Keywords
ABO blood group - subarachnoid hemorrhage - cerebral vasospasm - delayed cerebral
ischemia - Pakistan
Introduction
Subarachnoid hemorrhage (SAH) refers to the extravasation of blood into the subarachnoid
space between the arachnoid membrane and the pia mater, which normally contains cerebrospinal
fluid. In 2010, the crude global incidence of SAH was 6.1 per 100,000 person-years.[1] Even though the mortality rate for SAH has dropped from 50% to 35% over time,[2] it is still considered a life-threatening condition and a neurological emergency.
Ruptured intracranial aneurysms (aSAH) account for approximately 80% of SAH cases.[3] aSAH has a mortality rate of 25 to 50%, and half of those who survive, suffer from
long-term neuropsychological effects and decreased quality of life.[4] Cerebral vasospasm after an interventional or surgical procedure to secure the aneurysm
leads to delayed cerebral ischemia (DCI), which is the primary source of this morbidity
and mortality in patients of aSAH. Risk factors for DCI may include being female,
a Hunt and Hess grade of 4 to 5, modified Fisher grade (mFG), and a World Federation
of Neurosurgical Societies grade of IV to V.[5]
[6] Smoking, pre-existing hyperglycemia, pre-existing hydrocephalus, early systematic
inflammatory syndrome, a history of diabetes mellitus, and procedure- and patient-related
factors such as medication on admission, postoperative steroid use, and patient age
have been previously assessed as predictors of vasospasm.[7]
[8]
In addition to these factors, the ABO blood group system may also have a role in the
onset of DCI. Ever since ABO blood groups were discovered in the early 1900s, various
associations between blood type and diseases such as cancer, cardiovascular diseases,
hematologic disorders, metabolic disorders, cognitive disorders, and circulatory diseases
have been found,[9] and this is attributed to the variation in hemostaseologic properties of all four
blood group subtypes. Plasma vWF levels are approximately 25% lower in O individuals
than non-O individuals,[10] which may lead to excessive bleeding as vWF is crucial for the action of coagulation
factor VII. Data also show that SAH, followed by vasospasm, resulted in a 40% decrease
in NO secretion by eNOS.[11] NO works essentially as a vasodilator relaxing the smooth muscle and inhibiting
platelet aggregation, and is, therefore, significant for maintaining the cerebrovascular
tone. A possible relationship between endothelial ABO antigen and eNOS includes a
genetic linkage with other genes regulating vasoconstriction, vasodilation, or a direct
association of blood type antigen and vascular reactivity may also exist.[12]
[13]
The role of the ABO blood group as a risk factor for hemorrhage has been demonstrated
by prior studies.[14] Previously, the AB blood type has been positively associated with a higher risk
of ischemic stroke.[15] Similarly, the blood group AB has been documented to have a higher presence in non-aneurysmal
SAH than aSAH (OR 2.45) within an ethnically German cohort.[16] However, there is disagreement in the scarce literature evaluating the role of the
ABO blood group in DCI and clinical outcomes following aSAH, with one study demonstrating
no link,[17] and two others studies showing patients with blood group O[18] and those with blood group A[19] to have a higher incidence of DCI.[18]
[19] A large difference between these studies and possibly the reported results may be
a geographic or ethnic variation amongst the cohorts of these two studies, the first
being White German,[17] and another being a Hispanic and African American urban minority in New York.[18] Similarly, the which showed a higher incidence of DCI following aSAH in patients
with blood group A was carried out in England.[19] Our study then aimed to add to this debate and explore whether an association exists
and if yes, what is the association between the ABO blood group and the development
of DCI and subsequent functional outcomes after aSAH.
Materials and Methods
Study Design and Participants
This study was conducted at the Department of Neurosurgery, Allama Iqbal Medical College,
Jinnah Hospital Lahore, Pakistan. Ethical approval for this study was obtained from
our institutional review board. A retrospective analysis was made on data collected
from consecutive patients presenting with aSAH at a tertiary supraregional neurosurgery
department over 5 years. All patients were operated on by the senior author (Ashraf,
N). The participants were divided into two groups according to their ABO blood group
type: “O” type and “non-O” type.
The main inclusion criteria were all patients over the age of 18 years with aSAH.
In addition, patients with non-aneurysmal etiology and incomplete follow-up or clinical
information were excluded from the study.
