Abstract

Background

Post-ROSC patients in the prehospital setting are challenging to manage due to cerebral agitation, ventilatory failure and airway compromise. To allow safe transfer to definitive care, these patients often require prehospital emergency anesthesia (PHEA). However, haemodynamic instability and myocardial dysfunction renders this cohort of patients susceptible to hypotension, arrythmia and further cardiac arrest. We aimed to investigate the haemodynamic stability of a ketamine-rocuronium PHEA regime in a cohort of post-ROSC patients by comparing heart rate (HR) and mean arterial pressure (MAP) prior to induction and subsequently at 3,6 and 9 minutes.

Methods

A retrospective case review was conducted for all post-ROSC patients who received PHEA between 1st October 2015 and 31st May 2019. Patients over the age of 18 were included if they suffered a non-traumatic out of hospital cardiac arrest and received ketamine-rocuronium based PHEA. A repeated measures ANOVA evaluated changes in HR and MAP at 3,6 and 9 minutes compared to the pre-induction value.

Results

55 patients were included for analysis. Whilst the HR did not statistically differ following induction (p=0.260), the mean arterial pressure was significantly lower at 9 minutes when compared to the pre-induction value (p=0.0382). Prior to 9 minutes, no significant changes in mean arterial pressure were noted between pre-induction and 3 minutes (p=0.705), pre-induction and 6 minutes (p=0.738), nor between 3 and 6 minutes (p=0.233).

Conclusion

We demonstrated that whilst Ketamine induction provides relative haemodynamic stability compared to other agents, Ketamine is associated with delayed haemodynamic compromise. We hope this serves as an important reminder to clinicians that in this catecholamine deplete cohort of patients, Ketamine may not be as haemodynamically stable as often described.

Keywords:Out of Hospital Cardiac Arrest; Cardiopulmonary Resuscitation; Anesthesia; Emergency Medicine; Ketamine

Abbreviations:PHEA: Prehospital Emergency Anesthesia; HR: Heart Rate; MAP: Mean Arterial Pressure; ED: Emergency Departments; ROSC: Return of Spontaneous Circulation; SOP: Standard Operating Procedure; NIBP: Non-Invasive Blood Pressure; IQR: Interquartile Range

Introduction

After a return of spontaneous circulation (ROSC) following a prehospital cardiac arrest, reperfusion of ischaemic organs leads to an array of complications including myocardial dysfunction, anoxic brain injury and impaired vasoregulation; known as the post-cardiac arrest syndrome [1-3]. This is in addition to the precipitating insult responsible for the cardiac arrest. Prehospital clinicians are required not only to support and manage these complex and often unstable patients but are required to expedite definitive treatment of the underlying pathophysiology [4], often by bypassing local Emergency Departments (ED) and conveying the patient to a centre capable of primary percutaneous coronary intervention [5,6]. Post-ROSC patients can present a challenge in the prehospital environment due to cerebral agitation, ventilatory failure and airway compromise often requiring prehospital emergency anaesthesia (PHEA) to facilitate onward transfer to definitive care [7]. However, this is a high-risk procedure as haemodynamic instability and myocardial dysfunction renders this cohort of patients fragile and susceptible to hypotension, arrythmia and further cardiac arrest [8,9].

Whilst the efficacy and haemodynamic stability of various agents for PHEA in trauma patients have been well described [10,11], evidence for the optimal induction agent for post-ROSC patients is minimal with many services creating protocols based on the extrapolation of data from in-hospital post-ROSC patients, prehospital trauma patients and expert opinion. Miller et al [8] described a prehospital anaesthetic regime of midazolam, fentanyl and rocuronium used by two different UK prehospital physician led teams for post-ROSC anaesthesia but found that whilst this regime offered minimal episodes of post-induction hypertension and changes in heart rate, this was at the expense of hypotension following induction [8].

For post-ROSC PHEA, Magpas Air Ambulance physicianparamedic teams utilize a ketamine-rocuronium regime similar to the commonly utilized fentanyl-ketamine-rocuronium regime used in trauma PHEA [10,11]. We aimed to investigate the haemodynamic stability of ketamine induction in post-ROSC patients in comparison with the data presented by Miller et al [8]. We aimed to measure haemodynamic stability by comparing heart rate (HR) and mean arterial pressure (MAP) prior to induction and subsequently at 3,6 and 9 minutes. We sought to identify the frequency of hypertension, hypotension, tachycardia and bradycardia compared to the pre-induction values. We also sought to measure compliance against the standard operating procedure (SOP) and hoped to identify factors predictive of deviation in practice.

