3.7 Prediction of RR from Neonatal cardiovascular Parameters in Preterm Neonates using multiple linear regression
In predicting or estimating RR from neonatal cardiovascular measurements in pre-term neonates, a multiple linear regression analyses were carried out and the report is shown in (Table 3). Notably, It was observed that the significant correlation (r=3.013, p=0.0027; p<0.05) was obtained when neonatal MAP were entered in multiple linear regression analysis. Collectively, on multiple linear regression analysis, MAP had significant correlation with RR (p<0.05), whereas SBP and DBP had no significant correlation with RR (p>0.05). Therefore, the most important significant correlation was obtained from MAP (MAP=3.013, p=0.0027; p<0.05). Therefore, the best regression model to predict RR of neonate was=- RR (BPM)= 18.21 +[0.011×MAP] (r=3.013, p=0.0027; p<0.05).
Table 3
Multiple Regression that Estimates RR from SBP, DBP and MAP in Preterm Neonates, Given Body Weight
RR
|
Reg. Eqn.
|
SEM
|
t-stat
|
p-value
|
R2
|
Remark
|
|
Y = b0 + b1x1 + b2x2
|
|
|
|
|
|
Y
|
37.65 + (0.006) * SBP + (-1.09) * BW
|
3.8098
|
9.882
|
5.8577
|
0.0201
|
Insignificant
|
Y
|
37.43 + (0.011) * DBP + (-1.03) * BW
|
5.7466
|
6.514
|
5.9685
|
0.0200
|
Insignificant
|
Y
|
18.21 + (0.011) * MAP + (-0.06) * BW
|
3.1084
|
3.013
|
0.0027
|
0.0011
|
Significant
|
SBP = Systolic Blood Pressure, DBP = Diastolic Blood Pressure, MAP = Mean Arterial Pressure and RR = Respiratory rate Y = Predicted Neonatal Serum Variable (SBP, DBP or MAP), b 0 = regression intercept (constant), b1 = regression coefficient for predicted marker (Constant), b2 = regression coefficient for Pre-term neonate’s body weight (Constant), x1 = known levels of assayed marker [SBP, DBP or MAP], and x2 = Known Neonate’s Body Weight (Kg).
3.8 Prediction of RR from Neonatal cardiovascular Parameters in Term Neonates using multiple linear regression
In predicting or estimating RR from neonatal cardiovascular measurements in pre-term neonates, a multiple linear regression analyses were carried out. Notably, It was observed that the significant correlation (r=7.021, p=0.0500; p<0.05) was obtained when neonatal SBP were entered in multiple linear regression analysis. Collectively, on multiple linear regression analysis, SBP had significant correlation with RR (p<0.05), whereas DBP and MAP had no significant correlation with RR (p>0.05). Therefore, the most important significant correlation was obtained from SBP (SBP= r=7.021, p=0.0500; p<0.05). Interestingly, the best regression model to predict RR of term neonate was=- RR (BPM)= 41.05 +[0.012× SBP] (r=7.021, p=0.0500; p<0.05) (Table 4).
Table 4
Multiple Regression that Estimates RR from SBP, DBP and MAP in Term Neonates, Given Body Weight
RR
|
Reg. Eqn.
|
SEM
|
t-stat
|
p-value
|
R2
|
Remark
|
|
Y = b0 + b1x1 + b2x2
|
|
|
|
|
|
Y
|
41.05 + (0.012) * SBP + (-1.10) * BW
|
4.1016
|
7.021
|
0.0500
|
0.1121
|
Significant
|
Y
|
38.41 + (0.001) * DBP + (-1.11) * BW
|
5.0117
|
6.120
|
0.9001
|
0.3252
|
Insignificant
|
Y
|
33.10 + (0.000) * MAP + (-1.00) * BW
|
3.2324
|
4.214
|
0.1002
|
0.0144
|
Insignificant
|
SBP = Systolic Blood Pressure, DBP = Diastolic Blood Pressure, MAP = Mean Arterial Pressure and RR = Respiratory rate Y = Predicted Neonatal Serum Variable (SBP, DBP or MAP), b 0 = regression intercept (constant), b1 = regression coefficient for predicted marker (Constant), b2 = regression coefficient for Pre-term neonate’s body weight (Constant), x1 = known levels of assayed marker [SBP, DBP or MAP], and x2 = Known Neonate’s Body Weight (Kg).
