Fig. 1.
Study Cohort. Details of the inclusion and exclusion criteria are outlined at each step while assembling the study cohort. In total 2,965 patients were included in the final analysis.
Table 1.
Patient baseline, procedural, and encounter characteristics
| Overall N = 2965 | Absence of Inotropes N = 1973 | Presence of Inotropes N = 992 | ASD | |
|---|---|---|---|---|
| Age, years | 67 [59, 74] | 66 [58, 73] | 69 [61, 75] | 0.25 |
| Male | 2166 (73) | 1499 (76) | 667 (67) | 0.20 |
| ASA Classification | 0.36 | |||
| 2 | 56 (2) | 47 (2) | 9 (1) | |
| 3 | 1591 (54) | 1163 (60) | 428 (43) | |
| 4 | 1308 (44) | 758 (38) | 550 (55) | |
| 5 | 4 (0) | 2 (0) | 2 (0) | |
| Missing | 6 (0) | 3 (0) | 3 (0) | |
| Diabetes | 881 (30) | 523 (27) | 358 (36) | 0.21 |
| Hypertension | 2386 (81) | 1554 (79) | 832 (84) | 0.13 |
| Chronic Kidney Disease | 401 (14) | 207 (11) | 194 (20) | 0.26 |
| Acute Renal Replacement Therapy | 54 (2) | 10 (1) | 44 (4) | 0.26 |
| Tracheostomy | 40 (1) | 6 (0) | 34 (3) | 0.23 |
| CABG | 1602 (54) | 1051 (53) | 551 (56) | 0.05 |
| Aortic valve procedure | 1078 (36) | 752 (38) | 326 (33) | 0.11 |
| Mitral valve procedure | 763 (26) | 382 (19) | 381 (38) | 0.43 |
| Tricuspid valve procedure | 160 (5) | 41 (2) | 119 (12) | 0.40 |
| Pulmonic valve procedure | 6 (0) | 3 (0) | 3 (0) | 0.03 |
| CABG and valve procedure | 467 (16) | 252 (13) | 215 (22) | 0.24 |
| Thoracic aortic procedure | 434 (15) | 297 (15) | 137 (14) | 0.04 |
| Other cardiac procedure | 650 (22) | 352 (18) | 298 (30) | 0.29 |
| Redo Sternotomy | 9 (0) | 5 (0) | 4 (0) | 0.03 |
| Cardiopulmonary Bypass Time, minutes | 131 [97, 169] | 121 [93, 154] | 155 [114, 195] | 0.58 |
| LVEF | 62 [55, 67] | 63 [58, 68] | 58 [43, 65] | 0.66 |
| Normal, >= 50% | 2490 (84) | 1822 (92) | 668 (67) | |
| Low, <50% | 475 (16) | 151 (8) | 324 (33) | |
| LVEDD | 49 [44, 54] | 48 [43, 53] | 51 [45, 56] | 0.30 |
| Dilateda | 460 (15) | 236 (12) | 224 (23) | |
| Normala | 2220 (75) | 1523 (77) | 697 (70) | |
| Smalla | 285 (10) | 214 (11) | 71 (7) | |
| 30 Day Morality | 10 (0) | 3 (0) | 7 (1) | 0.09 |
| ICU Length of Stay, hours | 28 [23, 53] | 25 [22, 40] | 55 [30, 110] | 0.55 |
[i] Data is presented as median [quartile 1, quartile 3] and frequency (percentage) depending on variable type. Abbreviations: Absolute standardized difference (ASD), American Society of Anesthesiology (ASA), Coronary artery bypass grafting (CABG), left ventricle ejection fraction (LVEF), left ventricle end-diastolic diameter (LVEDD), Intensive Care Unit (ICU). aLVEDD is categorized as follows: Small (male: < 42; female: < 37), Normal (male: 42 to 58; female: 37 to 52), Dilated (male: > 58; female: > 52)
Table 2.
