Fig 1.
Illustration demonstrating performance of the passive leg raise test.
Table 1.
Demographic data.
Fig 2.
Hemodynamic responses to passive leg raising during hospitalization (pre) and following discharge (post).
HU: head-up (HU values are the average among 3rd~5th minutes); LR: leg raise (LR values are those values coincident with the largest SV); LR5: leg raise at the 5th minute; SR: supine rest (SR values are the average among 3rd~5th minutes). CD: change difference from HU to LR; CR: change ratio from HU to LR = CD/HU). (A) SV(stroke volume) at HU, LR, LR5 and SR; (B) CD of SV; (C) CR of SV. Likewise, (D)~(F) HR: heart rate; (G)~(I) CO: cardiac output; (J)~(L) SVCom: stroke volume compliance; (M)~(O) TPR: total peripheral resistance; (P)~(R) SBP: systolic blood pressure; (S)~(U) DBP: diastolic blood pressure.
Fig 3.
Receiver operating characteristic analysis for exercise intolerance by passive leg raise tests following coronary revascularization.
(A) Receiver operating characteristic curve of the SVLR/body weight during 1st PLRT in predicting poor aerobic fitness at 1st CPET with VAT<3METs. A cut-off value of 1504·10−3 ml/kg resulted in sensitivity of 0.954, specificity of 0.593 and the area under the curve of 0.822 for the measurement. (B) ROC curve of the COLR/body weight during 1st PLRT in predicting poor endurance at 2nd CPET with peak VO2<5METs. A cut-off value of 68.3 mL/min/kg resulted in sensitivity of 0.781, specificity of 0.773, and the AUC of 0.814 for the measurement.
Table 2.
Correlation matrix of multiple variables at 1st PLRT vs. VO2VAT at 1st submaximal CPET.
Table 3.
Area under curve of hemodynamic variables during 1st PLRT and VO2VAT at 1st CPET.
Fig 4.
Receiver operating characteristic curve analysis between CILR during 1st PLR vs. exercise intolerance during 2nd CPET.
ROC curve of the CILR during 1st PLRT in predicting poor aerobic fitness at 2nd CPET with VAT < 5METs is shown. AUC of the measurement is 0.749. A cut-off value of 3.9 liter/min/meter2 resulted in sensitivity of 0.779 and specificity of 0.625. The positive and negative predictive values are 0.78 and 0.61 respectively in the study cohort. CILR: cardiac index during leg raise
Table 4.
Correlation matrix of hemodynamic variables at 1st PLRT vs. peak VO2 at 2nd maximal CPET
Table 5.
Area under curve of hemodynamic variables during 1st PLRT and peak VO2 at 2nd CPET.
Fig 5.
Validity of bio-reactance-based NICOM (non-invasive cardiac output monitor) was verified against echocardiography by biplane Simpson method under apical four-chamber view and long-axis view.
Scatter plot of SV and CO between NICOM and echocardiography were demonstrated in A and C. Agreement using Bland-Altman plot between SV and CO corresponding to NICOM and echocardiography were shown in B and D. Graph B and D plot the respective differences between NICOM and echocardiography (y-axis) against the means of these two methods. The dark solid horizontal lines in each Bland-Altman plot represent average bias whereas the dotted lines stand for average bias ± 1.96 standard deviation (95% upper and lower limit). The average bias and standard deviation are 2.9 and 7.6 ml in B; and 0.29 and 0.54 liter/min in D.