3.9 Prediction of OSL from Neonatal cardiovascular Parameters in Term and Pre-term Neonates using multiple linear regression
In predicting or estimating OSL from neonatal cardiovascular measurements in pre-term neonates, a multiple linear regression analyses were carried out. However, on multiple linear regression analysis, SBP, DBP and MAP had no significant correlation with OSL (p>0.05). Moreso, no regression model was noted to predict OSL of term and pre-term neonate (Table 5 and 6).
Table 5
Multiple Regression that Estimates OSL from SBP, DBP and MAP in Pre-Term Neonates, Given Body
OSL
|
Reg. Eqn.
|
SEM
|
t-stat
|
p-value
|
R2
|
Remark
|
|
Y = b0 + b1x1 + b2x2
|
|
|
|
|
|
Y
|
87.19 + (-0.015) * SBP + (0.35) * BW
|
3.1253
|
12.243
|
1.2172
|
0.1517
|
Insignificant
|
Y
|
92.28 + (-0.038) * DBP + (0.62) * BW
|
4.0434
|
22.823
|
3.2011
|
0.0162
|
Insignificant
|
Y
|
61.12 + (-0.325) * MAP + (0.23) * BW
|
3.3223
|
13.253
|
1.2143
|
0.2352
|
Insignificant
|
Table 6
Multiple Regression that Estimates OSL from SBP, DBP and MAP in Term Neonates, Given Body Weight
OSL
|
Reg. Eqn.
|
SEM
|
t-stat
|
p-value
|
R2
|
Remark
|
|
Y = b0 + b1x1 + b2x2
|
|
|
|
|
|
Y
|
83.52 + (0.006) * SBP + (0.13) * BW
|
4.1232
|
11.224
|
0.2123
|
1.1321
|
Insignificant
|
Y
|
82.28 + (0.013) * DBP + (0.12) * BW
|
3.1221
|
10.122
|
0.4001
|
0.2262
|
Insignificant
|
Y
|
62.12 + (0.001) * MAP + (0.21) * BW
|
3.1224
|
9.232
|
0.2040
|
1.2522
|
Insignificant
|
SBP = Systolic Blood Pressure, DBP = Diastolic Blood Pressure, MAP = Mean Arterial Pressure and OSL = Oxygen Saturation Level; Y = Predicted Neonatal Serum Variable (SBP, DBP or MAP), b 0 = regression intercept (constant), b1 = regression coefficient for predicted marker (Constant), b2 = regression coefficient for Pre-term neonate’s body weight (Constant), x1 = known levels of assayed marker [SBP, DBP or MAP], and x2 = Known Neonate’s Body Weight (Kg). .
3.10 Prediction of OSL and RR from Neonatal pulmonary Parameters in Pre-Term Neonates using multiple linear regression
In predicting or estimating RR from neonatal pulmonary measurements in pre-term neonates, a multiple linear regression analyses were carried out. It was observed that the significant correlation (r=4.1760, p=0.0052; p<0.05) was obtained when neonatal OSL were entered in multiple linear regression analysis. Accordingly, on multiple linear regression analysis, OSL had significant correlation with RR (p<0.05), whereas RR had no significant correlation with OSL (p>0.05). Therefore, the most important significant correlation was obtained from OSL (OSL= r=4.1760, p=0.0052; p<0.05). Interestingly, the best regression model to predict RR of pre-term neonate was=- RR (BPM)= 99.44 +[0.018× OSL] (r=4.1760, p=0.0052; p<0.05) (Table 7).
