We conducted a single-centre, observational, before-and-after study at the Department of Neurology and Clinical Neurophysiological Explorations of Ibn Rochd University Hospital, Casablanca, Morocco. Ibn Rochd is a 1800-bed tertiary academic hospital affiliated to Hassan II University and serves a catchment area of approximately five million inhabitants in the Greater Casablanca metropolitan area. The neurology department operates the only 24/7 stroke alert pathway in the public sector of the city. Intravenous thrombolysis is delivered according to the local thrombolysis protocol, adapted from the AHA/ASA and ESO recommendations.

The study covered two consecutive periods, defined according to the opening of the in-hospital trauma centre on 1 August 2024:

Pre-trauma centre period: from 1 September 2023 to 31 July 2024.Post-trauma centre period: from 1 August 2024 to 30 September 2025.

The opening of the trauma centre was accompanied by the co-localisation of emergency, radiological and neurological teams in a single building, the implementation of a fast-track stroke circuit with priority access to a 1.5-T magnetic resonance imaging (MRI) scanner, and the standardisation of the in-hospital alert pathway under a unified electronic protocol.

All consecutive adult patients (≥18 years) admitted to the stroke-alert pathway between September 2023 and September 2025 were screened for eligibility. A total of 403 consecutive patients fulfilled the inclusion criteria and were included in the final analysis: 176 (43.7%) during the pre-trauma centre period and 227 (56.3%) during the post-trauma centre period. Patients with incomplete demographic, clinical or imaging data were excluded from the analysis.

Data were extracted retrospectively from the patients’ electronic medical records and from the prospective stroke alert registry of the department. The following variables were collected:

1) Demographic data: age, sex.

2) Vascular risk factors: hypertension, diabetes mellitus, dyslipidaemia, current smoking, atrial fibrillation, history of previous stroke or transient ischaemic attack.

3) Clinical severity: National Institutes of Health Stroke Scale (NIHSS) at admission, scored by a vascular neurology resident or attending neurologist.

4) Workflow intervals: onset-to-door time (from the last time the patient was known to be well to arrival at the hospital), door-to-needle time (from arrival to alteplase bolus), and onset-to-needle time (sum of both), all expressed in minutes.

5) Initial imaging modality: cerebral computed tomography (CT) or magnetic resonance imaging (MRI), with MRI protocols including diffusion-weighted imaging, fluid-attenuated inversion recovery, T2*-weighted gradient-echo and time-of-flight magnetic resonance angiography.

6) Final aetiological diagnosis (ischaemic stroke, intracerebral haemorrhage or stroke mimic), classified according to the topography and aetiology after completion of the diagnostic work-up.

7) Treatment delivery: administration of intravenous thrombolysis with alteplase (0.9 mg/kg, maximum 90 mg).

Continuous variables were summarised as mean ± standard deviation or median (interquartile range), according to their distribution. Normality was assessed before statistical analysis. Comparisons between the pre-trauma and post-trauma periods were performed using Student’s t-test for normally distributed continuous variables and the Mann-Whitney U test for non-normally distributed variables, including NIHSS scores and workflow time intervals. Categorical variables were compared using the Pearson χ² test or Fisher’s exact test, as appropriate. A two-sided p value < 0.05 was considered statistically significant. All analyses were performed using SPSS Statistics version 25 (IBM Corp., Armonk, NY, USA).

The study was conducted in accordance with the principles of the Declaration of Helsinki and the local regulations governing retrospective observational research. Given the retrospective and non-interventional nature of the study, the local institutional review board waived the requirement for individual written informed consent. All data were anonymised at the source.

Table 1. Baseline demographic and clinical characteristics of the study population, before and after the opening of the trauma centre.

NIHSS = National Institutes of Health Stroke Scale; SD = standard deviation; NS = not statistically significant (p > 0.05).

Between September 2023 and September 2025, a total of 403 consecutive patients were admitted to the thrombolysis alert pathway and included in the analysis: 176 (43.7%) before the opening of the trauma centre and 227 (56.3%) after. The mean age of the overall cohort was 65 ± 13.8 years, with a slight male predominance (52% men). The median NIHSS score at admission was 12 (interquartile range, 7 - 18), reflecting the moderate-to-severe stroke severity typical of patients evaluated in the thrombolysis alert pathway. The demographic and clinical baseline characteristics of the two groups are presented in Table 1.

