Systemic Inflammation Response Index and Systemic Immune-

inflammation Index for Predicting the Prognosis of Patients with

Aneurysmal Subarachnoid Hemorrhage

Seonyong Yun, MD,* Ho Jun Yi, MD,† Dong Hoon Lee, MD,‡ and
Jae Hoon Sung, MD‡

Objectives: Inflammatory response plays a pivotal role in the progress of aneurysmal

subarachnoid hemorrhage (aSAH). As novel inflammatory markers, systemic
inflammation response index (SIRI) and systemic immune-inflammation (SII) index
could reflect clinical outcomes of patients with various diseases. The aim of this
study was to ascertain whether initial SIRI and SII index were associated with prog-
nosis of aSAH patients. Methods: A total of 680 patients with aSAH were enrolled.
Their prognosis was evaluated with modified Rankin Scale (mRS) at 3 months, and
unfavorable clinical outcome was defined as mRS score of 3-6. Receiver operating
characteristic (ROC) curve analysis was performed to identify cutoff values of SIRI
and SII index for predicting clinical outcomes. Univariate and multivariate regres-
sion analyses were performed to explore relationships of SIRI and SII index with
prognosis of patients. Results: Optimal cutoff values of SIRI and SII index to dis-
criminate between favorable and unfavorable clinical outcomes were 3.2 £ 109/L
and 960 £ 109/L, respectively (P < 0.001 and 0.004, respectively). In multivariate
analysis, SIRI value  3.2 £ 109/L (odds ratio [OR]: 1.82, 95% CI: 1.46 3.24;
P = 0.021) and SII index value  960 £ 109/L (OR: 1.68, 95% CI: 1.24 2.74;
P = 0.040) were independent predicting factors for poor prognosis after aSAH.
Conclusions: SIRI and SII index values are associated with clinical outcomes of
patients with aSAH. Elevated SIRI and SII index could be independent predicting
factors for a poor prognosis after aSAH.
Key Words: Inflammation—Lymphocytes—Monocytes—Neutrophils—Subarachnoid
hemorrhage—Platelets
© 2021 Elsevier Inc. All rights reserved.

Introduction delayed cerebral ischemia (DCI) are known as predicting

factors for exacerbated prognosis of aSAH.2 5 Similar to
An aneurysmal subarachnoid hemorrhage (aSAH) can
other cerebrovascular diseases, aSAH can causes a pro-
result in disability and high mortality.1 Re-bleeding, poor
nounced peripheral inflammatory/immune response.
admission neurologic status, thick SAH, vasospasm, and
Systemic changes are also associated with outcomes of
patients with SAH.6,7 Although the exact mechanism has
From the *Department of Neurosurgery, Bucheon St. Mary’s Hospi-
not yet been fully revealed, inflammation is known to
tal, College of Medicine, The Catholic University of Korea, Bucheon,
Republic of Korea; †Department of Neurosurgery, Soonchunhyang plays an important role in the progress of aSAH by induc-
University Bucheon Hospital, Bucheon, Republic of Korea; and ing early brain damage, vasospasm, and subsequent
‡Department of Neurosurgery, St. Vincent’s Hospital, College of Medi- DCI.8 10 Prior clinical and experimental studies have
cine, The Catholic University of Korea, Suwon, Republic of Korea. shown that systematic inflammatory responses, such as
Received April 12, 2021; accepted April 28, 2021.
leukocytosis, thrombocytosis, and platelet activation are
Corresponding author (Ho Jun Yi) at: Department of Neurosurgery,
Soonchunhyang University Bucheon Hospital, 170, Jomaru-ro, associated with poor clinical outcome following
Bucheon, Gyeonggi-Do, 14584, Korea. E-mails: 431anarchy@naver. aSAH.11,12
com, 431anarchy@gmail.com. Alterations in peripheral blood cell composition are
1052-3057/$ - see front matter known as indicators of inflammatory responses in various
© 2021 Elsevier Inc. All rights reserved.
diseases.13,14 Prognosis predictivities of neutrophil-to-
https://doi.org/10.1016/j.jstrokecerebrovasdis.2021.105861

Journal of Stroke and Cerebrovascular Diseases, Vol. 30, No. 8 (August), 2021: 105861 1
2 S. YUN ET AL.

lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio was defined as IV-V. Clinical status contained symptom-
(LMR), and platelet-to-lymphocyte ratio (PLR) have been atic vasospasm and development of DCI diagnosed
reported in cerebrovascular diseases.15 17 Furthermore, according to previously published guidelines.2,24 Progno-
as more complicated inflammatory markers, systemic sis was evaluated with modified Rankin Scale (mRS) score
inflammation response index (SIRI) and systemic at 3 months after from ictus of aSAH. A favorable clinical
immune-inflammation (SII) index have been studied in outcome was defined as 3-month mRS score of 0-2. Radio-
various cancers, including brain tumor.18 22 To the best of logical variables included modified Fischer grade (thick
our knowledge, few studies have analyzed SIRI and SII SAH indicated a modified Fisher grade of III-IV), size
index in patients with aSAH. Zhang et al. have revealed (largest diameter) and location of aneurysm, and acute
the association between SIRI and the prognosis of aSAH, hydrocephalus. Laboratory findings, such as detailed
and Morgan et al. have reported clinical utility of SII index peripheral blood count, prothrombin time, activated par-
in aSAH.6,23 However, results of prior studies on aSAH tial thromboplastin time, total cholesterol, high-density
did not sufficiently demonstrated the usefulness of SIRI lipoprotein cholesterol, low-density lipoprotein choles-
or SII index as a potential predictor of prognosis or clinical terol, triglycerides, high-sensitivity
outcome. We hypothesize that SIRI and SII index as novel C-reactive protein, and erythrocyte sedimentation rate
biomarkers could better reflect the prognosis of aSAH were obtained using blood collected upon admission at the
patients than NLR, LMR, and PLR. Therefore, the objec- emergency department. NLR was calculated by dividing
tive of this study was to analyze associations of SIRI and the neutrophil count by the lymphocyte count, and LMR
SII index with prognosis of patients with aSAH, and com- value was obtained by dividing the lymphocyte count by
pare potential predictive values of SIRI and SII index to the monocyte count.17 PLR was calculated by dividing the
those of NLR, LMR and PLR for prognosis of SAH. platelet count by the lymphocyte count.25 SIRI and SII
index were defined as follows: SIRI = neutrophil count x
Materials and methods monocyte count / lymphocyte count, SII = platelet count x
neutrophil count/lymphocyte count.19,20
Study Population
This study was approved by the local Institutional Statistical analysis
Review Board of each participating center. From January
Continuous variables are presented as mean § stan-
2012 and December 2021, data of all consecutive patients
dard deviation (SD) or median with interquartile range
treated for SAH caused by aneurysm rupture, were ana-
(IQR). They were analyzed using Student’s t-test or
lyzed. Inclusion criteria were: (1) patients who underwent
Mann-Whitney U test. Categorical variables are expressed
surgical or interventional treatment for SAH caused by a
as frequency with percentage. They were analyzed with
ruptured aneurysm, (2) initiation of treatment for the rup-
x2 test or Fisher’s exact test. Variables with p < 0.20 in
tured aneurysm within 72 hours of ictus, (3) adequate
univariable analysis were entered into a backward multi-
blood sampling on admission for laboratory data before
variable logistic regression analysis to assess predicting
treatment, and (4) age of 18 to 85 years. Exclusion criteria
factors for prognosis and compute odds ratio (OR) with
were: (1) SAH caused by factors other than aneurysm,
95% confidence interval (CI) for each endpoint. Receiver
such as trauma and other cerebrovascular diseases, (2)
operating characteristic (ROC) analysis was performed to
patients who had not received treatment for ruptured
assess associations of inflammatory markers with clinical
aneurysm itself, (3) history of infection or stroke within
outcome, and determine optimal cutoff values of SIRI and
eight weeks, (4) underlying cancer, severe kidney or liver
SII index for predicting the prognosis. All data were ana-
disease, (5) history of auto-immune or hematologic disor-
lyzed using statistical software package SPSS 26.0 for
der, and (6) patients with inappropriate laboratory data
Windows (IBM, Armonk, NY, USA). Significant differ-
or loss of follow-up within three months.
ence was considered at two-tailed P-value  0.05.

Clinical and laboratory data

Results
aSAH was established by admission computed tomog-
Baseline characteristics
raphy angiography or magnetic resonance angiography.
The decision of treatment modality for aSAH (open sur- Clinical characteristics of patients are summarized in
gery or endovascular intervention) was made according (Table 1). A total of 640 patients were dichotomized
to our policy and each neurosurgeon’s preference. Demo- according to the 3-month mRS score (0 2, favorable; 3 6,
graphic data such as age, sex, risk factors, and past medi- unfavorable). The favorable clinical outcome group had
cal history (including hypertension, diabetes mellitus, 490 patients with a mean age of 56.9 § 13.4 years, includ-
smoking, and dyslipidemia) were obtaines. Initial neuro- ing 326 (66.5%) female patients. The unfavorable clinical
logic status was evaluated with Glasgow Coma Score outcome group had 190 patients with a mean age of 55.2
(GCS) and Hunt-Hess (H-H) grade. A high H-H grade § 12.4 years, including 138 (72.6%) female patients. There
PROGNOSTIC VALUES OF SIRI, SII INDEX IN SAH 3

Table 1. Baseline characteristics of patients with aneurysmal subarachnoid hemorrhage.

