Fat percentage (PG) is a measure that defines the proportion of fat that exists in the body in reference to body weight. It has been studied in the field of sports since it provides information regarding the athlete's performance and health. Regarding sports performance, PG is a measure that is related to better cardiorespiratory capacity and strength, which directly impacts the athlete's performance. In terms of health, an increase in PG leads to significant metabolic changes that generate physiological transformations that trigger, in the long term, non-communicable diseases. In Paralympic sport, people who have lower limb amputations have a greater risk of presenting alterations in their body composition. There are different methods that estimate body composition accurately. The gold standard is Dual Energy X-ray Absorptiometry (DEXA), but it is high cost and difficult to access. For this reason, accessible methods have been developed such as anthropometry, which, through different formulas, accurately estimates the PG in athletes. However, this method assumes a symmetrical distribution in the individual's body, making it a non-objective method for people with amputation. Two authors derived formulas for athletes with unilateral amputation, and reported good predictive performance compared to DEXA. However, a formula for the Colombian population that determines PG in athletes with unilateral amputation and is compared with the gold standard DEXA method has not been studied. Aim. Determine the validity of the formulas reported to determine the fat percentage in athletes with unilateral lower limb amputation in Colombia. Methodology. Observational, analytical, cross-sectional study. Results: All formulas significantly underestimated PG compared to DEXA. The formulas that were closest to the DEXA result were Durnnin and Womersley (difference = 3.86%, p value <0.001). When deriving the anthropometric formula, the variables of age, sex, BMI adjusted by the Amputee Coalition, wingspan and sum of three skinfolds (biceps, abdomen, thigh) were entered as potential predictors in the step-by-step multiple linear regression analysis. . A statistically significant model was obtained by comparing the formula designed with the DEXA method (R 2=0.85, p value <0.001). Conclusion: the anthropometric equations designed for the non-amputated population are not accurate to be applied in athletes with unilateral lower limb amputation to determine GP. However, regarding the formulas validated for the conventional population, the Durnnin and Womersley formula is the one that most closely resembles the results of the Gold standard. As a solution to this problem, the first approach was carried out in which a formula was derived that allows determining the PG, taking into account as fundamental variables the skin folds (biceps, abdomen and thigh), BMI adjusted by the Amputee Coalition and the wingspan. . However, cross-validation of our formula needs to be carried out so that health professionals can use it.
