Relationships between capnogram parameters by mainstream and sidestream techniques at different breathing frequencies

Abstract

Capnography, routinely used in operating rooms and intensive care units, reveals essential information on lung ventilation and ventilation-perfusion matching. Mainstream capnography directly measures CO₂ in the breathing circuit for accurate analysis and is considered a reference technique. Sidestream capnography, however, analyzes gas away from the patient leading to potentially less accurate measures. While these methodological differences impact the capnogram indices in mechanically ventilated patients, such assessments during spontaneous breathing are essentially lacking. Accordingly, we aimed to compare mainstream and sidestream capnography in spontaneously breathing subjects, focusing on differences in capnogram shape and dead space indices at various respiratory rates. Simultaneous mainstream and sidestream time and volumetric capnography were performed on spontaneously breathing adults (n = 35). Measurements were performed during controlled low (10/min), medium (12/min), and high (20/min) breathing rates as a challenge. Correlation and Bland-Altman analyses were used to assess trends and agreements between time and volumetric capnography indices obtained by the mainstream and sidestream techniques, including end-tidal CO₂ (ETCO₂), shape factors reflecting the slopes of phases 2 and 3, and anatomical and physiological dead space fractions. ETCO₂ and physiological dead space measured by mainstream and sidestream techniques showed excellent correlations (r > 0.90, p < 0.001 for all breathing rates) and agreements. While strong correlations and moderate agreements were evidenced in the parameters reflecting the late phase of expiration (phase 3 slope and exhaled CO₂ volume), these relationships were weaker for indices related to the early phase of expiration (phase 2 slope, anatomical dead space). Changing breathing frequency caused significant alterations in all capnography parameters, which were detectable by both mainstream and sidestream techniques. Sidestream capnography cannot substitute the more accurate mainstream technique for measuring the absolute values of shape factors and ventilation dead space fractions. However, sidestream capnography is also able to detect and track changes in uneven alveolar emptying, ventilation-perfusion matching and ventilation dead space fraction in spontaneously breathing subjects.

Keywords: Dead spaces; Main- and sidestream capnography; Spontaneous breathing; Time and volumetric domains.

Fig. 1

Patient Flow Chart: A total of 36 participants were enrolled in the study, with selection criteria including informed consent, being over 18 years of age, and the ability to cooperate. All volunteers were eligible for the intervention; however, one patient was excluded due to a lack of appropriate cooperation.

Fig. 2

Normalized phase 2 and 3 slopes obtained from time (Sn2_T, Sn3_T) and volumetric capnograms (Sn2_V, Sn3_V), Fowler’s anatomical (VnDF) and Bohr’s physiological (VnDB) dead space fractions measured with both mainstream (empty boxes) and sidestream (hatched boxes) techniques in spontaneously breathing subjects (n = 35) at low (blue), medium (white) and high (red) breathing frequencies. #: p < 0.05, $: p < 0.01, *: p < 0.001.

Fig. 3

Correlations between the absolute values of time and volumetric capnography indices measured by mainstream (MS, horizontal axis) and sidestream capnography (SS, vertical axis). Linear regressions for low (blue), medium (black) and high (red) breathing frequencies are shown as continuous lines, grey dashed lines indicate the lines of identity. Pearson correlation coefficients (r) and p values for each breathing rate are shown separately in each panel with the corresponding colors. *: r values p < 0.05 vs. low, #: r values p < 0.05 vs. medium. ETCO₂: end-tidal CO₂; Sn2_T and Sn3_T: normalized phase 2 and 3 slopes in time capnogram, respectively; Sn2_V and Sn3_V: normalized phase 2 and 3 slopes in volumetric capnogram, respectively; VnDF: Fowler (anatomical) dead space fraction; VnDB: Bohr (physiological) dead space fraction.

Fig. 4

Correlations between the changes in time and volumetric capnography indices to alterations of breathing frequency measured by mainstream (MS, horizontal axis) and sidestream capnography (SS, vertical axis). Linear regressions for the transitions from medium to low (black), high to medium (green) and high to low (magenta) breathing frequencies are shown as continuous lines, grey dashed lines indicate the lines of identity. Pearson correlation coefficients (r) and p values for each change in breathing rate are shown separately in each panel with the corresponding colors. *: r values p < 0.05 versus normal to low, #: r values p < 0.05 versus high to medium. ETCO₂: end-tidal CO₂; Sn2_T and Sn3_T: normalized phase 2 and 3 slopes in time capnogram, respectively; Sn2_V and Sn3_V: normalized phase 2 and 3 slopes in volumetric capnogram, respectively; VnDF: Fowler (anatomical) dead space fraction; VnDB: Bohr (physiological) dead space fraction.