Non-invasive assessment of shunt and ventilation/perfusion ratio in neonates with pulmonary failure

Abstract

Aims-To make non-invasive measurements of right to left (R-L) shunt and reduced ventilation/perfusion ratio (V A /Q) in neonates with pulmonary failure and to examine sequential changes in these variables after treatment. Methods-Twelve neonates with pulmonary failure were studied. They ranged in gestational age from 24 to 37 (median 27) weeks and were 1-39 (median 4) days old. Shunt and reduced V A /Q were derived from their eVects on the relation between inspired oxygen pressure (PIO 2 ) and arterial oxygen saturation measured with a pulse oximeter (SpO 2 ). Pairs of PIO 2 v SpO 2 data points were obtained by varying PIO 2 in a stepwise fashion. A computer algorithm based on a model of pulmonary gas exchange fitted a curve to these data. With PIO 2 on the abscissa, an increase in shunt produced a downward movement of the curve, whereas reducing V A /Q to < 0.8 shifted the curve to the right. The right shift gives a variable that is inversely related to V A /Q, the PIO 2 − P c O 2 diVerence, where P c O 2 is mixed capillary oxygen pressure. A reduced ventilation/perfusion ratio (V A /Q) and an increased shunt cause quite diVerent, but characteristic, eVects on the relation between inspired oxygen pressure (PIO 2 ) and arterial oxygen saturation (SaO 2 ). 1 2 A reduced V A /Q in the absence of shunt reduces both alveolar oxygen (PAO 2 ) and oxygen saturation but increasing the PIO 2 washes out alveolar nitrogen, restores PAO 2 , and corrects desaturation. In contrast, an isolated shunt reduces oxygen saturation, but increasing PIO 2 does not correct this because the shunted blood is never exposed to the increased PIO 2 . The nonshunted blood is almost fully saturated, and even administering 100% oxygen adds only a small amount of dissolved oxygen, which cannot compensate for the eVect of the shunt. Results-TenIt may be important in neonates to diVerentiate between a reduced V A /Q and a shunt. For example, in an animal model of surfactant deficiency when surfactant was administered into the airways either as a bolus or as an aerosol, V A /Q worsened with the former but improved with the latter.3 There may also be a prognostic significance of diVerent degrees of

Non-invasive measurement of reduced ventilation:perfusion ratio and shunt in infants with bronchopulmonary dysplasia: a physiological definition of the disease

“…These independent effects on gas exchange can be represented in the form of plots of SpO 2 against PIO 2 (fig 1). [8][9][10][11][12] At sea level (1 atm), PIO 2 (kPa) is the same as the inspired oxygen percentage. The curve relating alveolar (mixed capillary) PO 2 to oxygen saturation in the ideal lung represents the shape of the oxyhaemoglobin dissociation curve.…”

“…[7][8][9][10][11] We have applied this method to analyse the gas exchange abnormalities in infants with BPD and used these observations to model an improved approach to the definition of BPD, which measures the severity of gas exchange impairment.…”

A simplified method for deriving shunt and reduced VA/Q in infants

“…The same data series was analysed with a more complex DOS computer algorithm which has been used to measure gas exchange in patients of all age groups under widely varying circumstances 1 2 5 6 7. The new method produced results that agreed well with the more complicated algorithm.…”

“…They commented that “these models fit the patient data identically” 13. Later we varied P I O 2 and measured SpO 2 in neonates with pulmonary failure and showed that 10 out of 12 had both large shunts (median 19.9%) and considerable right shifts (median 19.8 kPa), some being equivalent to a V A /Q <0.1 7. Other workers using the invasive arterial nitrogen gradient method to measure V A /Q directly had shown, in similar groups of neonates, that a considerable reduction in V A /Q explained about 30% of the oxygen gradient 14 15.…”

“…We used this method to measure shunt and right shift in a variety of clinical situations in adults and infants 1 2 5 6 7. A limitation is that the algorithm sometimes cannot fit a curve to the data points if they do not describe enough of the steep and plateau portions of the curve, or if the saturation values on the plateau are 100%.…”