Periosteal thickening as a manifestation of trauma in infancy

C/t,/
PERIOSTEAL

1982

THICKENING AS A MANIFESTATION TRAUMA IN INFANCY

EMILYTuFTS,M.D.,* Departments

014S-2:34/82/0~359-WsO3.0010 Copyright 0 1982 Pergamon Press Ltd.

of Pediatrics

EUGENEBLANK,

and Radiology,

OF

M.D., AND DIANADICKERSON

Oregon Health Sciences University,

Portland,

Oregon 97201

Abstract-This paper reports the findings in a study of the incidence of periosteal elevation in children and its possible relationship to child abuse. Two separate sets of radiographs of the skeleton of children were taken for a variety of diagnostic purposes. The suspected abuse set consisted of 59 radiographs taken for suspected child abuse. The mean age for all patients examined was 1.22 years whereas, the mean age for patients with cortical thickening was 0.5 years, suggesting that cortical thickening occurs in a relatively young population. Cortical thickening was assessed by reason for examination. A significant difference (p = .05) was detected with 618 (75%) of the patients with cortical thickening from the suspected child abuse groups and only 218 (25%) of the patients from the seizure-diagnostic category. The two infants who were noted to have periosteal elevation but were not suspected of abuse had experienced unusual circumstances, one was wearing an abduction splint and the other was a severely hypotonic premature. It is our impression that periosteal thickening is not a normal finding in infants and does not represent a consequence of normal infant care practices. In each case in which cortical thickening was detected, there was evidence to suggest that the child had experienced abnormal or rough handling. As a result of these findings, we believe that cortical thickening of the long bones detected on radiograph is an indication of child abuse. RCsum&Les auteurs ont etudie la frequence de I’epaississement p&iostC ou cortical chez les petits enfants et ont tent6 de faire une relation entre cette trouvaille radiologique et les s&ices physiques. Pour cela, ils ont examine des radiographies du squelette appartenant a deux groupes d’enfants: un groupe d’enfants temoins qui avaient CtC vus pour Cpilepsie et un groupe d’enfants chez qui on soupconnait fortement des s&ices. Ce deuxieme groupe comprenait 59 radiographies. L’lge moyen pour le collectif global des patirnts etait de I ,22 annee, mais I’Lge moyen pour les patients presentant un epaississement cortical n’etait que de 0.5 ar,i,ee. Cela suggere que I’epaississement cortical ne se produit que chez les enfants tres jeunes. On a trouve une difference significative entre les deux groupes. Six patients sur huit, c’est-a-dire le 75% dans le groupe des enfants chez qui on soupconnait des s&ices et deux patients sur huit, c’est-a-dire 25% seulement des enfants Cpileptiques presentaient un epaississement cortical. Les deux enfants du groupe temoin qui avaient un Cpaississement ptriostt, avaient vecu des circonstances particulieres: I’un portait une attelle d’abduction et I’autre Ctait un premature tres severement hypotonique. Les auteurs ont I’impression que I’epaississement perioste n’est pas une trouvaille normale chez les nourrissons et ne survient pas lorsqu’un nourrisson est I’objet de soins normaux. Chaque fois qu’on trouve un epaississement cortical, il y a d’autres trouvailles qui suggerent que I’enfant a subi des manipulations anormales ou brutales. En conclusion, les auteurs croient que I’epaississement cortical des OS longs tel qu’on le voit sur les radiographies est un signe suggestif de maltraitance d’enfants.

PERIOSTEAL ELEVATION and thickening of the long bones of infants raises questions of trauma and excessive rough handling. Caffey [l] initially raised this issue in the late 1940s. Since that time, additional investigators have commented on periosteal reactions, metaphyseal chip fractures, and fractures of the long bones as indicators of the battering of infants. Our interest in this issue was sharpened when a patient, believed by our staff to have been abused, was released from jurisdiction. The interpretation of periosteal thickening as a normal finding was a primary factor influencing the decision to terminate wardship. It is the purpose of this paper to report the findings of a study of the incidence of periosteal elevation in children and its possible relationship to child abuse. *Reprint requests: Emily Tufts, M.D., Associate Professor of Pediatrics, Department of Pediatrics, University. 3181 S.W. Sam Jackson Park Road, Portland, “I’ 97201. Presented

at Third

International

Congress

on Child Abuse and Neglect, 359

Amsterdam,

Oregon Health Sciences

The Netherlands,

April

1981.

360

Emily Tufts, Eugene Blank, and Diana Dickerson

MATERIALS

AND METHODS

Two separate sets of radiographs of the skeletons of children were evaluated. The first set (Sequential Set) consisted of 89 consecutive radiographs taken during the period August 1979 to August 1980 at the Oregon Health Sciences University (OHSU) for a variety of purposes, including determination of bone age, growth delay, tumor, trauma, seizures, endocrine evaluation, child abuse, and unstated reasons (Figure 1). The second set of radiographs (Suspected Abuse Set) consisted of 59 radiographs taken during the interval 1970 to 1980 at the OHSU for suspected child abuse. Films and records were reviewed for each of the 148 subjects. Information concerning the presence or absence of cortical thickening on radiologic examination, site examined, purpose of examination, and patient age was obtained. Although bone fractures were detected in a number of examinations, this study will focus exclusively on the finding of cortical thickening for these sub_jects.