Variables and Outcomes
The key exposure was ABO blood group type classified into either “O” type or “non-O”
type, similar to Unda et al.[18] Other variables included in our study were patient demographics (gender, age, BMI,
hypertension, and smoking status), clinical variables (DCI, mFG, WFNS grade, and Glasgow
Outcome Scale at discharge), EVD placement, and size of the aneurysm (< 5 mm or > 5 mm).
All patients underwent neurosurgical clipping as a treatment modality. Forty-six of
the aneurysms were part of the posterior circulation, and the rest were part of the
anterior circulation. Posterior circulation aneurysms were excluded as they were transferred
to another unit for endovascular treatment; thus, we did not have their data. After
an aSAH diagnosis, DSA was performed on all patients. All patients were assessed at
hospital discharge.
The primary outcome of our study was the development of DCI. The clinical manifestation
of vasospasm is referred to as DCI, which is defined as and accepted to be, the presence
of a focal neurological impairment and/or a decrease in consciousness lasting for
at least an hour with no other apparent cause.[18]
[19]
[20] All patients with clinically suspected vasospasm underwent CT head. The formal definition
for clinical deterioration has been previously defined as “the occurrence of focal
neurological impairment (such as hemiparesis, aphasia, apraxia, hemianopia, or neglect),
or a decrease of at least 2 points on the Glasgow Coma Scale (either on the total
score or on one of its components [eye, motor on either side, verbal]). This should
last for at least 1 hour, is not apparent immediately after aneurysm occlusion, and
cannot be attributed to other causes through clinical assessment, CT or MRI scanning
of the brain, and appropriate laboratory studies.”[20]
All patients had clinical and neurological examinations performed by the neurosurgical
team throughout their stay and at discharge. Functional outcome at discharge, as measured
by GOS was our secondary outcome. GOS 4 to 5 was taken as a “good outcome,” and GOS
1 to 3 was taken as a “bad outcome.”
Statistical Analysis
Analysis was conducted using SPSS version 26.0 (IBM SPSS Statistics, New York, United
States). Descriptive statistics were calculated for all variables included in our
study. Associations of the variables with the ABO blood group type and DCI were explored
by Pearson's chi-square test (or Fisher's exact test if necessary) for categorical
variables and Student's t-test for continuous variables. Chi-square continuity correction was applied where
appropriate. Any variable found to be significantly associated with both ABO group
type and DCI (p < 0.05) was included as a potential confounder in the multivariable logistic regression
model used to calculate the effect of ABO type on the development of DCI. Moreover,
all variables found to be significantly associated with DCI (p < 0.05) were also chosen to be included in the model to identify independent predictors
of outcome (DCI). Forward selection was used to arrive at the final stepwise logistic
regression. A variable was only included in the model if it improved the model checked
by the Akaike information criterion (AIC),[21] and if the Omnibus likelihood ratio test was significant. Odds ratio (OR) and 95%
confidence intervals (CIs) were calculated. A p-value less than 0.05 was considered significant.
Results
Participants and Descriptive Data
A total of 175 patients were included in our study. There was an approximately equal
number of males and females. The majority of patients were nonsmokers (including former
smokers) (59.4%), hypertensive (59.4%), and the mean age was 53.59 years (SD 11.26)
and a mean BMI of 28.77 kg/m2. The frequency distribution of “O” type and “non-O” type blood groups was 34.3% and
65.7%, respectively. Fifty-one patients (29.1%) needed EVD placement, and 26 patients
(14.9%) suffered from DCI in our study ([Table 1]).
Table 1
Demographic and clinical characteristics of patients after aSAH
|
Parameter
|
Count (n = 175)
|
n (%)
|
|
Gender
|
|
Male
|
86
|
49.1
|
|
Female
|
89
|
50.9
|
|
Age (mean and SD)
|
53.59 (mean)
|
11.26 (SD)
|
|
BMI (mean and SD)
|
28.77 (mean)
|
2.29 (SD)
|
|
ABO type blood group
|
|
“O” type
|
60
|
34.3
|
|
“Non-O” type
|
115
|
65.7
|
|
Smoker
|
|
Current
|
71
|
40.6
|
|
Non-smoker (including former)
|
104
|
59.4
|
|
Hypertension
|
|
Yes
|
104
|
59.4
|
|
No
|
71
|
40.6
|
|
Delayed cerebral ischemia (DCI)
|
|
No
|
149
|
85.1
|
|
Yes
|
26
|
14.9
|
|
Modified Fisher grade (mFG)
|
|
Grade 0–2
|
54
|
30.9
|
|
Grade 3–4
|
121
|
69.1
|
|
WFNS grade
|
|
Grade 1–2
|
120
|
68.6
|
|
Grade 3–5
|
55
|
31.4
|
|
Outcome at discharge (Glasgow Outcome Scale)
|
|
GOS 4–5 (good outcome)
|
122
|
69.7
|
|
GOS 1–3 (bad outcome)
|
53
|
30.3
|
|
EVD placement
|
|
No
|
124
|
70.9
|
|
Yes
|
51
|
29.1
|
|
Size of aneurysm
|
|
< 5 mm
|
51
|
29.1
|
|
> 5 mm
|
124
|
70.9
|
Association of Demographic and Clinical Variables with ABO Type Groups
[Table 2] shows the distribution of variables across the “O” type and “non-O” type groups.