Variables Favorable (n = 490, 72.1%) Unfavorable (n = 190, 27.9%) P value

Demographics
Age, mean § SD 56.9 § 13.4 55.2 § 12.4 0.270
Female, n (%) 326 (66.5) 138 (72.6) 0.278
Hypertension, n (%) 258 (51.4) 108 (56.8) 0.370
Diabetic mellitus, n (%) 142 (29.0) 44 (23.2) 0.280
Smoking, n (%) 118 (24.1) 44 (23.2) 0.749
Dyslipidemia, n (%) 206 (42.0) 74 (38.9) 0.603
Treatment modality, n (%)
Surgery 124 (25.3) 47 (24.7) 0.714
Intervention 366 (74.7) 143 (75.3) 0.826
Clinical characteristics
Glasgow Coma Score, mean § SD 13.2 § 2.1 11.8 § 2.7 0.014
Hunt-Hess (H-H) grade, median [IQR] 3 [2-4] 3 [2-4] 0.578
High H-H grade (H-H grades IV-V), n (%) 103 (21.0) 68 (35.8) 0.012
Symptomatic vasospasm, n (%) 28 (5.7) 34 (17.9) 0.003
Delayed cerebral ischemia, n (%) 52 (10.6) 61 (32.1) 0.001
Radiologic characteristics
Modified Fisher grade, median [IQR] 2 [1 4] 3 [1 5] 0.005
Thick SAH, n (%) 131 (26.7) 91 (47.9) 0.008
Acute hydrocephalus, n (%) 127 (25.9) 61 (32.1) 0.087
Aneurysm size (mm), mean § SD 6.0 § 4.1 5.9 § 5.3 0.637
Aneurysm locations, n (%)
Anterior circulation 427 (87.1) 165 (86.9) 0.872
Posterior circulation 63 (12.9) 25 (13.1) 0.901
Laboratory findings, mean § SD
High-sensitivity C-reactive protein, mg/L 0.60 § 1.02 0.69 § 1.21 0.246
Erythrocyte sedimentation rate, mm/h 16.1 § 15.1 19.9 § 17.3 0.205
Total cholesterol, mg/dL 180.7 § 46.8 173.2 § 43.6 0.307
High-density lipoprotein cholesterol, mg/dL 50.8 § 22.3 49.6 § 19.2 0.494
Low-density lipoprotein cholesterol, mg/dL 107.1 § 36.0 105.3 § 33.2 0.512
Triglyceride, mg/dL 137.1 § 109.2 127.5 § 101.4 0.304
Prothrombin time, sec 14.1 § 6.5 13.6 § 6.6 0.678
Activated partial thromboplastin time, sec 36.1 § 13.1 36.9 § 13.8 0.527
Red blood cells, £ 1012/L 4.12 § 0.52 4.27 § 0.67 0.578
White blood cells, £ 109/L 8.42 § 5.11 8.63 § 5.85 0.328
Neutrophils, £ 109/L 5.97 § 4.23 6.42 § 5.88 0.043
Lymphocytes, £ 109/L 1.64 § 1.30 1.42 § 1.21 0.174
Monocytes, £ 109/L 0.70 § 0.51 0.79 § 0.59 0.320
Platelets, £ 109/L 224 § 119 231 § 115 0.427
NLR 3.6 § 2.5 4.5 § 3.2 0.029
LMR 2.3 § 1.9 1.8 § 1.6 0.031
PLR 137.1 § 103.7 159.3 § 121.4 0.081
SIRI, £ 109/L 2.6 § 2.0 3.6 § 3.3 0.004
SII index, £ 109/L 815 § 508 1044 § 841 0.007
Data are presented as n (%), mean § SD (standard deviation) and median [IQR = interquartile range].
Statistically significant differences are highlighted in bold.
Abbreviations: Favorable vs unfavorable clinical outcome, modified Rankin Scale (mRS) score at 3 months 0 2 vs 3 6; NLR, neutrophil-
to-lymphocyte ratio; LMR, lymphocyte-to-monocyte ratio; PLR, platelet-to-lymphocyte ratio; SIRI, systemic inflammation response index;
SII index, systemic immune-inflammation index.

were no significant differences in co-morbidities or treat- P = 0.012). The incidence of vasospasm and DCI was also
ment modality between the two groups. The unfavorable significantly higher in the unfavorable group than in the
group showed poorer initial clinical status with lower favorable group. (vasospasm: 17.9% vs 5.7%, P = 0.003;
mean GCS score and higher rate of high H-H grade than DCI: 32.1% vs 10.6%; P = 0.001). The unfavorable group
favorable group (GCS: 11.87 § SD 2.7 vs 13.2 § SD 2.1, had a higher median value of modified Fisher grade (3
P = 0.014; high H-H grade rate: 35.8% vs 21.0%, [IQR: 1-5] vs. 2 [IQR; 1 4], P = 0.005) and a higher ratio of
4 S. YUN ET AL.

thick SAH (47.9% vs 26.7%, P = 0.008) than the favorable Associations of SIRI and SII index with clinical and
group. In laboratory findings, the unfavorable group had laboratory data
significantly higher mean values of neutrophil (6.42 §
Patients were divided according to cutoff values of SIRI
5.88 £ 109/L vs 5.97 § 4.23 £ 109/L, P = 0.043) and NLR
and SII index. Elevated SIRI group ( 3.2) showed higher
(4.5 § 3.2 vs 3.6 § 2.5, P = 0.029) than the favorable group.
rate of high H-H grade (P = 0.012), more incidence of vaso-
However, the mean LMR value was lower in the unfavor-
spasm and DCI (P = 0.005 and 0.001, respectively), higher
able outcome than in the favorable outcome group (1.8 §
rate of thick SAH (P = 0.008), and higher value of median
1.6 vs 2.3 § 1.9, P = 0.031). Mean values of SIRI and SII
3-month mRS (P = 0.011), than the lower SIRI group
index were significantly higher in the unfavorable group
(<3.2 £ 109/L). Similarly, more vasospasm and DCI,
than in the favorable group (SIRI: 3.6 § 3.3 £ 109/L vs 2.6
higher rate of thick SAH, and poor 3-month mRS were
§ 2.0 £ 109/L, P = 0.004; SII index: 1044 § 841 £ 109/L vs
found in patients with elevated SII index ( 960 £ 109/L)
815 § 508 £ 109/L, P = 0.007).
than in those with lower SII index (<960 £ 109/L; Table 3).

