Keywords cervical compressive myelopathy - autonomic dysfunction - heat rate variability -
propofol - etomidate - hypotension
Introduction
Autonomic dysfunction (AD) is well-recognized after traumatic spinal cord injury;
however, very few studies have reported the incidence of AD in cervical compressive
myelopathies (CCM).[1 ]
[2 ] Hypotension is the most frequent adverse hemodynamic event noted during the initial
intraoperative phase and is frequently linked to poor perioperative outcomes.[3 ]
[4 ]
[5 ] Reduction in arterial blood pressure below the lower limit of the vascular autoregulation
curve may result in heart, brain, and kidney ischemia.[3 ]
[6 ] AD has been recognized as one of the variables that assist in predicting postinduction
hypotension.[7 ] Such patients require inotropic support, vasopressors, or other treatments after
induction to maintain normotension. Heart rate variability (HRV), the physiological
variations of the changes in heartbeats, is a simple objective tool to diagnose AD.[8 ] It is vital to categorize AD patients preoperatively by performing an objective,
easy, and bedside test like HRV and opt for anesthetic agents carefully to avoid the
risk of hypotension-induced spinal cord ischemia.[9 ]
[10 ]
[11 ]
Anterior and posterior cervical decompressive procedures are commonly performed neurosurgical
procedures for CCM resulting from degenerative or traumatic etiologies.[12 ] Prevention of hypotension during anesthesia induction is crucial to maintain spinal
cord perfusion in these patients with compromised AD. Thus, in patients with CCM,
HRV testing may facilitate anesthesiologists to forecast hypotension, thereby optimizing
the anesthetic technique for the prevention of hypotension-induced spinal cord ischemia.
Although propofol is the most commonly used induction agent, it is associated with
hypotension.[13 ]
[14 ] Etomidate, although a less commonly used agent, has minimal cardiovascular side
effects.[15 ] Etomidate, as an induction agent, reportedly has more cardiovascular stability than
propofol in vulnerable patients.[16 ] However, there are no studies to appraise the suitability of etomidate for induction
in patients with impaired AD.
The primary objective of the study was to compare the hemodynamic profile between
propofol and etomidate induction in CCM patients requiring decompressive procedures.
The incidence of pain on injection and myoclonus were also compared between the two
groups as a secondary objective. We hypothesized that etomidate is preferred over
propofol for anesthesia induction to minimize hypotension in patients with CCM having
AD.
Methods
A single-center, randomized, double-blinded, prospective study conducted from May
2020 to December 2021 included patients with CCM who visited a super specialty care
center based in Bengaluru, India. The study was approved by the institutional ethics
committee as per the Indian Council of Medical Research (ICMR) guidelines (MSRMC/EC/AP-04/03–2020).
Inclusion criteria considered were subjects aged between 18 and 70 years, patients
fulfilling American Society of Anesthesiologists (ASA) physical status 1 to 3, and
those requiring elective anterior or posterior cervical decompression. The exclusion
criteria considered were heart rate or rhythm abnormalities or use of medications
known to alter them, sepsis/recovered from sepsis (<6 months), diabetes, degenerative
neurological disease (e.g., Parkinson's disease), complete spinal cord injury, and
patients requiring re-exploration procedures. This prospective study was registered
with Clinical Trials Registry, India before recruitment of cases (CTRI/2021/01/030207).
Informed valid written consent was sought from patients in the language comprehensible
to them before enrolment.
The sample size was calculated based on a randomized controlled trial in patients
undergoing cardiac surgery to assess hemodynamic profiles with etomidate versus propofol,[16 ] where propofol caused a 34% greater reduction in “MAP (mean arterial pressure)—time
integral” from baseline after induction of anesthesia than etomidate (p < 0.009). To achieve the power of 90%, a level of significance of 5% (two-sided),
and 5% loss to follow-up, our study required a minimum sample size of 30 patients
in each group.[16 ] Patients who underwent HRV testing were randomized to the etomidate and propofol
groups in the ratio of 1:1 based on computer-generated random allocation after obtaining
informed consent. As per the random allocation number, the clinicians involved in
the trial induced the patients with etomidate or propofol.