Ischaemic strokes accounted for the majority of the final diagnoses. The middle cerebral artery (MCA) territory was the predominant topography (70% of patients), followed by vertebrobasilar strokes (11%), stroke mimics (10%) and intracerebral haemorrhages (9%). Stroke mimics included most frequently focal epileptic seizures, hypoglycaemia, conversion disorders and complicated migraine episodes. The distribution of final diagnoses was similar between the two periods (data not shown).

Hypertension was by far the most prevalent modifiable vascular risk factor (43% of patients), in keeping with North African epidemiological data. Diabetes mellitus was reported in 28% of patients, dyslipidaemia in 21% and current tobacco smoking in 18%. Atrial fibrillation, either previously known or diagnosed during the in-hospital work-up, was present in 10% of patients. The vascular risk factor profile did not differ significantly between the two study periods (Table 1).

The opening of the trauma centre was associated with a major shift in first-line imaging practice. Before the opening, CT was the predominant initial modality, performed in 65% of activations, while MRI was performed in 35%. After the opening, this ratio reversed, with MRI used in 87% of activations and CT in only 13% (p < 0.001) (Table 2). This shift towards multiparametric MRI as the first-line modality enabled a more refined characterisation of the ischaemic lesion, the identification of the diffusion-FLAIR mismatch, the visualisation of intra-arterial occlusion on magnetic resonance angiography, and the immediate exclusion of intracranial haemorrhage.

Pre-hospital and in-hospital workflow times before and after the opening of the trauma centre are summarised in Table 2. Onset-to-door time decreased numerically from 200 to 189 minutes (p = 0.19) and onset-to-needle time from 266 to 264 minutes (p = 0.93), but neither change reached statistical significance. The median door-to-needle time remained virtually unchanged (70 versus 69 minutes), and substantially exceeded the 60-minute benchmark recommended by international guidelines.

The proportion of patients ultimately treated with intravenous thrombolysis increased significantly between the two periods, from 16% before the opening of the trauma centre to 28% after (p = 0.005)—a relative increase of 75% (Table 2). This improvement was driven primarily by a more frequent confirmation of acute ischaemic lesions on MRI in patients who would otherwise have remained ineligible because of diagnostic uncertainty.

Among patients who did not receive intravenous thrombolysis, the most common reasons were presentation beyond the therapeutic time window and the presence of major contraindications to thrombolytic therapy. These reasons remained the principal causes of treatment exclusion during both study periods. The higher thrombolysis rate observed after the opening of the trauma centre was mainly associated with the increased use of MRI-based patient selection, which improved diagnostic confidence and facilitated identification of eligible patients.

Table 2. Workflow times, imaging strategy and intravenous thrombolysis rate before and after the opening of the trauma centre.

CT = computed tomography; MRI = magnetic resonance imaging.

In this single-centre, before-and-after study including 403 consecutive patients admitted to the thrombolysis alert pathway of a North African tertiary academic centre, the opening of a dedicated in-hospital trauma centre was associated with two main structural changes: 1) a major shift towards MRI as the first-line imaging modality (from 35% to 87%, p < 0.001), and 2) a near-doubling of the intravenous thrombolysis rate (from 16% to 28%, p = 0.005). However, these gains were not accompanied by a significant reduction in the door-to-needle time, which remained stable around 70 minutes and persistently above the 60-minute international benchmark.

Our findings are consistent with the international literature showing that organisational changes in stroke pathways have a major impact on the proportion of patients ultimately treated with reperfusion therapies. Mouthon-Reignier et al. reported that the implementation of a direct-admission stroke pathway in Tours, France, was associated with a significant reduction in the onset-to-needle time and an increase in the intravenous thrombolysis rate, attributed to the suppression of unnecessary intermediate transfers and a streamlined in-hospital workflow [16]. Tsivgoulis and colleagues, in a pooled analysis of thrombolysis registries, demonstrated that earlier tissue-plasminogen-activator administration was associated with faster reperfusion and improved 90-day functional outcome [9].

