Clinical study
Association between elevated cerebrospinal fluid D-dimer levels and delayed cerebral ischaemia after aneurysmal subarachnoid haemorrhage

https://doi.org/10.1016/j.jocn.2020.04.041Get rights and content

Highlights:

  • CSF D-dimer levels are elevated in patients that develop DCI.

  • CSF plasminogen is not predictive of DCI.

  • CSF D-dimer/plasminogen ratio is predictive of DCI.

Abstract

Delayed cerebral ischaemia (DCI) after aneurysmal subarachnoid haemorrhage (aSAH) is a major contributor to morbidity and mortality. It is currently not possible to reliably predict patients at risk of DCI after aSAH. The aim of this study was to quantify cerebrospinal fluid (CSF) D-Dimer and plasminogen levels and to investigate any association with development of DCI. Cerebrospinal fluid (CSF) samples collected from 30 patients within 72 h post-aSAH (n = 13 DCI and n = 17 non-DCI patients) were analysed. DCI was diagnosed when angiographic vasospasm was detected in the presence of new onset neurological deficit. Enzyme-linked immunosorbent assays were used to quantify D-dimer concentrations while western blotting was used to quantify plasminogen levels. Significant differences in CSF proteins between DCI and non-DCI cohorts were verified using Mann-Whitney test. Sensitivity and specificity of these proteins for detecting DCI was examined using a ROC curve and verified with a Fischer’s exact test. CSF levels of D-dimer within 72 h post aSAH were significantly elevated in DCI patients (54.29 ng/ml, 25.35–105.88 ng/ml) compared to non-DCI patients (26.75 ng/ml, 6.9–45.08 ng/ml) [p = 0.03]. In our sample population, D-dimer levels above 41.1 ng/ml had a sensitivity of 69.2% and specificity of 75% for predicting DCI. CSF levels of plasminogen (DCI: 0.50 signal-intensity/μl, 0.20–0.73 signal-intensity/μl, non-DCI: 0.28 signal-intensity/μl, 0.22–0.54 signal-intensity/μl) did not differ between the DCI and non-DCI cohort (p > 0.05). Our study suggests that elevated D-dimer in the first 72 h after aSAH may be a potential predictive biomarker for DCI.

Introduction

The incidence of non-traumatic subarachnoid haemorrhage (SAH) in Australia is 10.3 cases per 100,000 person-years [1] and it is associated with significant morbidity and mortality. Delayed cerebral ischaemia (DCI), a post-ictus complication, is partly responsible for the high mortality and morbidity rates of aneurysmal subarachnoid haemorrhage (aSAH). Typically occurring between days 4 and 12 post-aSAH in approximately 46% of patients [2], DCI is often reversible but can also lead to infarction and significant morbidity and even death [3] despite maximal therapy (Table 1).

Although DCI has historically been associated with reversible cerebral angiographic vasospasm, current research shows that both can occur independently of each other [3] and a number of aetiologies have been proposed including microvascular dysfunction [4], microthrombosis [5], [6], spreading depolarization [7] and burden of clot in the subarachnoid space [8], [9]. Fisher et al., in 1980, first established that greater amounts of blood in the subarachnoid space, seen on CT, correlated with higher incidence of angiographic vasospasm [8]. The lysis of clots (fibrinolysis) begin within hours after SAH and complete clearance of blood by haemolysis from the CSF can range from 6 to 30 days [10]. Several metabolites of haemolysis have been associated with angiographic vasospasm including oxyhaemoglobin, eicosanoids and free radicals [11]. Hence, proteins in the fibrinolytic and haemolytic pathway may be potential biomarkers of delayed cerebral ischaemia.

Plasminogen is an inactive plasma zymogen produced in the liver. Normally the blood brain barrier excludes large proteins like plasminogen but in the event of SAH, plasminogen enters the subarachnoid space with arterial blood and is incorporated into the fibrinous meshwork complex. Plasminogen is activated primarily by plasminogen activators, forming plasmin. In SAH, the source of plasminogen activators is believed to be the arterial blood entering the subarachnoid space which is made conducive by the breakdown of blood brain barrier, inflamed meningeal blood vessels and platelets. Plasmin, then mediates several pathways that lead to fibrinolysis [12]. As plasmin is highly potent, it is normally stringently regulated by several mechanisms in the plasma [12], although the nature of these regulations in the subarachnoid space remain largely unknown. With fibrinolysis, several degradation products are released, including vasospastic factors like oxyhaemoglobin [11] and fibrin degradation product D-dimer, which is a surrogate marker of fibrinolytic activity [13].

Regulators of the fibrinolytic pathway have shown to be potential biomarkers of DCI [14], [15], however plasminogen itself has not been studied. CSF concentrations of D-dimer has also previously shown promise as a biomarker [16] although it was not investigated in the early days immediately after aSAH. Hence, it is useful to study the CSF concentrations of these potential biomarkers in the early stages post-aSAH.

