Secondary Logo

Journal Logo

Respiratory Gas Analysis—Technical Aspects

Jaffe, Michael B., PhD

doi: 10.1213/ANE.0000000000002384
Technology, Computing, and Simulation: Narrative Review Article
Continuing Medical Education

A technology-focused review of respiratory gas analysis, with an emphasis on carbon dioxide analysis, is presented. The measurement technologies deployed commercially are highlighted, and the basic principles and technical concerns of infrared spectroscopy and mainstream versus sidestream gas sampling are discussed. The specifications of particular interest to the clinician, accuracy and response time, and the related standard, with typical values for a capnometer, are presented. Representative time and volumetric capnograms are shown with the clinically relevant parameters described. Aspects of the terminology in present-day use and the need for clarity in defining what is a breath and an end-tidal value are reviewed. The applications of capnography of particular interest to the anesthesiologist are noted, and key references are provided. Ongoing developments with respect to respiratory gas analysis, and those that will impact it, are noted.

From Cardiorespiratory Consulting, LLC, Cheshire, Connecticut.

Published ahead of print November 30, 2017.

Accepted for publication June 28, 2017.

Funding: None.

The author declares no conflicts of interest.

Reprints will not be available from the author.

Address correspondence to Michael B. Jaffe, PhD, Cardiorespiratory Consulting, LLC, 410 Mountain Rd, Cheshire, CT 06410. Address e-mail to

The accurate measurement of respiratory gases and delivery of therapeutic gases are vital for the proper management of the patient during anesthesia. The technologies used to measure these gases have leveraged both physical and chemical properties of the gas (Table 1). The focus of this article will be to discuss in general the technologies behind these gas measurements, with specific details provided with respect to capnography and infrared spectroscopy.

Table 1

Table 1

The measurement of many of these gases dates to the 19th century. For example, quantitative physical measurements of expired carbon dioxide (referred to as carbonic acid at the time) using infrared absorption can be traced to experiments in the 1860s in physicist John Tyndall’s laboratory.1 The applicability of these measurements to clinical care was recognized in the “popular” scientific literature of the time. With improvements in technology and advances in scientific understanding, physiologists and clinicians began to study aspects of the expiratory respiratory gases, laying the foundations of physiological concepts, such as dead space and alveolar gas. The commercial availability of rapid infrared analyzers after World War II allowed the measurement of respiratory gases, such as carbon dioxide, to transition from the physiology research laboratory to the “bedside.”2

Back to Top | Article Outline


A wide range of technologies have been developed to measure respiratory and therapeutic gases (Table 1) based on a variety of physical and chemical properties of the gases. For the measurement of anesthetic agents, N2O and CO2, infrared spectroscopy is widely used and, as such, will be the only technology further discussed.

Infrared absorption spectroscopy relies on the asymmetric modes of molecular vibration and the associated absorption in the 3- to 12-micron range to allow the quantification and differentiation of CO2, N2O, and the halogenated anesthetic agents. The infrared analyzer bench (Figure 1) usually comprises an infrared source (either a broadband or a narrowband source), filters at a wavelength at which the gas or gases being measured are not absorbed (ie, reference filter), filters at other wavelengths at which the specific gas or gases being measured are strongly absorbed (ie, data filter), and infrared-sensitive detectors (eg, thermal, microphonic, or photonic). The location of the data and reference filters in the bench can vary considerably with commonly used variants including fixed filters at the detector side and moving filters on a spinning wheel at the source side in use. The ratio of these measurements at these 2 different wavelengths (ie, ratiometric), using either formulas or lookup tables, allows the determination of the gas partial pressure. Corrections are often applied to these measurements, given that the infrared measurements are affected by the pressure, temperature of the sample gas, and presence of other gases. Increased pressure and/or temperature, or increased concentration of other gases, causes increased intermolecular collisions to occur, resulting in the widening of the absorption band. These device-independent effects, termed collision or pressure broadening are a complex function of the pressure and presence of other gases and often compensated for with the use of nominal correction values. Additionally, the overlap of absorption bands with other gases present in the mixture, termed cross interference, may also have significant impact on the measurement. For example, the impact of a N2O absorption band overlapping with the CO2 band at 4.24 microns is device dependent and can be minimized with the use of narrowband infrared sources or detector filters.

Figure 1

Figure 1

A respiratory gas monitor (RGM), by definition (per standard),3 may be either diverting (ie, sidestream) or nondiverting (ie, mainstream). A diverting RGM transports a portion of the respired gases from the sampling site, through a sampling tube, to the sensor, whereas a nondiverting RGM does not transport gas away from the sampling site. The difference between mainstream (nondiverting) gas analysis and sidestream (diverting) gas analysis can be viewed as direct near-instantaneous, high fidelity measurement of gas concentration versus a remote lower fidelity measurement delayed typically by a few seconds.