Our 28% post-intervention thrombolysis rate compares favourably with the rates reported from other low- and middle-income countries (often below 5% - 10% of all incident strokes), although it remains below the 30% - 40% achieved in some high-volume European and North American comprehensive stroke centres [11][21]. The shift from CT- to MRI-based selection deserves particular attention. Although CT remains the most widely used and guideline-endorsed first-line modality because of its near-universal availability and shorter acquisition time, MRI with diffusion-weighted imaging offers higher sensitivity for the detection of acute ischaemic lesions, allows the identification of the diffusion-FLAIR mismatch in patients with unknown time of onset, and reliably excludes haemorrhagic transformation [22][23]. In the setting of a dedicated, co-located trauma centre with prioritised access to MRI, the additional acquisition time can be minimised and the diagnostic accuracy substantially improved.

The most important negative finding of our study is the absence of any reduction in the door-to-needle time. Despite the improvement in imaging strategy and the increase in the thrombolysis rate, the median time from hospital arrival to alteplase bolus remained at approximately 70 minutes, well above the 60-minute target endorsed by AHA/ASA and ESO and the 45-minute target proposed for high-performing centres [10]-[12]. Several plausible explanations may be advanced. First, the increased use of MRI, while improving diagnostic accuracy, may lengthen the imaging-to-needle interval, particularly when MRI is performed sequentially rather than in a parallel workflow. Second, the in-hospital alert pathway, although co-located, may still rely on sequential rather than parallel processes, with the neurological assessment, biological work-up, treatment decision and drug preparation performed one after the other. Third, ad-hoc shortages of dedicated stroke nurses or attending neurologists during nights and weekends may further prolong in-hospital delays.

Several actionable strategies may be proposed to reduce the door-to-needle time at our centre. The Helsinki and Melbourne models have demonstrated that median door-to-needle times below 20 minutes are achievable through a bundle of interventions including pre-hospital notification by emergency medical services, direct admission to the imaging suite (“CT scan first” or “MRI first”), parallel workflow with early laboratory testing on the ambulance, alteplase preparation in the imaging suite, and routine team debriefings [13][17][24]. The systematic recording of workflow times and individualised feedback to all team members have also been associated with sustained improvements in performance [18][25]. The introduction of a digital alert platform connecting the emergency medical services, the trauma centre and the on-call neurologist would further strengthen the chain.

Beyond the local relevance, our results illustrate the potential of structural reorganisation as a high-yield, cost-effective strategy to improve stroke care in low- and middle-income countries. The combination of a single co-located emergency-neurology-radiology platform with a streamlined in-hospital protocol allowed a near-doubling of the thrombolysis rate without any change in the underlying patient population or in the hospital’s catchment area. These findings support the recommendation of the World Stroke Organization and the African Stroke Organization to prioritise the development of organised stroke pathways and dedicated stroke units in resource-limited settings [19][20].

This study has several strengths. The cohort is one of the largest from a Moroccan academic centre to evaluate the impact of a structural reform on the local thrombolysis pathway. Data collection was based on a prospective in-hospital registry, with consecutive inclusion of all patients admitted to the alert pathway, limiting selection bias. The before-and-after design allowed a direct evaluation of the impact of a single, well-defined organisational intervention.

Our work also has limitations that should be acknowledged. First, this is a single-centre observational study, which limits the generalisability of the findings to other settings. Second, the before-and-after design cannot exclude residual confounding from secular trends, including changes in patient awareness or in pre-hospital triage. Third, we did not collect 90-day functional outcome data (modified Rankin Scale), which precludes a direct evaluation of the impact of the reform on patient-centred outcomes. Fourth, we did not assess the rate of mechanical thrombectomy, which is currently being implemented at our centre and will be the focus of a subsequent analysis. Future prospective studies, ideally including a multicentre comparison and a follow-up of clinical outcomes at three months, are needed to confirm and extend our findings. In addition, because of the before-and-after study design, we cannot exclude the influence of concurrent organisational changes, including increasing staff experience, improved MRI accessibility, optimisation of stroke workflows, or changes in referral patterns, which may also have contributed to the observed improvements.