Limited data suggests ultrasound enhancing agent (UEA) use is associated with changes in clinical management and lower mortality in intensive care unit (ICU) patients. We conducted a retrospective observational study to determine if contrast echocardiography (vs non-contrast echocardiography) is associated with differences in length of stay (LOS) and subsequent resource utilization in the ICU setting. The Premier Healthcare Database (Charlotte, NC) was analyzed to identify patients receiving Definity vs. no use of contrast during the initial rest transthoracic echocardiogram (TTE) in an ICU setting. The primary outcomes of interest were subsequent TTE and transesophageal echocardiography (TEE) during the index hospitalization, and ICU LOS. Propensity scoring was used to statistically model treatment selection to minimize selection bias. A total of 1,538,864 patients from 773 hospitals were identified as undergoing resting TTE in the ICU with use of DEFINITY in 51,141 (3.3%) patients and no contrast agent use in 1,487,723 (96.7%) patients. After adjusting for patient, clinical, and hospital characteristics, patients in the Definity cohort were less likely to undergo a subsequent TTE or TEE as compared to those in the no contrast cohort (odds ratio = 0.704 for TTE, odds ratio = 0.841 for TEE; p < 0.0001 for both). Adjusted mean ICU LOS for the Definity cohort was shorter than that of the no contrast cohort (4.59 vs 4.15 days, p < 0.0001). In conclusion, Definity-enhanced echocardiography in the ICU setting (in comparison with non-contrast TTE) is associated with lower rates of subsequent TTE and TEE during the index hospitalization, and shorter ICU LOS.
Ultrasound enhancing agents (UEA) have been approved by the United States Food and Drug Administration for use in patients with suboptimal echocardiograms, to opacify the left ventricle and to improve endocardial delineation. , A recent guidelines and standards document published by the American Society of Echocardiography recommends use of UEAs in all technically difficult patients hospitalized in intensive care units (ICU) “to more quickly and accurately diagnose life-threatening conditions, and to reduce the need for downstream diagnostic testing”. These recommendations are based on previous studies, which (1) demonstrated an association with lower mortality in ICU patients who underwent transthoracic echocardiography (TTE) with an UEA in comparison with patients who underwent unenhanced echocardiography, and (2) showed evidence for reduced downstream testing and costs savings in patients undergoing TTE with an UEA. The objective of this study was to evaluate ICU length of stay, subsequent resource utilization, and management changes in ICU patients undergoing contrast echocardiography versus unenhanced echocardiography.
Methods
A retrospective observational study was conducted using the Premier Healthcare Database. The database is a large, geographically diverse, U.S. hospital administrative database containing patient-level, hospital-level, and payer-level data. All data in the database are statistically de-identified and compliant with the Health Insurance Portability and Accountability Act. The database includes information on patient demographics; hospital characteristics; coded diagnoses; and day-of-service-stamped billed services, including medications, laboratory tests performed, and diagnostic and therapeutic procedures for more than 660 million patient encounters, or one in every five hospital encounters in the nation. Patients can be followed across inpatient and outpatient encounters through a unique identifier, and treatments and costs tracked within a single hospital system that shares the same billing infrastructure. Adult inpatients ≥18 years of age with at least one Current Procedural Terminology code or hospital charge master description for rest TTE on a service day that indicates performance on an ICU day and discharged between January 1, 2009, and September 30, 2015, were identified. An ICU day(s) was identified as a day(s) in which ICU room and board charges were applied and included stepdown units, coronary care units, medical ICUs, and surgical ICUs. For patients with multiple qualifying hospitalizations during the study period, the first qualifying hospitalization was considered the index hospitalization, and was selected for use in the study. Patients were then classified into two cohorts based on use of the Definity contrast agent versus no use of any contrast agent during the first rest TTE (index rest TTE) as identified in the ICU charge master data (Definity cohort and no contrast cohort). Patients who expired during the index hospitalization and those who had a contrast agent other than Definity or an unidentifiable contrast agent from the charge master data during the index rest TTE, were excluded from the study.
