The Use of Catheter Mount Will Result in More Reliable Carbon Dioxide Monitoring under Fluid Exposing Conditions

Introduction Capnometer can be readily malfunctioned by fluid exposure during treatment of critically ill patients. This study aimed to determine whether placing capnometer distant from the endotracheal tube by connecting direct connect catheter mount (DCCM) is effective in yielding reliable end-tidal carbon dioxide (ETCO2) by reducing capnometer malfunctioning caused by water exposure. Methods In 25 healthy adults, a prospective, open label, crossover study was conducted to examine the effect of DCCM in mainstream and microstream capnometers under water exposing conditions. The primary endpoint was the comparison of ETCO2 between proximal DCCM (pDCCM) and distal DCCM (dDCCM). Results For mainstream capnometers, mean ETCO2 was significantly (p < 0.001) higher in dDCCM compared to pDCCM under water exposing conditions (29.5 vs. 19.0 with 5 ml; 33.8 vs. 21.2 with 10 ml; mmHg). Likewise, for microstream capnometers, ETCO2 was greatly higher (p < 0.001) in dDCCM compared to pDCCM (30.5 vs. 13.9 with 5 ml; 29.9 vs.11.4 with 10 mL; mmHg). ETCO2 measured by dDCCM was reliable in microstream settings, whereas it was unreliable in mainstream (correlation coefficient 0.88 vs. 0.27). Conclusions Application of DCCM onto the capnometer setting seems to be effective in reducing capnometer malfunctioning under fluid exposing conditions, which is obvious in microstream capnometer by producing more reliable ETCO2.


Introduction
Capnometer is a device that measures end-tidal carbon dioxide (ETCO 2 ) by infrared sensor [1,2]. Capnometer has been widely used to monitor proper placement of endotracheal tube (ETT) and status of ventilation [3,4]. According to the 2010 American Heart Association (AHA) and European Resuscitation Council (ERC) guidelines, the use of capnometer is recommended during cardiopulmonary resuscitation (CPR), in order to evaluate quality of CPR and detect recovery of spontaneous circulation (ROSC) in intubated patients [5,6]. During CPR, while low ETCO 2 (<10 mmHg) represents poor quality of CPR, a dramatic increase of ETCO 2 (up to 35-40 mmHg) indicates the occurrence of ROSC [6][7][8]. Accurate assessment of ETCO 2 is therefore essential to monitor placement of ETT and quality of CPR.
Nevertheless, reliable ETCO 2 measurement by capnometer is difficult in clinical settings. ETCO 2 measurement becomes readily unreliable when capnometer is vulnerable to fluid exposure through ETT, mostly in cases of wet lung conditions [1,4]. In wet lung conditions, fluid produced by patients is present in ETT, which interferes the infrared sensor on capnometer with detecting CO 2 absorption at 4.3 휇m wavelength, resulting in under-or oversensing ETCO 2 values. Furthermore, the sampling line connected to sidestream capnometer (i.e., microstream capnometer) is often occluded by condensed particles of fluid sourced from patients [1,4]. Therefore, clinical settings of which patients 2 Emergency Medicine International have wet lung conditions often render capnometer exposed to fluid through ETT, leading to higher chances of capnometer malfunctioning.
To achieve accurate ETCO 2 measurement, it is imperative for capnometer to avoid fluid contact and in that perspective, placing capnometer away from ETT is necessary. The direct connect catheter mount (DCCM, RT021 catheter mount, Fisher & Paykel Healthcare Ltd., Auckland, NZ) is a tubing system commonly inserted between breathing circuit and ETT to provide this connection with flexibility and a resultant reduction in extubation risk [9]. However, it can also be placed between ETT and capnometer to make them separate away, which could help capnometer avoid fluid contact supplied from patients.
The purpose of this study was to investigate whether placing capnometer away from ETT via DCCM insertion is effective in yielding reliable ETCO 2 by protecting capnometer against water contact. We hypothesized that the use of DCCM will reduce capnometer malfunction, leading to reliable ETCO 2 monitoring.  Figure 1).