Mainstream RGMs, consisting of a sample cell and gas analysis bench, “sample” the expired or inspired gases and locate the measurement optics and electronics at or near the airway (ie, proximally). This location results in a “crisp” waveform that reflects, in real-time, the partial pressure of the gas within the airway. On the other hand, sidestream devices aspirate a sample of gas from an adapter placed in a breathing circuit or via a patient interface, such as a mask or nasal cannula. This sample passes through a small-bore tube (typically 6–8 feet in length) and through a water handling accessory (eg, water trap) and/or filters prior to reaching the gas analysis bench. This remote analysis results in a delay that is dependent on sampling flow rate and volume of gas pathway to bench and a dampening of the waveform that can, under some circumstances, be significant for certain extremely time critical applications.4

Sidestream technology continues to improve; however, vigilance must persevere toward water removal and differing conditions at the sampling site and sample cell (ie, temperature and humidity), as well as the effect on waveform fidelity due to mixing of the sample gas as it is drawn through the sampling system.

Back to Top | Article Outline


Table 2

Table 2

Two of the most important specifications of an RGM to consider are measurement accuracy and system response time (see Table 2 for selected specifications). Required minimum measurement accuracy, defined as the “quality which characterizes the ability of an RGM to give indications approximating to the true value of the quantity measured,”3 is specified in the RGM standard as volume fraction percentage. Total system response time is generally defined as the time from the step function change in gas levels at the sampling site to the time at which 90% of a final gas reading is reached. Total response time may be further subdivided into delay time and rise time. Delay time only applies to sidestream systems and depends primarily on the sampling flow rate and the volume of the sampling tube. Rise time, typically reported as the time between the 10% and 90% of the final value, is an important factor for both the signal fidelity and the maximum breathing frequency that can be measured from the gas waveform. With higher breathing frequencies (eg, >60 breaths/min) and longer rise times (eg, >200 milliseconds), the waveform is more likely to appear highly damped and falsely low end-tidal values may be reported. It is important to note that when comparing devices from different manufacturers, the reported delay and rise time should take into account the effect of the sampling accessories (nasal cannula, airway sampling adapter, etc).

Back to Top | Article Outline


The nomenclature used for RGMs and their waveforms have evolved since their introduction into clinical use over the last few decades. The most well known of the RGMs for many years, the carbon dioxide analyzer, is generally referred to as either a capnometer, a device that measures and displays numerical values of carbon dioxide, or a capnograph, a device that measures and displays numerical values of carbon dioxide and includes a waveform tracing. The term “capnogram” denotes the time-based waveform, which represents the variation of carbon dioxide in either partial pressure or gas fraction units over time. This differentiates it from the “volumetric capnogram,” the variation of carbon dioxide (generally in gas fraction) over volume in which the inspiratory portion of the curve is typically not shown. Both types of capnograms are described by 3 phases associated with the source of the expiratory gases: (1) gas from the dead space; (2) the transition between dead space and alveolar gas; and (3) gas from sequential emptying of the alveolar volumes (Figure 2A). Even though time and volumetric capnograms may appear similar in shape, care is advised when interpreting the slopes of the phases of the time-based capnogram (Figure 2B).

Figure 2

Figure 2

From the time-based capnogram, estimates of respiratory rate, end-tidal CO2, and inspiratory levels of CO2 may be accurately determined and are commonly reported (Table 3). Coupling capnography with flow (and volume) permits the estimation of relevant parameters (Table 3), such as anatomic and physiological dead space ratios, CO2 elimination, and pulmonary capillary blood flow. These parameters allow insight into cardiopulmonary disorders, including adult (acute) respiratory distress syndrome, chronic obstructive pulmonary disease, asthma, and pulmonary embolism. Viewing the changes in carbon dioxide as a function of volume, rather than time, allows for interpretation of the reported values and changes in those values, in a context consistent with well-known physiological concepts.

Table 3

Table 3

The most well-known capnographic parameter, the end-tidal CO2 value, is actually one of the least understood. It is often expected to equate to the arterial value but, in fact, is dependent on how it is measured and the patient’s physiology (diffusion, ventilation, dead space, cardiac output), with an end-tidal arterial gradient usually involved.