Predictive values of SIRI and SII index for prognosis of Multivariate analysis of prognostic factor for SAH
SAH
In binary logistic regression analysis, high H-H grade
In ROC analysis, 3.2 £ 109/L was determined as an (IV-V) (OR: 2.16, 95% CI: 1.50 3.68, P = 0.011), vasospasm
optimal cutoff value of SIRI to discriminate between (OR: 2.60, 95% CI: 1.56 4.04, P = 0.007), DCI (OR: 3.02,
favorable and unfavorable clinical outcomes at 3 months 95% CI: 1.74 5.28, P = 0.002), and thick SAH (modified
(area under the curve [AUC]: 0.785, 95% CI: 0.773 0.836, Fisher grade, III IV) (OR: 2.33, 95% CI: 1.51 3.89,
P < 0.001). The optimal cutoff values of SII index to pre- P = 0.008) were independently associated with unfavor-
dict the prognosis was found to be 960 £ 109/L (AUC: able clinical outcomes. Among inflammatory parameters,
0.702, 95% CI: 0.638 0.765, P < 0.001; Fig. 1). AUCs of value of SIRI ( 3.2 £ 109/L) (OR: 1.82, 95% CI:
NLR and LMR levels were 0.592 (95%CI, 0.542 0.644; 1.46 3.24, P = 0.021) and SII index ( 960 £ 109/L) (OR:
P = 0.011) and 0.581 (95%CI, 0.531 0.633; P = 0.037), 1.68, 95% CI: 1.24 2.74, P = 0.040) were independent pre-
respectively. Therefore, SIRI and SII index had higher dictors of unfavorable prognosis after aSAH (Table 4).
accuracies for predicting 3-month prognosis of aSAH
than levels of NLR, LMR, and PLR (Table 2). Discussion
In this study, we found that SIRI and SII index were
associated with prognosis of aSAH. Patients with unfa-
vorable outcomes had higher values of SIRI and SII index
than patients with unfavorable outcomes. Predictable val-
ues of SIRI and SII for prognosis of aSAH were superior
than those of other inflammatory markers such as NLR,
LMR, and PLR. After dichotomization of patients with
the optimal cutoff values of SIRI (3.2 £ 109/L) and SII
index (960 £ 109/L), patients with high SIRI and SII index
showed poor prognosis than those with low SIRI and SII
index. In addition, SIRI value  3.2 £ 109/L and SII index
value  960 £ 109/L were independent predictors of
unfavorable prognosis.
SIRI is a marker of systemic inflammation. Its useful-
ness as a predictor of prognosis has been revealed in vari-
ous types of cancer.19 Topkan et al. have reported
independent and significant associations of a low value of
SIRI ( 1.78) with longer progression free survival and
overall survival in patients with glioblastoma multiforme
of brain.21 Recently, Zhang et al. have performed an anal-
Fig. 1. Receiver operating characteristic (ROC) curve analysis for optimal ysis for inflammatory markers including SIRI in patients
cut-off values of systemic inflammation response index (SIRI) and systemic with SAH.23 and reported that patients with poor clinical
immune-inflammation (SII) index to predict the prognosis of patients with outcome (3month Glasgow outcome scale [GOS] = 1 3)
an aneurysmal subarachnoid hemorrhage. Optimal cut-off values of SIRI
have higher median SIRI value than those with good clini-
and SII index for unfavorable clinical outcome (modified Rankin Scale score
of 3 6 at 3-month) were 3.2 £ 109/L and 960 £ 109/L, respectively (area
cal outcomes (GOS = 1 4) (4.80 [IQR]: 3.15 6.72] vs 3.11
under the curve [AUC]: 0.785, 95% CI: 0.773 0.836; P < 0.001 and [IQR]: 2.10 4.65, P = 0.001). The optimal cutoff value of
AUC: 0.702, 95% CI: 0.638 0.765; P < 0.001, respectively). SIRI was identified as 4.105 £ 109/L (AUC = 0.671). SIRI
PROGNOSTIC VALUES OF SIRI, SII INDEX IN SAH 5

Table 2. Associations of inflammatory markers with clinical outcome.

