This study hypothesized that there is no difference between energy expenditure measured by indirect calorimetry (IC) and that estimated by predicted formulas compared with the actual intake of children with spastic cerebral palsy (CP). Fifteen children aged 3 to 18 years with spastic CP and associated complications were recruited. IC was used to measure mean energy expenditure (MEE) compared with 3 predicted equations for energy expenditure (PEE), including body surface area (BSA), the recommended daily allowance (RDA), and an equation designed specifically for patients with CP. Friedman and paired t tests were used to examine the variance between PEE and MEE. Intraclass correlation coefficient (ICC) was used to explore the correlation between MEE and PEE. The pretest and posttest core temperatures were compared using the Wilcoxon signed rank test. Mean ± standard deviation MEE was 800.5 ± 295.7 kcal/d; BSA was 1,213.4 ± 171.2 kcal/d; RDA was 1,928.1 ± 341.0 kcal/d; and CP was 1,603.1 ± 215.8 kcal/d. The actual diet intake provided 935.3 ± 222.9 kcal/d. Post hoc analysis revealed a significant difference between mean MEE and PEE (P < .001) but not mean actual intake (P = .128). In addition, the ICC of MEE vs PEE was 0.635 at a 95% confidence interval, indicating a weak correlation. In addition, mean pretest body temperature was 36.4°C ± 1°C, and mean posttest body temperature was 35.8°C ± 2°C. The study showed that MEE was significantly different from PEE, but not from actual intake. This warrants further exploration to develop a population-specific PEE for children with spastic CP.