Prediction of Callus Subsidence in Distraction Osteogenesis Using Callus Formation Scoring System: Preliminary Study.

Abstract

To develop the scoring system which describes both quality and quantity of callus formation to predict the callus subsidence. Forty-seven bony segments with an average lengthening of 5.17 ± 2.83 cm were included. The score was calculated based on the amount of callus classified in five patterns and the summation with the density of the callus classified in four patterns; the total score was 9. Bony subsidence >10% or >10° angulation were considered significant. We analyzed all of the data to find the most appropriate score that would prevent callus subsidence <10% and prevented angulation of the regeneration bone <10 degrees. Data was analyzed by using the receiver operating characteristic (ROC) curve. An area under the curve of 0.9-1 indicated an excellent test, 0.8-0.9 indicated a good test, 0.7-0.8 indicated a fair test, 0.6-0.7 indicated a poor test, and 0.5-0.6 indicated a fail test. The appropriate score for Ilizarov removal was selected from the highest sensitivity and specificity. Twenty-two tibia segments and 25 femur segments were included. The mean of bone lengthening was 5.17 ± 2.83 cm (range, 1.6-13.5 cm) and the mean of percentage lengthening was 16.58% ± 10.03% (range, 4.63%-56.84%). The mean distraction period was 5 months. The average months of follow-up for measurement of bony subsidence was 4.2 months. Mean subsidence was 21.06% (1.54%-57.44%). The mean of callus subsidence was 1.29 ± 1.17 cm (range, 0.03-4.72 cm). There were 32 segments (68%) with callus subsidence greater than 10% and 15 segments (32%) with subsidence less than 10%. The callus subsidence ranged from 0.3 mm to 4.72 cm, with 68% of bony fragments having significant subsidence. Type 5 callus diameter was statistically significant (P < 0.0001) in preventing callus subsidence compared to the other types. Type 4 callus density was statistically significant in preventing callus subsidence compared to the other types (P < 0.0001). The ROC curve with area under the curve 0.961 and sensitivity 0.933 showed that a callus scoring system score >7.5 was effective in preventing significant callus subsidence. When using score 8 as a result from the ROC curve, 73.3% of bony fragment subsidence was <10% with sensitivity 93.3 and specificity 83.2. Callus diameter 81%-100% and callus density type 4 could prevent significant callus subsidence. Based on the results of the present study we suggest using callus score > 8 to determine the time of Ilizarov removal.