Area under the curve 95% confidence interval P value

NLR 0.592 0.542 0.644 0.011
LMR 0.581 0.531 0.633 0.037
PLR 0.563 0.512 0.613 0.101
SIRI, £ 109/L 0.785 0.773 0.836 <0.001
SII index, £ 109/L 0.702 0.638 0.765 <0.001
Data are presented as n (%), mean § SD (standard deviation) and median [IQR = interquartile range].
Statistically significant differences are highlighted in bold.
Abbreviations: NLR, neutrophil-to-lymphocyte ratio; LMR, lymphocyte-to-monocyte ratio; PLR, platelet-to-lymphocyte ratio; SIRI, sys-
temic inflammation response index; SII index, systemic immune-inflammation index.

had a better predictive ability for 3-month outcome of is known as a parameter reflecting poor prognosis of can-
patients with aSAH than other inflammatory markers, cers.20,26 In addition, higher level of SII index is a predict-
including NLR and PLR. Patients with elevated SIRI levels ing factor for poor outcomes in brain pathologies such as
had higher rates of unfavorable outcome. SIRI was an inde- sinus thrombosis, intracerebral hemorrhage, and
pendent predictor of poor outcome at 3 months after aSAH glioma.22,27,28 Morga et al. have performed a retrospective
(OR: 2.656, 95% CI: 1.156 6.103, P = 0.021).23 Despite dif- cohort study for SII index in patients with aSAH6 and
ferences in detailed parameters, these findings were very found that patients with aSAH show higher median SII
similar to results of the present study. In our study, the index values than patients without aSAH (1095.70 [IQR]:
unfavorable clinical outcome group (3 months mRS = 3 6) 762.43 2053.06] vs. 388.12 [IQR]: 388.09 498.44,
had higher mean value of SIRI than the favorable group P < 0.001). Those with a poor initial neurologic status of
(3.6 § 3.3 £ 109/L vs 2.6 § 2.0 £ 109/L, P = 0.004). In ROC aSAH (GCS = 3 13 and H-H grade = 3 5) had a higher
analysis, the optimal cutoff value of SIRI was found to be SII index value than those with a good initial neurologic
3.2 £ 109/L (AUC = 0.785). Patients with higher SIRI status (GCS = 14 15 and H-H grade 0 2). In addition,
showed more occurrence of vasospasm and DCI with worse outcome at discharge (GOS 1 3) was associated
higher value of median mRS. In multivariate analysis, with higher median PLR value (GOS 4-5 vs 1 3 = 130.75
higher value of SIRI ( 3.2 £ 109/L) was an independent [106.00 164.62] vs 324.31 [321.11 327.50], P = 0.032), but
predictor for unfavorable prognosis at 3 months after not associated with SII index (P = 0.131).6 Our study
aSAH (OR: 1.82, 95% CI: 1.46 3.24, P = 0.021). showed slightly different findings. In the present study,
Like SIRI, the SII index was also first studied for its patients with unfavorable 3-months mRS had higher
prognostic role in many malignancies. Increased SII index mean SII index value than the favorable group (1044 §

Table 3. Associations of SIRI and SII index with baseline characteristics.

Variables SIRI (£ 109/L) SII index (£ 109/L)

< 3.2 (n = 345)  3.2 (n = 335) P value < 960 (n = 392)  960 (n = 288) P value
Age 57.7 § 13.2 55.1 § 10.9 0.279 55.6 § 12.9 57.5 § 13.4 0.389
Female 235 (68.1) 229 (68.3) 0.880 261 (66.6) 203 (70.5) 0.417
High H-H grade 71 (20.6) 100 (29.9) 0.012 93 (23.7) 78 (27.1) 0.054
Vasospasm 24 (7.0) 38 (11.3) 0.005 30 (7.6) 32 (11.1) 0.009
DCI 33 (9.6) 80 (23.8) 0.001 51 (13.0) 62 (21.5) 0.004
Thick SAH 94 (27.2) 128 (38.2) 0.008 117 (29.8) 105 (36.5) 0.029
Hydrocephalus 93 (26.9) 95 (28.3) 0.307 114 (29.1) 74 (25.7) 0.142
mRS 2 [0 4] 3 [1 5] 0.011 2 [0 4] 3 [1 5] 0.012
WBC, £ 109/L 8.31 § 3.81 8.65 § 4.01 0.209 8.41 § 3.92 8.57 § 4.02 0.249
Neu, £ 109/L 5.64 § 3.82 6.56 § 5.17 0.034 5.81 § 3.77 6.48 § 4.92 0.044
Lym, £ 109/L 1.71 § 1.14 1.44 § 1.12 0.041 1.61 § 1.09 1.53 § 1.11 0.067
Mono, £ 109/L 0.73 § 0.50 0.72 § 0.53 0.508 0.72 § 0.56 0.73 § 0.54 0.439
Platelet, £ 109/L 225 § 102 226 § 111 0.725 223 § 92 229 § 114 0.267
Data are presented as n (%), mean § SD (standard deviation) and median [IQR = interquartile range].
Statistically significant differences are highlighted in bold.
Abbreviations: SIRI, systemic inflammation response index; SII index, systemic immune-inflammation index; High H-H grade, Hunt-Hess
grade IV-V; DCI, delayed cerebral ischemia; Thick SAH, subarachnoid hemorrhage with a modified Fisher grade of III-IV; mRS, modified
Rankin Scale; WBC, white blood cell; Neu, neutrophil; Lym, lymphocyte; Mono, monocyte.
6 S. YUN ET AL.