Methods

Post-ROSC PHEA Standard Operating Procedure

The local standard operating procedure (SOP) dictates the required standard for delivery of PHEA following cardiac arrest. Post-ROSC patients receive 1mg.kg-1 rocuronium and a variable ketamine dose dependent on their mean arterial pressure (MAP) and clinical presentation. ‘Lively’ post-ROSC patients with a MAP >70mmHg receive 1mg.kg-1 and obtunded or haemodynamically unstable patients with a MAP <70mmHg receive 0.5mg.kg- 1. The dose utilized for induction remains at the discretion of the treating physician. Regardless of the initial bolus induction dose of ketamine, all patients receive a ketamine maintenance infusion of 1mg.kg.hr-1. Minimum monitoring includes HR, noninvasive blood pressure (NIBP), end tidal CO 2 and 3 lead ECG obtained via a Lifepak 15 (Physio Control, Redmond, WA, USA). Intermittent measurements such as NIBP are set to automatically cycle at 3-minute intervals. Upon induction of anaesthesia, the ‘event mark’ feature of the Lifepak 15 is used to timestamp the provision of drugs prior to laryngoscopy. Whilst certain features such as minimum monitoring are mandatory, the drug regime and doses used for PHEA can be deviated upon based upon the clinical assessment. This can be decided by the individual team or following advice from the 24/7 consultant telephone support.

Data Recording

Following all missions, a paper patient report form is filled out as a summary of the clinical interventions provided. Full documentation of drugs, intervention and clinical decision making is then completed on an online database which allows classification of missions for audit purposes. Rationale for deviation from the PHEA SOP is documented within the electronic database and is reviewed for clinical governance purposes. Furthermore, the patient report form completed at scene is also scanned to the database and observations from the Lifepak 15 are downloaded and electronically stored.

Data Collection

A retrospective case review via an electronic database (HEMSbase - Medic One Systems, London, UK) search was conducted for all patients who achieved a ROSC and received PHEA between 1st October 2015 and 31st May 2019 (32 months). Patients were included if they were over the age of 18, had suffered a non-traumatic cardiac arrest and achieved a subsequent ROSC, received a ketamine-based induction of PHEA and had sufficient haemodynamic observations available for analysis. Sufficient haemodynamic data was defined as a HR and MAP within 3 minutes prior to induction and subsequently at 3, 6 and 9 minutes. A further pre-induction BP was sought (if available) between 3-6 minutes prior to induction if available to assess haemodynamic stability prior to induction. Required observations from eligible patients were downloaded, anonymised and stored in an Excel (Microsoft, Redmond, WA, USA) spreadsheet alongside nonidentifiable demographic data and cardiac arrest statistics. PHEA drug regime, doses and times were also recorded alongside the patient’s estimated weight.

Statistics, Data Analysis and Sample Size

Statistical analysis was conducted using GraphPad Prism version 9 (San Diego, California, USA). Statistical significance was deemed present if p<0.05 (two-tailed). Continuous variables are expressed as mean ± 1 standard deviation, or as median, interquartile range (IQR) and range. Categorical variables are expressed as counts and percent. Haemodynamic stability was assessed using a repeated measures one-way analysis of variance (ANOVA) for both MAP and HR. The Greenhouse-Geisser correction was utilised if sphericity assumptions were violated according to Mauchly tests [12]. If a statistically significant difference was found, post hoc tests were utilised (Tukey’s test) to provide intragroup analysis.

Furthermore, the incidence of hypertension/hypotension and tachycardia/bradycardia were examined. In keeping with the definitions used by Miller et al [8], 2 classifications of hypertension/hypotension were considered. A change in MAP to either <70mmHg or >110mmHg from a previously normotensive value constituted a hypotensive and hypertensive response respectively. Furthermore, a 20% decrease in the pre-induction value was considered to be undesirable. Undesirable changes in heart rate were also examined, defined as a change in HR to either >100 or <50 beats.min-1 from a value previously between 50-100 beats.min-1. Alternatively, a 20% increase from the pre-induction HR was identified as abnormal. To ensure a sufficient sample size to power a repeated measures one-way ANOVA, 48 patients were sought to achieve a medium effect size (0.5) with a power of 0.8 and significance level of p<0.05.

Ethics

Utilizing the NHS Health Research Authority guidance, this work was defined as a service evaluation and was therefore exempt from ethical approval. Local approval was sought, and permission was granted by the information governance officer. Sponsorship was granted by the Magpas Clinical Governance Committee. No funding was applied for or received.

