Foot & Ankle International - September 2019 - 1073

1073

Daniels et al	
Table 2.  Demographic and Clinical Factors for the Treatment Cohorts After Subclassification Using Propensity Scores.a
Variable

rhPDGF-BB/β-TCPCollagen (n = 132)

Autograft
(n = 167)

P Valueb

70 (53.0)
62 (47.0)
53.5 ± 14.7
35 (26.5)
97 (73.5)

86 (51.5)
81 (48.5)
56.2 ± 14.0
58 (34.7)
109 (65.3)

.816
 
 
 
.133
 

58 (43.9)
60 (45.5)
8 (6.1)
0
4 (3.0)
2 (1.5)

68 (40.7)
65 (38.9)
12 (7.2)
1 (0.6)
6 (3.6)
15 (9.0)

31 (23.5)
52 (39.4)
3 (2.3)
6 (4.6)
21 (15.9)
19 (14.4)
31.2 ± 6.1
63 (47.7)
68 (51.5)

46 (27.5)
59 (35.3)
0
9 (5.4)
17 (10.2)
36 (21.6)
31.4 ± 5.7
99 (59.3)
68 (40.7)

71 (53.8)
61 (46.2)

84 (50.3)
83 (49.7)

.103
 
 
 
 
 
.131
 
 
 
 
 
 
 
.061
 
.562
 
 

120 (90.9)
12 (9.1)
72.0 ± 22.3
49.8 ± 26.3
43.3 ± 17.2
50.6 ± 18.4
30.8 ± 8.4

145 (86.8)
19 (11.4)
69.9 ± 22.0
51.7 ± 26.4
43.6 ± 16.8
50.0 ± 15.1
31.2 ± 8.3

.569
 
 
 
 
 
 

Gender
 Male
 Female
Age, yc
  ≥65 years
  <65 years
Primary diagnosis
  Posttraumatic arthritis or deformity
 Primary/osteoarthritis
  Rheumatoid arthritis
  Ankylosing spondylitis
  Congenital or acquired deformity
 Other
Joints treated
  Ankle fusion
  Subtalar fusion
  Calcaneocuboid fusion
  Talonavicular fusion
  Double fusiond
  Triple arthrodesise
BMI, kg/m2 c,f
  ≥30
  <30
Ever smoked
 Yes
 No
Preoperative clinical outcome measures
  VAS pain weightbearing, >40g
  VAS pain weightbearing, ≤40g
  VAS pain weightbearingc,g
  VAS pain at fusion sitec
  AOFAS Ankle-Hindfoot Scale total scorec
  FFI total scorec
  SF-12 PCSc

Abbreviations: AOFAS, American Orthopaedic Foot & Ankle Society; BMI, body mass index; FFI, Foot Function Index; rhPDGF-BB/β-TCP,
recombinant human platelet-derived growth factor BB homodimer; SF-12 PCS, Short-Form 12 Physical Component Summary Score; VAS, visual
analogue scale.
a
The PS matching technique is designed to match the treatment groups such that P values for between-group comparisons of covariates are large.
Comparisons of covariates that were not part of the PS matching technique or comparisons using different methodologies than what were used in the
PS matching (eg, using binary versions of covariates in place of continuous data) may result in P values that are smaller than those found in the final
PS matching. Nevertheless, this table shows that all preoperative clinical outcome measures examined were similar for both groups. All values are
presented as n (%) or mean ± SD.
b
Fisher's exact test for categorical variables; 2-sample t test for continuous variables.
c
Results are based on generalized linear model with the following factors: treatment (rhPDGF-BB/β-TCP-collagen vs autograft) and propensity score
quintiles.
d
Combination of any two of the following joints: subtalar, talonavicular, and calcaneocuboid.
e
Subtalar, talonavicular, and calcaneocuboid joints.
f
Data missing for 1 patient in the rhPDGF-BB/β-TCP-collagen treatment group.
g
Data missing for 3 patients in the autograft control group.

via analysis of covariance, adjusted for PS strata.
Noninferiority was determined by comparing the appropriate upper or lower confidence limit to the desired

noninferiority threshold. For radiographic outcome rates,
AOFAS scores, and SF-12 scores, comparisons were made
to a lower confidence limit, while for VAS pain and FFI
Foot & Ankle International - September 2019

Table of Contents for the Digital Edition of Foot & Ankle International - September 2019
