Error analysis of combined measurements of regional ventilation and V/Q ratio using positron emission tomography

We measured the regional pulmonary ventilation/perfusion ratio with intravenous 13N in saline solution, using positron emission tomography, in eight normal healthy volunteers, 10 patients with asthma and 10 patients with chronic obstructive pulmonary disease. Regional pulmonary ventilation was also measured with 19Ne, and this information was used to (i) correct ventilation/perfusion ratios obtained with 13N in regions with impaired ventilation and (ii) calculate regional perfusion, Q. In this paper we analyse the errors inherent in these measurements. Dead space ventilation and focal inhomogeneities in ventilation and/or blood flow will affect the transport of both 19Ne and 13N, but in different ways. Combining the two methods, values of blood flow obtained tend to overestimate Q in normal supine subjects by some 6% in the non-dependent part of the lung and to underestimate Q by a similar amount in the dependent part. In patients with air flow obstruction, blood flow is underestimated in regions with low ventilation/perfusion ratios and overestimated in regions with high ventilation/perfusion ratios. In the groups of patients studied, errors may reach a level of 10-20%. The limitations of the method in regions with extensive gas flow and/or blood flow heterogeneities are discussed on a theoretical basis and in some extreme conditions blood flow may be overestimated by a factor of two to three.