Skip to main content

Table 2 Subgroup analysis stratified by different locations and different steel plate sites

From: The 3-dimensional miniplate is more effective than the standard miniplate for the management of mandibular fractures: a meta-analysis

Indicator Group Sample size Test of association Model Test of heterogeneitya,b
Cases Control RR (95% CI) Z P value Chi2 P value I 2 (%)
Fracture regions
 Infection Angle 127 88 0.56 [0.17, 1.89] 0.93 0.35 Fixed 3.28 0.35 8
Other 212 166 0.88 [0.42, 1.84] 0.33 0.74 Fixed 3.89 0.69 0
 Malocclusion Angle 55 56 0.40 [0.10, 1.60] 1.30 0.20
Other 212 166 0.44 [0.23, 0.83] 2.53 0.01 Fixed 1.91 0.75 0
 Hardware failure Angle 127 88 0.22 [0.07, 0.63] 2.82 0.005 Fixed 0.53 0.97 0
Other 133 87 0.99 [0.16, 5.99] 0.01 0.99 Fixed 1.12 0.29 11
 Nonunion/malunion Angle 117 78 1.33 [0.14, 12.33] 0.25 0.80
Other 117 71 0.65 [0.13, 3.30] 0.52 0.60
 Wound dehiscence Other 108 62 0.38 [0.09, 1.49] 1.39 0.16 Fixed 0.85 0.36 0
Other 108 62 0.38 [0.09, 1.49] 1.39 0.16 Fixed 0.85 0.36 0
 Paresthesia Angle 20 20 2.00 [0.21, 18.69] 0.61 0.54
Other 167 121 0.73 [0.27, 1.99] 0.61 0.54 Fixed 0.42 0.81 0
The size of the steel plate
 Infection 4 or 6 hole 114 114 0.46 [0.18, 1.16] 1.64 0.10 Fixed 1.03 0.91 0
Other 225 140 1.30 [0.53, 3.21] 0.57 0.57 Fixed 4.81 0.44 0
 Malocclusion 4 or 6 hole 114 114 0.46 [0.23, 0.90] 2.26 0.02 Fixed 1.68 0.64 0
Other 135 108 0.37 [0.12, 1.14] 1.74 0.08 Fixed 0.03 0.86 0
 Hardware failure 4 or 6 hole 55 55 0.67 [0.11, 3.89] 0.45 0.65 Fixed 1.49 0.22 33
Other 205 120 0.23 [0.08, 0.66] 2.75 0.006 Fixed 0.60 0.96 0
 Nonunion/malunion 4 or 6 hole 39 39
Other 195 110 0.86 [0.24, 3.17] 0.22 0.82 Fixed 0.27 0.61 0
 Wound dehiscence 4 or 6 hole 50 50 0.14 [0.02, 1.12] 1.85 0.06 Fixed 0.00 1.00 0
Other 175 90 0.65 [0.18, 2.36] 0.65 0.51 Fixed 0.39 0.82 0
 Paresthesia 4 or 6 hole 59 59 0.83 [0.28, 2.48] 0.33 0.74 Fixed 0.02 0.88 0
Other 128 82 0.99 [0.21, 4.79] 0.01 0.99 Fixed 1.03 0.31 2
  1. RCT randomized controlled trial, PCS prospective controlled study, RCS retrospective controlled study
  2. aRandom effects model was used when the P value for heterogeneity test <0.05, otherwise the fixed effects model was used
  3. b P value <0.05 is considered statistically significant for Q statistics