Section snippets

Patient population

Ethics approval for this retrospective cohort study was obtained from Melbourne Health Human Research Ethics Committee (MH project number 2012.50). Patients presenting to the Royal Melbourne Hospital with confirmed aSAH from 2012 till 2018 were included as part of an ongoing database. Consecutive patients were recruited with informed consent, either from themselves or from their next of kin. Aneurysmal SAH was confirmed by blood on computed tomography or xanthochromia on lumbar puncture.

Patient characteristics

In this patient cohort, 47% were male and 53% were female. Median post-aSAH day of CSF collection was 2 and the mean age of patients was 58.6. Although the median admission WFNS score was higher for patients who developed DCI (non-DCI was grade 2, DCI was grade 3), gender, age and day of CSF collection were similar across both cohorts (DCI and non-DCI). A binary logistic regression analysis revealed that none of these variables significantly contributed to the incidence of DCI.

Discussion

Although the exact pathogenesis of DCI remains unclear, it has long been recognised that the higher the blood volume in the subarachnoid space the higher the risk of subsequent DCI and it is believed that toxic blood breakdown products such as oxyhaemoglobin and haem contribute to the development of DCI via oxidative stress [17], [18], [19], [20] and inflammation [4]. Fibrinolysis is one of the factors that determines the rate of resolution of subarachnoid clot and therefore the exposure to

Conclusion

Our study suggests CSF D-dimer levels and D-dimer/plasminogen ratio within the first 72 h post-aSAH are significantly elevated in patients that later develop DCI. We found that CSF levels of plasminogen within the same time frame did not correlate with incidence of DCI. In our sample population, D-dimer and D-dimer/plasminogen ratio shows good sensitivity and specificity for predicting DCI, however a larger prospective study is needed to validate these preliminary findings. Additionally, our

Declaration of Competing Interests

None.

Funding

  • SJ, AA and JZ acknowledge funding from JT Reid Charitable Trust and Neuroscience Foundation for materials used in this research.

  • JM acknowledges Fellowship support from the Australian National Health and Medical Research Council.

Acknowledgements

We would like to thank Dr Dalia Ponce Garcia for assistance with SDS-PAGE, western blotting, and imaging densitometry.

References (23)

  • L. Lai et al.

    Incidence of subarachnoid haemorrhage: An Australian national hospital morbidity database analysis

    J Clin Neurosci

    (2012)
  • J.I. Weitz et al.

    A test in context: D-dimer

    J Am Coll Cardiol

    (2017)
  • J.I. Suarez et al.

    Aneurysmal Subarachnoid Hemorrhage

    N Engl J Med

    (2006)
  • M.D.I. Vergouwen et al.

    Definition of delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage as an outcome event in clinical trials and observational studies

    Stroke

    (2010)
  • J.R. Geraghty et al.

    Delayed cerebral ischemia after subarachnoid hemorrhage: beyond vasospasm and towards a multifactorial pathophysiology

    Curr Atheroscler Rep

    (2017)
  • B. Friedrich et al.

    Experimental subarachnoid hemorrhage causes early and long-lasting microarterial constriction and microthrombosis: an in-vivo microscopy study

    J Cereb Blood Flow Metab

    (2012)
  • M. Sabri et al.

    Mechanisms of microthrombosis and microcirculatory constriction after experimental subarachnoid hemorrhage

    Acta Neurochir Suppl

    (2013)
  • J. Woitzik et al.

    Delayed cerebral ischemia and spreading depolarization in absence of angiographic vasospasm after subarachnoid hemorrhage

    J Cereb Blood Flow Metab

    (2012)
  • C.M. Fisher et al.

    Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning

    Neurosurgery

    (1980)
  • H. Kasuya et al.

    Thrombin activity in CSF after SAH is correlated with the degree of SAH the persistence of subarachnoid clot and the development of vasospasm

    Acta Neurochir (Wien)

    (1998)
  • R. Macdonald et al.

    Cerebral vasospasm

    (2001)
  • Cited by (6)

    • Comparative proteomic analysis of ventricular and cisternal cerebrospinal fluid in haemorrhagic stroke patients

      2023, Journal of Clinical Neuroscience
      Citation Excerpt :

      A final biomarker found in our analysis was d-dimer. This finding builds on previous work showing potential predictive properties of this marker with differential early d-dimer values in aSAH patients with and without DCI [26]. Our findings show definitively the absence of d-dimer in a single patient sample, patient 8 who had an intra-parenchymal haemorrhage with no subarachnoid component.

    • Cerebral vasospasm after subarachnoid hemorrhage: Developing treatments

      2021, Brain Hemorrhages
      Citation Excerpt :

      Thus, a considerable number of predictive biomarkers for cerebral vasospasm has been investigated. These biomarkers include miRNA, mRNA, fibrin degradation product, soluble receptor, and calcium binding protein.144–149 For example, a prospective transcriptomic analysis of blood samples from 44 SAH patients revealed that sphinogosine-1-phosphate receptor 4 (S1PR4) mRNA level was significantly higher in patients who developed vasospasm following SAH than in patients who did not.147

    View full text