TABLE 3.

Optic nerve head vessel density optical coherence tomography angiography (OCTA) adjusted linear mixed model estimates (%).

VD region	All	Disc	Peripapil	SUP Hemi	INF Hemi	NS	NI	IN	IT	TI	TS	ST	SN
Intercept	62.4	67.0	64.3	65.9	62.8	47.2	51.4	61.3	64.8	49.5	70.1	60.0	59.0
Gender			4.6	4.4	4.6	5.7
MAP_0
Age
Diagnosis
POAG (vs. normal)	−12.9		−15.0	−13.6	−15.9	−13.6	−13.1	−17.1	−26.0	−7.8	−9.3	−19.9	−15.7
NTG (vs. normal)	−8.7		−10.0	−8.5	−12.0	−7.7	−7.9	−14.4	−20.7			−14.2	−6.3
NTG (vs. POAG)	4.3		5.0	5.2		5.9	5.1			4.5	6.2		9.4
Time
Gender*time											−3.1
MAP_0*time
MAPdiff*time
Age*time
Diagnosis*time
POAG (vs. normal)
NTG (vs. normal)
NTG (vs. POAG)

Only significant responses (p < 0.05) are reported. Blank cells indicate a non-significant result. The intercept represents the inferred VD for male healthy participants. If significant, the other variables show an additional effect on the intercept VD. E.g., adding the value for gender results in the inferred VD for female healthy participants. The variable time represents the overall effect of the handgrip test. The combination variables refer to how a certain variable influenced the VD change during the handgrip test. LMM variable changes can only be separately considered bearing in mind that the other variables -theoretically- need to remain constant. VD, vessel density; MAP_0, mean arterial pressure at baseline; POAG, primary open angle glaucoma; NTG, normal tension glaucoma; MAPdiff, difference of mean arterial pressure during examination; Peripapil, peripapillary mean; SUP, superior; Hemi, hemisphere. INF, inferior; NS, NI, IN, IT, TI, TS, ST, and SN as explained in the section “2 Materials and methods”.