Early amantadine treatment reduces the risk of death in patients with large hemisphere infarctions:a Chinese hospital-based study

Large hemispheric infarction (LHI), which usually results from occlusion of the internal carotid artery or proximal proximal middle cerebral artery (MCA), is one of the most devastating condition with high mortality and disability rate among acute ischemic stroke (AIS)patients [1‐3]. Until recently, no pharmacological strategies have been proven effective by randomized controlled trials (RCTs) [ 4]. Although decompressive hemicraniectomy (DHC) within 48 h has been proven to benefit LHI patients with malignant brain edema (MBE) [5], only highly selected cases would be eligible for DHC based on the strict eligibility criteria in the DHC trials [6]. Furthermore, because of its invasive nature and the need for multidisciplinary cooperation, DHC is also underused worldwide. Although there is a lack of evidence, many neuropharmacological therapies are still used off-label due to the lack of effective treatment in conservatively treated LHI patients.

Amantadine was first synthesized more than 50 years ago and initially developed as an antiviral agent [7]. Further work has demonstrated the effectiveness of amantadine as an anti-parkinsonism agent [8, 9], as well as a treatment option for neuroleptic-induced extrapyramidal symptoms [10, 11] and neuroleptic malignant syndrome [12]. Amantadine hydrochloride is one of the most frequently prescribed drugs for patients with prolonged impaired consciousness after traumatic brain injury (TBI) [13]. The mechanism of action of amantadine is still not fully understood, but it is thought to act as an N-methyl-D-aspartate (NMDA) receptor antagonist increasing dopamine synthesis and release in the striatum [14]. Sufficient synaptic dopamine levels are necessary for many physiological functions including motivation, motor control, emotion, and cognitive processing [15]. The result of a small sample randomized trial in patients with TBI-associated diffuse axonal injury suggested that amantadine was effective in improving neuro-recovery and it was well tolerated at a dosage of 200 mg/day with no serious adverse side effects [16]. In a multicenter trial, amantadine at doses of 200–400 mg/day has successfully been demonstrated effective and safe in accelerating the pace of functional recovery in patients with severe TBI [17]. A clinical-experimental study conducted in Russian reported that amantadine exhibited significant restoration of consciousness and better regress of neurological deficit on the first day of AIS [18].

So far, there is a lack of studies investigating amantadine for the treatment of patients with non-traumatic brain injuries such as severe stroke. The present study aimed to explore whether amantadine treatment is associated with better clinical outcomes in conservatively treated LHI patients.

During the 3-year study period, a total of 1574 AIS patients were screened. None of the patients administered intra-arterial revascularization procedures. We excluded 1416 patients. Finally, 31 LHI patients treated with amantadine combined with ST and 127 patients treated with ST were recruited to the current study. A flow diagram of included and excluded patients is provided in Fig. 1.

The baseline characteristics of LHI patients between groups are presented in Table 1. Patients treated with amantadine had a shorter prehospital delay (median: 2 vs. 10 h, p < 0.001), a higher baseline NIHSS score (21.71 ± 4.76 vs. 17.49 ± 5.84, p < 0.001) and a higher rate of dominant hemisphere involvement (67.74% vs. 45.67%, p = 0.028). The two groups did not differ in age, sex, baseline GCS score, baseline blood pressure and serum glucose, vascular risk factors, and stroke etiology (all p > 0.05).

After PS-matching, we identified two subgroups of 72 LHI patients that were balanced for all baseline characteristics, including 27 treated with amantadine combined with ST and 45 treated with ST. Relative multivariate imbalance in the form of the L1 measure was smaller (0.463 vs. 0.649) and no covariate had standardized mean differences of greater than 0.1 after PSM. Supplemental Fig. 1 also showed that covariate balance was massively improved in the matched dataset. As shown in Table 2, there was no significant difference in the baseline characteristics between groups in the matched dataset (all p > 0.05).

For the acute phase treatment of LHI, the two groups did not show differences in receiving thrombolysis, mechanical ventilation, and osmotic agents before and after PSM (all p > 0.05, Table 3). The mean duration of amantadine treatment was 28.71 days before PSM and 30.96 days after PSM. Outcomes of LHI patients between groups are shown in Tables 3 and 4.

