We conducted a 9-month prospective cohort study at the Neurology, and Physical Medicine and Rehabilitation departments of the Fann University Hospital Center in Dakar. Our study population consisted of patients with language disorders following a stroke, seen in the outpatient clinic and inpatient services. From this study population, we included all clinical stable patients verbal expressive language disorders of vascular origin confirmed by medical imaging, within two weeks of the stroke event. These patients constituted the intervention group (GBI). These patients were followed in neurovascular care for 3 months plus or minus 10 days, without speech rehabilitation (GNBI).

Patients with a history of persistent motor aphasia of vascular origin and those with severe cognitive impairment were excluded.

Aphasia was assessed using the LAST score.

The Language Screening Test (LAST) is a 15-item test that screens for aphasia. The maximum score is 15, with each item scored as 1 (correct) or 0 (incorrect). The LAST has two subscores: expression and reception. The expression index measures naming, repetition, and automatic speech. The maximum score is 8 points. The receptive index measures picture recognition and verbal instructions. The maximum score is 7 points.

LAST b score

Aphasia was defined by a total score of less than 15.

The first step was to recruit all patients who met the inclusion criteria and had no exclusion criteria.

After consent, all included patients (GBI) were referred to the Physical and Rehabilitation Medicine department by the end of the fourth week, where they were randomly divided into two groups, according to the language rehabilitation method used: speech therapy (GBI1), and music therapy (GBI2). The third group, with no therapeutic intervention, was made up of patients who had been undergoing neurovascular consultation for 3 months without language rehabilitation (GBI).

For music therapy, based on an interview, patients or their relatives were asked to name 3 of their favorite songs, and to provide us with a recording of each song. A first rehearsal was carried out with the patient, to identify which song was easiest to repeat. For patients who didn’t have any favorite songs, we suggested singing the national anthem or a well-known Koranic or Christian song. The music therapy session were continued at home. Patients had to repeat the chosen song once a day, three times a week for a minimum of three months. Telephone calls (twice a month) were made to monitor and reinforce the music therapy practice. Speech therapy sessions were also conducted for up to three months.

Informed consent was obtained from all patients prior to inclusion. Data collected was anonymized to ensure patients’ confidentiality.

Data analysis

Data were collected on data sheets, entered in a Microsoft Excel spreadsheet and analyzed with SPSS 26.0. Qualitative variables were expressed as frequencies and numbers. Quantitative variables were expressed as mean with standard deviation. The chi-square test was used to compare qualitative variables. A bivariate analysis was performed to determine the different prognostic factors associated with language recovery in the music therapy group. The threshold of significance was considered to be a value of p < 0.05.

A total of 55 patients were included ( Figure 1 ).

The mean age of our patients was 61.2 +/− 14.5 years, with a median of 65 years, a minimum of 16 years and a maximum of 89 years. The most affected age range was 50 - 70 (56.4%). Patients over 70 accounted for 21.8%. Other age groups, 30-50 and under 30, accounted for 16.4% and 5.5% respectively. The sex ratio was 1.2 and 61.8% had never been to school.

Aphasia-related data

Aphasia etiologies were dominated by ischemic stroke (87.3%). The main types of aphasia found were Broca’s aphasia (41.8%), global aphasia (36.4%) and Pierre Marie’s major aphasia (12.7%). Conduction aphasia and transcortical aphasia were found in 3.6% each.

Forty-six (46) patients were evaluated at 1 month, including 20 in music therapy, 10 in speech therapy and 16 in GNBI. Mean scores in GBI1 were 8.7 at one month, compared with 11.4 in GBI2 and 6.3 in GNBI. Eighty percent (80%) of speech therapy patients were aphasic, compared with 90% of music therapy patients and 87.2% in the non-rehabilitation group, at 1 month. At 3 months, the study population consisted of 38 patients, including 13 in music therapy, 9 in speech therapy and 16 in GNBI. Mean scores in GBI1 were 10.55 at three months, compared with 13.41 in GBI2 and 6 in GNBI. In the speech therapy group, 66.7% had aphasia, compared with 87.5% in the non-rehabilitation group and 46.2% in the music therapy group (p = 0.023) ( Figure 2 ).