Respiratory rate derived from the capnogram is generally determined as the time between successive expiratory and inspiratory transitions of each breath. On face value, the determination of these transitions seems simple, but, in practice, it can be quite complicated, and obtaining appropriate breath criteria is dependent on the clinical environment and application.5

The CO2 and flow waveforms may each be measured via a sidestream and/or mainstream mode. However, to calculate anything other than carbon dioxide elimination, at least one of these waveforms needs to be in-line and proximal (mainstream). The various commercial offerings have included measurements attained using proximal flow with distal CO2 (eg, sidestream sampled or at the exhalation port), proximal CO2 with distal flow, and proximal flow and CO2. With each approach, potential error sources include the effects of compressibility and condensation, complexity of signal alignment, and ability to estimate accurate end-tidal and volumetric values. Of these approaches, the use of proximal flow and gas measurements, as with an integrated CO2/flow airway adapter, can help ensure that these sources of error are minimized.6 Software and breathing circuit additions have allowed volumetric capnography devices to extend their applicability to maneuver-based measurements, such as partial rebreathing cardiac output, lung recruitment, and functional residual capacity.

Back to Top | Article Outline


In respiratory medicine as in other fields, the measurement of a seemingly simple quantity is complicated by reality. Given the use of RGMs to measure respiratory rate, it is critical to have quantitative criteria defining a start and/or end of a breath. For example, with a capnogram, this is straightforward using a threshold-crossing method to determine the end of expiration in a patient receiving mechanical ventilation, but may not be in a patient monitored via nasal cannula gas sampling who is receiving procedural sedation and analgesia and is making light breathing efforts. As such, some manufacturers have employed algorithms of various degrees of sophistication to more reliably identify the end of breath in capnograms and permit a more robust estimate of respiratory rate.5

The common meaning of an end-tidal value is complicated by the need to report a value over a wide range of breathing and circuit conditions. For example, with a capnogram, the value that appears at the end of expiration (ie, end-tidal) as detected by the waveform may not adequately reflect a quantity close to the alveolar value. To compensate for this, manufacturers often report the greatest value during expiration. Note that this value could be biased high in a breathing circuit due to the rebreathing of gases from the apparatus dead space. Also, this value could be biased low in systems with combined nasal gas sampling and oxygen delivery due to the dilution of the sampled gas by oxygen delivered throughout the breathing cycle.

The choice of the most appropriate approach to sample and analyze a patient’s expiratory gases continues to be debated and is dependent on technology, equipment availability, the patient interface (eg, endotracheal tube, nasal cannula, or mask), and the clinical environment. Early commercial capnometers used in anesthesia (eg, 1950) employed a breathe-through configuration (ie, mainstream) but were rapidly replaced with catheter sampling (ie, sidestream) due to the size of the infrared gas analyzer interfering with the clinician’s view of the patient’s face.2 Only in the last few decades have advances in technology allowed the mainstream infrared gas analyzer to be reduced in size to address the early concerns of size and weight. Interfaces to the patient’s airway and the pros and cons of each with respect to gas sampling can easily be the subject of a paper unto itself. A properly inflated endotracheal tube cuff and good facemask seal can help minimize leaks, assure proper inspiratory gas delivery to the patient, and allow for all of the expired gas volume to be available for analysis. With the use of a nasal cannula for gas sampling, only a small portion of the expired gas is collected which may be diluted by room air and/or delivered oxygen. Additionally, manufacturer-specific designs, the placement of the gas collecting portion of the cannula (eg, prongs) on the patient, and the relative amounts of oral/nasal breathing can impact the measured values and must be considered.

Sidestream and mainstream gas analysis each pose their own set of unique water handling challenges. When warm humid air contacts a cooler surface, decreasing its temperature to the dew point, water condenses. This condensation may occur on the windows of a mainstream sample cell and could cause scattering of the infrared radiation and artifactual absorption. Manufacturers have applied different solutions, including hydrophilic coatings, heaters, and structural features, to reduce the potential impact. In addition, devices with a ratiometric design often can effectively compensate for the effect of this water. The transport of a gas sample through a long tube results in greater equilibration with the ambient temperature and additional condensation. In a breathing circuit, this condensation, in combination with patient secretions, can contaminate the sample line and possibly cause a blockage. To reduce the impact in sidestream systems, water vapor-permeable tubing (eg, NAFION; Perma Pure LLC, Lakewood, NJ), filters, and water traps may be employed. The use of additional components in the sidestream gas path and the associated connections require added vigilance with respect to the possible leaks from these connections or from cracked filters or adapters.