Study variables included patient, clinical, index visit, and hospital characteristics. Patient characteristics included age, sex, race and/or ethnicity, and payer insurance coverage. Clinical characteristics included Charlson Comorbidity Index and major medical comorbidities of obesity, hypertension, diabetes, hyperlipidemia, heart failure, angina pectoris, and prior myocardial infarction. Patient clinical severity (minor, moderate, major, or extreme) was captured by the 3M All Patient Refined Diagnosis Related Group (APR-DRG) severity of illness and risk of mortality levels. Index visit characteristics included admission type; discharge disposition; admitting physician specialty; discharge year; type of ICU where patients received the rest index TTE; number of ICU days prior to index rest TTE; and use of parenteral inotropes (dobutamine, milrinone), parenteral unfractionated heparin, low-molecular-weight heparin (enoxaparin, dalteparin), and parenteral pressors (epinephrine, norepinephrine, dopamine, phenylephrine) on the day of index rest TTE. Hospital characteristics of bed size, U.S. census division, teaching status, and urban and rural populations served were recorded.
Primary outcomes of interest were presence of subsequent TTE (rest and stress) and subsequent TEE following index rest TTE during the index hospitalization and ICU LOS. TTE and TEE were determined by Current Procedural Terminology codes. ICU LOS was determined based on day-level room and board charges for ICU use and inclusive of all time spent in the ICU during the index hospital stay.
Changes in the use of parenteral inotropes, anticoagulants, and vasopressor medications following the index TTE were explored as secondary outcomes. Medication use data was obtained from the charge master. A newly started parenteral inotrope, anticoagulant, or vasopressor was defined as presence of a drug within the drug category on the day of index TTE or one day after index TTE that was not present on the day prior to index TTE. Discontinuation of a parenteral inotrope, anticoagulant, or vasopressor was defined as no presence of a drug within the drug category on the day of index TTE or one day after index TTE that was present on the day prior to index TTE.
Patient, clinical, index visit, and hospital characteristics and measured outcomes for the overall study population and for Definity and no contrast cohorts were reported. Mean, standard deviation, median, and interquartile ranges were reported for continuous data and counts and percentages were reported from categorical variables. Continuous variables were compared using Student’s T-test and categorical variables were compared using Chi-Square analysis.
Prior to comparing outcomes between the Definity and no contrast cohorts, inverse probability of treatment weighting using the propensity score was used to statistically model treatment selection to minimize selection bias. Patient, clinical, index visit and hospital characteristics were included in the logistic model to generate propensity score and backward elimination with a selection stay level of significance of 0.10 was performed to adjust the model. The propensity score, with a potential range between zero and one, represents a summary value of the covariates for each patient reflecting the propensity of a given patient to receive Definity. The inverse probability of treatment weighting of the Definity cohort was equal to the inverse of the propensity score, and the inverse probability of treatment weighting for the no contrast cohort was equal to the inverse of 1 minus the propensity score. The balance of baseline covariates was examined between the cohorts weighted by the inverse probability of treatment weighting.
Subsequently, weighted multivariable logistic regression was constructed to compare subsequent TTE and subsequent TEE between the Definity and no contrast cohorts. Weighted general linear model with Gamma variance and log link was constructed to compare ICU LOS between the two cohorts. Covariates used for adjustments include patient, clinical, index visit and hospital characteristics and medication changes. The multivariable models were built using backward elimination with a selection stay level of significance of 0.10.
An alpha of <0.05 was considered statistically significant. All analyses were performed using SAS software (version 9.4).
Results
A total of 1,538,864 patients from 773 hospitals were identified as receiving resting TTE in the ICU with use of Definity contrast agent in 51,141 (3.3%) patients and no contrast agent use in 1,487,723 (96.7%) patients. Baseline patient, clinical, index visit, and hospital characteristics are presented in Table 1 . During the index hospitalization, 9.8% of all patients underwent a subsequent TTE following the index rest TTE, and unadjusted analysis revealed that there was not a statistically significant difference between the Definity and no contrast cohorts (9.7% vs. 9.8%, p = 0.4161). Additionally, 4.4% of all patients underwent a subsequent TEE following the index rest TTE, with patients in the Definity cohort more likely to undergo a subsequent TEE than the no contrast cohort (5.1% vs. 4.4%, p < 0.001).
