Blood markers of cerebral vasospasm after aneurysmal subarachnoid hemorrhage: a narrative review of current literature

Abstract

Introduction:

There have been growing efforts to identify predictors of Cerebral Vasospasm to facilitate earlier diagnosis and establish patient predictive models. This review discusses the implications of serum electrolytes, glycemic indices, and inflammatory markers in predicting the occurrence of aSAH-induced CV.

Methods:

The literature was reviewed across PubMed, Scopus, Web of Science, Google scholars, the National Center for Biotechnology Information, the Centers for Disease Control and Prevention, the World Health Organization, and the National Institutes of Health. Studies were included within the following criteria: (a) focused on CV in human and (b) were published in peer-reviewed journals.

Results:

Studies included in this review comprised retrospective analyses, prospective studies, and systematic reviews. The literature suggests that hyponatremia is associated with the presence of CV post-aSAH. It is frequently encountered in the early phase postictally, emanating from the syndrome of inappropriate antidiuretic hormone secretion and cerebral salt wasting syndrome. Hyperglycemia, leukocytosis, elevated C-reactive protein levels and elevated cerebrospinal interleukin-6 levels also appear to be associated with the occurrence of CV post-aSAH.

Conclusions:

The studies included in this review suggest that the presence of hyponatremia, hyperglycemia, leukocytosis, and elevated CRP and CSF IL-6 levels are associated with the occurrence of CV post-aSAH. Further research is required to scrutinize temporal relations and causality of described markers in the setting of aSAH-induced CV. Establishing a panel of biomarkers will facilitate the development of patient predictive models. Such models may stratify patient groups, institute therapeutic and endovascular interventions and ultimately reduce vasospasm-associated morbidity and mortality to improve patient outcomes.

Introduction

Cerebral vasospasm (CV), the delayed, temporary narrowing of intracranial arteries, is a common sequela after aneurysmal subarachnoid hemorrhage (aSAH). Discounting initial ictus, CV is a major source of death and neurological decline after aSAH[1]. Most frequently, CV occurs in a biphasic pattern 3–4 days post-aneurysmal rupture and peaks within 6–10 days post-bleed[2].

HIGHLIGHTS

• We discuss the implications of serum electrolytes, glycemic indices, and inflammatory markers in predicting the occurrence of aSAH-induced CV.

• Hyponatremia is associated with the presence of CV post-aSAH. It is frequently encountered in the early phase postictally, emanating from the syndrome of inappropriate antidiuretic hormone secretion and cerebral salt wasting syndrome.

• Leukocytosis, elevated C-reactive protein levels and elevated cerebrospinal interleukin-6 levels also appear to be associated with the occurrence of CV post-aSAH. The prognostic significance of glucose measurements in predicting CV remains unclear.

CV is classified as radiographic or symptomatic. Radiographic CV, detected in up to 70% of patients, is defined as vascular narrowing identified on digital subtraction angiography, transcranial Doppler and computed tomography (CT)[3]. Symptomatic CV describes a new focal deficit and/or reduced consciousness which cannot be attributed to other causes by means of clinical assessment, CT or magnetic resonance imaging scanning of the brain, and appropriate laboratory studies[3]. While 70% demonstrate arterial narrowing on imaging (radiographic CV), only 20–30% develop symptoms (symptomatic CV)^[3,4]. Radiographic CV may progress to symptomatic CV if there is significant vascular narrowing[3].

At present, treatment interventions for CV are inconsistent and/or ineffective. They may also be associated with significant risks, including stroke and cardiovascular compromise. Given the complexity associated with CV and its delayed onset, there have been growing efforts to identify blood markers of CV. Such markers may be used to predict the occurrence of CV early, institute prompt treatment, and facilitate the development of patient predictive models to risk stratify patients. In this review of the published literature, we aim to investigate the current understandings of non-radiological blood markers of CV after aSAH.