Association of LVEF and LVEDD with the presence of ICU inotrope therapy
| Odds Ratio (95% CI) | P Value | |
|---|---|---|
| LVEF, unadjusted | 0.50 (0.46 to 0.54) | <0.001 |
| LVEDD, unadjusted | 1.36 (1.26 to 1.47) | <0.001 |
| LVEF, adjusted | 0.45 (0.41 to 0.50) | <0.001 |
| LVEDD, adjusted | 1.18 (1.07 to 1.31) | 0.001 |
[i] Adjusted models include the following prespecified variables: age (centered and scaled), sex, ASA, presence of diabetes, presence of hypertension, presence of chronic kidney disease, cardiopulmonary bypass time (centered and scaled), and presence of redo sternotomy. As LVEF and LVEDD have been centered and scaled, odds ratio are interpreted per unit standard deviation. Abbreviations: left ventricle ejection fraction (LVEF), left ventricle end-diastolic diameter (LVEDD).
Table 3.
Association of LVEF and LVEDD with hours of ICU inotrope therapy
| Effect Estimate (95% CI) | P Value | |
|---|---|---|
| LVEF, unadjusted | 0.88 (0.81 to 0.96) | 0.003 |
| LVEDD, unadjusted | 0.97 (0.89 to 1.05) | 0.41 |
| LVEF, adjusted | 0.75 (0.68 to 0.82) | <0.001 |
| LVEDD, adjusted | 0.96 (0.88 to 1.06) | 0.45 |
[i] Only patients with ICU inotrope administrations were included. Adjusted models include the following prespecified variables: age (centered and scaled), sex, ASA, presence of diabetes, presence of hypertension, presence of chronic kidney disease, cardiopulmonary bypass time (centered and scaled), and presence of redo sternotomy. LVEF and LVEDD were centered and scaled. Effect estimates have been exponentiated and are interpreted as percent changes per unit standard deviation. Abbreviations: left ventricle ejection fraction (LVEF), left ventricle end-diastolic diameter (LVEDD).
Table 4.
Association of LVEF and LVEDD with ICU hours
| Effect Estimate (95% CI) | P Value | |
|---|---|---|
| Multivariable Interaction | ||
| LVEF, unadjusted | 0.88 (0.86 to 0.91) | <0.001 |
| LVEDD, unadjusted | 0.94 (0.92 to 0.97) | <0.001 |
| LVEF * LVEDD, unadjusted | 0.97 (0.94 to 0.99) | 0.01 |
| Multivariable Interaction | ||
| LVEF, adjusted | 0.89 (0.87 to 0.92) | <0.001 |
| LVEDD, adjusted | 0.97 (0.94 to 0.99) | 0.02 |
| LVEF * LV, adjusted* | 0.98 (0.95 to 0.99) | 0.03 |
[i] Adjusted models include the following prespecified variables: age (centered and scaled), sex, ASA, presence of diabetes, presence of hypertension, presence of chronic kidney disease, cardiopulmonary bypass time (centered and scaled), presence of redo sternotomy, acute renal replacement therapy, and presence of tracheostomy. LVEF and LVEDD were centered and scaled. Effect estimates have been exponentiated and are interpreted as percent changes per unit standard deviation. Abbreviations: left ventricle ejection fraction (LVEF), left ventricle end-diastolic diameter (LVEDD).
Fig. 2.
Interaction Between LVEF and LVEDD on ICU Length of Stay. The association between left ventricular end-diastolic diameter (LVEDD) and the natural log (ln) of ICU length of stay is reported for varying categories of left ventricular ejection
Supplemental Table 1:
Sensitivity analysis of inotrope therapy and ICU length of stay outcomes after adjusting for body surface area
| Effect Estimate (95% CI) | P Value | |
|---|---|---|
| Presence of ICU Inotrope Therapy | ||
| LVEF | 0.45 (0.41 to 0.50) | <0.001 |
| LVEDD | 1.19 (1.07 to 1.32) | 0.001 |
| Hours of ICU Inotrope Therapy | ||
| LVEF | 0.75 (0.68 to 0.82) | <0.001 |
| LVEDD | 0.97 (0.88 to 1.06) | 0.47 |
| ICU Hours | ||
| LVEF | 0.89 (0.87 to 0.92) | <0.001 |
| LVEDD | 0.97 (0.94 to 1.00) | 0.03 |
| LVEF * LV | 0.98 (0.95 to 1.00) | 0.03 |