Key-words: Jugular venous oxygenation - propofol - traumatic brain injury
Introduction
Traumatic brain injury (TBI) in the present day scenario is a major public health
problem resulting in long-term disability and death especially in young adults. Primary
injury or initial impact results in initiating an inflammatory cascade, edema formation,
and excitotoxicity thus causing increase in intracranial pressure (ICP) and decrease
in cerebral perfusion pressure (CPP).[[1 ]],[[2 ]] Secondary systemic insults occurring thereafter in the form of hypoxia, hypotension,
hypercarbia, hyperglycemia, hyperthermia, anemia, etc., are preventable.[[3 ]],[[4 ]] Neuroanesthesiolgists are actively involved in resuscitation of TBI patients, thereby
playing an imperative role in prevention secondary brain injuries and improving their
overall outcome. The drugs used for induction and maintenance of anesthesia have a
direct effect on cerebral blood flow (CBF), cerebral metabolic requirement of oxygen
(CMRO2), ICP, and CPP.[[5 ]] Jugular venous oxygen saturation (SjVO2) is an indirect assessment of cerebral
oxygenation, which reflects the global cerebral balance between cerebral oxygen demand
and supply.[[6 ]],[[7 ]] It provides an early diagnosis of ischemia resulting from either intracranial or
systemic causes. The anesthetic agents can provide neuroprotection by maintaining
an adequate balance between cerebral oxygen demand and supply. Various inhalational
anesthetic agents have demonstrated promising results as cerebral protectants through
their preconditioning effect in animal models.[[8 ]] The superiority of one anesthetic agent over the other has yet not been established
despite the difference in their neurophysiological properties due to lack of substantial
evidence favoring one over the other. Sevoflurane and propofol are both being increasingly
used in neurosurgery due to their property of rapid onset and emergence from anesthesia
aiding in early assessment of neurosurgical status following surgery. However, the
intravenous anesthetic agents maintain the CBF-metabolism coupling as against the
inhalational counterparts which disrupt it at higher maximum additive concentration
(MAC) values and thus increase cerebral blood volume and thereby, ICP, which may prove
detrimental to the already compromised brain function.[[9 ]] Thus, intravenous anesthetic agents are given a preference over inhalational anesthetics
to provide adequate brain relaxation by reducing intracranial blood volume in moderate-to-severe
TBI. Sevoflurane may prove to be equally beneficial in maintaining cerebral oxygenation
and hemodynamics compared to propofol in TBI patients and may have an additional advantage
of providing neuroprotection by virtue of its preconditioning effects.
Interleukin-6 (IL-6) is a pro-inflammatory cytokine that is sensitive for brain injury
and can be easily detected in serum.[[10 ]] The anesthetic agents may have anti-inflammatory properties and consequently neuroprotective
effects which can influence the serum IL-6 levels.
Our primary aim was to compare the effects of sevoflurane and propofol on cerebral
oxygenation measured using jugular venous oximetry. Secondary aims were to determine
the effect of these drugs on intraoperative heart-rate, blood pressure, brain relaxation
score, and attenuation of cerebral inflammatory response by analyzing IL-6 in moderate-to-severe
TBI patients undergoing decompressive craniectomy.
Methods
This prospective, randomized, double-blind study was conducted after approval from
Institution Ethics Committee and registration with the clinical trial registry of
India (Regn. no-CTRI/2018/02/012139). Written informed consent was obtained from participants'
relatives and all the procedures were performed in accordance with the Helsinki declaration.
The CONSORT recommendations for reporting randomized trials were followed [[Figure 1 ]].
Figure 1: Consort diagram
We enrolled 42 American Society of Anesthesiologists physical status IE-IIE patients
aged 18–60 years with severe head injury who underwent decompressive hemicraniectomy.
Patients with mild-to-moderate head injury (Glasgow Coma Scale >8), patients in shock
(systolic blood pressure [SBP] <90 mm of Hg) even after resuscitation, any comorbidity
other than hypertension or diabetes (i.e., coronary artery disease, chronic obstructive
pulmonary disease, hepatic, or renal impairment) were excluded. Randomization was
done using computer-generated random numbers table and patients were divided equally
into two groups - Group P received propofol and Group S received sevoflurane for maintenance
of anesthesia.
Each patient underwent preanesthetic check-up prior to shifting the patient inside
the operation theatre. Preinduction monitoring included electrocardiography, noninvasive
blood pressure, entropy, neuromuscular transmission, pulse oximetry (SpO2), and invasive
blood pressure using 20 G intra-arterial catheter in radial artery. Patients were
induced with fentanyl 2 μg/kg and propofol 1–2 mg/kg administered in titrated doses.
