Introduction
Subarachnoid hemorrhage is a type of hemorrhagic stroke, whose incidence is associated,
in ∼ 80 to 85% of cases, with the rupture of an intracranial aneurysm, characterized
by an aneurysmal subarachnoid hemorrhage (aSAH).[1]
[2] Aneurysmal subarachnoid hemorrhage has a worldwide incidence of 9 per 100,000 individuals
per year, represents 5% of strokes, has high mortality and disability rates, and its
general prognosis depends on the volume of initial bleeding, the occurrence of rebleeding,
and the degree of delayed cerebral isquemia.[3]
Among the most important risk factors for the development of brain aneurysms are high
blood pressure, smoking, alcoholism, family history of the aneurysm in first-degree
relatives, and female gender. In addition to these, autosomal dominant polycystic
kidney disease is relevantly associated with the formation of aneurysms, and other
conditions such as Marfan syndrome, Ehlers-Danlos syndrome type IV, neurofibromatosis,
and fibromuscular dysplasia also have a weak association.[3]
As for the rupture, the location, size, and type of aneurysms are relevant to the
increased risk. Aneurysms > 1 cm and located in the posterior circulation, especially
at the top of the basilar artery, in the posterior cerebral artery, in the vertebrobasilar
distribution, and the origin of the posterior communicating artery, were associated
with a greater chance of rupture. Furthermore, saccular aneurysms account for 90%
of aneurysm morphology and correspond to the most common cause of aSAH.[3]
In addition to the acute effects resulting from aSAH, and the mortality rate of ∼
35%,[2] patients with aSAH may present secondary complications, which result in a worse
prognosis and are associated with neurological sequelae such as cognitive alteration,
and motor and/or behavioral deficit.[4] The main complications include rebleeding, hydrocephalus, vasospasm, cerebral edema,
and late cerebral ischemia.[5]
[6] Among them, early rebleeding is the most frequent in the first 24 hours, being associated
with mortality rates of 50 to 80%.[7]
Factors associated with worse clinical outcomes and incidence of complications after
aneurysmal subarachnoid hemorrhage are older age, higher grades on the World Federation
of Neurologic Surgeons (WFNS) scale at admission, larger aneurysms located in the
posterior circulation, intraventricular hemorrhage, hematoma intracerebral injection
and history of arterial hypertension, acute myocardial infarction, liver disease or
previous subarachnoid hemorrhage.[8] In addition, biomarkers and electrolytes correlated with an unfavorable evolution
in aSAH, such as the glucose-potassium ratio, the glucose-phosphate index, sodium
concentration, and plasma glucose, stand out.
Therefore, the present study aims to review the influence of glycemia and natremia
on the propensity to develop complications, the worst prognosis, and the risk of mortality
in patients with aneurysmal subarachnoid hemorrhage.
Material and Methods
This is an integrative literature review. The present study had as its guiding question:
“Do changes in blood glucose values or in plasma sodium concentration influence in-hospital
morbidity and mortality of patients with aneurysmal subarachnoid hemorrhage?” based
on the PECOS strategy: P
atient – people with aneurysmal subarachnoid hemorrhage; E
xposure – alterations in glycemia (hyper or hypoglycemia or glycemic variations) or plasma
sodium concentrations (hyper or hyponatremia); C
ontrol – Normoglycemic and with normal sodium levels; O
utcome – increased mortality; S
tudies – clinical trials, prospective and retrospective cohorts, case-control, and systematic
reviews were included.
Inclusion Criteria
As for the inclusion criteria, complete articles were selected for the present work,
published between 2017 and 2022, written in English, and belonging to the following
types of study: clinical trial, prospective or retrospective cohort, case-control
study, or systematic review. In addition, in their methodology, the studies should
have performed at least one measurement of blood glucose (in the form of mean glucose
per glycated hemoglobin or measurements of daily capillary glucose) or natremia of
patients hospitalized for aSAH. Finally, the selected articles should present a statistical
correlation between changes in glycemia or natremia with in-hospital mortality; or
else demonstrate differences in clinical presentations or prognosis in patients with
aSAH and glycemic and natremia alterations in comparison to aSAH patients with normoglycemic
and normal sodium levels.
