A commonly occurring complication of stroke is the poststroke cognitive impairment (PSCI) refers to a range of syndromes that meet the diagnostic criteria for cognitive disorders within 6 months after the clinical event of a stroke. These vary from conditions like “poststroke cognitive impairment no dementia” (PSCIND) to “poststroke dementia” (PSD), which not only enhances patient mortality but also severely impairs their normal daily life and social ability. Prevention and early intervention are, therefore crucial for alleviating PSCI.

Distant ischemic adaptation or remote ischemic conditioning (RIC) is a new method for the prevention and treatment of cerebrovascular disease. In principle, RIC applies transient ischemic stimulation to one organ to induce an endogenous protective response against severe ischemic injury in another distant organ. For example, remote limb ischemic conditioning (RLIC) applies intermittent interruptions of blood supply to non-vital body organs (the limb) to protect vital organs (the heart or brain) from ischemic injury (Brandli, 2015; Pan et al., 2016).

In general, RIC is divided into distinct types based on the time sequence of stimulation and organ injury: distant or remote ischemic preconditioning (RIPreC), distant ischemia-reperfusion adaptation or RIPerC, and distant ischemia adaptation of remote ischemic postconditioning (RIPostC) (Hess et al., 2015) refers to the protective response induced by transient ischemic stimulation of a distant organ after ischemia-reperfusion injury.

Multiple studies have indicated that RLIC reduces the risk of cerebral infarct and intracranial atherosclerotic stenosis, and decelerates cognitive deterioration in patients with cerebral small-vessel disease (Gonzalez et al., 2014; Hougaard et al., 2013; Meng et al., 2012, 2015; Nikkola et al., 2015; Wang et al., 2017a). However, the impact of RLIC on PSCI has not been clearly defined. The present study analyzes the impact of RLIC performed during the acute phase of ischemic stroke on PSCI to elaborate its mechanism of action. For the evaluation of cognitive function, we used the Montreal Cognitive Assessment (MoCA) scale for neuropsychological evaluation and P300 for objective evaluation. Previous studies show that P300 can be used to identify the early stages of dementia and to distinguish between dementia and depression-related pseudodementia. The efficacy of drugs to improve cognitive function can also be reliably evaluated by P300, which is an effective tool for clinical evaluation of cognitive function (Fabrizio et al., 2011; Parra et al., 2012).

Early recognition and intervention against PSCI in stroke patients are crucial. It is known that 17-92% of stroke patients experience mild cognitive impairment within 3 months, while 6-32% of patients with PSCI progress to dementia (Pasi et al., 2012). While the specific rate of incidence of PSCI may vary in different countries, regions, ethnicities, and diagnostic systems, there is scientific consensus that the risk of PSCI remains very high in stroke patients and is being observed to rise (Pendlebury et al., 2009; Qu et al., 2015; Sun et al., 2014; Yu et al., 2013). PSCI detrimentally impacts patient compliance in physical rehabilitation training, severely hindering limb function restoration and reducing the quality of life and survival of the patient.

Remote limb ischemic conditioning is a nonpharmacological, noninvasive treatment that promises clinical transformation (Meng et al., 2015). Its biggest advantages are the absence of definite adverse effects and a greater chance of patient acceptance. In a recent clinical trial, Hougaard et al. (2013) investigated the feasibility of RLIC in prehospital rescue for patients with acute ischemic stroke and found that RLIC remarkably reduced the risk of cerebral infarct. Further, the Remote Ischemic Conditioning After Stroke Trial study assessed patients within 24 hours of an ischemic stroke who received four cycles of RLIC. Their findings revealed that RLIC improved patients’ National Institute of Health stroke scale (NIHSS) scores in 90 days (England et al., 2017). It has also been demonstrated that brief, repetitive, bilateral arm ischemic preconditioning combined with current standard medical management can reduce stroke recurrence in patients (< 80 years) with symptomatic atherosclerotic intracranial arterial stenosis (SIAS) (Meng et al., 2012).

Moreover, it significantly improved cerebral perfusion and metabolism and reduced focal vascular resistance in the stenotic area. In a later study, the team found that RLIC was also safe and well-tolerated in elderly patients with SIAS (> 80 years) who were not candidates for endovascular interventions, while RLIC could lower the recurrence rate of ischemic events (stroke and transient ischemic attack) (Meng et al., 2012). Together, these studies proved that RLIC is a safe, feasible, and remarkably efficacious intervention for cerebral infarct prevention.