Vecuronium (0.1 mg/kg) was used for tracheal intubation and lignocaine 1.5 mg/kg was
administered around 90 s prior to laryngoscopy to prevent intubation response. All
patients were administered a continuous infusion of fentanyl at the rate of 1 μg/kg/h
for intraoperative analgesia. After intubation, both the groups were mechanically
ventilated with oxygen-air mixtures (50:50) and an I/E ratio of 1:2. The tidal volume
was set to 6–8 ml/kg and the respiratory rate was adjusted to maintain a PaCO2 (partial
pressure of carbon dioxide in arterial blood) value between 32 and 35 mm Hg. Anesthesia
was maintained by continuous infusion of propofol 75–150 μg/kg/h in Group P and with
sevoflurane 0.8–1 MAC in Group S to titrate the entropy values between 40 and 60.
For jugular bulb oximetry, a 5 Fr, 15 cm long central venous catheter was inserted
in the side of predominant injury or right internal jugular vein in case of diffuse
axonal injury requiring decompression along the retrograde direction toward the jugular
bulb by Seldinger's technique. The length of insertion was estimated from the point
of insertion to the mastoid prominence or till the resistance was felt near the mastoid
at which point the catheter was pulled back by 1 cm. Confirmation of correct placement
was done using C-arm by lateral neck radiography to position the tip of the catheter
between C1 and C2 vertebra. The blood samples were withdrawn slowly at a rate not
more than 2 mlmin-1. Sampling of jugular bulb blood was done three times, and the
values were recorded to find any episodes of desaturation (<50%) or hyperemia (>75%).
Baseline SjVO2 values were recorded just after insertion of catheter into the jugular
bulb. Arterial and jugular bulb blood samples were withdrawn simultaneously after
catheter insertion (baseline value), at the end of surgery, and 12 h after completion
of surgery.
The muscle relaxation was achieved with intermittent boluses of vecuronium (0.01mg/kg)
in intraoperative period to maintain a train of four count <2. The packed red blood
cells were administered in case the hematocrit readings were below 30. All patients
received mannitol (0.5 g/kg), phenytoin (5mg/kg), and antibiotics as per the institutional
protocol.
The hemodynamic parameters (heart rate [HR], SBP, diastolic blood pressure, and mean
arterial pressure [MAP]) were recorded prior to induction of anesthesia as baseline
values, intraoperatively at every 10 min interval till the end of surgery and 12 h
after completion of surgery. The MAP was kept above 65 mm of Hg. In case of hypotension
(MAP <65 mm Hg), crystalloid fluid bolus of 3–5 ml/kg was given initially followed
by boluses of intravenous mephentermine 3 mg or phenylephrine (50–100 μg). If MAP
persisted at <65 mm of Hg for more than 5 min, an infusion of nor-adrenaline was started
at the rate of 0.05–0.1 μg/kg/min. In case of hypertension (MAP >110 mmHg), intravenous
2–3 esmolol boluses (0.3–0.5 mg/kg) were administered. In case of any bradycardia
of <50/min, intravenous atropine (0.5 mg) was administered. All rescue drugs used
to maintain hemodynamic were recorded and urinary bladder was catheterized in all
the patients to monitor intraoperative urine output.
Interleukin-6 assay
The blood samples for IL-6 were taken twice, one sample before induction of anesthesia
which was considered the baseline value and other two, at the end of surgery. Levels
of IL-6 were measured in duplicate in plasma by enzyme-linked immunosorbent assay
method. A volume of 3–5 ml of blood sample was taken and centrifuged to obtain an
adequate amount of serum which was then stored at −20°C–−40°C and was analyzed in
duplicate by TECAN (Infinite M200 pro, 2014) analyzer.
The surgery was conducted by an experienced neurosurgeon (>2 year experience in neurosurgery)
who was blinded to the agent used for maintenance of anesthesia and were asked to
assess and grade the brain relaxation after the elevation of bone flap using a four-point
grade.[[11 ]] Grade I - excellent with no brain swelling, Grade II - minimal but acceptable brain
swelling, Grade III - moderate brain swelling but no specific change required in management,
and Grade IV - severe brain swelling requiring some intervention such as change in
position, a further reduction in PaCO2, additional dose of mannitol and/or furosemide.