Exclusion Criteria
The established exclusion criteria were duplicate publications on platforms and search;
publications inconsistent with the purpose of this research; publications with strong
biases that compromise the reliability of the article and publications in languages
other than English. In addition, case report studies, narrative reviews, and animal
experimentation. Studies that used glycemic monitoring by cerebral microdialysis,
whose focus of discussion was the drug treatments of glycemic alterations or natremia,
and articles that report on ventriculoperitoneal shunt were also excluded.
Search and Identification of Articles
The search for articles was performed on the PubMed platform, limiting the results
to articles published in the last 5 years (2017–2022) in English. To survey the articles,
the following descriptors were used: Glucose; Blood Glucose; Hyperglycemia; Hypoglycemia; Sodium; Hypernatremia; Hyponatremia;
Potassium; Hyperkalemia; Hypokalemia and their respective Entry Terms separated by the Boolean operator OR. Then we used
the Boolean AND operator to include the Aneurysmal Subarachnoid Hemorrhage descriptor and its Entry Terms.
Results and Discussion
General Description of Results
Based on the described methodology, 106 results were found. After analyzing the titles
and abstracts, 40 articles were included for a full reading. Of these, the following
were excluded: three because of the impossibility of accessing the complete material,
another three because they did not present clear relationships with the objective
of the research (glucose and sodium), in addition to three Keywordsget, is of factors
for shunt or ventriculoperitoneal shunt. Furthermore, one study for not stratifying
hypoglycemia, hypoxia, and hypotension as a cause for worse outcomes or mortality,
and another for not using exposure and control groups to assess the risk of hyponatremia.
It was also decided to remove two items that contained their investigations associated
with ischemic stroke and nonaneurysmal subarachnoid hemorrhage, which prevented their
adequate participation in the present study. Ultimately, one element was removed as
it focused on treatment impact. Thus, in the final analysis of the results, 26 articles
were considered, 14 referring to sodium ([Fig. 1]) and 12 to glucose ([Fig. 2]).
Fig. 1 Included articles referring to the prognostic value of sodium in aSAH.
Fig. 2 Included articles referring to the prognostic value of glucose in aSAH.
Sodium Change
Primarily, it should be noted that dysnatremia events and fluctuations in serum sodium
ion values are extremely prevalent in patients with aSAH. The main event is hyponatremia,
present in ∼ 30 to > 50%[9] of the patients, followed by hypernatremia found between 31 and 33.6% of the cases
and variations in sodium values in ∼ 39.3% – considering the value of 12 mmol/L.[10]
Regarding prognosis, the role of such an ion is complex. Studies indicate that both
hypernatremia and hyponatremia, or even variation in serum sodium concentration, are
more common in patients with unfavorable evolution – including deaths, neurological
sequelae, or a higher level of dependence – in-hospital or in later periods of up
to 6 months.[10]
[11]
[12]
However, the assessment of independent predictors for patient outcomes is more restrictive.
The main factor described is hypernatremia, in which there are correlations of values > 145 mmol/L
in the first 2 weeks with higher mortality during hospitalization or with a worse
prognosis, in general, when considering the value of 146 mmol/L.[11] Specifically, Sokół et al.,[13] established that values > 155 mmol/L were an independent predictor of mortality,
with a specificity of 97.8% and a sensitivity of 47.6%. Furthermore, concentrations > 155 mmol/L
were associated with unfavorable outcomes within 3 months.[10] More broadly, a study showed that sodium in the blood, in addition to rebleeding,
is an independent risk factor, leading to poor prognosis.[14]
The role of hyponatremia as a predictor is controversial. In most articles, it is
demonstrated that it has no prognostic value in the short or long term, especially
in terms of mortality.[10]
[11]
[15]
[16] Kieninger et al.[17] even demonstrated that the rate of poor outcome at discharge from the intensive
care unit (ICU), 6 months after the bleeding event, was significantly lower in patients
with moderate hyponatremia (125–129 mmol/L), allowing only a limited conclusion, as
the diagnosis of hyponatremia regularly led to early and elaborate measures to achieve
rapid normalization of the sodium level and maintain normonatremia in the later course
of ICU treatment. A pathophysiological hypothesis proposed by Tam et al.[10] would be the ability of brain neurons to adapt to the situation of hyponatremia,
in a self-regulation mechanism, managing to reach a state of functional stability
in less than 24 hours.