STUDY RESULTS The overall incidence of cortical thickening for all patients examined was 5.4% (g/148 patients). Of those patients examined specifically for suspected child abuse, the incidence of cortical thickening was 10.2% (6/59 patients) compared with 2.2% (2/89 patients) for patients examined for all other reasons (p = .0344).

60

50 TOTAL SUBJECTS CORTICAL THICKENING 40

30

20

I

10

0

ti-

/, 7-9

10 - 12

13 - 15

AGE IN MONTHS Figure 1. Incidence of Thickening by Age Group

16 - 18

GT 18

361

Petiosteal thickening

Evaluation

by Age

The mean age for all patients examined was 1.22 years, while the mean ages for patients evaluated for suspected abuse versus all other reasons was 1.35 years and 1.17 years, respectively (p = .2932). Mean age for patients found to have cortical thickening present on radiographs was 0.5 years compared with a mean age of 1.26 years for patients free of cortical thickening (p = .0385). When age was compared in patients with the presence or absence of cortical thickening, no significant difference was detected for patients less than six months of age (p = .8578). It is of interest to note that 6/8 (75%) of the patients with cortical thickening are found in this age group. When the age margin was extended to one year, a borderline level of significance (p = .09) was detected in mean age for the presence or absence or cortical thickening. These data suggest that cortical thickening occurs in a relatively young population (Figure 1). Evaluation

by Reason for Examination

Cortical thickening suspected child abuse, docrine-metabolic, and thickening was detected

was assessed by reason for examination using the following categories: bone age-growth delay, limp-trauma-infection, seizures-diagnostic, enunknown reasons (Figure 2). A significant difference (p = .05) in cortical between these groups with 6/8 (75%) of the patients with cortical thickening

60

50 TOTAL SUBJECTS CORTICAL THICKENING

tnIl 40

30

20

10

0

I

TRAUMA

T!v!II DIAG

I- , TUMOR

UNSPECIF END/MET

Figure 2. Incidence of Thickening

by Exam Reason

GROWTH

ABUSE

Emily Tufts, Eugene Blank, and Diana Dickerson

362

Table 1. Reason for Examination Incidence of Cortical Thickening

Number of Cases

Reason

58 28 10 I 9 21 14

Child abuse Growth delay Tumor ;frauma Diagnostic Endocrine Unknown

6 0 0 0 2 0 0

Incidence w/o Cortical Thickening

(10.3%) (0 %) (0 %) (0 %) (22.2%) (0 %) (0 %)

52 28 10 7 7 21 14

( 89.7%) (100 %) (100 %) (100 %) ( 77.8%) (100 o/o) (100 o/c)

from the suspected abuse group and 2/8 (25%) of the patients from the seizures-diagnostic The patient distribution is described in Table I.

Evaluation

group.

by Site of Examination

When cortical thickening was evaluated by site of radiograph examination, no significant differences were observed (p = .2509). These data revealed that of the 8 patients with cortical c

50

TOTAL SUBJECTS CORTICAL THICKENING 40

30

20

10

0

c

I

ARM

!

I

UNKNOWN

I

OTHER

LEG

HALF SKELETON

EXAMINATION SITE Figure 3. Incidence of Thickening

by Exam Site

WHOLE LEGS + ARMS SKELETON

Periosteal thickening

363

thickening, 2 (25%) were noted in legs only examinations, 1 (12.5%) in an examination of both legs and arms, 4 (50%) in an examination of the entire skeleton, and 1 (12.5%) in an undesignated site (Figure 3).

PATIENT Sequential 1.

2.

3.

2. 3.

4.

5.

Cortical Thickening

A. H. was a premature infant with multiple congenital anomalies. At two months of age, radiographs done to investigate the possibility of congenital dislocation of the hip revealed cortical thickening as an incidental finding. D. C. had worn an abduction harness for six weeks and was referred for evaluation of his hips. Radiographs revealed cortical thickening of the medial aspects of both femurs. Four months later at age six months the thickening was still present and noted to be laminated. A. D., age seven months, was referred for evaluation of a “knot” on her arm, easy bruising and a human bite mark on her abdomen. Radiographs revealed healed fractures of the humerus and tibia as well as cortical thickening of the right femur. The authorities were notified.

Suspected 1.

