Effect of skin temperature on multifrequency bioelectrical impedance analysis

Rani Gudivaka, Dale Schoeller, Robert F. Kushner*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

86 Scopus citations


This study assessed the effects of changes in skin temperature on multifrequency bioimpedance analysis (MF-BIA) and on the prediction of body water compartments. Skin temperature (baseline 29.3 ± 2.1°C) of six healthy adults was raised over 50 min to 35.8 ± 0.6°C, followed by cooling for 20 min to 26.9 ± 1.3°C, by using an external heating and cooling blanket. MF- BIA was measured at both distal (conventional) and proximal electrode placements. Both distal and proximal impedance varied inversely with a change in skin temperature across all frequencies (5-500 kHz). The change in proximal impedance per degree centigrade change in skin surface temperature was ~60% of distal impedance. The change in measured impedance at 50 kHz erroneously increased predicted total body water (TBW) by 2.6 ± 0.9 liters (P < 0.001) and underpredicted fat mass by 3.3 ± 1.3 kg (P < 0.0001). Computer modeling of the MF-BIA data indicated changes in predicted water compartments with temperature modifications; however, the ratio of extracellular water (ECW) to TBW did not significantly change (P < 0.4). This change in impedance was not due to a change in the movement of water of the ECW compartment and thus probably represents a change in cutaneous impedance of the skin. Controlled ambient and skin temperatures should be included in the standardization of BIA measurements. The error in predicted TBW is < 1% within an ambient temperature range of 22.3 to 27.7°C (72.1-81.9°F).

Original languageEnglish (US)
Pages (from-to)838-845
Number of pages8
JournalJournal of Applied Physiology
Issue number2
StatePublished - Jan 1 1996


  • extracellular water
  • total body water

ASJC Scopus subject areas

  • Physiology
  • Physiology (medical)


Dive into the research topics of 'Effect of skin temperature on multifrequency bioelectrical impedance analysis'. Together they form a unique fingerprint.

Cite this