Corrections of Enghoff's dead space formula for shunt effects still overestimate Bohr's dead space.

Dead space ratio is determined using Enghoff's modification (VD(B-E)/VT) of Bohr's formula (VD(Bohr)/VT) in which arterial is used as a surrogate of alveolar PCO₂. In presence of intrapulmonary shunt Enghoff's approach overestimates dead space. In 40 lung-lavaged pigs we evaluated the Kuwabara's and Niklason's algorithms to correct for shunt effects and hypothesized that corrected VD(B-E)/VT should provide similar values as VD(Bohr)/VT. We analyzed 396 volumetric capnograms and arterial and mixed-venous blood samples to calculate VD(Bohr)/VT and VD(B-E)/VT. Thereafter, we corrected the latter for shunt effects using Kuwabara's (K) VD(B-E)/VT and Niklason's (N) VD(B-E)/VT algorithms. Uncorrected VD(B-E)/VT (mean ± SD of 0.70 ± 0.10) overestimated VD(Bohr)/VT (0.59 ± 0.12) (p < 0.05), over the entire range of shunts. Mean (K) VD(B-E)/VT was significantly higher than VD(Bohr)/VT (0.67 ± 0.08, bias -0.085, limits of agreement -0.232 to 0.085; p < 0.05) whereas (N)VD(B-E)/VT showed a better correction for shunt effects (0.64 ± 0.09, bias 0.048, limits of agreement -0.168 to 0.072; p < 0.05). Neither Kuwabara's nor Niklason's algorithms were able to correct Enghoff's dead space formula for shunt effects.