The mean BMI was significantly higher in the “non-O” type group (29.07 ± 2.24) than
in the “O” type group (28.20 ± 2.28). In addition, the “O” type group had a significantly
higher incidence of DCI (25.0%) as compared with the “non-O” type group (9.6%). No
other variables had any significant association with the ABO type groups.
Table 2
Distribution of demographic and clinical variables between ABO type groups
|
Parameter
|
ABO type blood group
|
p-Value
|
|
“O” type
|
“Non-O” type
|
|
Count (n = 60)
|
n (%)
|
Count (n = 115)
|
n (%)
|
|
Gender
|
1.000
|
|
Male
|
29
|
48.3
|
57
|
49.6
|
|
Female
|
31
|
51.7
|
58
|
50.4
|
|
Age (mean and SD)
|
51.57 (mean)
|
12.09 (SD)
|
54.65 (mean)
|
10.70 (SD)
|
0.085
|
|
BMI (mean and SD)
|
28.20 (mean)
|
2.28 (SD)
|
29.07 (mean)
|
2.24 (SD)
|
0.016*
|
|
Smoker
|
0.707
|
|
Current
|
26
|
43.3
|
45
|
39.1
|
|
Non-smoker (including former)
|
34
|
56.7
|
70
|
60.9
|
|
Hypertension
|
0.707
|
|
Yes
|
34
|
56.7
|
70
|
60.9
|
|
|
No
|
26
|
43.3
|
45
|
39.1
|
|
|
Delayed cerebral ischemia (DCI)
|
0.012*
|
|
No
|
45
|
75.0
|
104
|
90.4
|
|
|
Yes
|
15
|
25.0
|
11
|
9.6
|
|
|
Modified Fisher grade (mFG)
|
0.727
|
|
Grade 0–2
|
17
|
28.3
|
37
|
32.2
|
|
|
Grade 3–4
|
43
|
71.7
|
78
|
67.8
|
|
|
WFNS grade
|
0.573
|
|
Grade 1–2
|
39
|
65.0
|
81
|
70.4
|
|
|
Grade 3–5
|
21
|
35.0
|
34
|
29.6
|
|
|
Outcome at discharge (Glasgow Outcome Scale)
|
0.645
|
|
GOS 4–5 (good outcome)
|
40
|
66.7
|
82
|
71.3
|
|
|
GOS 1–3 (bad outcome)
|
20
|
33.3
|
33
|
28.7
|
|
|
EVD placement
|
1.000
|
|
No
|
43
|
71.7
|
81
|
70.4
|
|
|
Yes
|
17
|
28.3
|
34
|
29.6
|
|
|
Size of aneurysm
|
1.000
|
|
< 5 mm
|
17
|
28.3
|
34
|
29.6
|
|
|
> 5 mm
|
43
|
71.7
|
81
|
70.4
|
|
* p < 0.05; ** p < 0.01; *** p < 0.001.
Association of Demographic and Clinical Variables with DCI
[Table 3] shows that the mean BMI was significantly lower in the patients that suffered DCI.
A significantly larger proportion of patients with DCI had higher grades on the modified
Fisher scale (grades 3–4) and WFNS scale (grades 3–5) than those who did not suffer
from DCI. In addition, there was a significantly higher number of EVD placements in
patients with DCI (61.5%) than in those who did not (23.5%). The rest of the variables
did not show any statistically significant results.