| Overall | Definity | No Contrast | p-values | |||||
|---|---|---|---|---|---|---|---|---|
| (N=1,538,864) | (N=51,141) | (N=1,487,723) | Definity vs. No Contrast | |||||
| Unique Patients | 1,538,864 | 51,141 | 1,487,723 | |||||
| Unique Hospitals | 773 | 423 | 772 | |||||
| Age (years) | Mean | 66.06 | 64.65 | 66.11 | <.0001 | |||
| Std Dev | ±15.74 | ±13.84 | ±15.80 | |||||
| Median | 67 | 65 | 67 | |||||
| IQR | 56 | 78 | 56 | 75 | 56 | 78 | ||
| 18-34 | 60,910 | 4.00% | 1,266 | 2.50% | 59,644 | 4.00% | <.0001 | |
| 35-44 | 83,338 | 5.40% | 2,782 | 5.40% | 80,556 | 5.40% | ||
| 45-64 | 522,816 | 34.00% | 20,394 | 39.90% | 502,422 | 33.80% | ||
| 65-74 | 360,169 | 23.40% | 13,644 | 26.70% | 346,525 | 23.30% | ||
| 75 + | 511,631 | 33.20% | 13,055 | 25.50% | 498,576 | 33.50% | ||
| Male | 815,102 | 53.00% | 32,916 | 64.40% | 782,186 | 52.60% | <.0001 | |
| Female | 723,344 | 47.00% | 18,225 | 35.60% | 705,119 | 47.40% | ||
| Unknown | 418 | 0.00% | 0 | 0.00% | 418 | 0.00% | ||
| Race/Ethnicity | White | 1,090,718 | 70.90% | 37,360 | 73.10% | 1,053,358 | 70.80% | <.0001 |
| Black | 200,252 | 13.00% | 5,567 | 10.90% | 194,685 | 13.10% | ||
| Other | 247,894 | 16.10% | 8,214 | 16.10% | 239,680 | 16.10% | ||
| Not Hispanic or Latino | 896,073 | 58.20% | 34,938 | 68.30% | 861,135 | 57.90% | <.0001 | |
| Hispanic or Latino | 91,157 | 5.90% | 1,835 | 3.60% | 89,322 | 6.00% | ||
| Other | 551,634 | 35.80% | 14,368 | 28.10% | 537,266 | 36.10% | ||
| Payor Type | Commercial – Indemnity | 68,958 | 4.50% | 2,344 | 4.60% | 66,614 | 4.50% | <.0001 |
| Managed Care | 240,093 | 15.60% | 9,674 | 18.90% | 230,419 | 15.50% | ||
| Medicaid | 150,918 | 9.80% | 5,124 | 10.00% | 145,794 | 9.80% | ||
| Medicare | 934,945 | 60.80% | 29,964 | 58.60% | 904,981 | 60.80% | ||
| Self Pay | 74,868 | 4.90% | 2,095 | 4.10% | 72,773 | 4.90% | ||
| Other | 69,082 | 4.50% | 1,940 | 3.80% | 67,142 | 4.50% | ||
| Charlson Comorbidity Index (CCI) | Mean | 2.74 | 3.03 | 2.73 | <.0001 | |||
| Std Dev | ±2.21 | ±2.17 | ±2.21 | |||||
| Median | 2 | 3 | 2 | |||||
| IQR | 1 | 4 | 1 | 4 | 1 | 4 | ||
| Charlson Comorbidity Index (CCI) Group | 0 | 193,456 | 12.60% | 3,926 | 7.70% | 189,530 | 12.70% | <.0001 |
| 1-5 | 1,181,763 | 76.80% | 40,805 | 79.80% | 1,140,958 | 76.70% | ||
| 6-10 | 152,931 | 9.90% | 6,057 | 11.80% | 146,874 | 9.90% | ||
| 11-15 | 10,549 | 0.70% | 347 | 0.70% | 10,202 | 0.70% | ||
| 16-20 | 164 | 0.00% | 6 | 0.00% | 158 | 0.00% | ||
| 21-25 | 1 | 0.00% | 0 | 0.00% | 1 | 0.00% | ||
| Obesity | 282,219 | 18.30% | 17,427 | 34.10% | 264,792 | 17.80% | <0.001 | |
