Comparison of Four Methods for Calculating Diffusing Capacity by the Single Breath Method: Discussion
Our main findings are as follows: (1) the singlebreath Deo was largest using the ESP timing method compared with JM, 3PIT, and Og methods (in decreasing order of mean Deo); (2) the difference between Deo using Og, JM, and 3PIT methods was small in patients with normal pulmonary function and became less in patients with obstruction, restriction, or low Deo; and (3) the reproducibility of Deo for replicate maneuvers was equivalent among the four methods. The differences among JM, Og and ESP timings reside in differences in definition of start and end of breath hold (Fig 1). Using backward elimination of nonsignificant terms, only IT and ET remained a significant term in the multiple regression models comparing Deo by Og, ESP, and JM (Table 4). The ESP gives the largest Deo in all patient groups because it uses the shortest breath hold time, start of breath hold being generally later than Og and JM, and end of breath hold at beginning of sampling. The Og is smallest in patients with normal expiratory flows because the Og inspiratory time is always larger than the others, due to the earlier start of breath hold. In normal individuals the differences between JM and both Og and ESP definitions for end of breath hold were insignificant. canadianneighborpharmacy.com
However, because of the influence of expiratory slowing on increasing expiratory time (Fig 3), the JM-Og Deo difference became less in patients with reduced FEVj (Fig 2, R2 = 0.40). There was a lower correlation between the Og—ESP Deo difference and FEVj (R2 = 0.26), since both methods defined end of breath hold the same, and expiratory slowing had a smaller effect on inspiratory time than expiratory time (Table 5). An ideal timing protocol would produce a calculated Deo that is nearly independent of breath-hold time, assuming uniformity of Dco/Va ratios within the lung. When patients reduce their breath-hold time from the standard 10 s to 5 s, Deo using the Og timing protocol increases, whereas both the JM and three equation methods are less affected. Though we did not address variations in breath-hold timing in each patient in our study, in several patients use of the 3PIT method significantly reduced the variability in calculated Deo between maneuvers (Fig 4). Disregarding such technically unsatisfactory tests, there were no significant differences in the maneu-ver-to-maneuver variability in calculated Deo among the Og, JM, ESP, or 3PIT methods (Table 3). Previous comparisons of Og and either the JM or ESP timings involved normal volunteers or epidemiologic studies of young people with relatively normal pulmonary function. Our results in the subgroup of patients with normal pulmonary function compare favorably with those earlier reports: Deo using ESP times averaged 8.4 percent larger than Deo using classic Og timing. The Deo by the three-equation technique was 3.5 percent larger compared with Og method, and 1.6 percent smaller compared with JM. Compared with these studies, we found a larger difference in Deo by Og compared with JM (Og 5.7 percent less than JM) in patients with normal pulmonary function. This difference may in part be related to differences in the expired sample volume, though Leech et al did not state the sample volume they obtained using a manually operated switching valve. The fixed volume of 500 ml that we used is at the lower end of the range recommended by ATS; use of a larger volume would increase the JM breath-hold time, reducing the difference between JM and Og Deo.
Figure 4. Example spirogram from two diffusing capacity maneuvers in one patient. Note the prolonged inspiratory time in the second maneuver, bottom. The difference in inspiratory time resulted in the large difference in calculated Deo using Ogilvie or ESP timing protocols, but smaller difference in Deo using 3PIT or Jones-Meade methods. This technically unsatisfactory test is shown to illustrate the potential advantage of the 3PIT method, although data are not included in averages in Tables 2 and 3.