Table 3
Association of demographic and clinical variables with the incidence of DCI
|
Parameter
|
Delayed cerebral ischemia (DCI)
|
p-Value
|
|
No
|
Yes
|
|
Count (n = 149)
|
n (%)
|
Count (n = 26)
|
n (%)
|
|
Gender
|
0.759
|
|
Male
|
72
|
48.3
|
14
|
53.8
|
|
|
Female
|
77
|
51.7
|
12
|
46.2
|
|
|
Age (mean and SD)
|
53.87 (mean)
|
11.43 (SD)
|
52.04 (mean)
|
10.31 (SD)
|
0.447
|
|
BMI (mean and SD)
|
28.94 (mean)
|
2.31 (SD)
|
27.81 (mean)
|
1.88 (SD)
|
0.019*
|
|
Smoker
|
0.375
|
|
Current
|
63
|
42.3
|
8
|
30.8
|
|
|
Non-smoker (including former)
|
86
|
57.7
|
18
|
69.2
|
|
|
Hypertension
|
1.000
|
|
Yes
|
89
|
59.7
|
15
|
57.7
|
|
|
No
|
60
|
40.3
|
11
|
42.3
|
|
|
Modified Fisher grade (mFG)
|
0.037*
|
|
Grade 0–2
|
51
|
34.2
|
3
|
11.5
|
|
|
Grade 3–4
|
98
|
65.8
|
23
|
88.5
|
|
|
WFNS grade
|
0.001**
|
|
Grade 1–2
|
110
|
73.8
|
10
|
38.5
|
|
|
Grade 3–5
|
39
|
26.2
|
16
|
61.5
|
|
|
Outcome at discharge (Glasgow Outcome Scale)
|
0.772
|
|
GOS 4–5 (good outcome)
|
105
|
70.5
|
17
|
65.4
|
|
|
GOS 1–3 (bad outcome)
|
44
|
29.5
|
9
|
34.6
|
|
|
EVD placement
|
< 0.001***
|
|
No
|
114
|
76.5
|
10
|
38.5
|
|
|
Yes
|
35
|
23.5
|
16
|
61.5
|
|
|
Size of aneurysm
|
0.971
|
|
< 5 mm
|
44
|
29.5
|
7
|
26.9
|
|
|
> 5 mm
|
105
|
70.5
|
19
|
73.1
|
|
* p < 0.05; ** p < 0.01; *** p < 0.001.
ABO Type Blood Group as a Risk Factor for DCI
A stepwise logistic regression model showed that after adjusting for BMI, mFG, WFNS
grade, and EVD placement, “O” type blood group was an independent risk factor for
DCI, greatly increasing the risk of DCI as compared to “non-O” type groups, OR = 3.27,
95% CI: 1.21–8.82, ([Table 4]). In addition, BMI was a potential confounder ([Tables 2] and [3]), and the rest of the variables included in the multivariable analysis were significant
predictors of DCI ([Table 3]).
Table 4
Logistic regression with DCI as dependent variable
|
Predictor
|
Unadjusted OR (95% CI)
|
Adjusted OR (95% CI)
|
|
ABO type blood group
|
|
“Non-O” type
|
Referent
|
Referent
|
|
“O” type
|
3.15 (1.34–7.40) p = 0.008
|
3.27 (1.21–8.82) p = 0.019
|
Discussion
This study explored the relationship between ABO blood type and the onset of DCI,
linking DCI onset significantly and positively with the presence of “O” blood type.
Our study population had 34.3% individuals with “O” blood type and 65.7% with “non-O”
blood type. It is also important to state that there were no statistically significant
differences among baseline characteristics between “O” and “non-O” patients except
for BMI.
After adjusting the BMI, mFG, WFNS grade, and EVDs by regression, ABO blood type was
an important independent predictor for DCI. According to a retrospective analysis
by Dubinski et al,[16] AB blood type is more frequent in non-aneurysmal subarachnoid hemorrhages than in
aneurysmal etiology patients.[16] However, the same authors also found no significant influence of ABO blood type
on cerebral vasospasm onset, SAH-associated intracerebral hemorrhage, or delayed infarction.[17] But their results were suggestive of a potential clinically relevant association
of the ABO blood group and hence warranted further investigation. A study carried
out by Richards et al[19] demonstrates that blood group A patients have a significantly higher risk of developing
DCI following aSAH, and a suggested trend of patients with non-O blood type at a higher
risk of developing DCI as well.[19] However, Unda et al's[18] study conflicted with this, demonstrating that O blood type significantly increases
the risk of DCI onset with an odds ratio 2.5 times higher relative to non-O blood
group patients.[18] However, this finding was in an ethnic minority of Hispanics and African Americans
living in New York.[18] Unda et al[18] also document non-statistically significant poorer functional outcomes in those
with O blood group.[18] Our study also documents a 3.15 increase in the odds of developing DCI for those
with 'O' blood group.
