This study was conducted in three phases:

The first phase involved creation of a Spanish adaptation of the Melodic Intonation therapy; the second phase consisted of a pilot non-randomized intervention study to analyse its feasibility for use with 4 Spanish patients with poststroke nonfluent aphasia; the third phase consisted of a pilot randomised, crossover, intervention pilot trial in a different set of patients with poststroke nonfluent aphasia (N=20)

Inclusion Criteria:

Patients diagnosed with nonfluent aphasia due to unilateral stroke in the left hemisphere without neuroimaging evidence of lesions in the right hemisphere who fulfilled the following criteria:

• The time elapsed since the stroke exceeded 6 months.

• Patient had received a standard program of conventional speech therapy after stroke.

• Persistent nonfluent aphasia with the following characteristics:

  • Severely restricted language, which might be limited to meaningless stereotypy; unlike that observed in verbal tasks, the patient might produce some real and relevant words when singing familiar songs.
  • Poor repetition, even for single words.
  • Moderately preserved language comprehension.
  • The nonstereotyped language was produced with a slurring of speech.
  • The total score for the repetition did not exceed the 70th percentile in the Boston Diagnostic Aphasia Examination. The score was obtained from the average score in three areas: repetition of words and repetition of phrases; listening comprehension exceeding the 15th percentile of Boston Diagnostic Aphasia Examination; and from the average score obtained in three areas: word comprehension, commands and complex ideational material.

• The patient was motivated, emotionally stable and had good attention.

• Signed informed consent was provided.

Exclusion Criteria:

• patients with a history of a previous stroke other than the index event or with any clinical condition (e.g., short life expectancy, coexisting disease) or other characteristics that precluded appropriate follow-up in the study
• patients participating in any therapeutic intervention clinical trials evaluating poststroke recovery
• use of psychotropic drugs that interfere with patient evaluation.

A neuropsychologist blinded to the patients' clinical characteristics and the group to whom the patient was allocated evaluated outcome variables.