Randomized, placebo-controlled, double-blind, pilot trial to investigate safety and efficacy of Cerebrolysin in patients with aneurysmal subarachnoid hemorrhage

After establishing the diagnosis of a ruptured intracranial aneurysm by either computed (CT) tomography angiography or catheter angiography, patients were randomly assigned to receive either intravenous Cerebrolysin (intervention group) or normal saline infusions (placebo group). In prospective studies focused on ischemic stroke, Cerebrolysin dosages varied from 10 to 50 ml per day [32, 34, 37, 44, 45]. According to two phase III RCTs, a daily dose of 30 ml Cerebrolysin was administered for a duration of 10 to 21 days [32, 44]. It was decided to adopt a similar daily dose for 14 days in order to cover the period where DCI most frequently occurs after SAH [46]. Subjects in the intervention group received 30 ml of IV Cerebrolysin per day for 14 days. Eight-hourly 30-min IV infusions consisting of 10 ml of Cerebrolysin diluted with saline to a total volume of 100 ml was given. Subjects in the control group received 100 ml of saline every 8 h for the same period. Allocation was performed according to a predefined block randomization plan generated by the Statistical Package for the Social Sciences version 22.0 (SPSS Inc., Chicago, IL, US). A block size of 10 was used and group allocation within each block was conducted with a 1:1 ratio. Assignment instructions were sealed in envelopes and opened after subject recruitment. Patients and study outcome assessors were blinded to group identity while clinicians directly involved in their medical management were aware. Since Cerebrolysin carries a yellow tint, infusion bags were wrapped in opaque plastic and amber-coloured IV tubing were utilized for both study groups in order to mask the infusate administered.

All recruited subjects were treated according to the latest 2012 American Heart Association (AHA)/ American Stroke Association (ASA) Guidelines for the Management of Aneurysmal SAH [47]. In the initial phase all patients were treated at the neurocritical care unit with adherence to Class I and IIa recommendations for the management of DCI: four-hourly doses of Nimodipine 60 mg was administered for 21-days, euvolemia was maintained, induction of hypertension was performed if DCI was diagnosed and intra-arterial vasodilator therapy was performed in patients with symptomatic cerebral vasospasm unresponsive to hypertensive therapy [47]. When considered fit for post-stroke rehabilitation, all subjects were enrolled into a minimum two-week standardized early inpatient mobilisation physiotherapy program and occupational therapy for basic activities of daily living training.

Data from clinical records, operation notes, medication dispensing records, laboratory and radiological investigations were collected by an independent neurosurgeon, with 5 years of specialist experience, without knowledge of the subject’s group assignment. Clinical severity of SAH was classified according to the modified World Federation of Neurosurgical Societies (WFNS) grading scale. This categorization is regarded as a core data element for SAH trials according to the National Institutes of Health (NIH)/ National Institute of Neurological Disorders and Stroke (NINDS) Common Data Elements Project (CDE) [48, 49]. Good-grade WFNS was defined as grade I or II, i.e. patients presenting with a GCS of 14 or 15. The Acute Physiology and Chronic Health Evaluation (APACHE II) score, an independent predictor for in-hospital mortality for SAH patients, was also calculated [50]. The Charlson Comorbidity Index (CCI), a validated prognostic system for ischemic stroke patients, comprising of a weighted score of 17 comorbidities based on the International Classification of Diseases, Ninth Revision, was determined [51, 52]. The degree of SAH on the first CT brain scan was evaluated according to the modified Fisher’s grading and Hijdra scoring systems [53, 54]. The Hijdra system consists of a semiquantitative assessment of the amount of blood identified in 14 regions of interest with a score of > 22 being an independent predictor for poor mRS functional outcome at six-months [55].