The HRV testing for autonomic function was performed using Vagus HRV (Recorders and
Medicare Systems-RMS, India) a day before surgery. The patients were made to rest
in the supine position for 10 minutes. Electrocardiographic (ECG) leads were connected
as per standards, and lead II ECG was recorded continuously for 10 minutes with the
patients' eyes open. The data were extracted using Vagus HRV apparatus. The ECG was
analyzed using RMS Vagus HRV software (RMS, India). Time-domain parameters measured
were standard deviation of NN (the number of RR interval differences) intervals (SDNN
ms), root mean square of successive RR interval differences (RMSSD ms), and percentage
of successive RR intervals that differ by >50 ms (pNN50%). Frequency-domain variables
measured were low-frequency (LF) power in ms2 (0.04–0.15 Hz), high-frequency (HF) power in ms2 (0.15–0.4 Hz), total power (ms2 ), and the ratio of LF to HF power (LF/HF%).
The HRV-based classification into normal and abnormal groups was done based on HRV
metrics and norms ([Fig. 1 ]).[17 ] The current study majorly considered the LF/HF ratio as an indicator of sympathovagal
balance. Therefore, short-term recordings of 10 minutes were conducted, and 1.1 was
considered the cutoff value between normal and deviated LF/HF ratio. Statistical analysis
was done based on LF/HF ratio >1.1 to 11.6 (normal HRV) and LF/HF ratio <1.1 (abnormal
HRV).[17 ] While analyzing the hemodynamic data, the two groups were subdivided into four groups,
namely, the propofol group with or without AD and the etomidate group with or without
AD based on HRV.
Fig. 1 CONSORT 2010 diagram depicting the patient enrolment and subgrouping.
On the day of surgery, all subjects received 500 mL of Ringer's lactate in the preoperative
room before shifting to the operation room. ASA standard monitors were connected before
induction. After lignocaine infiltration, radial artery was cannulated for invasive
blood pressure monitoring with a 20-gauge cannula under ultrasound guidance. Baseline
values were recorded, and preoxygenation was done for 3 minutes. The etomidate group
received intravenous (IV) fentanyl 2 µg/kg, etomidate 0.3 mg/kg, vecuronium 0.1 mg/kg,
followed by tracheal intubation and intermittent positive pressure ventilation (IPPV)
with oxygen, nitrous oxide, and sevoflurane to achieve minimum alveolar concentration
(MAC) 1.0. The propofol group received propofol 2 mg/kg, and the rest of the protocol
remained the same. Both induction agents were administered over a period of 30 to
60 seconds. A propofol preparation of 1% emulsion with a combination of medium chain
triglyceride (MCT) and long chain triglyceride (LCT) and etomidate of 0.2% emulsion
containing MCT were used for induction.
The heart rate and invasive blood pressures were continuously recorded and noted at
preinduction, postinduction, laryngoscopy, and 1, 3, 5, 10, and 15 minutes postintubation
(PI). Occurrence of pain at injection and any myoclonic movements were also recorded.
Hypotension, defined as a reduction in MAP <60 mm Hg or 20% of baseline, was treated
with ephedrine 6 mg IV bolus. The ephedrine boluses were repeated if the hypotension
did not settle in 60 seconds. Patients with bradycardia (<50 beats per minute) received
atropine 0.6 mg IV. Total ephedrine and atropine used were noted, and adverse effects
were carefully monitored. Pain on injection was recorded on a 4-grade scale, with
0 = no pain, 1 = verbal complaint of pain, 2 = withdrawal of the arm, and 3 = both
verbal complaint and withdrawal. Myoclonus in patients was recorded on a scale of
0 to 2 (0 = no myoclonus, 1 = minor myoclonus movement, and 2 = major myoclonus).
While analyzing the data, the patients were subgrouped as the etomidate group with
normal preoperative HRV, etomidate group with impaired preoperative HRV, propofol
group with normal preoperative HRV, and propofol group with impaired preoperative
HRV. Intraoperative drop in blood pressure at different time intervals after induction
and total usage of vasopressors to treat this hypotension was recorded. The data were
collected till MAP returned to normal or till 15 minutes postinduction or whichever
was longer.
Statistical analysis: Descriptive and inferential statistical analyses were performed. Continuous variables
were presented as mean ± SD (minimum–maximum) and categorical variables as percentages.
Significance was assessed at a 5% level of significance. The following two assumptions
on data were made, that is, dependent variables should be normally distributed and
two samples drawn from the population were random. Independent Student's t -test (two-tailed, independent) was used to find the significance of study parameters
on a continuous scale between two groups (intergroup analysis) on metric parameters.
Chi-squared/Fisher's exact test was used to find the significance of the study parameters
on a categorical scale between two or more groups in a nonparametric setting for qualitative
data analysis. VassarStats online tools were used for all the statistical computation,
and the line graphs were plotted using Excel 2019. Cramér's V effect was used to find
the effect size. Since the sample size was <30 for each subgroup, a nonparametric
test was used for the comparison.