Methods

The literature was reviewed across PubMed, Scopus, Web of Science, Google scholars, the National Center for Biotechnology Information, the Centers for Disease Control and Prevention, the World Health Organization, and the National Institutes of Health. The search was limited to papers published in English over the past 25 years that were focused on the following keywords for research in different combinations: “cerebral vasospasm,” “subarachnoid hemorrhage,” “aneurysm,” “sodium,” “glycemic indices,” “leukocyte count,” “CRP,” and “interleukin.” Studies were included within the following criteria: (a) focused on CV in human and (b) were published in peer-reviewed journals. Non-English studies and studies with animal subjects were excluded. Figures were created using Biorender.

Results

There were 43 studies included in this review, of which 15 evaluated the prognostic significance of serum sodium levels for predicting CV, 6 evaluated glycemic indices, further 6 evaluated leucocyte counts, and 15 evaluated inter-leukins and CRP levels.

Sixteen studies were conducted in North America, 1 in South America, 11 in Europe, and 15 in Asia. Most studies were conducted retrospectively (n = 28, 65.1%), 14 prospectively, and 1 was a systematic review.

Overview of extracted data is presented in Tables 1-4.

Table 2.

Summary of studies assessing the association between glycemic indices and development of cerebral vasospasm (CV) after aneurysmal subarachnoid hemorrhage (aSAH)

Study	Country	Research design	Definition of parameters (mg/dL)	Patient number (M, F)	Outcomes	P value
McIntyre et al, 2021[20]	USA	Retrospective analysis	Hyperglycemia, >160	217 (75, 142)	2-fold increase in CV risk amongst those with a minimum blood glucose measurement <90 mg/L	P = 0.015
Nadeich et al, 2010[21]	USA	Retrospective analysis	Moderate hypoglycemia, <80 Severe hypoglycemia, <40	172 (57,115)	Lower nadir glucose is observed in radiographic and symptomatic CV	Radiographic CV: P = 0.01 Symptomatic CV: P = 0.04
Dumont et al, 2009[22]	USA	Retrospective analysis	Diabetes mellitus, HbA1c >6.5 mmol/L	113 (42, 71)	No association between hyperglycemia and symptomatic CV	P = 0.166
Frontera et al, 2005[23]	USA	Prospective analysis	Hyperglycemia, >200	281 (210,71)	No association between hyperglycemia and symptomatic CV	P = 0.361
Badjatia et al, 2005[24]	USA	Retrospective analysis	Hyperglycemia, >140	352 (106, 246)	Increasing blood glucose values are associated with the occurrence of symptomatic CV after aSAH	P = 0.001
Charpentier et al, 1993[25]	France	Retrospective analysis	Hyperglycemia, >216	244 (98, 146)	Hyperglycemia is predictor of symptomatic CV	P < 0.2

F, female; M, male.

Table 3.

Summary of studies assessing the association between leukocyte counts and development of cerebral vasospasm (CV) after aneurysmal subarachnoid hemorrhage (aSAH)

Study	Country	Research design	Definition of leukocytosis (× 109/L)	Patient number and sex (M, F)	Outcomes	P value
Buce-Satoba et al, 2022[26]	Latvia	Retrospective analysis		117 (48, 69)	SAH-CV patients demonstrated higher WBC and CRP compared to control	WBC, P = 0.01 CRP, <0.001
Shrestha et al, 2022[27]	Nepal	Retrospective analysis	>12	90 (30, 60)	No correlation between leukocytosis and vasospasm occurrence	P = 0.43
Bacigaluppi et al, 2020[28]	Italy	Prospective analysis	NR	42 (13, 29)	Early rise in leukocyte identified in patients developing vasospasm	P = 0.0018
Mijiti et al, 2016[29]	China	Retrospective analysis	NR	378 (205, 173)	Admission leucocyte count >11 × 109/L predictor of CV occurrence	P = 0.007
Chou et al, 2011[30]	USA	Prospective analysis	NR	55 (24, 31)	Elevated blood leukocyte counts 0–14 days post-aSAH are associated with vasospasm	P = 0.0003
McGirt et al, 2003[31]	USA	Retrospective analysis	>15	224 (NR)	Leukocytosis tripled risk of symptomatic CV occurrence.

