Predictors for Hamstring Graft Diameter in Anterior Cruciate Ligament Reconstruction

Abstract
The ability to accurately predict the diameter of autograft hamstring tendons has implications for graft choice and fixation devices used in anterior cruciate ligament (ACL) reconstruction. To determine whether simple anthropometric measurements such as height, mass, body mass index (BMI), age, and gender can be used to accurately predict the diameter of hamstring tendons for ACL reconstruction surgery. Cohort study (prevalence); Level of evidence, 2. The authors conducted medical record reviews and telephone interviews of 106 consecutive patients with ACL reconstruction using quadrupled semitendinosus-gracilis autograft from 2004 to 2006. Data included anthropometric measurements (height, mass, gender, and age at the time of surgery). Hamstring diameter was obtained using cylindrical sizers in 0.5-mm increments and recorded in the patient's surgical record. Correlation coefficients (Pearson r) and stepwise, multiple linear regression were used to determine the relationship between the outcome variable (hamstring graft diameter) and the predictor variables (age, gender, height, mass, and BMI). Independent sample t tests were used to compare hamstring graft diameter between genders. Hamstring graft diameter was related to height (r = .36, P < .001), mass (r = .25, P = .005), age (r = —.16, P = .05), and gender (r = —.24, P = .006) but was not related to BMI (P > .05). Height was a statistically significant prediction variable (R2 = .13, P < .001). From the current data, a regression equation was calculated that suggested that a patient <147 cm (58 in) tall is likely to have a quadrupled hamstring graft diameter <7 mm in diameter (graft size = 2.4 + 0.03 × height in cm). Women had significantly smaller hamstring graft diameters (7.5 ± 0.7 mm) than did men (7.9 ± 0.9 mm, P = .01). Of the parameters studied, height was the best predictor of hamstring tendon diameter, particularly in women.

This publication has 11 references indexed in Scilit: