The decision to surgically lengthen the hamstring muscle group in a child with spastic cerebral palsy is based, in part, on the range of motion measurements made in the clinic, particularly, the popliteal angle. This measurement is often repeated when the child is under anaesthesia, before surgery. We were interested to know whether a systematic difference exists between the popliteal angles measured under these conditions, and to understand under which condition the measurement is more reliable. The effect of anaesthesia on the measurement of range of motion has not been reported previously. We repeated the measurement of popliteal angle in 15 children with spastic cerebral palsy (30 limbs, mean age: 8 years 3 months), on the evening before surgery when they were conscious, and again when they were under anaesthesia. The means of measurements made under the two conditions were not significantly different (p=0.17, paired t-test). Repeated measurements of popliteal angle when the patient was conscious were, however, less reliable than those made under anaesthesia (standard error of 15 and 5 degrees, respectively). This resulted in poor agreement between measurements made on the same patient under the two conditions (95% limits of agreement were -19 and 25 degrees). Our study supports the common practice of surgeons of reassessing hamstring length under anaesthesia before single level and multilevel surgeries and incorporation of these measurements into their decision-making.