F, female; M, male; NR: not recorded.

Table 1.

Summary of studies assessing the association between serum sodium levels and development of cerebral vasospasm (CV) after aneurysmal subarachnoid hemorrhage (aSAH)

Study	Country	Research design	Definition of parameters, mEq/L	Patient number (M, F)	Outcomes	P value
Helliwell et al, 2023[5]	USA	Retrospective analysis	Hypernatremia, >145 Hyponatremia, <135	320 (83, 237)	No independent associations found between hyponatremia or hypernatremia and vasospasm	Hypernatremia, P = 0.07 Hyponatremia, NR
Chua et al, 2022[6]	USA	Retrospective analysis	Hyponatremia, <135	271 (96, 175)	No association between serum sodium levels in the first 2 weeks post-aSAH and radiographic CV	P = 0.59
Rumalla et al, 2021[7]	USA	Retrospective analysis	NR	8364 (2420, 5925)	Hyponatremia is associated with greater incidence of CV	P < 0.05
Alabbas et al, 2020[8]	Saudi Arabia	Retrospective analysis	Hyponatremia, <135	45 (22, 23)	Hyponatremia is more common in vasospasm and most likely to occur in first 4 days	P = 0.016
Escamilla-Ocañas et al, 2020[9]	USA	Retrospective analysis	Hyponatremia, <135	164 (36, 128)	CV incidence higher in the hyponatremia group (65.1%) compared to the normonatremic (28.5%) group	P < 0.001
Ridwan et al, 2019[10]	Germany	Prospective analysis	Hyponatremia, <135	101 (31, 70)	Hyponatremia at day 1–7 post-rupture showed no significant influence on occurrence of symptomatic CV	NR
Uozumi et al, 2017[11]	Japan	Retrospective analysis	Hyponatremia, <131	97 (65, 32)	Decrease in serum sodium occur 1 day prior to symptomatic CV and is predictive of CV	NR
Bales et al, 2016[12]	USA	Retrospective analysis	Hyponatremia, <135	198 (58, 140)	Sodium variability of 6–12 mEq/L associated with vasospasm-induced cerebral infarction	P = 0.001
Mapa et al, 2015[13]	USA	Systematic review	NR	NR	Hyponatremia is associated with vasospasm and duration of hospitalization	NR
Vrsajkov et al, 2012[14]	Serbia	Retrospective analysis	Hyponatremia, <135	82 (50, 32)	62% of patients in hyponatremia group had CV compared to 38% in normonatremia group	P = 0.03
Zheng et al, 2011[15]	China	Retrospective analysis	Hyponatremia, <135	124 (100, 24)	Cerebral infarction is associated with late-onset hyponatremia in high grade patients	P = 0.01
Igarashi et al, 2007[16]	Japan	Retrospective analysis	NR	67 (24, 44)	Negative sodium balance on day 1 post-ictal predict symptomatic vasospasm	P < 0.05
Chandy et al, 2006[17]	USA	Retrospective analysis	Hyponatremia, <135	106 (43, 63)	Patients with hyponatremia were 2.6	P = 0.023
					times more likely to develop CV in setting of aSAH
Qureshi et al, 2002[18]	USA	Prospective analysis	Hyponatremia, <135	298 (NR)	Hypo- and hypernatremia not associated with symptomatic CV	Hypernatremia: P = 0.2
			Hypernatremia, >145			Hyponatremia: P = 0.6
Morinaga et al, 1992[19]	Japan	Retrospective analysis	Hyponatremia, <130	121 (NR)	Increased incidence of aSAH-induced CV in patients with hyponatremia	P < 0.01

F, female; M, male; NR: not recorded.

Table 4.