Global functional outcome was evaluated by an independent assessor, a registered nurse, at 30 days, 3 months and 6 months after ictus using the following instruments: the Extended Glasgow Outcome Scale (GOSE), mRS and BI. The primary endpoint was favourable GOSE performance, defined as grades 5 to 8 (moderate disability to good recovery) at 6 months after ictus. Multi-dimensional secondary study endpoints were also assessed: mRS, BI, neurocognitive function and quality of life at 30 days, three- and six-months. In particular, mRS was considered as a highly recommended outcome measure for disability according to the CDE Project working group and favourable recovery was defined as a score of 0–3 (asymptomatic to moderate disability) [49, 56]. A BI cut-off score of > 75 was also utilized to define a favourable outcome [57]. Neurocognitive performance was evaluated by the Montreal Cognitive Assessment (MOCA), another highly recommended CDE Project outcome metric, and the Neurobehavioral Cognitive State Examination (NCSE) [49, 56]. Quality of life (QoL) was appraised by adopting the Chinese version of the 36-item Short Form Health Survey questionnaire (SF-36) and the Stroke-specific QOL Scale (SS-QOL). The occurrence of DCI, cerebral vasospasm and radiological evidence of cerebral infarction were determined by an independent neurosurgeon and neuro-radiologist. DCI was defined as a decrease in GCS of > 2 points or the development of focal neurological deficit for at least 1 h not related to post-treatment complications, rebleeding, hydrocephalus, infection, electrolyte/ metabolic disturbances or seizures according to Vergrouwen et al. and recommended by a recent systematic review of standardized SAH outcomes [5, 58]. According to our institution’s management protocol, seizures were detected by performing bedside electroencephalography (EEG) for all patients with delayed clinical deterioration or if they had a history of seizures on presentation. Cerebral vasospasm was defined as angiographically detectable moderate-to-severe arterial narrowing not attributable to atherosclerosis, catheter-induced spasm or vessel hypoplasia. A transcranial Doppler ultrasound reading of the middle cerebral artery (MCA) with a mean velocity of > 120 cm/sec or a Lindegaard ratio, MCA: internal carotid artery (ICA) mean velocity, of > 3 was considered indicative of cerebral vasospasm [6]. SAH-related cerebral infarction was defined as demonstrable CT or MRI evidence within 6 weeks of ictus, that was absent on scans performed 24 to 48 h after aneurysm occlusion and was not judged to be a complication of neurosurgical intervention [5]. Finally, 30-day mortality, three- and six-month mortality were also recorded.

Cerebrolysin-related severe adverse effects (SAEs) are rare. They are defined as hypersensitivity reactions such as anaphylactic shock, seizures and acute renal failure [59]. Other AEs are generally infrequent, transient and mild: agitation, headache, vertigo, gastrointestinal symptoms such dyspepsia, diarrhoea, constipation, nausea and vomiting [59].

Patients assigned to receive Cerebrolysin were monitored by the treating clinician for any changes in vital signs as well as in their general physical and neurological examinations. Laboratory tests were evaluated for abnormalities attributable to Cerebrolysin. If SAEs occurred the decision for premature trial termination was made by the study investigators.

Since the treatment effects of Cerebrolysin in aneurysmal SAH are unknown, to determine the sample size for this pilot study, trials that focused on ischemic stroke were used as a reference. The largest phase III trial for Cerebrolysin in ischemic stroke determined that a sample size of 990 subjects, with an α-level of 0.025 (one-sided) and 90% power, was required to detect treatment superiority over standard care alone for functional performance at 3 months [44]. By adopting a Bayesian decision-theoretic approach, it was proposed that phase II trials should have a sample size approximately 0.03 times that of a subsequent phase III study, which in this case would be 30 (= 990 × 0.03) subjects in total [60]. However, it was ultimately decided to increase the sample size to 50 subjects since Sim et al. advocated that this number was the absolute minimum required to decide on whether to proceed with a main RCT with a 5% two-tailed α-level and 80% power [61]. All analyses were performed on a modified intention-to-treat basis where the last available observed outcome measure was carried forward to handle missing data. In addition to assessing dichotomized six-month GOSE and mRS outcomes as either favourable or unfavourable, their ordinal character lends itself to proportional odds analysis. This approach was described by Weir et al. to have superior efficiency compared to the conventional binary analysis and was therefore also utilized in this study [62]. Prespecified subgroup analyses were performed for age (> 65 or < 65 years), pre-existing hypertension, modified WFNS grade (good-grade: I, i.e. an admitting GCS of 15 or II, i.e. GCS of 14 versus poor-grade: III, i.e. GCS 13; IV, i.e. GCS 7–12 or V, i.e. GCS 3–6), modified Fisher grading (I, II or III, IV), aneurysm location (anterior or posterior circulation) and treatment modality (endovascular therapy or microsurgical clipping). The reporting of this study was in accordance with the recommendations outlined in the Consolidated Standards of Reporting Trials (CONSORT) statement [63]. Statistical tests included logistic regression, the chi-squared test, the independent-samples t-test, the Wilcoxin Mann-Whitney U-test and proportional odds analysis. An α-level of 0.05 was used to define statistical significance. Tests were performed by either using Statistical Package for the Social Sciences software version 20.0 (SPSS Inc., Chicago, Illinois, USA) or R version 3.3.2 (R Foundation for Statistical Computing, Vienna, Austria).