Set Who Demonstrated

SUMMARIES

Abuse Set Who Demonstrated

Cortical Thickening

H. H., age 10 months, was referred with a fracture of the femur which allegedly occurred during a family fight. The parents were heavy drug users, and after a year of intensive community agency involvement and no improvement in parenting skills, termination of parental rights was planned for H. H. and his three siblings. M. K., age 7 months, was admitted with bilateral femoral fractures. This injury and a sibling’s black eye were never explained. The father was an alcoholic and beat his wife. C. C., age 2 months, was thin, had scattered bruises, scabies, and a club foot. Radiographs showed fractures of the sixth, seventh, eighth, ninth, and tenth ribs. Later, the father confessed to squeezing the baby to make him stop crying, was convicted and imprisoned. C. R. had a swollen thigh and bruises. The mother reported that the father hit the baby to make her stop crying. Radiographs revealed a fracture of the femur. The father also had a fractured femur at the time of the baby’s injury. G. L., age 18 months, sustained a Salter I slipped capital femoral epiphysis when he stepped in a hole. He had previously been hospitalized for failure to thrive. His mother was unstable and had trouble with alcohol and drugs.

SPECULATION

AND RELEVANCE

Because of ethical considerations involved in performing skeletal radiographs on normal children, we have opted to study the incidence of periosteal elevations through examination of sequential skeletal surveys conducted at the OHSU Department of Pediatric Radiology. Patient age and the reason for examination proved to be significant factors in explaining cortical thickening. If periosteal thickening was a normal finding in infants, one would expect to detect its presence routinely in radiographs of infants. Our study results do not support the supposition that periosteal thickening is a common finding among normal infants. Rather, our study suggests that the incidence of cortical thickening is significantly higher among those patients examined for suspected abuse than among those patients examined for all other combined reasons. It is interesting to note that the mean age of children demonstrating cortical thickening, 0.5 years, is consistent with the theory of loose attachment of periosteum to the bone in the very

364

Emily Tufts, Eugene Blank, and Diana Dickerson

young infant with resulting traumatic subperiosteal hemorrhage. Moreover, the presence of cortical thickening was more readily detected in “frog-leg” views than in the anterior-posterior views. Two infants who were noted to have periosteal elevation but were not suspected of abuse were found to have experienced circumstances which were outside the normal range of infant routine, and which may have contributed to undue trauma or torsion resulting in the thickening of the periosteum. Infant D. C. with cortical thickening had been wearing an abduction splint because of questionable congenital dislocation of the hip. The position assumed by an infant in the abduction splint is identical with that required for optimal diapering. One might suppose that the exasperated parent, further irritated by the messy infant, may too forcefully abduct the hips to diaper the infant. Infant A. H. with cortical thickening was a severely hypotonic premature. It is possible that his bones were not protected by normal muscle mass and tone. Dr. C. Henry Kempe [2] in 1962 called attention to the periosteal and epiphyseal lesions which he ascribed to traction and torsion of the extremities-convenient handles to grasp a child. Dr. Caffey [3] and Dr. Silverman [4] described traumatic involucra as not being fracture dependent, but rather due to indirect acceleration/deceleration traction forces which disrupt the vessels as they pass from periosteum to cortical bone. The resulting subperiosteal hemorrhage forms fibrous bone and is ultimately resorbed. Snedecor et al. [5] describes peristeal calcification and new bone formation in neonates following difficult obstetrical manipulations involving traction and twisting of the extremities. Woolley and Evans, [6] after an extensive study of infants treated for trauma, concluded that “a history of injury in any category of skeletal damage may be readily obtained, elicited only with difficulty, or not confirmed at all.” They felt that there was little evidence for unusual fragility of bone and that in those disorders leading to increased fragility of bone, there should be no confusion as to the responsible clinical condition.

CONCLUSIONS It is our impression that periosteal thickening is not a normal finding in infants and does not represent a consequence of normal infant care practices. In each case in which cortical thickening was detected, there was evidence to suggest that the child had experienced abnormal or rough handling of some kind. As a result of these findings, we believe that cortical thickening of the long bones detected on radiograph is an indicator for child abuse.

REFERENCES 1. CAFFEY,

J., Multiple fractures in the long bones of infants suffering from chronic subdural hematoma. Am. J. 56: 163 - I73 ( 1946). 2. KEMPE, C. H., SILVERMAN, F. N., STEELE, B. F., DROEGEMUELLER. W., and SILVER, H. K. The battered child syndrome. JAMA 181:17-24 (1962). 3. CAFFEY, J., On the theory and practice of shaking infants. Am. J. Dis. Child. 124: 161- 169 (1972). 4. SILVERMAN, F. N., Unrecognized trauma in infants: The battered child syndrome of Ambroise Tardieu. Radiology Roentgen

104:337-353 (1972). 5. SNEDOCOR, S. T., Obstet.61:385-387 6. WOOLLEY, P. V., JAMA 158539-543

KNAPP, R. E., WILSON,

H. B., Traumatic ossifying periostitis of the newborn.

Surg. Gynecol.

(1935).

EVAN, (1955).

W. A. Significance

of skeletal lesions in infants resembling

those of traumatic

origin.