Clinical determinants of the outcome of manual stretching in the treatment of congenital muscular torticollis in infants. A prospective study of eight hundred and twenty-one cases.

BACKGROUND The natural history of congenital muscular torticollis and the outcome of different treatment modalities have been poorly investigated, and the results of treatment have varied considerably. METHODS The main objective of this prospective study was to evaluate the outcomes of 821 consecutive patients with congenital muscular torticollis who were first seen when they were less than one year old, were treated with a standardized program of manual stretching, and were followed for a mean of 4.5 years. Before treatment, the patients were classified into one of three clinical groups: (1) palpable sternomastoid tumor, (2) muscular torticollis (thickening and tightness of the sternocleidomastoid muscle), and (3) postural torticollis (torticollis but no tightness or tumor). RESULTS Of the 821 patients, 452 (55%) had a sternomastoid tumor; 276 (34%), muscular torticollis; and ninety-three (11%), postural torticollis. Multivariate analysis of the outcomes showed that (1) the duration of treatment was significantly associated with the clinical group (p < 0.0001), a passive rotation deficit of the neck (p < 0.0001), involvement of the right side (p < 0.0001), difficulties with the birth (p < 0.009), and age at presentation (p < 0.0001); (2) the overall final assessment score was associated with the rotation deficit (p = 0.02), age at presentation (p = 0.014), and duration of treatment (p < 0.0001); and (3) subsequent surgical treatment was required by 8% (thirty-four) of the 452 patients in the sternomastoid tumor group compared with 3% (eight) of the 276 patients in the muscular torticollis group and 0% (none) of the ninety-three patients in the postural torticollis group. CONCLUSIONS This large prospective study demonstrated that controlled manual stretching is safe and effective in the treatment of congenital muscular torticollis when a patient is seen before the age of one year. The most important factors that predict the outcome of manual stretching are the clinical group, the initial deficit in rotation of the neck, and the age of the patient at presentation. Surgical treatment is indicated when a patient has undergone at least six months of controlled manual stretching and has residual head tilt, deficits of passive rotation and lateral bending of the neck of >15 degrees, a tight muscular band or tumor, and a poor outcome according to our special assessment chart.