Results

Eighty-six patients received PHEA following a return of spontaneous circulation from a non-traumatic cardiac arrest. 31 patients were excluded; one received a ketamine and midazolam induction; six patients were excluded as they had suffered a traumatic cardiac arrest and were incorrectly categorized and one patient was excluded as they were under 18 years of age. A further 23 patients had insufficient haemodynamic data documented leaving 55 patients with one pre-induction and three post-induction MAPs and HRs available for analysis. Full baseline characteristics for each patient are available in Table 1.

Of the 55 patients included for analysis, 41 (75%) were male and 36 (65%) presented in VF. The average weight and age were 83 kilograms and 57 years respectively. Patients received an average of 2 shocks prior to ROSC and of the patients who received 1:10,000 adrenaline, the median dose prior to ROSC was 1mg. Successful intubation was achieved in 54 (98%) patients as a ‘plan A’ approach and one patient had their airway managed successfully via an iGel (Intersurgical, UK) to the Emergency Department. Full descriptive and demographic data is presented in Table 1.

The median dose (IQR [Range]) of ketamine used for induction was 80mg (50-100 [30-200]), with a median dose relationship of 1mg.kg-1 (0.5-1 [0.5-2]), in keeping with the median weight described in Table 1. Twenty-five patients (46%) received an induction regime in keeping with the SOP for post-ROSC patients, however 27 (49%) received an induction dose larger than recommended. Only 3 (5%) patients received an induction dose smaller than recommended. 100% of rocuronium doses were compliant with the SOP. To assess pre-induction stability, 37 patients had 2 MAPs recorded prior to induction, 3 minutes apart. Furthermore, 36 patients had 2 heart rates recorded prior to induction. The mean difference between the two MAPs and HRs collected (-6 and -3 minutes) were 3mmHg (95% CI: -6-13, p=0.515) and 3beat.min-1 (95% CI: -6-12, p=0.522) respectively, suggesting relative haemodynamic stability prior to induction.

Mean Arterial Pressure

The MAP before and after the induction of anesthesia are shown in Table 1. A repeated measures one-way ANOVA with the Geisser-Greenhouse correction applied identified a statistically significant difference between the MAP before and after induction (F (2.678,141.9) = 4.068, p=0.0108)). Tukey’s test for post-hoc analysis identified that whilst the MAP decreased over time, only the 9-minute MAP was statistically significantly different from the pre-induction value (p=0.0382). There was no statistically significant difference between the pre-induction MAP and the MAP at 3 minutes (p=0.705) or between the pre-induction MAP and the MAP at 6 minutes (p=0.738) (Figure 1). Twenty-five (45%) patients became hypotensive following induction. Twentyfour (44%) patients had a >20% decline in MAP and 8 (15%) patient’s MAP fell to <70mmHg from a previously normal preinduction value. Seven of the 8 patients whose MAP fell <70mmHg also demonstrated a >20% drop in MAP. Nine patients became hypertensive (MAP >110mmHg) compared to a previously normal pre-induction value. When combining both hypotensive and hypertensive changes, only 21 (38%) patients received a haemodynamically stable induction.

Heart Rate

The average HR prior to, and at 3,6 and 9 minutes after induction is shown in Table 2. A repeated measures one-way ANOVA using the Greenhouse-Geisser correction identified no statistically significant difference in HR over time (F (1.961, 104.0) = 1.364, p=0.260) (Figure 2). Eighteen (33%) patients became tachycardic (by either definition) following induction. Fourteen (7%) patients had an increase >20% from their pre-induction rate and 14 patients (7%) heart rates rose above 100 beats.min-1 despite a previously normal value. Ten patients met both definitions, however three of these patients had episodic ventricular tachycardia (pulsed) and therefore met both definitions. No patients developed a new bradycardia post-induction (Table 3).

GCS; Glasgow Coma Score, VF; Ventricular Fibrillation, VT; Ventricular Tachycardia, PEA; Pulseless Electrical Activity, ETT: Endotracheal Tube, SGA; Supraglottic Airway, FMV; Facemask Ventilation, FONA; Front of Neck Access, C&L; Cormack-Lehane, STEMI; ST Elevation Myocardial Infarction, RBBB; Right Bundle Branch Block, LBBB; Left Bundle Branch Block.

Discussion

We sought to investigate the haemodynamic stability of post- ROSC ketamine PHEA from medical prehospital cardiac arrests and found that only 21 (38%) of patients had a haemodynamically stable induction, with a statistically significant decrease in MAP nine minutes following induction. However, we did find haemodynamic stability prior to this point, with no statistical or clinically significant changes in blood pressure or heart rate. These results are in keeping with the results described by King et al with a fentanyl-ketamine-rocuronium regime for post-ROSC patients 9 and by Ter Avest et al [10] using a fentanyl-ketamine-rocuronium regime for trauma PHEA [10]. To the best of our knowledge, this is the first haemodynamic description of a ketamine-rocuronium regime in post-ROSC patients, without the use of fentanyl as an adjunct.

