Plasma angiopoietin-2 is associated with age-related deficits in cognitive sub-scales in Ugandan children following severe malaria

Table 3 Association between Angpt-2 level and longitudinal age-adjusted cognitive z-scores over two years follow-up in children with a severe malaria episode at < 5 years of age

	N (obs), N	Unadjusted LME Model¹		Adjusted LME Model^1,2
	N (obs), N	Beta (95% CI)	p	Beta (95% CI)	p	Significant^†
Children < 5 years of age at time of testing (MSEL)
Fine motor	935, 295	− 0.34 (− 0.59, − 0.09)	0.0087	− 0.30 (− 0.53, − 0.06)	0.0131	†
Gross motor	944, 295	− 0.33 (− 0.60, − 0.07)	0.0140	− 0.28 (− 0.55, − 0.02)	0.0338	†
Receptive language	935, 295	− 0.42 (− 0.68, − 0.16)	0.0016	− 0.31 (− 0.56, − 0.06)	0.0149	†
Expressive language	915, 294	− 0.29 (− 0.55, − 0.04)	0.0234	− 0.32 (− 0.56, − 0.07)	0.0120	†
Visual reception	932, 295	− 0.43 (− 0.70, − 0.16)	0.0021	− 0.41 (− 0.67, − 0.15)	0.0021	†
Children ≥ 5 years of age at time of testing (KABC-2)
Planning³	–			–	–
Learning	186, 129	− 0.43 (− 0.76, − 0.10)	0.0104	− 0.39 (− 0.74, − 0.05)	0.0253	†
Sequential processing	186, 129	− 0.38 (− 0.84, 0.07)	0.0992	− 0.27 (− 0.72, 0.17)	0.2302
Simultaneous processing	182, 127	− 0.19 (− 0.48, 0.10)	0.2018	− 0.34 (− 0.61, − 0.08)	0.0101	†

^†p < 0.05 in model after correction for multiple comparisons (n = 5 for children < 5 years of age; n = 3 for children ≥ 5 years of age)
¹Linear mixed effects (LME) model where observations within subject were correlated using a subject-specific intercept, and time points as categorical variables
²Models included age, gender, height-for-age z-score, socio-economic status, home environment, parental education, preschool education of the child (any), presence of coma, number of seizures during admission, and acute kidney injury
³Planning is assessed in children ≥ 7 years of age, so results are not available for children < 5 years of age at the time of the severe malaria episode

ISSN: 1475-2875