Important differences to consider include that the study by Unda et al[18] was from New York and had only 11.3% Caucasians, with mainly a cohort of Hispanic
patients. Richards et al.[19] had an 87.4% Caucasian population from England, whilst our cohort was entirely Pakistani
patients. Literature has shown that racial differences can be significantly influential
in pathologies though their role in individual diseases and outcomes are likely to
vary substantially and thus should be further investigated in specific cases.[22]
[23]
[24] For example, it has previously been proposed that race-specific ABO antigens may
also lead to a disparity in the rates of aneurysms and intraventricular hemorrhages,[25]
[26] therefore while DCI onset may be associated with the ABO blood group, further investigations
and better study methodology are required to answer this with certainty whilst accounting
for differences in race and geography. Equally important to account for are treatment
paradigms, which for aneurysm securing varied amongst our study, and that of Richards
et al[19] who reported endovascular intervention in 89.6% of their cohort. Unda et al[18] cohort underwent neurosurgical clipping in 33.8% of cases and endovascular coiling
in 66.2% of cases. Treatment paradigms for neurovascular surgery differ between centers
in the same country, let alone between continents; this is also partly due to socioeconomic
reasons. In our setup, all patients undergo neurosurgical clipping due to the lack
of affordability for endovascular coils in government sector neurosurgery departments.
Therefore, the findings of these studies, including ours, may not be generalizable
across demographics.
Richards et al's[19] study, which showed blood group A to be associated with a higher risk of DCI after
aSAH, hypothesized that this could be contributed to by the finding that blood group
A patients have a higher risk of developing hyperlipidemia, atherosclerosis, and heart
failure.[19]
[27] They postulated that the presence of atherosclerotic disease in the cerebral blood
vessels, even if subclinical or insignificant may contribute to the vasospasm as these
patients would be less likely to tolerate hypoperfusion relative to their O or B blood
group counterparts.[19] Richards et al[19] also postulated that the relative degree of vasospasm and hypoperfusion required
to induce clinically significant cerebral ischemia might be less if the cerebral vasculature
already has pre-existing atherosclerotic plaques causing stenosis.[19] Similarly, Unda et al,[18] who found blood group O patients, have a higher risk of DCI following aSAH, support
their results by stating evidence that reduced levels of vWF and factor 8 in plasma
of blood group O patients lead to excessive bleeding in the subarachnoid space. However,
these authors acknowledged the converse evidence that shows increased levels of factor
8 contribute to peripheral vasospasm[28]
[29]
[30] though Unda et al[18] postulate this evidence may not be translatable to cerebral vasculature after aSAH.
Both theories are thought-provoking and provide a solid rationale for studying this
topic in further detail. Whether the factor that causes disparity in different ethnic
groups in regard to the ABO blood group system and DCI is genetic, environmental or
both, works in isolation or conjunction with the type of the blood group in particular
ethnicities and/or environments needs to be evaluated in the laboratory. This is vital
to address the recent question of the role of the ABO blood group in aSAH. It will
add to our knowledge of aSAH pathophysiological sequelae and, at large, the role of
the ABO blood group in vascular disorders.