The present study determined whether amantadine treatment is associated with better clinical outcomes in conservatively treated LHI patients. To the best of our knowledge, this is the first study to investigate the treatment effect of amantadine in patients with severe stroke. Although it was not a randomized study, and there might be some subtle differences in clinical presentation which would have influence on the decisions of treating clinicians to prescribe amantadine or not. However, in the present study, the amantadine group had a higher baseline NIHSS score (21.71 ± 4.76 vs. 17.49 ± 5.84) and higher rate of dominant hemisphere involvement (67.74% vs. 45.67%) than the ST group before PSM. The amantadine group also had a higher rate of brain edema (93.55% vs. 31.50%) and hemorrhagic transformation (45.16% vs. 24.41%). Neither supports that treating clinicians tend to prescribe amantadine in the subjects most likely to survive. Our results suggested that early amantadine treatment was associated with significantly lower rates of in-hospital and 3-month death and nonsignificant trend of favorable outcomes in conservatively treated LHI patients, after adjusting for age, baseline NIHSS score, and other confounders which might influence the decisions to prescribe amantadine, or in the final baseline characteristics-matched dataset. In our study, a total of 30 patients died during hospitalization and 60 patients died at 3-month follow-up. The most common cause of 3-month death was brain herniation in both groups (6 patients in the amantadine group and 26 patients in the ST group), followed by pneumonia (3 patients in the amantadine group and 13 patients in the ST group). For financial concerns, religion, and values, some critically ill patients were given up treatment by their family members and died within a few days after discharge, especially those patients from rural areas. In the present study, 6 patients in the amantadine group and 24 patients in the ST group died after discharge, however, among those patients, 5 (83.33%) in the amantadine group and 10 (41.47%) in the ST group had withdrawal of care. Thus, our results could not be explained by that amantadine resulted in more wakefulness then perhaps families would have withdrawn care less often making it appear that amantadine resulted in lower mortality. Meanwhile, most of the survivors with severe disabilities were taken home directly or opted for hospice care, rather than seeking further speech therapy and physiotherapy, which could explain why the follow-up data from LHI patients recovering from such a devastating condition are so close to their discharge [27].

Stroke is the second most common cause of death worldwide and the leading cause of death in China [29]. Ischemic stroke, which accounts for approximately 80% of all strokes, is a serious disease with a complex pathophysiology. Various agents that may interfere with each step of the ischemic cascade can, in theory, be developed as a candidate for stroke treatment [30]. Two major strategies have been developed to treat ischemic stroke: recanalization (including thrombolysis and mechanical thrombectomy) and neuroprotection [31]. During the last three decades, thousands of neuroprotective agents have been tested in animal models with reports of experimental efficacy, and nearly 200 neuroprotection clinical trials are ongoing or have been completed, but none has been proven effective in clinical trials [32]. Despite the failure of most neuroprotective drugs, neuroprotection has never been abandoned especially with the help of improved preclinical testing and clinical trial design [33, 34].

LHI, which usually results from a failure in successful reperfusion of occlusion of the internal carotid artery or proximal MCA, is a devastating disease with high mortality rate [1‐3]. Although DHC has been proven to benefit LHI patients with malignant course [5], only highly selected patients younger than 60 years of age would be eligible for DHC [6]. Until recently, no pharmacological strategies have been proven effective by clinical trials [4]. Despite the lack of evidence, many neuropharmacological therapies are still used off-label in clinical practice due to the lack of effective treatment in conservatively treated LHI patients.