In contrast, Rumalla et al.[18] and Escamilla-Ocañas et al.[9] demonstrated that hyponatremia was significantly associated with cerebral vasospasm,
a complication that very often precedes aSAH events.[9]
[18] Escamilla-Ocañas et al.[9] also reported that poor clinical results and longer hospital stays and ICU were
significantly more evident in the hyponatremic group, compared with the normonatremic
group. Such data are, in part, supp partlyidwan et al. (2019),[15] who found a correlation between the development of hyponatremia at any time during
hospitalization and longer duration of hospital stay; however, with no impact on ICU
stay.
Furthermore, some authors, when considering specific periods, such as the decrease
in sodium between 14 and 21 days after hospitalization[15] or values below the usual levels, such as 132 mmol/L,[11] found worse outcomes related to hyponatremia within 1 year or at hospital discharge,
respectively. Another impacting factor that can be considered is the treatment of
such hydroelectrolytic conditions, which usually receive a more aggressive treatment
of fluid replacement.[10]
Cohen et al.[19] mention that there are several possible explanations for the discrepancy observed
in the influence of hyponatremia on morbidity and mortality from aSAH, including the
use of variable outcome measures, with only a few studies reporting the long-term
neurological status; differences in the definition of hyponatremia (which can be categorized
as present or absent or by varying threshold values and a varying number of samples)
and reports from analyzes and models that do not take into account duration or severity
(for example, temporal changes over an admission).
In turn, the variation in serum sodium concentration is increasingly an important
item to be evaluated in patients with aSAH. Increased sodium variability was associated
with a longer hospital stay[19] and, between the 1st and 3rd days, it was associated with higher in-hospital mortality.[16] Also, Tam et al.[10] detected that a variation > 12 mmol/L of sodium during hospitalization could be
associated with a worse patient outcome, represented by states 1 to 3 of the Glasgow
Outcome Scale in the 3 months following aSAH. Furthermore, within[14] days of an aSAH episode andan6 months of hospital discharge, it has been associated
with the development of symptomatic vasoespasm[20] and poor neurological outcome,[19] respectively, with progressive decreases in serum sodium being found to precede
such complications.
It is speculated that this fluctuation is caused by a response to hormones, such as
those related to stress, which begins to show inappropriate secretion after brain
damage associated with aSAH.[19]
[20]
Furthermore, Yoshikawa et al.,[21] when analyzing patients with low-grade aSAH who were ≥ 70 years old and who received
treatment, demonstrated that the mean absolute daily difference in the normal plasma
sodium level was significantly associated with the modified Rankin Scale scores in
3 and 12 months after aSAH.
Conversely, a study showed that there was no significant difference in serum sodium
levels, over the first 14 days post-aSHA, in patients who later developed vasospasm,
compared with those who did not, and that, in terms of outcomes neurological, functional,
and mortality factors, changes in sodium levels over time were not associated with
these outcomes.[22]
Finally, some results support the use of tests referring to natremia more comprehensively,
since serum sodium values can be considered as an independent factor for mortality[16] and its variations as predictors of a worse neurological condition at hospital discharge
and within 6 months.[10]
[11]
Glucose Change
At the time of aneurysm rupture, it may result in an increase in glucose within 72 hours
after the onset of bleeding.[7] As a result, in addition to hyponatremia events in individuals with aSAH, as already
mentioned, we may have some changes in glucose.
In the study by Jia et. al,[23] it was described that 4,804 (70.9%) patients who suffered a spontaneous subarachnoid
hemorrhage had hyperglycemia. Of these, a higher in-hospital mortality rate was identified
for patients with more severe hyperglycemia in whom the odds of in-hospital mortality
were higher, significantly higher in patients with moderate hyperglycemia (odds ratio
[OR]: 2.61; 95% confidence interval [CI]: 1,52–3.06) and higher in patients with severe
hyperglycemia (OR: 3.18; 95% CI 2.24–4.53;
p < 0.001). The mortality of these patients may be related to changes in blood glucose.
