Table 2.

Trend analysis of blood gas values over time during the 5% CO₂ inhalation.

Outcome	Variables	Regression coefficient (β)	95% CI	p values
Base deficit	Intercept	0.64	−2.15; 3.43	<0.0001
	Baseline	0.74	0.47; 0.99	0.001
	Time elapse (h)	−0.61	−1.11; −0.11	0.045
	Thompson encephalopathy score	1.40	0.28; 2.53	0.021
Lactate	Intercept	1.14	−2.14; 4.42	<0.0001
	Baseline	0.58	0.21; 0.94	0.010
	Time elapse (h)	−0.55	−0.84; −0.26	0.001
	Thompson encephalopathy score	0.88	−0.49; −2.24	0.172
pH	Intercept	2.73	−0.04; 5.51	<0.0001
	Baseline	0.62	0.18; 1.07	0.016
	Time elapse (h)	−0.00	−0.01; 0.00	0.260
	Thompson encephalopathy score	−0.03	−0.06; 0.01	0.135

A repeated-measures linear mixed-effect model was run for trend analysis of base deficit, lactate, and pH during the 5% CO₂ inhalation. The baseline corresponds to the last measured value prior to the start of CO₂ inhalation. The time in hours since the start of CO₂ inhalation baseline blood gas values and Thompson encephalopathy score (low as 0, medium as 1, high as 2) were included as variables with fixed effects. See text for details

Mathematical modeling yielded the following equations:

Base deficit (mmol/L) = 0.64 + (Baseline × 0.74) − (Time elapse in hours × 0.61) + (Thompson encephalopathy score × 1.40)

Lactate (mmol/L) = 1.14 + (Baseline × 0.58) − (Time elapse in hours × 0.55) + (Thompson encephalopathy score × 0.88)

pH = 2.73 + (Baseline × 0.62) − (Time elapse in hours × 0.00) − (Thompson encephalopathy score × 0.03)