A study was conducted to assess the effect of changes in sample pH and
pO2 on the Accutrend, ExacTech Companion, Glucometer II, One Touch II
, and Refloflux II blood glucose meters. Venous blood was tonometered
after the addition of strong acid or alkali to produce 10 samples with
a pH range of 6.54 to 7.73 but with a similar mean pO2 (9.09 +/- 0.09
6 (SEM) kPa) and PCO2 (5.52 +/- 0.024 kPa). The ExacTech showed a > 15
% deviation from the value obtained at pH 7.40 below pH 6.95 and abov
e pH 7.85 (r = 0.69, p = 0.026). Ten further samples were made with a
pO2 range of 2.0 to 33.6 kPa but mean pH (7.28 +/- 0.012) and pCO2 (7.
98 +/- 0.25 kPa) kept alike. The ExacTech was accurate at pO2 = 11.5 k
Pa but deviated by > 15 % below pO2 = 5.8 kPa and above 22.9 kPa (r =
-0.95, p = 0.00002). The Accutrend and One Touch II required a pO2 cha
nge of 42 and 45 kPa to achieve the same error (r = -0.82 and r = -0.6
7 respectively). Thus caution is required in the interpretation of Exa
cTech glucose measurements in severely acidotic or hypoxic patients.