Reinforcing this analysis, it was also identified in the study by Sun et. al[24] with 119 patients with aSAH who had a high admission glycemic interval (aGG) ≥30 mg/dL
(66.4%), is associated with markers of disease severity and hospital outcomes, strengthening
the concept that is an indicator of physiological response to stress to aSAH. Furthermore,
aGG outperformed admission glucose in predicting in-hospital mortality and was equally
accurate in the discerning poor composite outcome.
It is also notable that discussions in works regarding glucose variation and its relationship
with the occurrence of vasospasm.[25]
[26] The study by Matano et al[25] including 333 patients, made a statistical correlation between ischemia due to cerebral
vasospasm and glucose/potassium ratio (p < 0.0001), glucose (p = 0.016), and potassium (p = 0.0017). The glucose/serum potassium ratio was elevated in cerebral vasospasm (Spearman
r = 0.1207; p = 0.0279).
Brief-form vasospasm is a common complication after aSAH and is a major contributor
to the high morbidity and mortality rate of the disease. The pathophysiology of vasospasm
is not well understood and probably involves an interaction between blood products,
vasoactive substances, and inflammatory cascades.[27]
In addition to the glucose/potassium ratio, the serum glucose-phosphate index is a
potential marker of severity and poor outcomes for patients with aSAH.[28] Higher blood glucose levels were identified in patients with rebleeding, in addition
to patients in the group of rebleeding who had a significantly higher glucose/potassium
ratio than patients without rebleeding.[12] Furthermore, Zhang et al[29] reported that the glucose-phosphate ratio was significantly correlated with vasospasm
(r ¼: 0.581; p < 0.001) and DCI (r ¼: 0.523; p < 0.001), resulting in an unfavorable prognosis.
As evidenced, high blood glucose levels on admission are associated with aSAH severity
and worse evolution. A recent study addressed that cerebral vasospasm exacerbated
by hyperglycemia may be a potential mechanism for the poor neurological outcomes observed.[29]
The correlation of hyperglycemia with a poor prognosis has several scientific explanations;
some experimental studies indicate that hyperglycemia induces apoptosis, while others
claim that hyperglycemia increases the production of superoxide, damaging the blood-brain
barrier, causing cerebral edema. Another explanation is that hyperglycemia impairs
different components of innate immunity, leading to a systemic anti-inflammatory response.[20] Another explanation would be about cerebral vasospasm exacerbated by hyperglycemia
being a potential mechanism for poor neurological outcomes.[29]
On the other hand, the occurrence of hypoglycemia in patients with aSAH is associated
with unfavorable neurological outcomes and risk of vasospasm.[30]
[31] Therefore, glucose variability, both hyper and hypoglycemia, may be correlated with
hospital mortality or with a poor prognosis in the long term in patients with aSAH.[32]
Studies identified that previous hyperglycemia (diabetes mellitus [DM]) HAS does not
seem to affect the neurological status from admission or the outcome at 6 months.
However, hyperglycemia affects these elements, as it is probably a reflection of an
acute brain injury.[33]
[34] Therefore, this information suggests that the unfavorable prognosis is more related
to post-aSAH hyperglycemia, but more studies should be performed for the discussion
of the relationship between the prognosis of patients who have DM and who have suffered
HAS.
Limitations
The present study has several limitations. Primarily, as this is secondary research,
the reliability of the information depends on the quality of the primary data. To
minimize this bias, the authors tried to stick to the methodology of each work, to
include in the results only items with technical quality and scientific rigor. Second,
due to the adopted style of integrative and nonsystematic review, certain studies
may have escaped the scope of the evaluation. However, it is worth mentioning that
the objective is to obtain and synthesize the most recent evidence on the subject
and, for this, the recommendations for care in methodological preparation and selection
criteria were followed.
In addition, the articles included distinctions in terms of location, sample number,
laboratory evaluation method, time of collection, and the scale for analyzing the
result – such as the Glasgow Outcome Score and the Modified Rankin Scale, among others.
However, the authors chose not to be very judicious about the complementary test used
– provided that at least one measurement of serum sodium or blood glucose was performed
– or the assessment scale to allow gathering the largest possible sample size and
summarizing the current understanding of the impact of glycemia and natremia in mortality
and neurological outcome of patients after aSAH. Thus, the present findings must be
evaluated with caution, and systematic reviews and future meta-analyses will be necessary
to determine with more precision the correct period to request laboratory evaluation
and the values with greater influence on the prognosis.