Results
Out of 80 patients initially evaluated for enrolment eligibility, 20 were excluded
before allotment due to various reasons (anticipated difficult airway requiring awake
fiberoptic intubation [AFOI], multiple ectopics noted during HRV recording, refusal
to consent after HRV recording, etc.). The 60 selected patients were randomized into
two groups using a computer-generated list (100 numbers with a block of 5) of 30 each
to receive either etomidate or propofol. The anesthetists attending the cases were
blinded to HRV parameters and analysis. The CONSORT 2010 diagram depicting the patient
enrolment and subgrouping (etomidate with normal and abnormal HRV and propofol with
normal and abnormal HRV) is shown in [Fig. 1 ].
The demographic and the diagnostic variables between etomidate and propofol groups
are shown in [Table 1 ]. The age, gender, height, and body mass index (BMI) were comparable between the
groups. The patients with spondylosis with canal stenosis were higher in the abnormal
HRV group, and it was significant in abnormal HRV group who received propofol (p = 0.002). The details of time and frequency domains of HRV between etomidate and
propofol groups are listed in [Table 2 ]. The corresponding number of patients noted with abnormal SDNN (ms), RMSSD (ms),
NN50, total power (ms2 ), low-frequency power (ms2 ), high-frequency power (ms2 ), and LF/HF ratio on HRV analysis were 51 (85.0%), 41 (68.33%), 40 (66.66%), 35 (58.33%),
40 (66.66%), 43 (71.66%), and 31 (51.67%), respectively.
Table 1
Demographic and diagnostic details between etomidate and propofol group
(N = 60)
Etomidate (N = 30)
Propofol (N = 30)
HRV normal (N = 11)
HRV abnormal (N = 19)
p -value
HRV normal (N = 18)
HRV abnormal (N = 12)
p -value
Age
51.0 ± 17.13
45.84 ± 16.15
0.42
47.39 ± 16.54
48.75 ± 12
0.81
Gender (M/F)
7/4
10/9
0.25
11/7
4/8
0.60
Height
161.46 + 6.8
161.79 + 11.05
0.93
161.44 + 9.49
160.75 + 10.40
0.85
BMI
24.98 + 2.31
24.48 + 1.40
0.46
25.58 + 1.78
23.97 + 1.18
0.07
Spondylosis with canal stenosis
3 (27.27%)
10 (52.63%)
0.18
2 (11.11%)
7 (58.33%)
0.002
OPLL with compression
6 (54.5%)
3 (15.78%)
0.02
5 (27.77%)
3 (25%)
0.45
Spondylolisthesis with compression
2 (18.18%)
631.57%)
0.43
11 (61.11%)
2 (16.66%)
0.07
Levels of compression
2–4
4–6
5 (45.45%)
9 (47.36%)
0.91
10 (55.55%)
6 (50%)
0.78
6 (54.54%)
10 (52.63%)
0.915
8 (44.44%)
6 (50%)
0.58
Abbreviations: BMI, body mass index; HRV, heart rate variability; OPLL, ossified posterior
longitudinal ligament.
Note: Data expressed as mean + standard deviation and n (%).
Table 2
Heart rate variability (HRV) in etomidate and propofol group
HRV variables
Etomidate (mean ± SD)
Propofol (mean ± SD)
p -value
SDNN (ms)
150.03 ± 69.48
147.19 ± 58.97
0.38
RMSSD (ms)
85.10 ± 128.26
79.22 ± 106.27
0.38
NN50 (ms)
11.16 ± 4.20
9.62 ± 3.70
0.21
Total power (ms2 )
2,121.05 ± 1,315.76
2,606.38 ± 1,120.75
0.13
Low-frequency power (ms2 )
945.12 ± 597.86
954.45 ± 679.21
0.48
High-frequency power (ms2 )
932.28 ± 1,068.94
843.79 ± 450.16
0.34
LF/HF ratio
1.10 ± 0.34
1.21 ± 0.80
0.12
Abbreviations: HF, low frequency; HF, high frequency; NN50, successive RR intervals
that differ by >50 ms; RMSSD, root mean square of successive RR interval differences;
SDDN, standard deviation of NN intervals;
Heart rate changes were similar in all the groups and were not statistically significant.