Individuals with “O” blood type have lower mean levels of vWF followed by A, B, and
AB blood types. vWF is a highly glycosylated protein whose properties such as platelet
agglutination and susceptibility to proteolytic degradation are affected by removing
carbohydrate residues.[31] A, B, and H oligosaccharide structures have also been identified on the N-linked
glycan chains of human plasma vWF.[32]
[33] This gives rise to the plausibility of blood group determinants playing a role in
the processing, catabolism, or release of vWF, affecting its plasma concentration.[31] The odds ratio between certain vascular diseases and blood group types are also
similar to those predicted by the effect of ABO(H) on vWF levels.[12] The variation in plasma concentrations of the vWF–FVIII complex is clinically significant
because low levels increase the risk of deep vein thrombosis, peripheral vascular
disease, and myocardial infarction.[34]
Similarly, these low levels may also lead to excessive bleeding in the subarachnoid
space. A potential, though statistically insignificant, trend for an increased rebleeding
risk has also been observed by Richards et al[19] in patients of aSAH with the O+ blood group.[19] This bleeding may create an environment with excessive amounts of hemoglobin and
oxidized hem products that intercept high-affinity free NO rapidly,[35] and synthesize endogenous eNOS inhibitors.[12] The role played by endothelial nitric oxide synthase (eNOS), which regulates NO
production, is crucial in maintaining cerebrovascular tone. Low basal NO levels constantly
being released contribute to vascular relaxation and thrombosis. Due to eNOS inhibitors,
a decreased amount can lead to the inhibition of this relaxation, which may further
lead to cerebral vasospasm.[17]
In our study, significant factors linked with DCI with a p-value less than 0.05 apart from ABO blood type include BMI, mFG, WFNS grading system,
and EVD placement. According to CDC, BMI between 18.5 and less than 25 is considered
normal; BMI between 25 and less than 30 is considered overweight; BMI greater than
30 is considered obese. The overweight and obese groups may have an inverse relation
with mortality and functional outcome after SAH, partially supporting the obesity
paradox. Being overweight lends a protective effect after SAH, whereas obesity is
detrimental to death and poor outcome.[36] This may be because an extra amount of nutrients and is helpful until the harms
of obesity override these benefits. Some studies, however, directly link high BMI
with an increased risk of infarction.[37] In our study, the mean BMI of patients with and without DCI was 27.81 and 28.94,
respectively. This value may seem statistically significant but likely holds no clinical
significance as both fall in the same overweight category.
The mFG accounts for thick cisternal and ventricular blood. The risk for symptomatic
vasospasm increases progressively from grade 0 to 4 because it is directly linked
to the worsening of the aneurysm. For each increase in scale level, the odds of symptomatic
deterioration increased by almost 28%.[38] The WFNS grading system is the most effective grading system for the clinical condition
of SAH, primarily based upon the GCS and the presence of focal neurological deficits[39]; as the scale moves from grade 1 to 5, the severity of SAH increases as the GCS
score decreases and the presence of focal neurological deficits increases. This further
leads to an increased chance of vasospasm and, ultimately, DCI. The results of our
study, where 38.5% of patients with WFNS grades 1 to 2 and 61.5% of patients with
WFNS grades 3 to 5 suffered from DCI, support this. EVDs are placed to manage hydrocephalus
and intracranial pressure after subarachnoid hemorrhage by draining CSF.[40]
The percentage of patients discharged with a poor functional outcome was higher in
the O group (33.3%) versus the non-O group (28.7%). This, however, did not reach statistical
significance similar to Unda et al's study. Functional clinical outcomes and other
non-significant variables might be influenced by clinical, economic, socio-demographic,
and other factors. Therefore, ABO blood type would likely not singlehandedly impact
functional recovery after aSAH. However, it is associated with DCI, the most common
complication of aSAH and thus may play a role.
Limitations
The single-center retrospective nature of this study limits our findings as it may
lead to the inclusion of biases and potential confounders. The absence of data on
potential confounders might take away from the opportunity to investigate their associations.
Given the conflicting findings of our study, Unda et al[18] and Richards et al[19] study, the association between the ABO blood groups with DCI may be coincidental.
Further, a large confidence interval in our study adds to the uncertainty of our results.
The scarcity of literature to compare with is also a limitation. Individual studies
investigating this topic are limited by the small sample size, especially given that
the primary outcome has a low incidence. Multicenter data analysis would be an ideal
method to address this limitation. Larger data sets with different modalities to secure
the aneurysm would be beneficial to study if method of aneurysm occlusion (microsurgical
clipping versus endovascular coiling) is a contributing factor toward the development
of DCI. Another limitation in our study was that all aneurysms were of the anterior
circulation. This is because there is a single government-funded neurosurgery department
in our city where endovascular treatment is provided free of cost, and thus posterior
circulation aneurysms are referred to them. Thus, analyzing only anterior circulation
aneurysm data adds an important selection bias and again limits generalizability.
In addition, data for long-term outcomes after discharge has not been included in
this study. However, this study can still serve as a source of pilot data. Given that
the associations between ABO blood group and DCI onset following aSAH have been demonstrated
by two other studies, there is a strong rationale to investigate this topic further.
Conclusion
This study demonstrates an association between blood group O and a relative increase
in the development of DCI following aSAH. Given the limitations of our study and the
sparsity of relevant literature, there is a need to investigate the role of the association
between the ABO blood group system and DCI development following aSAH.
