This study showed that the pNfL concentration within 24 hours was an independent risk factor for LVO after an anterior circulation stroke, even after adjusted for potential influencing factors regarded as clinically relevant. In the present study, it was found that LVO patients exhibited higher pNfL and NIHSS scale than Non-LVO patients. Atrial fibrillation and the TOAST classification were also associated with LVO, and the patients with LVO tend to poorer outcome. The pNfL was positively correlated with severity of stroke, slightly correlated with age and the time to blood sampling time. The levels of pNfL showed significant diagnostic accuracy in discriminating patients with LVO from those without LVO, especially when combined with NIHSS. This is the first study that has investigated the pNfL levels in AIS patients with LVO.
NfL, as a protein exclusively expressed in neurons[22], can reflect the severity of neuron injury, and has potential application prospects in patient monitoring, observation and intervention research[21]. Based on the theory that stroke caused by LVO damaged nervous system more serious, so as to express more pNfL. The present study not only confirmed that the expression of pNfL in patients with LVO is higher than that in Non-LVO, but also proved that pNfL has high accuracy in the identification of LVO. Previous studies have shown that NfL expression was associated with dementia[23-26], small vessel disease[27], and other neurodegenerative diseases[28-31]. These results indicate that NfL is a specific marker of nerve injury, which can be highly expressed in a variety of nervous system diseases[32]. Subsequently, an increasing number of studies have demonstrated that NfL levels were associated with clinical characteristics and outcome in stroke patients[13, 17, 18, 21], and the CSF NfL increased months before the first dementia symptoms appeared, suggesting it might serve as a preclinical marker[33]. Interestingly, the higher NfL level was found in a transient ischemic attack (TIA) patient who developed an ischemic stroke 1 day after blood sampling[16]. It might suggest that sNfL releasing ischemic brain injury may have already started before symptoms became clinically apparent. In addition, NfL is related to the clinical severity[34] and can distinguish different nervous system diseases[35], which indicates that the degree of neuronal damage is related to the expression of NfL, and further indicates the feasibility of NfL in differentiating LVO.
The era of endovascular treatment for LVO, the identification and triage of LVO is more complex than ever. With the continuous development of guidelines, tools and systems, an objective and accurate evaluation method is needed to guide the treatment process and the allocation of medical resources. Although previous studies have shown that many LVO screen scales have moderate or even high accuracy[10], the unavoidable defects of LVO screen scales are lack of objectivity and consistency[36, 37]. Especially for inexperienced clinicians, the accuracy of evaluation is significantly lower than that of experienced doctors, and the first medical institution for patients with AIS is often not a comprehensive stroke center. Therefore, the exploration of objective biomarkers has never stopped. Proteins associated with acute LVO pathogenesis and endothelial function may appear in blood samples of AIS patients due to LVO, thus permitting development of blood-based biomarkers for its diagnosis and prognosis. Recent studies identified collateral circulation related biomarker[38] and cardiac biomarkers[39] could provide diagnostic aid to the existing modalities for AIS due to LVO. However, these markers are not neuron specific and may be affected by many factors. Atrial fibrillation, systolic blood pressure and admission NIHSS were also correlated with LVO[40, 41],which is consistent with this study. But the specificity of these indicators is worse. Therefore, NfL is superior to previous biomarkers in the identification of LVO, and the combination of NfL and previous biomarkers may further improve the diagnostic efficiency in the future.
It cannot be denied that the expression of NfL also has some influencing factors. First of all, as mentioned above, NfL may increase on the basis of other nervous system diseases. Therefore, the impact of other nervous system diseases must be excluded in the diagnostic process of using NFL. This study excluded patients with other possible neurological diseases and previous cerebral infarction. Secondly, the expression of NfL changed dynamically with time after stroke[42], so the effect of blood collection time on NfL cannot be ignored. This study showed that there was no significant difference in blood collection time between the two groups. At the same time, NfL was slightly correlated with blood sampling time, which may be related to the relatively concentrated blood collection of patients who are only included within 24 h of onset.
Several limitations to this study should be noted. First, retrospective studies are prone to selection biases. In any case, prospective studies are needed to determine the value of pNfL in making triage decisions for transport of patients with LVO. Future planned studies to expand the cohort, both at our institution and at multicenter, will help to reduce this variance. Second, the small sample size due to strict inclusion criteria raises the risk of chance findings. Third, we were limited to a cross-sectional analysis as longitudinal sNfL measurements were not available. Previous studies showed that sNFL in stroke patients were with a peak at day 7[21], at third week[42], between the acute phase and 3-month post-stroke[43]. More work should be carried out to determine the specific peak value. Finally, single biomarkers may not be sufficient, and multiple biomarkers combined with a machine-learning algorithm should be used to automatically diagnose and predict LVO.