Summary of studies assessing the association between interleukins (IL) and C-reactive protein (CRP) and the development of cerebral vasospasm (CV) after aneurysmal subarachnoid hemorrhage (aSAH)

Study	Country	Research design	Patient number (M, F)	Outcomes	P value
Szanto et al, 2024[32]	Hungary	Retrospective analysis	175 (69, 106)	CRP was higher in patients with CV	P = 0.07
Li et al, 2024[33]	Germany	Retrospective analysis	450 (145, 305)	Admission CRP was an independent risk factor for CV	P < 0.001
Lucke-Wold et al, 2022[34]	USA	Prospective analysis	4 (0, 4)	IL-6 peaked 3 days post-SAH and increased in aneurysmal blood and CSF	P < 0.05
Lenski et al, 2017[35]	Germany	Retrospective analysis	63 (21, 42)	Higher CSF IL-6 was observed in patients with vasospasm	P = 0.57
Wu et al, 2016[36]	China	Retrospective analysis	57 (33, 24)	CV patients showed significantly higher levels of IL-6 in CSF	P < 0.001
Kao et al, 2015[37]	Taiwan	Prospective analysis	53 (21, 32)	Aneurysmal IL-6 showed modest to moderate correlations with vasospasm grade	P = 0.038
Nam et al, 2011[38]	South Korea	Prospective analysis	24 (9, 15)	The activation index of IL-1 was higher in patients with symptomatic CV	P = 0.039
Ni et al, 2011[39]	China	Prospective analysis	46 (24, 22)	Patients who developed vasospasm had significantly higher levels of CSF IL-6 on days 1, 2, 3, 5, and 7 post-ictus	P < 0.05
Sarrafzadeh et al, 2010[40]	Germany	Prospective analysis	38 (11, 27)	Insignificant correlation between CSF IL-6 and secondary deterioration in aSAH patients	P > 0.05
Hendryk et al, 2004[41]	Poland	Retrospective analysis	161 (NR)	CSF IL-6 significantly increased in days 0–3 post-aSAH, and aggravated CV	P < 0.05
Osuka et al, 1998[42]	Japan	Prospective analysis	24 (10, 14)	Patients with symptomatic CV had significantly higher levels of CSF IL-6	P < 0.05
Zhong et al, 2017[43]	China	Prospective analysis	101 (31, 70)	Higher levels of CRP predict CV	P < 0.05
Romero et al, 2014[44]	Brazil	Prospective analysis	100 (56, 44)	CV significantly correlated with higher CRP levels	P < 0.0001
Hwang et al, 2013[45]	Korea	Prospective analysis	61 (15, 46)	CRP levels significantly higher on postoperative day 1,3 and 5 in patients with aSAH	Day 1: P = 0.097
					Day 3: P = 0.023
					Day 5: P = 0.03
Jeon et al, 2012[45]	Korea	Retrospective analysis	93 (33, 60)	CRP on day 1 and 2 postpreparative is an independent predictor for symptomatic CV	P < 0.05
Fountas et al, 2009[46]	Greece	Prospective analysis	41 (25, 16)	Patients developing angiographic CV had significantly higher CRP levels	P < 0.0001

F, female; M, male; NR, not recorded; CSF, cerebrospinal fluid.

Discussion

Predictors of cerebral vasospasm (Fig. 1)

Figure 1.

The discussed non-radiological blood markers. Blood markers of cerebral vasospasm occurrence following aneurysmal subarachnoid hemorrhage. Created by Rabii Aboulhosn using Biorender.

Serum sodium

Hyponatremia in the setting of aSAH is linked to cerebral salt wasting syndrome (CSWS) and syndrome of inappropriate antidiuretic hormone secretion (SIADH). CSWS is a phenomenon of excessive natriuresis and osmotic diuresis, which contributes to hyponatremia and decreased cerebral venous pressures[28]. It accounts for 55.1% of hyponatremia cases[15]. SIADH describes dilutional hyponatremia from water retention in euvolemic subjects. The ensuing hyponatremia causes cerebral edema and affects the microcirculation by increasing hematocrit and blood viscosity^[26,28].

Hyponatremia has been associated with the presence of CV post-aSAH.^{[9-11,13,15,16,19,47]} developing in the first 10 days following initial ictus. It may also precede the development of clinically detectable CV, occurring shortly prior (median 1.5 days)[9]. Uozomi et al[11]. reported a sequential decrease in serum sodium levels 1-day preceding CV, albeit eunatremia was maintained. This may suggest that fluctuations in serum sodium, in the context of eunatremia, may also serve as predictors of aSAH-induced CV. Indeed, it has been demonstrated that patients with CV exhibit greater serum sodium fluctuations than aSAH patients without CV[38].