Materials and Methods
Prior to the study, all participants were provided with 5 min oral instructions on breathing techniques. During inspiration, participants were asked to inhale through nose only, not allowing them to open their mouth and swallow water. During expiration, however, they were forced to occlude their nose and exhale forcefully only through mouth when exposed to the water-sprayed ETT. Participants were also asked to breathe regularly according to the metronome sound (15 beats per min), which was made by the Micro Metronome application (SPACEWARE Inc., Android apps on Google Play) installed on the smart phone (LG Optimus G Pro, LG Electronics, Seoul, Republic of Korea).
All participants (n=25) used one of the mainstream and microstream capnometers and then the other capnometer.
They were instructed to hold the ETT cuffs by their mouth and breathe for 2 min each time when the ETT was sprayed with 0 (baseline), 5, or 10 ml distilled water, in turn, which were given to the ETT just before connecting with pDCCM. Five min intervals were given for each time of breathing. The water sprayed within the ETT was allowed to contact capnometers freely when participants were exhaling. During crossovers, 30 min intervals were given to all participants. All ETCO 2 values measured after each time of breathing were recorded from different capnometers. All capnometers were calibrated regularly each time before initiating breathing tests of individual participant. The flow sampling rate in microstream capnometer was 50 ml/min. Sampling lines and adapters installed on capnometers were also changed accordingly ( Figure 2).
The primary endpoint is the comparison of ETCO 2 values measured in water exposing conditions by pDCCM and dDCCM in both mainstream and microstream settings. The secondary endpoint is the reliability of ETCO 2 values measured by dDCCM among three different water conditions in both mainstream and microstream settings.
. . Statistical Analysis. Prior to the experiments, minimum sample size was calculated based on our pilot study results. Briefly, when ETT was sprayed with 5 ml water, mean ETCO 2 values measured by pDCCM and dDCCM were 14.4 ± 19.9 mmHg and 29.2 ± 7.4 mmHg, respectively. Difference from dDCCM to pDCCM was 14.7 ± 20.6 mmHg. Sample size was calculated by Wilcoxon-signed rank test using Gpower software (version 3.1.7; Heine Heinrich University, Düsseldorf, Germany). By using 훼 = 0.05 and 훽 = 0.05, it was concluded that at least 21 participants are required.
SPSS software (version 20; IBM Corp, NY, USA) was used for data analysis. All variables were analyzed with Shapiro-Wilk normality test. ETCO 2 values measured in water sprayed conditions were analyzed by repeated measures ANOVA. The graphs showing individual ETCO 2 values monitored from every breath over time were represented as mean with standard errors (95% CI) using error bars. We analyzed the agreement of ETCO2 measurement between pDCCM and dDCCM using Bland-Altman plots. We also compared the agreement of ETCO2 measurement between mainstream and microstream capnometer. The intraclass Assessed for eligibility (n=25)  correlation coefficient (ICC) was calculated for ETCO 2 values of dDCCM to estimate interrater reliability among three different water conditions. All data are shown as means ± SD. p-value < 0.05 was considered significant.
. . Individual ETCO for Each Breath Measured by pDCCM and dDCCM. Individual ETCO 2 values for each breath over time measured by pDCCM and dDCCM are shown in Figure 3. For both mainstream and microstream capnometers when treated with 5 and/or 10 ml water sprays, ETCO 2 measurements by pDCCM were far lesser than those at baseline (0 ml water), whereas these measurements by dDCCM were very close to those at baseline. In the Bland-Altman plot, wider range of the 95% limits of agreement was shown in the water exposing conditions (5, 10mL in ETT) comparing with baseline (0mL in ETT) for both capnometers, which suggests inaccuracy of pDCCM under water exposing conditions [13]. Additionally, the microstream capnometer showed better  agreement between pDCCM and dDCCM than mainstream capnometer (Supplementatary Figure S1).

