We reviewed the hospital records of 371 children who were suspected of
having been physically abused. Our purpose was to determine specific
features that might be used to distinguish injuries resulting from abu
se from accidental injuries. Soft-tissue injuries were found in 341 (9
2 per cent) of these children. Ecchymoses were most common, accounting
for 555 (62 per cent) of 892 soft-tissue injuries, and very few had a
suspicious pattern. Only thirty-four (9 per cent) of the children had
a radiographically documented fracture, but radiography was performed
for only thirty-seven (10 per cent) of the patients and it rarely was
done unless a fracture was clinically obvious. The patterns of injury
were age-specific. The forty-four children who were nine months old o
r less had an average of only one soft-tissue injury; thirty soft-tiss
ue injuries involved the head or face, and seven (16 per cent) of the
children had a burn. These children were the most severely injured: tw
o (5 per cent) died and twenty (45 per cent) had a fracture. The sixty
-one children who were ten months to two years and eleven months old h
ad an average of two soft-tissue injuries; fifty-four soft-tissue inju
ries involved the head or face, and nine (15 per cent) of the children
had a burn. A fracture was found in eight (13 per cent) of these chil
dren. Two hundred and sixty-six children who were three years old or m
ore had an average of three soft-tissue injuries; 165 involved the hea
d or face, and only six children (2 per cent) in this group had a frac
ture. We believe that all children who might have been abused should b
e examined for soft-tissue injuries. In children who are nine months o
ld or less, any soft-tissue injury indicates possible abuse. In those
who are ten months to two years and eleven months old or less, multipl
e soft-tissue injuries or a burn (especially if the head and face are
involved) suggests the possibility of abuse. Whole-body skeletal radio
graphs should be made for children who are three years old. Soft-tissu
e injuries were frequently the only sign of abuse in children who were
three years old or more. These children should be examined with skele
tal radiographs when clinically indicated because of pain, swelling, o
r deformity.