Based on the HRV (LF/HF) analysis and the blood pressure response, the patients were
analyzed after subgrouping them into CCM patients with AD (n = 31, of which 19 received etomidate and 12 received propofol) and CCM patients without
AD (n = 29, of which 11 received etomidate and 18 received propofol). Hypotensive episodes
at 1, 3, 5, and 10 minutes were observed to be higher in the propofol group compared
with the etomidate group in patients both with and without AD. The episodes of hypotension
were observed to be higher at postintubation 3 (p = 0.02), 5 (p = 0.04), and 10 (0.06) minutes in propofol with AD group and at postintubation 5 minutes
in propofol without AD ([Fig. 2 ]). The observed effect size was 0.26, 0.37, and 0.056 for comparison of hypotension
in the abnormal HRV group at 3, 5, and 10 minutes, respectively. This indicates that
the magnitude of the difference between the observed data and the expected data was
medium at 3 and 5 minutes, whereas at the 10th minute, the difference was very minimal.
Fig. 2 Comparative analysis of hypotensive episodes among the groups.
Hypotensive episodes at various time points were significantly higher in patients
with abnormal HRV than in patients with normal HRV, requiring significantly higher
doses of ephedrine ([Figs. 3 ] and [4 ]). Comparative analysis of pain and myoclonus between the two groups demonstrated
that the etomidate group had lower grades of pain than propofol. The result was statistically
significant ([Table 3 ]), and myoclonus was noted higher in the etomidate group.
Fig. 3 Comparative analysis of hypotensive episodes between heart rate variability (HRV)
abnormal and HRV normal groups.
Fig. 4 Comparative analysis of ephedrine requirement between heart rate variability (HRV)
abnormal and HRV normal groups.
Table 3
Comparative analysis of pain and myoclonus with induction anesthesia
Etomidate (n = 30)
Propofol (n = 30)
p -value
Pain on injection
No pain
18 (60.0%)
12 (40.0%)
0.21[a ]
Verbal complaint of pain
12 (40.0%)
10 (33.33%)
0.07[a ]
Withdrawal of arm
0
8 (26.67%)
0.01[b ]
Myoclonus
No myoclonus
23 (76.67%)
28 (93.33%)
0.07[b ]
Minor myoclonus movement
7 (23.33%)
2 (6.67%)
a Chi-squared test.
b Fisher's exact test.
Discussion
The pathogenesis of CCM involves compression of the cervical spinal cord resulting
in dysfunction of ascending and descending tracts and spinal gray matter, causing
motor, sensory, and autonomic disturbances like bladder, bowel, and sexual dysfunctions.
There is a sparsity of literature on the incidence of cardiac AD in CCM patients although
it is well documented in traumatic spinal cord pathologies.[1 ] We assessed AD with various time- and frequency-domain HRV parameters. AD was observed
in 51.67% of patients when deviated LF/HF ratio was considered to define abnormal
HRV as in our study in CCM patients requiring decompression.
According to the Task Force of the European Society of Cardiology and the North American
Society of Pacing and Electrophysiology (1996), analysis of short-term HRV recordings
(5 minutes) is generally used to evaluate the pathophysiological correlation of autonomic
control with HRV.[18 ] Studies have noted vagal activity as a key contributor to the HF component.[19 ] Certain investigators have used LF/HF ratio to mirror sympathovagal balance or to
reflect sympathetic modulations. Hence, HRV is considered standard for the diagnosis/classification
of parasympathetic and sympathetic responses. RMS Vagus HRV software used in the current
study is a validated and standard tool for noninvasive testing of AD.[19 ] The study has primarily evaluated the LF/HF ratio, which mirrors both the components
of autonomic function.
The cardiac autonomic system plays an important role in the occurrence of hypotension
after induction of anesthesia in patients undergoing elective surgery.[17 ] The present study has evaluated the potential of HRV in predicting hypotensive CCM
patients during decompressive surgery and the superiority of etomidate over propofol
in preventing postinduction hypotension. The study has found that patients who encountered
hypotension postinduction (p = 0.02) and postintubation at 1 minute (p = 0.08), 3 minutes (p = 0.02), and 10 minutes (p = 0.07) were more among patients with AD compared with those without AD. In addition,
the ephedrine requirement was significantly higher in patients diagnosed with AD than
in those without AD.