Table 7
Multiple Regression that Estimates OSL and RR in Pre-Term Neonates, Given Body Weight
OSL or RR
|
Reg. Eqn.
|
SEM
|
t-stat
|
p-value
|
R2
|
Remark
|
|
Y = b0 + b1x1 + b2x2
|
|
|
|
|
|
Y (RR)
|
99.44 + (0.018) * OSL + (4.00) * BW
|
23.8130
|
4.1760
|
0.0052
|
0.0396
|
Significant
|
Y (OSL)
|
102.46 + (-0.038) * RR + (3.97) * BW
|
8.5161
|
12.032
|
2.7922
|
0.0398
|
Insignificant
|
Pressure and OSL = Oxygen Saturation Level, RR = Respiratory Rate; Y = Predicted Neonatal Serum Variable (OSL or RR), b 0 = regression intercept (constant), b1 = regression coefficient for predicted marker (Constant), b2 = regression coefficient for Pre-term neonate’s body weight (Constant), x1 = known levels of assayed marker [OSL or RR], and x2 = Known Neonate’s Body Weight (Kg).
3.11 Prediction of OSL and RR from Neonatal pulmonary Parameters in Term Neonates using multiple linear regression
In predicting or estimating RR from neonatal pulmonary measurements in term neonates, a multiple linear regression analyses were carried out. It was observed that the significant correlation (r=10.2529, p=00.0022; p<0.05) was obtained when neonatal OSL were entered in multiple linear regression analysis. Accordingly, on multiple linear regression analysis, OSL had significant correlation with RR (p<0.05), whereas RR had no significant correlation with OSL (p>0.05). Therefore, the most important significant correlation was obtained from OSL (OSL= r=10.2529, p=00.0022; p<0.05). Interestingly, the best regression model to predict RR of term neonate was=- RR (BPM)= 11046+[0.023× OSL] (r=10.2529, p=00.0022; p<0.05) (Table 8).
Table 8
Multiple Regression that Estimates OSL and RR in Term Neonates, Given Body Weight
OSL or RR
|
Reg. Eqn.
|
SEM
|
t-stat
|
p-value
|
R2
|
Remark
|
|
Y = b0 + b1x1 + b2x2
|
|
|
|
|
|
Y(OSL)
|
94.21 + (0.021) * RR + (3.01) * BW
|
18.3322
|
3.1244
|
0.1328
|
1.0103
|
Insignificant
|
Y(RR)
|
11046 + (0.023) * OSL + (1.01) * BW
|
10.2529
|
2.0125
|
0.0022
|
1.2633
|
Significant
|
Pressure and OSL = Oxygen Saturation Level, RR = Respiratory Rate; Y = Predicted Neonatal Serum Variable (OSL or RR), b 0 = regression intercept (constant), b1 = regression coefficient for predicted marker (Constant), b2 = regression coefficient for Pre-term neonate’s body weight (Constant), x1 = known levels of assayed marker [OSL or RR], and x2 = Known Neonate’s Body Weight (Kg).
3.12 Predictive Accuracy Measure of the Regression Model
Based on the predictive biomarkers of the regression model, the individual parameters (e.g. MAP and OSL) as indicated in model 1 had better predictive potential for pre-term neonate survival whereas the best predictor for term neonate are SBP and OSL (Model 2) as statistically correlated with RR (Table 9).
Table 9
Regression Model Predictive Accuracy Measurements for neonatal survival
MODEL 1:
Regression model for pre-term neonates
|
RR (BPM) = 18.21 + [0.011 × MAP] + (-0.06) x BW
RR (BPM) = 99.44 + [0.018 × OSL] + (4.00) x BW
|
MODEL 2:
Regression model for term neonates
|
RR (BPM) = 41.05 + [0.012 × SBP] + (1.10) x BW
RR (BPM) = 110.46 + [0.023 × OSL] + (1.01) x BW
|