Background
The sharp increase in health care costs over the past decade has prompted health care providers to reevaluate how diagnostic imaging is utilized. In response to the need for more rational use of imaging services, the American College of Cardiology Foundation and the American Society of Echocardiography have developed appropriate use criteria (AUC) for transthoracic echocardiography to guide its utilization. Although community and regional hospitals, such as Danbury Hospital, account for 85% of registered hospitals in the United States, very little is known about adherence to the AUC at these institutions.
Methods
The electronic medical records of 1,205 patients who underwent inpatient transthoracic echocardiography from January 1 to June 30, 2008, were retrospectively examined to determine the reasons for ordering the studies. The 2007 and 2011 AUC were used to classify indications as appropriate, inappropriate, or uncertain.
Results
Using the 2007 AUC, 86% of echocardiographic examinations were classified as appropriate. One percent had indications that were inappropriate, and there were no uncertain indications. Thirteen percent of studies were ordered for reasons not defined by the 2007 AUC. The most common appropriate indications were symptoms due to suspected cardiac etiology, initial evaluation after acute myocardial infarction, and acute chest pain with suspected myocardial ischemia. When evaluated using the 2011 AUC, appropriate and inappropriate indications increased to 97% and 2%, respectively. Ninety-three percent of undefined studies, using the 2007 AUC, could be classified using the 2011 guidelines.
Conclusions
Consistent with studies conducted at university hospitals, Danbury Hospital, a regional hospital, showed good adherence to the AUC. This suggests that the AUC are valuable across a large continuum of inpatient settings and can serve as an excellent guide for utilization and appropriateness.
Over the past decade, there has been a sharp increase in health care costs, leading to concerns about the sustainability of such growth. The United States spends approximately 16% of its gross domestic product on health care, of which one half is spent on hospital care and physician and clinical services combined. Diagnostic imaging services have increased at a faster rate than any other types of physician services, and according to Medicare data, transthoracic echocardiography (TTE) constitutes a significant proportion of cardiovascular imaging services. Health care providers and payers are being asked to reevaluate how these services are utilized and practice more cost effective health care.
In an effort to respond to the need for rational use of imaging services, the American College of Cardiology Foundation in conjunction with the American Society of Echocardiography and other imaging subspecialty societies developed appropriate use criteria (AUC) for TTE and transesophageal echocardiography using the RAND-UCLA appropriateness method. These AUC were anticipated to have a significant impact on physician decision making, as well as reimbursement policy, and to provide a framework of broad indications that are applicable to daily clinical practice.
Studies have examined the use of TTE in relation to these AUC, but little is known about how this service is utilized in community and regional hospitals. Previous studies were performed in hospitals with an average of 919 beds and approximately 45,313 admissions per year; in contrast, we chose to examine utilization in a typical regional teaching hospital. In 2009, community and regional hospitals accounted for 86% of the 5,795 hospitals registered in the United States and 95% of admissions. Thus, we sought to determine the pattern of use of TTE at a regional hospital to define the adherence to the AUC at this level.
Methods
We retrospectively examined a consecutive sample of patients who underwent inpatient TTE at Danbury Hospital (Danbury, Connecticut). The protocol was reviewed and approved by the Danbury Hospital institutional review board.
Study Population
Danbury Hospital is a 315-bed teaching hospital located in Fairfield County, Connecticut, with nearly 20,000 admissions per year. The hospital serves a population of 375,000 residents; 32% of Danbury residents are aged > 50 years, and 35% report minority status.
All inpatients undergoing TTE between January 1, 2008, and June 30, 2008, were eligible for inclusion in this study. A complete list of patients undergoing TTE within this period was obtained from the hospital’s daily admission census. Of the 1,205 inpatient studies performed during this period, 26 records were excluded because there was no documentation in the hospital record pertaining to the admission during which TTE were performed. After this exclusion, the total number of records reviewed was 1,179. There were no other exclusion criteria.