Whilst ketamine has been widely described as a haemodynamically stable agent due to sympathetic and adrenocortical stimulation increasing endogenous catecholamine and norepinephrine levels, it is thought that this effect is seldom seen in patients with limited myocardial reserve and an increased myocardial oxygen demand [13]. Waxman et al found that in critically ill patients, ketamine reduced left ventricular output in 50% of patients [14]. Furthermore, other studies have failed to identify positive haemodynamic changes in response to ketamine in critically ill patients [9,10,15]. This may be as ketamine relies on secondary sympathomimetic mechanisms to maintain haemodynamic stability, which are likely to be significantly obtunded or absent in a catecholamine deplete patients [13].

Whilst post-induction hypotension may be attributable to ketamine, it is possible that the haemodynamic changes described are due to the pre-existing haemodynamic instability seen in the inherently unstable post-ROSC cohort of patients, due to not only the underling aetiology but also the myocardial dysfunction and vasoplegia secondary to systemic hypoperfusion and reperfusion [1-3]. However, we did not identify any changes in MAP or HR between -6, -3 and +3-minutes surrounding induction suggesting relative baseline haemodynamic stability. Several actions at the point of induction may impact the patient’s blood pressure. Patients receive mechanical ventilation with PEEP which has the potential to decrease venous return, further exacerbating hypotension [16]. In addition to this, all patients receive a maintenance ketamine infusion of 1mg.kg.hr-1 following the induction bolus. However, the physiological effects of this are unlikely to manifest until approximately 30 minutes after initiation so it is unlikely to be responsible for the lowered MAP as observed at 9 minutes.

A common concern regarding the use of ketamine PHEA for post-ROSC patients is the possibility of a subsequent tachycardia due to sympathomimetic stimulation [13], increasing myocardial oxygen demand and worsening cardiac efficacy. Therefore, in post-ROSC PHEA, minimal changes in HR are advantageous. In our data, we found a non-statistically significant difference in pre and post induction HRs following ketamine PHEA. Whilst 18 patients demonstrated an increased heart rate, four of these were due to a rise in HR to greater than 100beats.min^-1 from a previously normal value (<100 beats.min^-1). However, the mean pre-induction HR was 100 beats.min-1 so a clinically insignificant increase (such as 99 beats.min-1 to 101 beats.min^-1) would count as a tachycardic response by this definition. However, this definition was used to produce comparative results with Miller et al [8]. In addition to minimal tachycardia, we found minimal rates of hypertension. Our low incidence of tachycardia and hypertension is in keeping with the findings of Miller et al, when using a fentanyl-midazolamrocuronium regime [8].

We also sought to describe compliance with the standard operating procedures for post-ROSC PHEA. The data shows that 49% of patients received a dose greater than recommended as per the SOP. Miller et al [8] found that the midazolam dose used for induction was often reduced, in keeping with dose ranges used for procedural sedation likely due to familiarity. It is possible a similar effect is seen with ketamine in these data, where patients received a larger dose in keeping with that recommended for trauma PHEA (2mg.kg^-1 or 1mg.kg^-1) as is the more common remit for ketamine. There may be an association between a post-induction MAP decrease and a higher than recommended dose of ketamine. However, it would be difficult to describe a dose-response relationship due to the significant variation in SOP compliance as a degree of clinical acumen is utilized to decide required doses and it may be that patients who were perceived to be relatively stable received a higher dose due to concern of possible awareness. However, the median GCS was 3 suggesting that a maximum dose of 0.5mg.kg-1 would be appropriate.

These findings are not without limitations; most notably those associated with a retrospective chart review. Furthermore, the data is reliant upon NIBP and ECG monitoring which may be subject to movement artefact associated with prehospital care. Furthermore, NIBP measurements can become unreliable in hypotensive tachycardic patients. In addition to this, atrial fibrillation, ectopic beats and other arrythmias are common in ischaemic post-ROSC patients which further skews the reliability of NIBP. Future studies should consider the use of intra-arterial monitoring. When auditing drug doses, it is important to remember ideal drug doses and SOP compliance were reviewed based on the MAP prior to induction. However, this doesn’t account for the clinical trend post-ROSC. A haemodynamically unstable patient may be optimised prior to induction and then have a reduced induction dose based on their clinical trend however our data won’t necessarily represent this prior instability.