Fentanyl infusion was stopped at the beginning of skin closure whereas the maintenance
agents were stopped following completion of skin closure. None of the patients were
extubated at the end of surgery and were shifted to neurosurgical intensive care unit.
Statistical analysis
The sample size of total 42 patients (21 in each group) was calculated on the basis
of a previous study,[[12 ]] taking an alpha error of 0.05% at 80% power and considering SjVO2 values <50% as
significant cerebral hypoperfusion. Statistical analysis was carried out using statistical
package for social sciences (SPSS Inc., Chicago, IL, USA, version 21.0). The continuous
data were presented as mean ± standard deviation for normally distributed data; two
groups were compared using Student's t-test. Categorical and nominal data were described
as proportions and Chi-square test or Fischer's exact test was used to look at significant
associations. Paired sample t-test was also used for baseline comparisons. A P < 0.05
was considered statistically significant.
Results
A total of 48 patients were assessed for decompressive hemicraniectomy for severe
TBI. Six patients were excluded due to refusal of consent and hemodynamic instability.
A total of 42 patients were finally analyzed [[Figure 1 ]].
Demographic data and other characteristics were comparable in both the groups [[Table 1 ]].
Table 1: Comparison of demographic data (n=21)
The neurological diagnosis of patients is described in [[Table 2 ]].
Table 2: Type of traumatic brain injury (n=21)
Values of SjVO2 were comparable between Group P and Group S at baseline, end of surgery,
and at 12 h after surgery [[Table 3 ]].
Table 3: Intergroup comparison of jugular venous oxygen saturation (n=21) (%)
HR was recorded prior to induction as well at various intervals intraoperatively.
The baseline (pre-induction) HR was comparable in both the groups (P = 0.867). The
difference in HR was not found statistically significant at any point of time and
was comparable in both the groups [[Figure 2 ]].
Figure 2: Comparison of intraoperative heart rate among the two groups
Baseline values of MAP were comparable in both the groups (P = 0.569), but after the
start of surgery, there was a significant difference (P < 0.05) in MAP at various
time intervals intraoperatively, end of surgery and even after 12 h of surgery. This
shows that MAP was on lower range in Group P as compared to Group S [[Figure 3 ]]. A statistically significant difference was found with mean IV fluid infused -
2800 ml in Group P and 2385.71 ml of fluid in Group S (P < 0.001). More intravenous
fluid requirement was seen in Group P as compared to Group S to maintain hemodynamics.
Figure 3: Comparison of intraoperative mean bp among the two groups
The total urine output was compared in both groups and a statistically significant
difference was found with total urine output values of 1352.38 ml in Group P and 1166.66
ml of fluid in Group S (P < 0.001) [[Table 4 ]].
Table 4: Comparison of intravenous fluids and urine output (n =21)
The brain relaxation was assessed by the operating surgeon who was blinded to the
randomization of groups before opening of dura and was noted using four-point grades.
most patients had a brain relaxation score of Grade-II in both the groups, none of
the patients had Grade-IV brain relaxation score. No statistical difference in brain
relaxation was noted between the groups (P = 0.626) [[Figure 4 ]].
Figure 4: Brain relaxation score comparison between the two groups
The values of IL-6 were compared within the groups and between the groups. Both the
baseline IL-6 levels and IL-6 levels at end of surgery were comparable between the
two groups. A significant difference was found between the baseline value and value
at the end of surgery in Group S (P = 0.040) [[Table 5 ]].
Table 5: Intergroup comparison of interleukin-6 Levels (in pg/ml) (n=21)
Discussion
Administering general anesthesia to the patients with moderate-to-severe head injury
is a challenge as these patients commonly have hemodynamic instability, neurological
complications as well as respiratory compromise. Resuscitation, prevention of further
damage to brain cells and providing neuroprotection are the goals of neuroanesthesia
in these patients posted for decompressive craniectomy. Optimal neuroprotective strategies
include appropriate patient positioning, management of systemic as well as cerebral
hemodynamics.[[13 ]] Anesthetic agents may provide neuroprotection by keeping a balance between cerebral
oxygen demand and supply.
In the study, SjVO2 was comparable in patients receiving either propofol or sevoflurane
at baseline value or at the end of surgery and 12 h after surgery. In patients receiving
propofol SjVO2 values were a little on higher side compared to those receiving sevoflurane,
though the difference was not statistically significant. Higher SjVO2 may be due to
decrease in CMRO2 by propofol. The values were similar at 12 h after surgery to that
of baseline values suggesting that both propofol and sevoflurane are short acting
agents and effects do not last longer.