| Hypertension | 1,123,743 | 73.00% | 39,629 | 77.50% | 1,084,114 | 72.90% | <0.001 | |
| Diabetes Mellitus | 562,881 | 36.60% | 24,169 | 47.30% | 538,712 | 36.20% | <0.001 | |
| Hyperlipidemia | 704,867 | 45.80% | 27,231 | 53.20% | 677,636 | 45.50% | <0.001 | |
| Heart Failure | 558,834 | 36.30% | 23,859 | 46.70% | 534,975 | 36.00% | <0.001 | |
| Angina Pectoris | 16,336 | 1.10% | 553 | 1.10% | 15,783 | 1.10% | 0.6574 | |
| Prior MI | 146,260 | 9.50% | 7,154 | 14.00% | 139,106 | 9.40% | <0.001 | |
| APR DRG Severity of Illness | Minor | 109,278 | 7.10% | 2,944 | 5.80% | 106,334 | 7.10% | <.0001 |
| Moderate | 319,721 | 20.80% | 9,312 | 18.20% | 310,409 | 20.90% | ||
| Major | 525,672 | 34.20% | 17,131 | 33.50% | 508,541 | 34.20% | ||
| Extreme | 584,188 | 38.00% | 21,754 | 42.50% | 562,434 | 37.80% | ||
| Undefined | 5 | 0.00% | 0 | 0.00% | 5 | 0.00% | ||
| APR DRG Risk of Mortality | Minor | 223,468 | 14.50% | 5,563 | 10.90% | 217,905 | 14.60% | <.0001 |
| Moderate | 327,803 | 21.30% | 9,674 | 18.90% | 318,129 | 21.40% | ||
| Major | 469,518 | 30.50% | 15,944 | 31.20% | 453,574 | 30.50% | ||
| Extreme | 518,070 | 33.70% | 19,960 | 39.00% | 498,110 | 33.50% | ||
| Undefined | 5 | 0.00% | 0 | 0.00% | 5 | 0.00% | ||
| Admission Type | Emergency | 1,123,805 | 73.00% | 32,447 | 63.40% | 1,091,358 | 73.40% | <.0001 |
| Urgent | 237,847 | 15.50% | 11,550 | 22.60% | 226,297 | 15.20% | ||
| Elective | 152,997 | 9.90% | 5,614 | 11.00% | 147,383 | 9.90% | ||
| Trauma | 14,698 | 1.00% | 859 | 1.70% | 13,839 | 0.90% | ||
| Other/Unknown | 9,517 | 0.60% | 671 | 1.30% | 8,846 | 0.60% | ||
| Discharge Status | Home | 939,434 | 61.00% | 30,785 | 60.20% | 908,649 | 61.10% | <.0001 |
| Transferred | 218,943 | 14.20% | 7,660 | 15.00% | 211,283 | 14.20% | ||
| SNF | 290,581 | 18.90% | 9,905 | 19.40% | 280,676 | 18.90% | ||
| Hospice | 64,649 | 4.20% | 1,749 | 3.40% | 62,900 | 4.20% | ||
| Other/Unknown | 25,257 | 1.60% | 1,042 | 2.00% | 24,215 | 1.60% | ||
| Physician Specialty | Cardiovascular Diseases (CD) | 188,236 | 12.20% | 9,924 | 19.40% | 178,312 | 12.00% | <.0001 |
| Family Practice (FP) | 118,784 | 7.70% | 2,462 | 4.80% | 116,322 | 7.80% | ||
| General Surgery (GS) | 44,934 | 2.90% | 1,548 | 3.00% | 43,386 | 2.90% | ||
| Hospitalist (HOS) | 324,170 | 21.10% | 12,137 | 23.70% | 312,033 | 21.00% | ||
| Internal Medicine (IM) | 518,881 | 33.70% | 13,131 | 25.70% | 505,750 | 34.00% | ||
| Other | 343,859 | 22.30% | 11,939 | 23.30% | 331,920 | 22.30% | ||
| Year of ICU Encounter | 2009 | 157,053 | 10.20% | 1,517 | 3.00% | 155,536 | 10.50% | <.0001 |
| 2010 | 190,803 | 12.40% | 3,631 | 7.10% | 187,172 | 12.60% | ||