Conclusion

We aimed to build on the data described by Miller et al [8] and King et al [9] describing the haemodynamic response to post- ROSC PHEA. We identified an undesirable drop in MAP 9 minutes following induction serving as a reminder that ketamine may not be as hemodynamically stable as often described in catecholamine deplete patients, such as those following an OOHCA. Whilst further research is required to further evaluate the haemodynamic changes of a ketamine-rocuronium induction in post-ROSC patients, we hope this data prompts clinicians not to be falsely reassured by immediate haemodynamic stability following a ketamine-based induction and to anticipate delayed haemodynamic compromise, allowing it to be managed expeditiously.

Acknowledgements

The authors would like to thank the staff and clinicians from both Magpas Air Ambulance and the East of England Ambulance Service for their exceptional care of these patients.

References

1. Reynolds JC, Lawner BJ (2012) Management of the Post-cardiac Arrest Syndrome. J Emerg Med 42(4): 440-449.
2. Adrie C, Adib-Conquy M, Laurent I, Mehran Monchi, Christophe Vinsonneau, et al. (2002) Successful Cardiopulmonary Resuscitation After Cardiac Arrest as a “Sepsis-Like” Syndrome. Circulation 106(5): 562-568.
3. Neumar RW, Nolan JP, Adrie C, Mayuki Aibiki, Robert A Berg, et al. (2008) Post-Cardiac Arrest Syndrome. Circulation 118(23): 2452-2483.
4. Lyon RM, Nelson MJ (2013) Helicopter emergency medical services (HEMS) response to out-of-hospital cardiac arrest. Scand J Trauma Resusc Emerg Med 21(1): 1.
5. Dumas F, Cariou A, Manzo-Silberman S, David Grimaldi, Benoît Vivien, et al. (2010) Immediate Percutaneous Coronary Intervention Is Associated with Better Survival After Out-of-Hospital Cardiac Arrest. Circ Cardiovasc Interv 3(3): 200-207.
6. Lemkes JS (2020) Coronary Angiography After Cardiac Arrest. Circulation 142(21): 2013-2015.
7. Lyon R, Nelson M (2012) Helicopter emergency medical services (HEMS) response to out-of-hospital cardiac arrest. Resuscitation 83: e96.
8. Miller M, Groombridge CJ, Lyon R (2017) Haemodynamic changes to a midazolam-fentanyl-rocuronium protocol for prehospital anaesthesia following return of spontaneous circulation after cardiac arrest. Anaesthesia 72(5): 585-591.
9. King C, Lewinsohn A, Keeliher C, Sarah McLachlan, James Sherrin, et al. (2022) Cardiovascular complications of prehospital emergency anaesthesia in patients with return of spontaneous circulation following medical cardiac arrest: a retrospective comparison of ketamine-based and midazolam-based induction protocols. Emergency Medicine Journal 39(9): 672.
10. Ter Avest E, Ragavan D, Griggs J, Dias M, Mitchinson SA, et al. (2021) Haemodynamic effects of a prehospital emergency anaesthesia protocol consisting of fentanyl, ketamine and rocuronium in patients with trauma: a retrospective analysis of data from a Helicopter Emergency Medical Service. BMJ Open 11(12): e056487.
11. Lyon RM, Perkins ZB, Chatterjee D, Lockey DJ, Russell MQ, et al. (2015) Significant modification of traditional rapid sequence induction improves safety and effectiveness of prehospital trauma anaesthesia. Crit Care 19(1): 134.
12. Kim HY (2015) Statistical notes for clinical researchers: A one-way repeated measures ANOVA for data with repeated observations. Rde 40(1): 91-95.
13. Morris C, Perris A, Klein J, Mahoney P (2009) Anaesthesia in haemodynamically compromised emergency patients: does ketamine represent the best choice of induction agent? Anaesthesia 64(5): 532-539.
14. Waxman K, Shoemaker WC, Lippmann M (1980) Cardiovascular effects of anesthetic induction with ketamine. Anesth Analg 59(5): 355-358.
15. Fouche PF, Meadley B, St Clair T, Alison Winnall, Paul A Jennings, et al. (2021) The association of ketamine induction with blood pressure changes in paramedic rapid sequence intubation of out-of-hospital traumatic brain injury. Academic Emergency Medicine 28(10): 1134-1141.
16. Carroll GC, Tuman KJ, Braverman B, W G Logas, N Wool, et al. (1988) Minimal Positive End-Expiratory Pressure (PEEP) May be “Best PEEP”. Chest 93(5): 1020-1025.