Discussion
This is a concept study that simulates fluid using water and evaluates the efficacy of DCCM in yielding reliable ETCO 2 under fluid exposing conditions. The key finding from this study is that adoption of DCCM can reduce capnometer malfunctioning particularly when fluid inflow occurs through the ETT. It was evident from the results that, under water exposing conditions (5 or 10 ml water sprays), ETCO 2 measurements were significantly lower in pDCCM than in dDCCM in both mainstream and microstream capnometers. With water, additionally, all ETCO 2 values measured by pDCCM were far less than baseline ETCO 2 (0 ml water spray), indicating that these levels of water caused pDCCM malfunctioning. On the contrary, ETCO 2 values obtained from dDCCM placing away from water sources were very similar to the baseline ETCO 2 , implying that securing some distance from patients (∼15 cm) [9] by DCCM installation seems to be effective in protecting against capnometer malfunctioning by lessening water vapor concentrations and inhibiting the capnometers from direct water contact.
We did not set partial pressure of carbon dioxide (PaCO 2 ) in arterial blood gas analysis as a reference standard. Since there could be wide gap between ETCO 2 and PaCO 2 under water exposing conditions, baseline ETCO 2 (0 ml water spray) was set to the reference standard. ETCO 2 monitoring is most reliable and accurate method to monitor success for tracheal intubation or CPR quality. However, under fluid exposing conditions, fluid could hinder obtaining reliable ETCO 2 in proximal ETT. Additionally, in Emergency Medicine International 5 cases of patients receiving CPR or having fluid, low ETCO 2 could be frequently observed. This study suggests that the use of DCCM could provide benefit to obtain more reliable ETCO 2 by using DCCM in those cases. It is of interest that, under water exposing conditions, the use of DCCM was more effective in microstream capnometer than in mainstream capnometer. It could be explained by the fact that the infrared sensor of mainstream capnometer was directly exposed to water, being vulnerable to being malfunctioning, whereas that of microstream capnometer was indirectly exposed to water via sampling lines connected to the opposite direction of gravity, implying that a continuous measurement of ETCO 2 is possible unless there is water condensation and/or direct water contact.
In this study, two types (mainstream vs. microstream) of capnometers were used for CO 2 monitoring. Depending on the use of gas sampling system, capnometers can be classified into two categories such as mainstream capnometer and sidestream capnography [1,14], and the microstream capnograph used in this study is a type of sidestream capnography [14]. Mainstream capnometer can measure ETCO 2 directly by using infrared sensor and does not need gas sampling system [1,3]. Both the sidestream and microstream capnographs can monitor CO 2 aspirated by sampling line [1,3]. Water vapor can cause condensation in sample lines which can thus interfere with CO 2 monitoring [1]. The occurrence of capnometer malfunctioning caused by water condensation is more often in sidestream capnograph compared to microstream capnograph. It is also known that, compared to mainstream capnometer, sidestream capnograph is more vulnerable to water itself [3]. It is the reason why mainstream capnometer and microstream capnograph were chosen for this study in assessing the level of capnometer malfunctioning under water exposing conditions. This study was performed under water exposing conditions, where two different water conditions were simulated by spraying 5 ml or 10 ml of distilled water into ETT. These water amounts were equivalent to one-third (5 ml) or twothirds (10 ml) of mechanical dead space of tracheal tube (15 ml), which was gauged by filling water into the tracheal tube.
From the study, we found that, in water exposing conditions, measuring ETCO 2 is more accurate when using DCCM. In spite of it, using dDCCM to monitor ETCO 2 is not the way traditionally recommended. Earlier studies indicated that ETCO 2 measurement should be performed at the proximal point of tracheal tube (pDCCM) connected to the ventilator circuit [15], in order to avoid possible inconsistency between PaCO 2 and ETCO 2 when measured by dDCCM. We found that a baseline ETCO 2 value for mainstream capnometers was significantly lower in dDCCM than in pDCCM. However, when measured by microstream, mean ETCO 2 values were not significantly different between pDCCM and dDCCM. We assume that it is originated from the difference of gas sampling technique in capnometer. A microstream capnometer withdraws a continuous sample of gas through a capillary tube from the patient's airway to the monitor and a water trap removes particles of water before measurement takes place. Hence, dDCCM of mainstream capnometer is more vulnerable to water vapor than microstream.
This study has several limitations. Firstly, use of DCCM might not guarantee the complete inhibition of capnometer malfunctioning during water exposing conditions, as installation of DCCM itself is not sufficient to block water contact completely, but rather it can help in reducing the capnometer malfunctioning by allowing a certain distance from the water sources. Secondly, in simulating fluid, pure water was used in this study to prevent ethical conflicts in healthy volunteers. The potential effect of other components of fluid on the infrared sensor of capnometer remains still unknown. Thirdly, when conducting this study, other resuscitation techniques (i.e., suctioning and bag valve mask ventilation) were excluded, as they can act as confounding factors when analyzing the sole effect of DCCM on capnometer malfunctioning. Fourthly, the CO 2 rebreathing effect possibly occurring during DCCM mounting onto the ETT was incompletely corrected. As 25 ml dead space of DCCM can increase the CO 2 rebreathing effect and elevate baseline ETCO 2 for both pDCCM and dDCCM, a possible gap still exists between ETCO 2 measurements obtained from this study and real CO 2 levels from exhales of the participants. However, to avoid rebreathing effect, participants were guided to exhale forcefully to reach maximal flow rate and tidal volume.

Conclusions
In conclusion, application of DCCM onto the capnometer setting seems to be effective in reducing capnometer malfunctioning under fluid exposing conditions, which is obvious in microstream capnometer by producing more reliable ETCO 2 . To demonstrate the efficacy of DCCM in real world patients under fluid exposing conditions, further clinical studies are needed.

Data Availability
All data and materials in this study are fully available without restriction.