Our study found that the patients with canal stenosis secondary to cervical spondylosis
had significantly abnormal HRV recordings (p = 0.003) as compared with patients diagnosed with OPLL and spondylolisthesis causing
cord compression. A study by Shindo et al reported the presence of AD in CCM secondary
to cervical spondylosis. This study group measured muscle sympathetic nerve activity
(MSNA), an indicator of sympathetic outflow to muscles, which was found to be significantly
reduced in patients with cervical spondylosis. The authors attributed the negative
correlation between burst incidence of MSNA and motor power to the posterior column
involvement.[20 ]
The current study has shown that HRV-based AD categorization may help better predict
postinduction hypotension. Several previous studies have analyzed the clinical use
of HRV in predicting hypotension.[21 ]
[22 ]
[23 ] Hanss et al studied HRV-directed severe hypotension in patients scheduled to undergo
elective cesarean delivery. They found that LF/HF is a useful tool to suggest prophylactic
therapy in patients at risk of hypotension after a subarachnoid block during cesarean
delivery.[21 ] In line with this finding, the present study has shown that a lower LF/HF ratio
is an indicator of sympathovagal imbalance. In a study on patients with human T-lymphotropic
virus type 1 (HTLV-1) associated myelopathy, the LF/HF ratio was found to be an indicator
of sympathovagal balance. This group demonstrated that the LF/HF ratio was further
reduced in patients with orthostatic hypotension, and it also correlated with cord
atrophy on imaging.[24 ]
Etomidate is not as commonly used as propofol for induction of anesthesia for various
reasons. The incidence of myoclonus has been reported as much as 50 to 80% after etomidate
induction,[25 ] and there is a high rate of transient adrenal insufficiency and mortality, especially
in patients with sepsis, which is debatable.[26 ]
[27 ] But etomidate has a favorable hemodynamic profile due to its unique lack of effect
on both the sympathetic nervous system and baroreceptor function[28 ]
[29 ] and its capacity to bind and stimulate peripheral α-2B adrenergic receptors with
subsequent vasoconstriction.[30 ] Hypotension occurring with propofol is mainly due to the reduction of sympathetic
activity causing vasodilation or its direct effect on vascular smooth muscles and
myocardial depression.[31 ]
[32 ]
Various authors have studied the effects of propofol and etomidate on the autonomic
nervous system (ANS) objectively. Wang et al studied a spectrogram derived by continuous
wavelet transform of electrocardiography and pulse photoplethysmography (PPG) signals
at baseline, early phase, and late phase after propofol induction. They found that
propofol administration resulted in reductions in instantaneous high frequency (HFi)
and low frequency (LFi) and increases in the LFi/HFi ratio and PPG amplitude. This
study demonstrated significant immediate changes in ANS activity that include temporally
relative elevation of cardiac sympathovagal balance and reduced sympathetic activity
after propofol.[33 ] Ebert et al studied changes in MSNA, forearm vascular resistance, and blood pressure
after propofol and etomidate administration. MSNA was reduced after propofol leading
to a reduction in forearm vascular resistance and significant hypotension, whereas
etomidate preserved these. Both cardiac and sympathetic bar slopes were maintained
with etomidate but were significantly reduced with propofol, especially in response
to hypotension.[34 ] The current study administered HRV testing preoperatively to all CCM patients requiring
decompression and found etomidate administration prevented postinduction hypotension
better as compared with propofol, especially in patients with abnormal preoperative
HRV. However, we did not perform HRV testing during or after the administration of
these induction agents.
Several studies have compared the effectiveness of etomidate over propofol in preventing
perioperative hypotension.[16 ]
[35 ]
[36 ] Comparison of the efficacy of etomidate over propofol in cardiac surgical patients
by Ladha et al showed that the adrenal suppression caused by etomidate can present
a challenge to the anesthesiologist in a variety of clinical settings, despite its
superior hemodynamic profile.[36 ] This finding is debatable, but the present study has excluded patients who could
have had adrenal suppression. Although not statistically significant, the overall
pain score was more in the propofol group and increased incidence of myoclonus in
the etomidate group. Several Indian studies have reported similar findings.[37 ]
[38 ]
[39 ]
The present study holds significant relevance, as there is very limited literature
evidence suggesting the potential of preoperative HRV in detecting AD in CCM patients.
In addition, the study has also highlighted the potential benefit of etomidate in
reducing the incidence of hypotension on induction in CCM patients with impaired HRV.
One of the major limitations of the current study is the single-center study design.
The number of patients belonging to each group was further reduced due to the categorization
of etomidate and propofol groups into HRV normal and HRV abnormal groups. Hence, the
power of the study calculated retrospectively was low (0.53). The correlation of preoperative
neurological deficits and chronicity of CCM with HRV analysis would give valuable
information. Large-scale, multicenter, randomized clinical trials are warranted to
corroborate.
Conclusion
We conclude that HRV-based AD categorization of CCM patients may assist in better
prediction of postinduction hypotension, and etomidate is preferred over propofol
for induction of anesthesia in these patients. Pain scores were higher following propofol
injection, and myoclonus incidence was higher after etomidate induction.