Chart Review
For each patient, the following were reviewed: emergency room reports, admission history and physical examination reports, discharge summaries, and specialist consultations. Two of the investigators (S.A.B. and I.M.) performed the review. Each investigator examined a similar number of patients. Demographics, principal diagnosis at discharge, reason for ordering TTE, classification using the AUC, and the indication number as listed in the AUC for TTE were documented for each patient.
Indication Determination and Classification
For each patient, all clinical information contained in the hospital record for the admission during which TTE was performed was reviewed. This included the admission history and physical examination, specialist consults, laboratory results, and discharge summaries. In determining the indication for ordering TTE, the history and physical examination conducted at the time of admission were first reviewed. If the reason was not stated, any specialty consults performed were reviewed, followed by the hospital course outlined in the discharge summary.
If the reason for ordering TTE was found, the study was then classified as appropriate, inappropriate, or uncertain on the basis of the 2007 AUC tables. If the reason for TTE was not found in the reports, a classification of unable to evaluate was assigned. When the indication for TTE was identified in the hospital records but not identified in the 2007 AUC, the test was classified as not defined in 2007 AUC ( Figure 1 ).
The revised 2011 AUC were applied to all studies for which indications for ordering TTE were found, to determine the impact that the revised guidelines would have on indication classification. If an investigator was unable to assign a classification, using the 2007 and 2011 AUC, both investigators met and reviewed the hospital chart for a consensus assignment.
Interobserver Variability
To determine interobserver variability, both investigators independently determined the appropriateness of TTE performed on 109 randomly selected patients who underwent TTE within the study period, and the results were compared.
Statistical Analysis
Interobserver variability was determined on a randomly selected subset of 109 patients who underwent TTE within the study period. Both investigators independently determined the appropriateness of TTE performed on these patients using the AUC, and agreement between the classifications was determined. Average values are expressed as mean ± SD. Chi-square or Fisher’s exact tests, as appropriate, were used to assess the differences between appropriateness levels for the 2007 compared with the 2011 AUC. Two-tailed P values < .05 was regarded as statistically significant. JMP version 8 (SAS Institute Inc., Cary, NC) was used for statistical analyses.
Results
The mean patient age was 71.2 ± 15 years (range, 20–101 years).
Appropriateness Classification
A total of 1,205 inpatient transthoracic echocardiographic studies were performed during the period under review. After the exclusion of 26 studies for which there were no medical records for the admission during which the study was performed, we reviewed the medical records for 1,179 studies. Of the 1,179 studies evaluated, we were able to determine the reasons for TTE in 92% ( n = 1,080); the remaining 99 (8%) were classified as unable to evaluate and thus were excluded from further analysis. Using the 2007 AUC, the majority of studies (86%) were determined to be appropriate ( Figure 1 ). Inappropriate indications were identified in 1% of studies, and there were no studies performed for uncertain indications. A total of 13% of studies were performed for reasons not defined by the 2007 AUC.
Appropriate Studies
The 2007 AUC categorize indications for TTE into six major categories. Table 1 shows the proportion of patients who underwent TTE in each category. Fifty-four percent of studies were appropriately performed for the general evaluation of the structure and function of the heart ( Table 1 ). Within this category, the three most common appropriate indications were symptoms due to suspected cardiac etiology (48%), initial evaluation of left ventricular function after acute myocardial infarction (27%), and prior testing concerning for heart disease (15%). Table 2 shows the most common appropriate indications across all the categories.
