [67%], aortic regurgitation [49%], and mitral regurgitation [49%]; P < .01) ( Table 4 ). Rates of timely follow-up TTE were also significantly higher when ordering physicians were cardiac practitioners (cardiologists and cardiovascular surgeons) compared with noncardiac practitioners (65% vs 45%, P = .009) ( Table 5 ). Compared with those without timely follow-up, patients receiving timely follow-up TTE were younger (66 ± 15 vs 71 ± 15 years, P = .002) and more likely to be male (61% vs 47%, P = .01). There was no significant difference in left ventricular ejection fraction (LVEF) or body mass index ( P > .05 for both; Table 3 ).
| Primary diagnosis (most severe lesion) | Patients ( n ) | Appropriate follow-up ( n ) | Appropriate follow-up rate (%) |
|---|---|---|---|
| Aortic stenosis | |||
| Moderate (2+) | 39 | 33 | 85 |
| Moderate/severe (3+) | 30 | 21 | 70 |
| Severe (4+) | 31 | 23 | 74 |
| All aortic stenosis | 100 | 77 | 77 |
| Aortic regurgitation | |||
| Moderate (2+) | 39 | 17 | 44 |
| Moderate/severe (3+) | 9 | 7 | 78 |
| Severe (4+) | 1 | 0 | 0 |
| All aortic regurgitation | 49 | 24 | 49 |
| Mitral stenosis | |||
| Moderate (2+) | 6 | 4 | 67 |
| All mitral stenosis | 6 | 4 | 67 |
| Mitral regurgitation | |||
| Moderate (2+) | 98 | 43 | 44 |
| Moderate/severe (3+) | 32 | 15 | 47 |
| Severe (4+) | 19 | 15 | 79 |
| All mitral regurgitation | 149 | 73 | 49 |
| All patients | 304 | 178 | 59 |
| Ordering physician specialty | Patients ( n ) | Appropriate follow-up ( n ) | Appropriate follow-up rate (%) | P ∗ |
|---|---|---|---|---|
| Cardiologist | <.01 | |||
| Noninvasive | 102 | 70 | 69 | |
| Electrophysiologist | 29 | 14 | 48 | |
| Interventional | 59 | 39 | 66 | |
| General | 8 | 4 | 50 | |
| Cardiovascular surgeon | 11 | 8 | 73 | |
| All cardiac practitioners | 209 | 135 | 65 | |
| Subspecialty internist | 18 | 8 | 44 | |
| Primary care physician/internist | 74 | 35 | 47 | |
| Other specialties | 3 | 0 | 0 | |
| All noncardiac practitioners | 95 | 43 | 45 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