Correcting serum sodium imbalances has been postulated to stabilize the microcirculation and therefore improve outcomes following aSAH[26]. Corticosteroids and hypertonic saline have been advocated in the setting of CSWS to reduce natriuresis and induce hypervolemia.^[29–31] The evidence on the use of corticosteroid use in reducing incidence of CV post-aSAH is conflicting, however. Mistry et al found a reduction in symptomatic CV in patients receiving corticosteroid treatment in a pooled meta-analysis[48]. A fixed-effects meta-analysis of four randomized control trials identified no statistically significant difference of mineralocorticoid treatment in reducing the incidence of symptomatic CV[49]. More recently, vasopressin receptor antagonists (vaptans), have offered promising treatment modalities for hyponatremia in SAH by inducing aquaresis and sodium retention. These agents have been demonstrated to improve outcomes 6 months post-ictus[50].

Overall, hyponatremia is often encountered in the early phase after aSAH, emanating from CSWS and SIADH. Increased serum sodium and urinary sodium excretion following aSAH can be considered as early surrogate markers to predict aSAH-induced CV. Furthermore, serum sodium levels have been demonstrated to predict the need for endovascular treatment of CV, and may therefore be exploited to identify suitable candidates for this intervention[51].

Glycemic indices

Hyperglycemia is a common manifestation post-aSAH, occurring in 70–90% of patients on admission[52]. It is thought to be a product of an increased sympathetic autonomic nervous system response, producing catecholamines and cortisol, and an inflammatory process which stimulates cytokines (including interleukin-1 and -6) and insulin resistance[53]. The ensuing hyperglycemia has been demonstrated to be associated with poor functional outcomes post-aSAH[47].

Shi and colleagues (2022) performed a meta-analysis of studies investigating the prognostic significant of hyperglycemia in patients with aSAH and identified that hyperglycemia was significantly associated with the development of cerebral vasospasm (P = 0.0002)[54]. Interestingly though, strict glycemic control offers minimal benefit in SAH, which may be attributable to the increased incidence of hypoglycemia in this setting.^[32–34]

The glucose-potassium ratio (GPR) has also been explored in the context of aSAH-induced CV. In aSAH, a catecholamine surge induces hypokalemia via stimulation of sodium/potassium-ATPase[35]. Matano et al[36]. examined a cohort of 333 aSAH patients and identified a positive correlation between GPR and CV severity. This is in accordance with the findings of Fujiki and colleagues[37] who demonstrate a relationship between GPR and vasospasm-induced cerebral infarction.

Leucocyte count

Dysregulated inflammation occurs within 72 hours of aneurysmal rupture, emanating from a reaction to extravascular blood, disrupted cerebral autoregulation, and release of products from severed brain tissue[55]. This is accompanied by cytokine-induced transmigration of peripheral leukocytes into the cerebrospinal fluid and brain. The ensuing robust systemic inflammatory response has been associated with poor neurological outcome in aSAH, including development of CV[56].

Early and persistent elevations in leukocytes may be predictive of CV occurrence following aSAH.^[56–60] The rise in leukocytes occurs as early as day 0 post-ictal[59]. Leukocyte levels of >11–15.4 × 10⁹/L have been suggested to predict CV insurgence^[58,59]. The differences in the amount of extravasated blood in aSAH could be related to the different recruitment of immune cell subsets. Similarly, neutrophil to lymphocyte ratios may also offer prognostic value in detecting CV and predicting poor functional outcome,^[39–44] with levels greater than 6.48 found to be associated with less favorable outcomes on ROC analysis.[41]

Interleukins and C-reactive protein

Interleukins (IL) have been suggested to be responsible for the delayed manifestation of CV following aSAH. In the setting of aSAH, IL-6 has been the most widely described. Cytokines fueling the inflammatory response in aSAH, including interleukin-1 and interleukin-6, have been identified as stimuli for c-reactive protein (CRP) synthesis[61].