Back to Top | Article Outline


As this is a technology-focused review, current and evolving applications of respiratory gas analysis will be left to textbooks on the subject7,8 and the clinical literature. With respect to carbon dioxide, the American Society of Anesthesiologists standards9 have included requirements for monitoring of exhaled carbon dioxide during general anesthesia and have expanded it in the last few years to include procedural sedation. Additionally, the applications of continuous capnography include airway management during varying levels of anesthesia, nasogastric tube placement, assessment of the effectiveness of chest compression during CPR, and monitoring during patient transfer and postoperatively and during acute exacerbations of respiratory disease. It is important to note that volumetric capnography is growing in acceptance and allows physiologically important parameters with diagnostic, screening, and therapeutic applications to be quantified.10

Back to Top | Article Outline


The future of respiratory gas analysis is promising. The recognition of the value of capnography has led to its inclusion in clinical standards for sedation and monitoring during advanced cardiopulmonary resuscitation. Clinical and commercial interest remains in leveraging the unique applications of oxygraphy,11 the measurement of the oxygen waveform. Smarter algorithms for existing parameters and new indices that include environment-specific optimizations are being introduced by manufacturers. Researchers are exploring new applications using novel features extracted from the waveforms in combination with machine-learning algorithms12 that may allow easier interpretation and improved safety by providing an earlier indication of physiological deterioration. To improve patient safety, standards development organizations (eg, Association for the Advancement of Medical Instrumentation/Underwriters Laboratories, International Organization for Standardization Technical Committee 121) have developed a framework for an integrated clinical environment,13 are in the process of developing a family of standards for safe medical device interoperability and a forensic data recorder, and are adding annexes to existing device standards, including RGMs, to help standardize the data that the RGM provides via its external interface, thus facilitating the development of new applications, such as closed-loop control.14 International standards continue to evolve forward in reducing the opportunity for misconnections between small-bore connectors, which may consequently impact RGMs and other applications by limiting the use of the standard luer fitting to intravenous applications and phasing in the use of alternative fittings.15 With the growth of mobile health, personal health and medical products using respiratory gas analysis have been introduced in the market or are in development, to provide new and improved tools and therapies for disease and lifestyle management to the patient and clinician.

Back to Top | Article Outline


Name: Michael B. Jaffe, PhD.

Contribution: This author wrote the manuscript.

This manuscript was handled by: Maxime Cannesson, MD, PhD.

Back to Top | Article Outline


1. Tyndall J. On Radiation. The “Rede” Lecture. 1865.London: Longman, Roberts & Green.
2. Jaffe MB. Infrared measurement of carbon dioxide in the human breath: “breathe-through” devices from Tyndall to the present day. Anesth Analg. 2008;107:890–904.
3. ISO 80601-2-55. Medical electrical equipment—Part 2-55: particular requirements for the basic safety and essential performance of respiratory gas monitors. 2011.
4. Weingarten M. Respiratory monitoring of carbon dioxide and oxygen: a ten-year perspective. J Clin Monit. 1990;6:217–225.
5. Jaffe MB. What is a “valid” breath? - Methodological issues. Anesth Analg. 2011;113(2S Suppl):5.
6. Jaffe MB, Orr JA. Challenges of on-airway respiratory measurements. IEEE Eng Med Biol Mag. 2010;29:44–52.
7. Gravenstein JS, Jaffe MB, Gravenstein N, Paulus D. Capnography: Clinical Aspects. 2011.2nd ed.Cambridge, UK: Cambridge University Press.
8. Ehrenwerth J, Eisenkraft JB, Berry JM. Anesthesia Equipment: Principles and Applications. 20132nd ed. Philadelphia, PA: Saunders; 768.
9. ASA Standards for Basic Anesthetic Monitoring, Standards and Practice Parameters (Approved by the ASA House of Delegates on October 21, 1986, and last amended on October 20, 2010 with an effective date of July 1, 2011)—viewed 7-18-12 ( Available at: Accessed August 8, 2017.
10. Suarez-Sipmann F, Bohm SH, Tusman G. Volumetric capnography: the time has come. Curr Opin Crit Care. 2014;20:333–339.
11. Gadhinglajkar SV, Sreedhar R, Unnikrishnan KP. Oxygraphy: an unexplored perioperative monitoring modality. J Clin Monit Comput. 2009;23:131–135.
12. Jaffe MB. Using the features of the time and volumetric capnogram for classification and prediction. J Clin Monit Comput. 2017;31:19–41.
13. ASTM F2761-09. Medical Devices and Medical Systems—Essential Safety Requirements for Equipment Comprising the Patient-Centric Integrated Clinical Environment (ICE)–Part 1: General Requirements and Conceptual Model. 2013.West Conshohocken, PA: ASTM International.
14. Weininger S, Jaffe MB, Rausch T, Goldman JM. Capturing essential information to achieve safe interoperability. Anesth Analg. 2017;124:83–94.
15. ISO 80369-1. Small-bore connectors for liquids and gases in healthcare applications—Part 1: General requirements. 2010.
© 2018 International Anesthesia Research Society