Brain swelling is independent of peripheral plasma cytokine levels in Malawian children with cerebral malaria

Malaria Journal

Table 1 Demographic characteristics of the study participants: univariate and multivariable logistic regression model showing the association between brain swelling and demographic characteristics of the study participants

Variable	Characteristic	No severe swelling	Severe swelling	p-value*	Unadjusted			Adjusted
Variable	Characteristic	No severe swelling	Severe swelling	p-value*	OR	95% CI	p-value**	OR	95% CI	p-value***
Gender	Female (%)	33 (37.5)	55 (62.5)	0.24	0.68	0.38; 1.22	0.195	1.04	0.28; 3.87	0.954
Gender	Male (%)	50 (46.73)	57 (53.27)
Age (months)	Median (IQR)	48 (31; 68)	42 (28.0; 65.5)	0.29	1	0.99; 1.00	0.523		ND
Hours in coma prior to admission	Median (IQR)	2.08 (1.79; 2.48)	2.20 (1.79; 3.04)	0.22	1.25	0.89; 1.75	0.191	1.31	0.64; 2.66	0.462
Parasitaemia (parasites log₁₀/μL)	Median (IQR)	10.83 (8.16; 11.78)	10.64 (7.46; 12.41)	0.66	1.03	0.93; 1.14	0.597	ND	ND

In this analysis brain swelling scores of 3 to 6 were considered to not have severe swelling and scores of 7 and 8 were considered to have severe brain swelling. Variables with p-value ≤ 0.2 in univariate analysis were included in the multivariate model. Group comparisons were done using Mann–Whitney U test except for age, where Fisher’s Exact test was used. Results for parasitaemia were log transformed (log₁₀)
ND not done
* Univariate association
** Simple logistic regression
*** Multivariable logistic regression

ISSN: 1475-2875