Currently, most studies on RLIC in vascular dementia have been limited to animal experiments. The bilateral common carotid artery occlusion model in rodents is considered the most valid animal model for simulating vascular cognitive impairment (VCI). In a study by Xu et al. (2011), 36 rats randomized into control group, bilateral carotid arteries occlusion group, and RLIC group, determined that neurocognitive function was significantly improved in RLIC group rats as compared to the occlusion model rats in 7-8 days (Xu et al., 2011). Interestingly, they found that RLIC could increase the expression of Bcl-2, reduce apoptosis of nerve cells, and improve nerve function. Interestingly, RLIC has also been suggested to improve cognitive function in cerebrovascular impairment via the induction of autophagy and increased expression of autophagy-related proteins (Wang et al., 2017b).

On the other hand, Khan et al. (2015) found that RLIC could reduce cognitive impairment via improved cerebral blood flow and reduced tissue injury in mice with vascular cognitive impairment (VCI). Additional evidence also points to the role of RLIC in improving cognitive impairment in patients following colon surgery and heart surgery (He et al., 2017; Hudetz et al., 2015). Wang et al. (2017b) applied a five-cycle RLIC to patients with cerebral small-vessel disease (cSVD) twice daily for 1 year and found that the treatment could decelerate cognitive function decline in patients with cSVD. Most importantly, no severe adverse events were reported following RLIC in any of these studies.

Under current standards, neuropsychiatric evaluation remains an important method for the recognition and diagnosis of cognitive impairment and a crucial tool for monitoring treatment efficacy and disease outcome. In PSCI screening, the MoCA scale is more sensitive than the Mini-Mental State Examination (MMSE) and is considered to have higher reliability and validity (Blackburn et al., 2013; Cumming et al., 2013). Our study has found that the MoCA score was higher, and the ADL score lower in the RLIC group than control group patients at 6 months poststroke. Specifically, MoCA scores were significantly higher in the domains of visuospatial and executive functioning, and attention, in RLIC group patients. Additionally, the incidence of PSCI was significantly lower in the RLIC group than in the control group. Overall, these results strongly indicate that RLIC improved cognitive function in patients following stroke and suggest its role as a key prevention strategy against PSCI.

P300 event-related potential is a sensitive test for specific cognitive activities, and its major monitoring indices are latency and amplitude. P300 latency is negatively correlated with the MMSE score, with shorter latencies associated with superior cognitive performance. Therefore, P300 latency can aid in the determination of a decline in cognitive function and therapeutic efficacy, serving as an effective clinical tool. Correlation analyses performed in our study consistently indicated that patients’ MoCA scores were negatively correlated with P300 latency and positively correlated with amplitude, respectively. These findings suggest a relationship between severe cognitive impairment, longer P300 latency, and lower P300 amplitude.

Further, P300 latency was negatively correlated with MoCA scores in the domains of visuospatial and executive functioning and attention, while P300 amplitude was positively correlated with these scores. These results thus suggest a link between severe damage in the stated domains with longer latency times and lower amplitudes. Together, our observation of significantly shorter P300 latency and significantly higher P300 amplitude in RLIC group patients as compared to control group patients, further proves that RLIC improves cognitive function following stroke.

The neuroprotective mechanisms of RLIC are complex, involving multiple factors and signaling pathways, the precise details of are still poorly understood. RLIC treatment alters the expression of multiple genes and proteins, including hypoxia-inducible factor-1α (HIF-1α), HIF-1β, the anti-apoptotic gene Bcl-2, and superoxide dismutase (SOD), all of which are observed to be elevated. RLIC has also been observed to inhibit the expression of the apoptotic protein, p53 (Jin et al., 2016; Salido et al., 2013; Tidball et al., 2016). These changes can affect a range of processes, such as improving the metabolism of nerve cells, inhibiting oxidative damage, and reducing neuronal apoptosis. The use of single-photon emission CT to evaluate the impact of RLIC on metabolism and blood flow restoration showed significantly increased blood flow at affected brain sites (Meng et al., 2012). This radiographically verified the correlation between the cerebral protective effects of RLIC and increased cerebral perfusion in the ischemic region. Our study similarly used TCD detection technique to assess cerebral blood flow to find that RLIC indeed increases mean blood flow in the middle cerebral artery and reduces cerebral artery and basilar artery PI, thereby improving the blood supply to the brain.