However, SjVO2 as a marker of cerebral oxygenation has its drawbacks. SjVO2 is a global
oxygenation parameter, which cannot detect regional ischemia. SjVO2 monitoring has
a high specificity but low sensitivity for the detection of regional ischemia. In
other words, a normal SjVO2 does not guarantee against regional ischemia but a low
SjVO2 is a definitive indicator of global ischemia and/or focal ischemia.[[14 ]] Nevertheless, we must emphasize that a low SjVO2 not always equates with cerebral
anoxia, but can indicate an increase in oxygen extraction, which may be an early warning
sign of possible ischemia. In this study, none of the SjVO2 values were less than
lower normal limit (<50%).
Intravenous agents are considered to produce more brain relaxation compared to inhalational
anesthetic agents. In our study, brain relaxation was assessed by the blinded experienced
neurosurgeon not involved in study, after elevation of bone flap, using brain relaxation
grades of I to IV and it was observed that the degree of brain relaxation was comparable
in the two groups. Chui et al., in the meta-analysis and systematic review compared
intravenous with inhalational agents in 1819 patients and observed that sevoflurane
is as good a maintenance agent as propofol to maintain brain relaxation at <1MAC and
controlled ventilation.[[15 ]]
Hypotension is one of the major causes of secondary brain injury and poor outcome
in patients with TBI. In our study, the MAP in Group P was found to be comparable
to Group S (P = 0.569) at the start of surgery. After induction of anesthesia, there
was fall in MAP significantly in Group P compared to Group S throughout the surgery.
This decrease in MAP may be attributed to decrease in systemic vascular resistance
caused by propofol. In contrary to our observation, Sneyd et al. observed increased
number of hypotensive episodes in sevoflurane group compared to propofol.[[16 ]] They justified their findings by demonstrating greater depth of anesthesia in sevoflurane
group as compared to propofol group. In our study, the depth of anesthesia was maintained
by titrating the doses of anesthetic agent's propofol and sevoflurane at an entropy
value of 40–60. The end tidal carbon dioxide, temperature, and fraction of inspired
oxygen were similar in both the groups.
The total intravenous fluid used was significantly higher in propofol group than sevoflurane
group (P < 0.001) which might be explained by increased fluid requirement to maintain
MAP values in the normal range in the group receiving propofol.
IL-6 is a pro-inflammatory biomarker and in our study, we estimated the levels of
IL-6 in serum observing that the values were comparable between both the groups. Intragroup
comparison revealed that the levels of IL-6 decreased significantly in the sevoflurane
group at the end of surgery compared to the baseline (P = 0.040). This may be explained
by the anti-inflammatory effects of sevoflurane which might demonstrate a beneficial
neuroprotective effect in patients with TBI. A similar beneficial effect of sevoflurane
was also demonstrated in the study by Potočnik et al. where the authors observed decreased
levels of IL-6 with sevoflurane compared to propofol in patients undergoing one lung
ventilation for lung surgery, thus predicting its anti-inflammatory role.[[17 ]] Markovic-Bozic et al. studied effect of propofol and sevoflurane on the inflammatory
response of patients undergoing craniotomy.[[18 ]] They found that neither propofol nor sevoflurane had any significant impact on
the occurrence of postoperative complications and suggested to incite future studies
to prove a potential medically important anti-inflammatory role of propofol in neuroanesthesia.
Limitations
We did not measure SjVO2 continuously, and hence, we could have missed a few desaturation
or hyperemic episodes during the intraoperative period which could have better elucidated
the effects of intravenous or inhalational agents as real-time monitor, which would
have aided in an earlier prevention of secondary brain insults. And second, we did
not follow-up these patients to assess their long-term outcomes in relation to the
secondary insults, that they might have had during perioperative period and hospital
stay. Furthermore, we did not assess their long-term outcome in terms of cognitive
dysfunctions, neurological morbidity and mortality to evaluate neuroprotective potential
of propofol or sevoflurane.
Conclusions
The effects of propofol and sevoflurane used for maintenance of anesthesia are similar
in regards to cerebral oxygenation as measured by jugular venous oximetry and brain
relaxation scores. Significant reduction in MAP by propofol needs a caution in the
intraoperative period for its judicious and careful use. Hypotension should be prevented
at all costs as it might be detrimental to the overall outcome of these patients.
Decrease in level of IL-6 at the end of surgery compared to baseline values may suggest
a neuroprotective potential of sevoflurane which needs to explore further by larger
randomized controlled studies on patients with severe TBI.