| AUC table number | Category | Percentage of all studies for which an indication was found ( n = 1,080) | |||||
|---|---|---|---|---|---|---|---|
| Appropriate | Inappropriate | Uncertain | |||||
| 2007 | 2011 | 2007 | 2011 | 2007 | 2011 | ||
| 1 | General Evaluation of Structure and Function | 53.5 | 47.2 ∗ | 0.5 | 1.5 ∗ | 0 | 0 |
| 2 | Cardiovascular Evaluation in an Acute Setting | 14.2 | 30.0 ∗ | 0.4 | 0.4 | 0 | 0.3 |
| 3 | Evaluation of Valvular Function | 2.3 | 3.0 | 0.3 | 0.5 | 0 | 0 |
| 4 | Evaluation of Intracardiac and Extracardiac Structures and Chambers | 3.6 | 3.3 | 0 | 0 | 0 | 0 |
| 5 | Evaluation of Aortic Disease | 0 | 0 | 0 | 0 | 0 | 0 |
| 6 | Evaluation of Hypertension, Heart Failure or Cardiomyopathy | 12.7 | 12.9 | 0 | 0 | 0 | 0.1 |
| 7 | Evaluation of Adult Congenital Heart Disease | 0 | 0.1 | 0 | 0 | 0 | 0 |
| Total | 86.3 | 96.5 ∗ | 1.2 | 2.4* | 0 | 0.4 | |
| AUC indications | Percentage of appropriate ( n = 933) |
|---|---|
| Indication #1: symptoms potentially due to suspected cardiac etiology | 29 |
| Indication #8: initial evaluation of left ventricular function after acute myocardial infarction | 17 |
| Indication #12: acute chest pain with suspected myocardial ischemia | 11 |
| Indication #2: prior testing concerning for heart disease | 9 |
| Indication #43: reevaluation of known heart failure | 7 |
| Indication #41: initial evaluation of known or suspected heart failure | 6 |
| Indication #6: sustained or nonsustained supraventricular tachycardia or ventricular tachycardia | 6 |
| Indication #11: evaluation of hypotension or hemodynamic instability | 4 |
| Indication #36: evaluation of pericardial conditions | 4 |
| Indication #31: initial evaluation of suspected infective endocarditis | 1 |
| Other ∗ | 6 |
∗ Evaluation of patient with known or suspected acute pulmonary embolism to guide therapy, evaluation for dyssynchrony in a patient being considered for cardiac resynchronization therapy, reevaluation of native valvular regurgitation in patients with a change in clinical status, initial evaluation of murmur, evaluation of respiratory failure with suspected cardiac etiology, evaluation for cardiovascular source of embolic event, reevaluation of left ventricular function after myocardial infarction during recovery phase when results will guide therapy, reevaluation of a patient with native valvular stenosis who has had a change in clinical status, evaluation of cardiac mass (suspected tumor or thrombus), initial evaluation of suspected hypertensive heart disease, assessment of known or suspected adult congenital heart disease, evaluation of known or suspected pulmonary hypertension, evaluation of suspected complication of myocardial ischemia or infarction, initial evaluation of known or suspected native valvular stenosis, initial evaluation of prosthetic valve for establishment of baseline after placement, reevaluation of patients with prosthetic valves with suspected dysfunction or thrombosis or a change in clinical status, patient with known implanted pacing device with symptoms possibly due to suboptimal pacing device settings, reevaluation of known hypertrophic cardiomyopathy in a patient with a change in clinical status, and baseline and serial reevaluations in patients undergoing therapy with cardiotoxic agents.
Of the 276 patients evaluated for symptoms due to suspected cardiac etiology the most common symptoms were cerebrovascular accident (CVA) or transient ischemic attack (TIA) (46%), shortness of breath or dyspnea (35%), and syncope (19%). One third of patients evaluated because of CVA or TIA had previously documented echocardiograms, of which seven (17%) were obtained within the preceding 6 months. Some 36% of patients evaluated because of syncope had previous echocardiograms, of which 32% were obtained in the preceding 6 months.
Heart failure evaluation accounted for 13% of appropriate studies. This included both the initial evaluation as well as reevaluation of symptomatic patients with known heart failure. Eleven percent of appropriate studies were performed in patients with acute chest pain with suspected myocardial ischemia.
Inappropriate Indications
Using the 2007 AUC, only 1% of transthoracic echocardiographic studies were performed for inappropriate indications. Inappropriate cases were limited to four indications: diagnosis of pulmonary embolism (33%), evaluation of left ventricular function with prior normal ventricular function evaluation within the past year in a patient with no clinical change (25%), evaluation of native and/or prosthetic valves in patients with transient fever but without evidence of bacteremia or new murmur (25%), and patients with isolated atrial premature contraction or premature ventricular contraction without other evidence of heart disease (17%).