In the present study, we found that early amantadine treatment was associated with a significantly lower death rate and nonsignificant trend of favorable outcomes in conservatively treated LHI patients. These results are similar to a clinical-experimental study that was carried out in patients with AIS in Russian and reported that amantadine sulfate (PK-Merz) exhibited significant restoration of consciousness and better regress of neurological deficit in the first hours of AIS [18]. There are two main reasons for the effects of amantadine. First, amantadine has profound NMDA receptor antagonist effects, which may contribute to its neuroprotective effects when administered in the first several hours or days after brain injury [16]. It is theorized that it can block the activation of glutamate receptors and other NMDA channels to inhibit the elevation of intracellular Ca²⁺ levels. Excess of intracellular Ca²⁺ plays a unique role in the ischemic pathophysiology by activation of a variety of Ca²⁺ dependent enzymes, leading to irreversible mitochondrial damage, inflammation, cytotoxic edema, necrotic and programmed cell death [35, 36]. Second, amantadine may promote dopaminergic activity by facilitating presynaptic release and blocking postsynaptic reuptake, which is related to the enhancement of endogenous recovery mechanisms. The favorable functional consequences of amantadine may reflect the strengthening of neurotransmission in the dopamine-dependent nigrostriatal, mesolimbic, and frontostriatal circuits that are responsible for mediating arousal, drive, attentional, and cognitive functions [37, 38]. A systematic review concluded that improvement in arousal and cognitive function had been observed in patients with TBI when amantadine was initiated 3 days to 5 months after brain injury [39]. Two case studies using positron emission tomography to evaluate the effects of amantadine on chronic TBI showed that amantadine treatment would result in arousal and executive function improvement and a significant increase in prefrontal cortical metabolism and a nonsignificant increase in striatal D2 dopamine-receptor availability [40, 41]. Results of the above studies support the role of amantadine in enhancing neurological rehabilitation through the dopaminergic system independent of its neuroprotective effects. In a multicenter trial, amantadine with a treatment duration of 4 weeks has successfully been demonstrated effective and safe in accelerating the pace of functional recovery in patients with severe TBI, and the investigator of the trial indicated that the response of amantadine is drug-dependent [17]. In our study, the mean duration of amantadine treatment was 28.71 days before PSM and 30.96 days after PSM, however, we did not find a significant effect in 3-month functional outcome. Since amantadine has the dual effect of neuroprotection and recovery enhancement, it is reasonable to choose a treatment duration of 4 weeks or longer in future research.

The seizure is a side effect of the biggest concern to clinicians when treating LHI patients with amantadine. However, poststroke seizures are not frequent in LHI patients with an incidence rate of 7.0% in our previous study [27]. In our study, LHI patients treated with amantadine did not suffer a higher rate of seizures before and after PSM. Amantadine’s effect on the seizure threshold is dose-dependent. It has been demonstrated that lower doses of amantadine may elevate the seizure threshold, while high doses (greater than 400 mg/day in adults) may induce seizures [42]. Similar dosages of amantadine to our study even may be an effective alternative treatment for some types of seizures [43, 44], which most likely results from its NMDA receptor antagonist effects [45].

It’s worth noting that so few LHI patients received intravenous thrombolysis and mechanical ventilation in the present study. This could be explained by that our study was a retrospective, Chinese hospital-based study enrolling LHI patients admitted from February 2012 to January 2015. In that period, the rate of thrombolysis in China is relatively low, and early CT hypodensity sign and severe stroke were considered as contraindications or relative contraindications to intravenous thrombolysis, so clinicians were less positive in performing intravenous thrombolysis in patients with anterior circulation large vessel occlusion. In the area where our hospital located, many families hesitate to accept invasive treatments such as mechanical ventilation due to financial concerns, religion, and other values. In our previous published study [27], which included LHI patients from October 2011 to September 2014, only 2.7% cases received thrombolysis, and only 11.7% cases were treated with mechanical ventilation. In the current study, we excluded LHI patients who had received DHC during hospitalization (who were more likely to receive mechanical ventilation), so the rate of cases receiving mechanical ventilation was even lower than that reported in our previous study. As a result, the effect of amantadine needs to be confirmed in LHI patients who had received reperfusion therapy and mechanical ventilation in further study.

The results of our study provide initial evidence that early amantadine treatment was associated with a significantly lower death rate and nonsignificant trend of favorable outcomes in conservatively treated LHI patients. Considering the limitations of observational study, randomized controlled trials with a large sample size may help provide a clearer picture of the utility of amantadine in LHI patients.