| 2011 | 230,022 | 14.90% | 5,857 | 11.50% | 224,165 | 15.10% | ||
| 2012 | 253,888 | 16.50% | 5,830 | 11.40% | 248,058 | 16.70% | ||
| 2013 | 253,054 | 16.40% | 9,731 | 19.00% | 243,323 | 16.40% | ||
| 2014 | 257,072 | 16.70% | 12,530 | 24.50% | 244,542 | 16.40% | ||
| 2015 | 196,972 | 12.80% | 12,045 | 23.60% | 184,927 | 12.40% | ||
| ICU Type [1] | Stepdown | 223,309 | 14.50% | 4,057 | 7.90% | 219,252 | 14.70% | <.0001 |
| Coronary Care Unit | 282,239 | 18.30% | 13,793 | 27.00% | 268,446 | 18.00% | ||
| Medical ICU | 83,866 | 5.40% | 2,416 | 4.70% | 81,450 | 5.50% | ||
| Surgical ICU | 47,480 | 3.10% | 1,533 | 3.00% | 45,947 | 3.10% | ||
| Other/Unspecified | 901,970 | 58.60% | 29,342 | 57.40% | 872,628 | 58.70% | ||
| ICU Type Stepdown | Yes | 223,309 | 14.50% | 4,057 | 7.90% | 219,252 | 14.70% | <0.001 |
| ICU Type CCU | Yes | 433,698 | 28.20% | 16,306 | 31.90% | 417,392 | 28.10% | <0.001 |
| ICU Type MCU | Yes | 136,443 | 8.90% | 3,664 | 7.20% | 132,779 | 8.90% | <0.001 |
| ICU Type SCU | Yes | 48,106 | 3.10% | 1,638 | 3.20% | 46,468 | 3.10% | 0.3099 |
| Bed Size | 1-149 | 148,433 | 9.60% | 2,959 | 5.80% | 145,474 | 9.80% | <.0001 |
| 150-249 | 237,466 | 15.40% | 4,639 | 9.10% | 232,827 | 15.60% | ||
| 250-349 | 269,067 | 17.50% | 10,156 | 19.90% | 258,911 | 17.40% | ||
| 350-449 | 247,120 | 16.10% | 7,953 | 15.60% | 239,167 | 16.10% | ||
| 450-549 | 206,527 | 13.40% | 9,562 | 18.70% | 196,965 | 13.20% | ||
| 550+ | 430,251 | 28.00% | 15,872 | 31.00% | 414,379 | 27.90% | ||
| Region | East North Central | 223,178 | 14.50% | 15,314 | 29.90% | 207,864 | 14.00% | <.0001 |
| East South Central | 101,398 | 6.60% | 5,858 | 11.50% | 95,540 | 6.40% | ||
| Middle Atlantic | 178,899 | 11.60% | 9,739 | 19.00% | 169,160 | 11.40% | ||
| Mountain | 54,004 | 3.50% | 643 | 1.30% | 53,361 | 3.60% | ||
| New England | 35,474 | 2.30% | 1,728 | 3.40% | 33,746 | 2.30% | ||
| Pacific | 202,816 | 13.20% | 2,718 | 5.30% | 200,098 | 13.40% | ||
| South Atlantic | 471,169 | 30.60% | 6,708 | 13.10% | 464,461 | 31.20% | ||
| West North Central | 90,924 | 5.90% | 5,062 | 9.90% | 85,862 | 5.80% | ||
| West South Central | 181,002 | 11.80% | 3,371 | 6.60% | 177,631 | 11.90% | ||
| Teaching Status | Non-Teaching | 868,699 | 56.50% | 19,069 | 37.30% | 849,630 | 57.10% | <.0001 |
| Teaching | 670,165 | 43.50% | 32,072 | 62.70% | 638,093 | 42.90% | ||
| Urbanicity | Rural | 173,829 | 11.30% | 4,909 | 9.60% | 168,920 | 11.40% | <.0001 |
| Urban | 1,365,035 | 88.70% | 46,232 | 90.40% | 1,318,803 | 88.60% | ||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
