The accuracy of 3D image navigation with a cutaneously fixed dynamic reference frame in minimally invasive transforaminal lumbar interbody fusion

Figures & data

Figure 1. (a) DRF attached to the sacral hiatus with sterile tape. (b) The DRF is positioned out of the way of the surgeon.

Table I.  Summary of demographic data.

No. of patients	20
Gender (female/male)	11/9
Mean age (years)	55.9 (range: 34–74)
Mean BMI	24.4 (range: 19.9–28.6)
Level of fusion:
1 level (4 screws)	19
2 levels (6 screws)	1
Total No. of screws	82

Figure 2. (a) Recorded navigation image and measured SAA and SAS of virtual screw location. (b) Second cb-CT scan image after screw insertion, and measured SAA of actual screw location

Table II.  Incidence of screw breach according to fusion level.

Level	Grade 0	Grade I	Grade II	Grade III	Total
L2	1	0	1	0	2
L3	10	1	0	1	12
L4	22	2	0	2	26
L5	30	0	0	0	30
S1	11	1	0	0	12
Total	74	4	1	3	82

Table III.  Intraclass correlation coefficient (ICC) values in the analysis of the virtual/actual SAA and SAS validity.

	ICC value	95% CI	p-value
Virtual SAA	0.620	0.429 ∼ 0.758	<0.001
Actual SAA
Virtual SAS	0.691	0.524 ∼ 0.806	<0.001
Actual SAS

Notes: SAA = screw axis angle in axial plane. SAS =screw axis angle in sagittal plane. Actual =in O-arm scan; Virtual =in navigation.