ICAM-1 is a key adhesion molecule for leukocytes crossing the blood-brain barrier (BBB). ICAM-1 expression is observed to significantly increase following cerebral ischemia. Leukocytes adhere to endothelial cells, transmigrate, and infiltrate into the extravascular brain parenchyma, becoming concentrated in regions of injury due to the binding of ICAM-1 proteins with corresponding receptors on leukocytes. This results in microvascular occlusion and production of oxygen-free radicals. In turn, cytotoxic enzymes are released, increasing BBB permeability, altering the vasoconstriction response, stimulating the release of inflammatory cytokines, and causing brain tissue damage. Meanwhile, microvascular occlusion aggravates cerebral ischemia, setting up a vicious cycle (Cheng et al., 2008; Shen et al., 2008).

As the most potent vasoconstrictor so far (Kwan et al., 2002), ET-1 is synthesized in vascular endothelial cells, smooth muscle cells, neurons, and glial cells, while its receptors are readily available in the cerebrovascular system. Cerebral ischemia and hypoxic conditions greatly increase the production and release of ET-1. This results in atherosclerosis due to the promotion of vascular smooth muscle proliferation and aggravated inflammation from enhanced neutrophil adhesion. These changes cause considerable damages to nerve cells and vascular endothelial cells, directly impacting synaptic plasticity and injuring the cholinergic system, and therefore affecting cognitive function (Masaki et al., 1991; Selnes and Vinters, 2006). Serum ET-1 levels are strongly negatively correlated with cognitive impairments (Briyal et al., 2011).

Serum ICAM-1 and ET-1 are also important markers of endothelial function (Armentero et al., 2011; Fang et al., 2011). Vascular endothelial cells regulate blood flow, control the permeability of plasma components, and influence the adhesion and aggregation of platelets and cytokines. Therefore, a healthy endothelium plays an important role in the regulation of cerebral vascular homeostasis (Michiels et al., 2003). The pathophysiological effects of vascular endothelial dysfunction on atherosclerosis have been confirmed by numerous studies (Yoshida et al., 2010). In the present study, we have also found that RLIC can reduce serum levels of ICAM-1 and ET-1 and improve endothelial function. We speculate that this favorable effect may be associated with the fact that RLIC improves blood supply to the brain and alleviates cerebral ischemia and hypoxia, which, in turn, reduces serum levels of ICAM-1 and ET-1. Pearson correlation analysis also suggested that the improvement of cerebral hemodynamics may also improve vascular endothelial function.

Finally, baPWV is also a sensitive index for the assessment of atherosclerosis (Sabayan et al., 2013). Increased arterial stiffness can lead to a decrease in arterial buffering function, while increased severity of cerebrovascular injury, cerebrovascular damage, and cerebral hypoperfusion may result in damage to neurons in the vulnerable areas of the brain (such as hippocampal gyrus neurons), leading to cognitive decline (Cavalcante et al., 2011; Najjar et al., 2008; O’Sullivan et al., 2003; Rothwell et al., 2010). In our study, RLIC was able to decelerate baPWV and reduce the degree of arteriosclerosis. This effect might also be associated with a reduction in ICAM-1 and ET-1 levels and improvement in vascular endothelial cell function following RLIC. Pearson correlation analyses on our data consistently suggested that vascular endothelial function can slow baPWV and reduce the degree of arteriosclerosis.

In conclusion, RLIC improved blood supply to the brain and endothelial function, alleviated degrees of arteriosclerosis, improved cognitive function following stroke, and reduced the incidence of PSCI. It especially improved cognitive domains linked to visuospatial, executive functioning, and attention. As such, RLIC can serve as an effective clinical strategy for PSCI improvement.

This research was approved by the ethical review board of Zhabei Central Hospital, Jing'an District, Shanghai, People’s Republic of China, code ZBLL2018062605. All participants gave written consent after being informed about the experimental procedures.

S.X. conceived this project and supervised the experiments. X.F. and S.X. wrote the paper. X.F., L.H., Z.W., L.W., X.D., Q.W., and S.X. performed the experiments and analyzed the data.

This work was supported by the Foundation of Municipal Health Commission of Jing’an District, Shanghai (2018MS11).

The authors declare no conflict of interest.