Table 4. Multivariate analysis for prediction of unfavorable clinical outcome.

Variables Univariate Multivariate

OR (95% CI) P value OR (95% CI) P value
Age (> 70yr) 1.82 (0.68 2.22) 0.337
High H-H grade 3.08 (1.60 5.24) 0.005 2.16 (1.50 3.68) 0.011
Vasospasm 3.38 (1.78 6.42) 0.003 2.60 (1.56 4.04) 0.007
Delayed cerebral ischemia 4.12 (1.84 8.78) 0.001 3.02 (1.74 5.28) 0.002
Thich SAH 3.11 (1.73 4.95) 0.004 2.33 (1.51 3.89) 0.008
NLR ( 4.0) 1.91 (1.07 3.13) 0.037 1.73 (0.91 3.07) 0.088
LMR (< 2.0) 1.82 (1.08 2.84) 0.048 1.54 (0.88 4.04) 0.103
PLR ( 145) 1.77 (0.59 4.13) 0.224
SIRI ( 3.2 £ 109/L) 2.60 (1.58 4.46) 0.009 1.82 (1.46 3.24) 0.021
SII index ( 960 £ 109/L) 2.22 (1.30 3.82) 0.013 1.68 (1.24 2.74) 0.040
Statistically significant differences are highlighted in bold.
Abbreviations: Unfavorable clinical outcome, modified Rankin Scale (mRS) at 3 month 3-6; OR, odds ratio; CI, confidence interval; High
H-H grade, Hunt-Hess grade IV-V; Thick SAH, subarachnoid hemorrhage with a modified Fisher grade of III-IV; NLR, neutrophil-to-lym-
phocyte ratio; LMR, lymphocyte-to-monocyte ratio; PLR, platelet-to-lymphocyte ratio; SIRI, systemic inflammation response index; SII
index, systemic immune-inflammation index.

841 £ £ 109/L vs 815 § 508 £ £ 109/L, P = 0.007). The prognosis of aSAH. Furthermore, SIRI (neutrophil x
optimal cut-off value for SII index was 960 £ 109/L. monocyte/lymphocyte count) and SII index (neutrophil x
Between high and low SII index groups, there was no dif- platelet/lymphocyte count) which include three compo-
ference in initial H-H grade. However, the high SII index nents are thought to be more effective than markers with
group had a higher median 3-months mRS (3 [1 5] vs 2 two components, including NLR (neutrophil/lympho-
[0-4], P = 0.012). Furthermore, SII index was indepen- cyte), LMR (lymphocyte/monocyte), and PLR (platelet/
dently associated with a poor prognosis of aSAH (OR: lymphocyte). In our results, prognostic predictivities of
1.68, 95% CI: 1.24 2.74, P = 0.040). SIRI and SII were superior to those of NLR, LMR, and
It has been reported that inflammatory response can PLR.
affect the course and prognosis of aSAH. Prior studies This study has several limitations. First, its retrospec-
have revealed the relationship between aSAH and various tive non-blinded study design might have induced sev-
inflammatory markers such as monoclonal antibodies eral bias and errors in data interpretation. Second, we
(CD-3,4,8), toll like receptor 4, macrophage migration did not include dynamic change of inflammatory
inhibitory factor, and lipoprotein-associated phospholi- markers with repeated measurements at different time
pase A2.29 32 However, it is difficult to apply these points. Third, history of past medication and other
markers clinically in terms of cost effectiveness and equip- inflammatory markers such as interleukins and throm-
ment setup. Therefore, research for easily accessible boxane that could affect results could not be obtained.
peripheral blood parameters has been widely conducted. However, every effort was made to adjust for the possi-
NLR, LMR, and PLR are well-known hematologic inflam- bility of spurious results.
matory markers. In patients with aSAH, elevated NLR
and PLR and decreased LMR are associated with vaso-
Conclusions
spasm, DCI, and poor clinical outcome.9,15,33,34 This phe-
nomenon can be explained by the role of each blood cell. Our study demonstrated the clinical importance of SIRI
An aSAH can stimulate inflammatory response with leu- and SII as novel inflammatory markers in patients with
kocytosis and platelet activation, subsequently inducing aSAH. Predictive values of SIRI and SII index for progno-
brain damage and vasospasm.11,35 Platelet can induce sis of aSAH were superior to those of NLR, LMR, and
thrombosis and concomitantly release inflammatory mol- PLR. SIRI and SII index at admission were found to have
ecules.12 Neutrophils can promote inflammatory strong associations with the prognosis of patients with
responses, induce damage to the blood-brain barrier, and aSAH. Elevated SIRI and SII index independently pre-
release inflammatory mediators.36 Monocytes can pro- dicted poor prognosis after aSAH. Further studies are nec-
mote inflammatory and pro-thrombotic pathways by essary to validate our findings and identify mechanisms.
interacting with platelets and endothelial cells.37 In con-
trast, lymphocytes play an important role in anti-inflam-
Acknowledgments
matory response.38 Inflammatory markers that can
simultaneously contain the actions of these complex cas- This work was supported by the Soonchunhyang Univer-
cades, are thought to be helpful for predicting the sity Research Fund.
PROGNOSTIC VALUES OF SIRI, SII INDEX IN SAH 7