Both preclinical and clinical studies demonstrate the association between IL-6 levels and the development of CV post-aSAH.^{[20–23,62,63]} Cerebrospinal fluid (CSF) IL-6 has been proposed to be a more reliable prognostic marker of CV post-aSAH than serum IL-6, as the latter exhibits pre-injury variations amongst patient cohorts. Patients with CSF IL-6 levels between 530 and 3100 pg/mL have an increased probability of developing CV[22].

The elevation of IL-6 levels in patients with aSAH is likely triggered by blood accumulating in the subarachnoid space[63]. IL-6 contributes to the development of CV by directly transmigrating into the arterial wall smooth muscle or indirectly, by inducing the activity of vasoconstrictors including ET-1[45]. Given its prominent role in neuroinflammation and its involvement in the pathophysiology of CV after aSAH, IL-6 represent a potential therapeutic target for reducing both the incidence and complications associated with aSAH-induced CV^[46,64].

CRP is a sensitive but non-specific inflammatory marker[65]. Furthermore, interpretation of CRP values in patients with aSAH may be complicated by the fact that some patients frequently have other intercurrent pathologies that can stimulate the acute phase response, including bacterial infections[27]. In aSAH, studies have identified CRP concentration to be an independent predictor of CV occurrence.

CRP has been analyzed in the preoperative and postoperative setting and, in both, has exhibited higher levels in the patients developing CV. The rise in CRP occurs promptly postictal, peaking at day 3[66]. As CRP levels could rise promptly after aneurysmal rupture, identification of a single absolute CRP level may not reflect disease severity in the acute phase. Consideration of regular CRP measurements may add prognostic value post-aSAH.

Future directions

Artificial intelligence and machine learning can learn from complex data to improve predictions and provide interferences. In the setting of aSAH, machine learning has been demonstrated superior capabilities compared to standard learning[67]. Patel et al[68]. developed a digital model for predicting CV following aSAH which integrated data from transcranial Doppler, Fisher score, Hunt Hess grade, aneurysm diameters and white blood cell counts. They identified a true positive and precision ratio of >70%.

A predictive model could facilitate more prompt diagnosis in the acute setting by identifying patients at risk before clinical signs and symptoms manifest. It can also be utilized to stratify patients into CV phenotypes and identify those who warrant closer monitoring or necessitate more aggressive intervention[69].

Limitations

The present research has a few limitations. Most studies were conducted retrospectively, which may predispose to bias. Similarly, it remains unclear from the literature whether the blood markers contribute to the development of CV or are a manifestation of this entity.

Non-blood markers of CV were also not discussed. Furthermore, this was a narrative review to summarize the current findings and no systematic review was conducted. Despite these limitations, this paper provides a summary of our current understanding of CV and aSAH to prompt future investigations.

Conclusion

The studies included in this review suggest that the presence of hyponatremia, hyperglycemia, leukocytosis, and elevated CRP and CSF IL-6 levels are associated with the occurrence of CV post-aSAH. They may therefore serve as a complementary prognostic tool alongside the commonly used clinical and radiological classification systems to help with patient risk stratification following aSAH. Further research in the form of randomized control trials or quasi-experimental designs are required to identify temporal relations and causality of described markers in the setting of aSAH-induced CV. Future research may also wish to consider whether the concurrent presence of described blood markers has combined, prognostic value for predicting the occurrence of vasospasm post-aSAH.

Ethical approval

Ethics approval was not required for this review article.

Consent

Informed consent was not required for this review article.

Sources of funding

None.

Author contributions

R.A.: writing the paper; M.M., B.C., A.A., S.R., and G.P.: study concept or design, data collection, and data analysis or interpretation.

Conflicts of interest disclosure

None.

Research registration unique identifying number (UIN)

None.

Guarantor

Bipin Chaurasia.

Provenance and peer review

Not commissioned, externally peer-reviewed.

Data availability statement

None.