Indications Not Defined by the 2007 AUC
Thirteen percent of transthoracic echocardiographic studies were performed for reasons not defined in the 2007 AUC. The most common undefined indications were atrial fibrillation (44%), elevated troponin I with no other evidence of acute myocardial injury (21%), and preoperative evaluation (9%) ( Figure 2 ).
Indication Classification Using the 2011 AUC
The 2011 AUC categorize indications for TTE into seven major categories. Table 1 shows the proportion of patients who underwent TTE in each category. Using these guidelines, 97% of studies performed were determined to be appropriate, with inappropriate indications identified in 2% of studies. Uncertain indications and indications not defined by the 2011 AUC were found in 1% of studies performed ( Figure 3 ).
Of the 933 studies categorized as appropriate using the 2007 AUC, two (0.2%) performed for the evaluation of presyncope were reclassified as inappropriate, while 99.8% remained appropriate. As shown in Table 3 , symptoms potentially related to suspected cardiac etiology (22%) and initial evaluation of ventricular function after acute coronary syndromes (15%) remained the two most common appropriate indications. Sustained or nonsustained atrial fibrillation, supraventricular tachycardia, or ventricular tachycardia accounted for 11% of appropriate transthoracic echocardiographic studies performed. Heart failure evaluation accounted for a total of 12% of appropriate studies; there were no studies done for the reevaluation of known heart failure with a change in clinical status or cardiac exam and a documented clear precipitating change in medication or diet.
| AUC indication | Percentage of appropriate ( n = 1,042) |
|---|---|
| Indication #1: symptoms potentially due to suspected cardiac etiology | 22 |
| Indication #24: initial evaluation of left ventricular function after acute coronary syndromes | 15 |
| Indication #5: sustained or nonsustained atrial fibrillation, supraventricular tachycardia, or ventricular tachycardia | 11 |
| Indication #21: acute chest pain with suspected myocardial infarction and nondiagnostic electrocardiographic results | 10 |
| Indication #10: prior testing concerning for heart disease | 10 |
| Indication #71: reevaluation of known heart failure without precipitating change in diet or medication | 6 |
| Indication #70: initial evaluation of known or suspected heart failure | 6 |
| Indication #19: evaluation of hypotension or hemodynamic instability | 4 |
| Indication #9: syncope | 4 |
| Indication #59: suspected pericardial conditions | 2 |
| Indication #7: clinical symptoms or signs consistent with a cardiac diagnosis known to cause lightheadedness, presyncope, or syncope | 2 |
| Other ∗ | 8 |
∗ Frequent premature ventricular contractions, evaluation of suspected pulmonary hypertension, suspected complication of myocardial ischemia or infarction, reevaluation of left ventricular function after acute coronary syndromes during recovery phase when results will guide therapy, respiratory failure or hypoxemia of uncertain etiology, known acute pulmonary embolism to guide therapy, severe deceleration injury or chest trauma with suspected cardiac injury, initial evaluation when there is reasonable suspicion of valvular or structural heart disease, reevaluation of known valvular heart disease with a change in clinical status, routine surveillance (≥1 year) or moderate or severe valvular stenosis without change in clinical status, initial evaluation of prosthetic valve for establishment of baseline after placement, initial evaluation of suspected infective endocarditis with positive blood cultures or a new murmur, suspected cardiac mass, suspected cardiovascular source of embolus, reevaluation of known pericardial effusion to guide management, initial evaluation of suspected hypertensive heart disease, initial evaluation or reevaluation after revascularization and/or optimal medical therapy to determine candidacy for device therapy and/or determine optimal choice of device, known implanted pacing device with symptoms possibly due to device complications or suboptimal pacing device settings, reevaluation of known cardiomyopathy with a change in clinical status, baseline and serial reevaluations in patients undergoing therapy with cardiotoxic agents, and initial evaluation of known or suspected adult congenital heart disease.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