Declaration of Competing Interest pathophysiology to new therapeutic strategies. Lancet

Neurol 2011;10:471-480.
None. 14. Sarrafzadeh A, Schlenk F, Meisel A, Dreier J, Vajkoczy P,
Meisel C. Immunodepression after aneurysmal subarach-
Ethical Approval/Informed Consent noid hemorrhage. Stroke 2011;42:53-58.
15. Tao C, Wang J, Hu X, Ma J, Li H, You C. Clinical value of
This study was carried after the approval of Institutional neutrophil to lymphocyte and platelet to lymphocyte
Review Board of all participating hospitals and with the writ- ratio after aneurysmal subarachnoid hemorrhage. Neuro-
crit Care 2017;26:393-401.
ten informed consent from legal guardians of patients.
16. Fan Z, Hao L, Chuanyuan T, Jun Z, Xin H, Sen L, et al.
Neutrophil and platelet to lymphocyte ratios in associat-
References ing with blood glucose admission predict the functional
outcomes of patients with primary brainstem hemor-
1. Rinkel GJ, Algra A. Long-term outcomes of patients with rhage. World Neurosurg 2018;116:e100-e107.
aneurysmal subarachnoid haemorrhage. Lancet Neurol 17. Oh SW, Yi HJ, Lee DH, Sung JH. Prognostic significance
2011;10:349-356. of various inflammation-based scores in patients with
2. Frontera JA, Fernandez A, Schmidt JM, Claassen J, War- mechanical thrombectomy for acute ischemic stroke.
tenberg KE, Badjatia N, et al. Defining vasospasm after World Neurosurg 2020;141:e710-e717.
subarachnoid hemorrhage: what is the most clinically rel- 18. Qi Q, Zhuang L, Shen Y, Geng Y, Yu S, Chen H, et al. A
evant definition? Stroke 2009;40:1963-1968. novel systemic inflammation response index (SIRI) for
3. Schatlo B, Fung C, Fathi AR, Sailer M, Winkler K, Daniel RT, predicting the survival of patients with pancreatic cancer
et al. Introducing a nationwide registry: the Swiss study on after chemotherapy. Cancer 2016;122:2158-2167.
aneurysmal subarachnoid haemorrhage (Swiss SOS). Acta 19. Wei L, Xie H, Yan P. Prognostic value of the systemic
Neurochir (Wien) 2012;154:2173-2178. discussion 2178. inflammation response index in human malignancy: A
4. Risselada R, Lingsma HF, Bauer-Mehren A, Friedrich meta-analysis. Medicine (Baltimore) 2020;99:e23486.
CM, Molyneux AJ, Kerr RS, et al. Prediction of 60 day 20. Yang R, Chang Q, Meng X, Gao N, Wang W. Prognostic
case-fatality after aneurysmal subarachnoid haemor- value of Systemic immune-inflammation index in cancer:
rhage: results from the International Subarachnoid Aneu- A meta-analysis. J Cancer 2018;9:3295-3302.
rysm Trial (ISAT). Eur J Epidemiol 2010;25:261-266. 21. Topkan E, Kucuk A, Ozdemir Y, Mertsoylu H, Besen AA,
5. Dijkland SA, Roozenbeek B, Brouwer PA, Lingsma HF, Sezen D, et al. Systemic inflammation response index pre-
Dippel DW, Vergouw LJ, et al. Prediction of 60-day case dicts survival outcomes in glioblastoma multiforme
fatality after aneurysmal subarachnoid hemorrhage: patients treated with standard stupp protocol. J Immunol
external validation of a prediction model. Crit Care Med Res 2020;2020:8628540.
2016;44:1523-1529. 22. Liang R, Li J, Tang X, Liu Y. The prognostic role of preoper-
6. Morga R, Dziedzic T, Moskala M, Slowik A, Pera J. Clini- ative systemic immune-inflammation index and albumin/
cal relevance of changes in peripheral blood cells after globulin ratio in patients with newly diagnosed high-grade
intracranial aneurysm rupture. J Stroke Cerebrovasc Dis glioma. Clin Neurol Neurosurg 2019;184:105397.
2020;29:105293. 23. Zhang P, Li Y, Zhang H, Wang X, Dong L, Yan Z, et al.
7. Rass V, Gaasch M, Kofler M, Schiefecker AJ, Ianosi BA, Prognostic value of the systemic inflammation response
Rhomberg P, et al. Systemic inflammatory response syn- index in patients with aneurismal subarachnoid hemor-
drome as predictor of poor outcome in nontraumatic sub- rhage and a Nomogram model construction. Br J Neuro-
arachnoid hemorrhage patients. Crit Care Med 2018;46: surg 2020:1-7.
e1152-e1159. 24. Vergouwen MD, Vermeulen M, van Gijn J, Rinkel GJ,
8. Dumont AS, Dumont RJ, Chow MM, Lin CL, Calisaneller Wijdicks EF, Muizelaar JP, et al. Definition of delayed
T, Ley KF, et al. Cerebral vasospasm after subarachnoid cerebral ischemia after aneurysmal subarachnoid hemor-
hemorrhage: putative role of inflammation. Neurosur- rhage as an outcome event in clinical trials and observa-
gery 2003;53:123-133. discussion 133-125. tional studies: proposal of a multidisciplinary research
9. Al-Mufti F, Amuluru K, Damodara N, Dodson V, Roh D, group. Stroke 2010;41:2391-2395.
Agarwal S, et al. Admission neutrophil-lymphocyte ratio 25. Xu JH, He XW, Li Q, Liu JR, Zhuang MT, Huang FF, et al.
predicts delayed cerebral ischemia following aneurysmal Higher platelet-to-lymphocyte ratio is associated with
subarachnoid hemorrhage. J Neurointerv Surg 2019;11: worse outcomes after intravenous thrombolysis in acute
1135-1140. ischaemic stroke. Front Neurol 2019;10:1192.
10. Fu C, Yu W, Sun L, Li D, Zhao C. Early cerebral infarction 26. Zhong JH, Huang DH, Chen ZY. Prognostic role of sys-
following aneurysmal subarachnoid hemorrhage: fre- temic immune-inflammation index in solid tumors: a sys-
quency, risk factors, patterns, and prognosis. Curr Neu- tematic review and meta-analysis. Oncotarget 2017;8:
rovasc Res 2013;10:316-324. 75381-75388.
11. Kasius KM, Frijns CJ, Algra A, Rinkel GJ. Association of 27. Li S, Liu K, Gao Y, Zhao L, Zhang R, Fang H, et al. Prog-
platelet and leukocyte counts with delayed cerebral ische- nostic value of systemic immune-inflammation index in
mia in aneurysmal subarachnoid hemorrhage. Cerebro- acute/subacute patients with cerebral venous sinus
vasc Dis 2010;29:576-583. thrombosis. Stroke Vasc Neurol 2020;5:368-373.
12. Frontera JA, Provencio JJ, Sehba FA, McIntyre TM, Now- 28. Trifan G, Testai FD. Systemic Immune-Inflammation (SII)
acki AS, Gordon E, et al. The role of platelet activation index predicts poor outcome after spontaneous supraten-
and inflammation in early brain injury following sub- torial intracerebral hemorrhage. J Stroke Cerebrovasc Dis
arachnoid hemorrhage. Neurocrit Care 2017;26:48-57. 2020;29:105057.
13. Macrez R, Ali C, Toutirais O, Le Mauff B, Defer G, Dir- 29. Suzuki H, Fujimoto M, Kawakita F, Liu L, Nakano F, Nishi-
nagl U, et al. Stroke and the immune system: from kawa H, et al. Toll-like receptor 4 and tenascin-C signaling
8 S. YUN ET AL.