Based on the LAST-b score

A statistically significant improvement was obtained by speech therapy in the naming domain of the LAST-b score (p = 0.0001). Music therapy was associated with a significant improvement in naming (p = 0.001), repetition (p = 0.001) and automatic series (p = 0.014), musical environment (p = 0, 016), aphasia etiology (p = 0.006), automatic series (p = 0.009). No statistically significant difference was observed for listening comprehension ( Table 1 ).

In our study population, we found a mean age of onset of aphasia of 61y (16y - 89y) +/− 14 years. Adults aged 50-60 years were the most affected, with a median age of 65 years. These results are similar to those of Sène et al. in 2008 in Senegal, who found a mean age of 58.6 years (28 years-86 years) [4] . Runarvot et al. in 2024, in Normandy, found a median age of 63 years similar to our series [10] . Our study population was dominated by men (54.5%), with a sex ratio of 1.2; Ossou-Nguiet et al. also found a male predominance.

Broca’s aphasia accounted for 41.6% of the study population, global aphasia for 36.4%, Pierre Marie aphasia for 12.7%, transcortical motor aphasia and conduction aphasia for 3.6%, and anomia for 1.8%. These results are similar to those of Ossou-Nguiet et al., Gnonlonfoun et al., who found Broca’s aphasia to be predominant [5] [11] . Pedersen et al. found a predominance of global aphasia (32%), followed by Broca’s aphasia (12%), transcortical motor aphasia (2%), Wernicke’s aphasia (16%), conduction aphasia (5%) and anomia (25%) [12] . This difference may be explained by the small sample size of our study.

In our series, patients receiving music therapy had better scores than those receiving speech therapy or no rehabilitation. Indeed, Godecke et al. had found a significant improvement in patients who had received aphasia-focused daily therapy compared with patients who had received usual therapy [12] . Mattioli et al. in 2019, in Italy, found speech therapy to be effective in the early and chronic phase (> 6 months) after stroke, with greater effects observed if intensive treatment is provided [13] .

These results may be explained by the reduced number of speech therapy sessions performed by patients in our series, due to the long distance between their homes and hospital. Furthermore, music therapy showed the greatest statistically significant improvement over time, suggesting a potentially greater efficacy of music therapy compared with other methods. These results are similar to those of Lim et al., who showed that music therapy improves language function in patients with subacute non-fluent aphasia [3] . Furthermore, such findings may indicate that there are other pathways involved in speech and singing, and that speech therapy and music therapy may be effective in treating aphasia by activating the brain via different pathways [3] . Our results are also similar to those of Liu et al. in 2022, who found that music therapy can improve functional communication, repetition and naming in patients with post-stroke aphasia, but does not significantly improve comprehension [14] . Belin et al. reported normal activation of language areas when patients benefited from music therapy [15] . Kil-Byung Lim et al. in 2013 found significant improvements in spontaneous speech, comprehension and naming in the subacute and chronic groups treated with music therapy. Hypotheses explaining the effectiveness of music therapy for aphasia include the fact that words can be pronounced more slowly when singing than when speaking. In addition, singing may offer more opportunities to distinguish phonemes made up of words and phrases due to syllable length. Non-fluent aphasic patients can become more fluent by slowly singing connected segments [3] . In addition, musical environment (p = 0, 016), aphasia etiology (p = 0.008) and automatic series (p = 0.009) were significantly associated with aphasia treated by music therapy at 3 months. Other prognostic factors associated with aphasia such as laterality, age, level of education, initial severity, lesion site and size were also found [7] , [9] , [12] . The difference in factors can be explained by the variation in the speech therapy techniques used [16] .