Mtc. Liang et S. Norris, EFFECTS OF SKIN BLOOD-FLOW AND TEMPERATURE ON BIOELECTRIC IMPEDANCE AFTER EXERCISE, Medicine and science in sports and exercise, 25(11), 1993, pp. 1231-1239
To examine the effects of increased skin blood flow (BF(sk)) and skin
temperature (T(sk)) on bioelectric impedance (BIA), 30 young males par
ticipated as subjects. All subjects underwent the following measuremen
ts: 1) BF(sk) and T(sk) for the chest, biceps, thigh, and calf; 2) bod
y composition using the BIA prediction equations and underwater weighi
ng (UW); and 3) triceps skinfold (SF(tri)) thickness. After the baseli
ne studies (M1) the subjects exercised for 30 min at approximately 83%
of maximal heart rate. The above measurements were repeated immediate
ly after exercise (M2), and at 1-h recovery from exercise (M3). Repeat
ed measures ANOVA showed that mean T(sk) and BF(sk), and for the four
measured sites were significantly increased from M1 to M2 (P < 0.05) a
nd decreased from M2 to M3 (P < 0.05). These changes did not affect BI
A measurement for resistance (R) and reactance between M1 and M2 (P >
0.05), and between M2 and M3 (P > 0.05). With regression analysis the
following independent variables were identified as strong contributing
factors for predicting fat-free mass (FFM): 1) SF(tri) and total body
water for M1 and M3; and 2) WT, SF(tri), and T(sk) of the chest for M
2. Two FFM equations have little multicolinearity (M1 and M2), e.g., t
hey have low root mean square errors (less-than-or-equal-to 2.6 kg), v
ery high values for R2 (greater-than-or-equal-to 0.94), and relatively
low values for coefficient of variations (less-than-or-equal-to 5.65%
). Differences (P < 0.05) were observed between the UW method and the
Lukaski equation for estimating FFM for M1 and M3, but not for M2. Usi
ng the Guo equation for estimating percent body fat (%BF), statistical
difference (P < 0.05) was observed for M2 only. We conclude that: 1)
exercise-induced changes in BF(sk) and T(sk) do not influence the BIA
study; and 2) after exercise the following independent variables were
strong predictors for estimating FFM: S2/R, SF(tri), weight, and T(sk)
of the chest.