in cerebral vasospasm and brain injuries after subarachnoid 34. Yi HJ, Lee DH, Sung JH. Inflammation-based scores are
hemorrhage. Acta Neurochir Suppl 2020;127:91-96. associated with the prognosis of patients with aneurys-
30. de Oliveira Manoel AL, Macdonald RL. Neuroinflamma- mal subarachnoid hemorrhage after neuro-intervention.
tion as a target for intervention in subarachnoid hemor- Curr Neurovasc Res 2020;17:676-685.
rhage. Front Neurol 2018;9:292. 35. Sehba FA, Mostafa G, Friedrich Jr. V, Bederson JB. Acute
31. Chen YH, Cheng ZY, Shao LH, Shentu HS, Fu B. Macro- microvascular platelet aggregation after subarachnoid
phage migration inhibitory factor as a serum prognostic hemorrhage. J Neurosurg 2005;102:1094-1100.
marker in patients with aneurysmal subarachnoid hem- 36. Herz J, Sabellek P, Lane TE, Gunzer M, Hermann DM,
orrhage. Clin Chim Acta 2017;473:60-64. Doeppner TR. Role of Neutrophils in Exacerbation of
32. Luo YG, Han B, Sun TW, Liu X, Liu J, Zhang J. The asso- Brain Injury After Focal Cerebral Ischemia in Hyperlipi-
ciation between serum adipocyte fatty acid-binding demic Mice. Stroke 2015;46:2916-2925.
protein and 3-month disability outcome after aneurys- 37. Ren H, Liu X, Wang L, Gao Y. Lymphocyte-to-monocyte
mal subarachnoid hemorrhage. J Neuroinflammation ratio: a novel predictor of the prognosis of acute ischemic
2020;17:66. stroke. J Stroke Cerebrovasc Dis 2017;26:2595-2602.
33. Wu Y, He Q, Wei Y, Zhu J, He Z, Zhang X, et al. The asso- 38. Si Y, Liu J, Shan W, Zhang Y, Han C, Wang R, et al. Asso-
ciation of neutrophil-to-lymphocyte ratio and delayed ciation of lymphocyte-to-monocyte ratio with total coro-
cerebral ischemia in patients with aneurysmal subarach- nary plaque burden in patients with coronary artery
noid hemorrhage: possible involvement of cerebral blood disease. Coron Artery Dis 2020;31:650-655.
perfusion. Neuropsychiatr Dis Treat 2019;15:1001-1007.