PEDIATRICS Vol. 108 No. 4 October 2001, pp. 1009-1011

EXPERIENCE AND REASON:
Femur Fracture in Infants: A Possible Accidental Etiology

	ABSTRACT

Top Abstract Introduction Discussion Conclusion References

Femoral fractures in nonambulating infants are generally felt to be attributable to abuse in the absence of significant trauma or underlying organic pathology. The investigation of such fractures includes a report to appropriate social service and law enforcement agencies, and legal involvement. This paper describes 2 nonambulatory infants who sustained identical oblique distal femoral metaphyseal fractures extending through the growth plate after playing in an infant stationary activity center called an Exersaucer. It is possible that the twisting motion provided by the Exersaucer (Evenflo, Picqua, OH) might be consistent with the generation of forces necessary to cause these fractures.

Femoral fractures are reported in 12% to 29% of physically abused children.¹ Femoral fractures in nonambulating infants, particularly spiral fractures, are believed to be highly specific for inflicted injury. This association has been replicated over time. In 1983, a retrospective chart review of all children <4 years of age seen at a university hospital for a femur fracture found that 30% were attributable to abuse.² The most common reason for suspicion of abuse was lack of any history of injury. In that study, the radiographs did not distinguish between abused and nonabused children, as the most common site of fracture for both was the mid-diaphysis. The authors noted that abused children had a higher incidence of fractures of the proximal diaphysis and the distal metaphysis. Another study published the same year retrospectively analyzed femur fractures in children <5 years of age. Among children <1 year of age, 70% of fractures were found to be attributable to abuse.³ In one study from 1991, 60% of fractures of the femur occurring in infants 1 year or younger were the result of abuse, whereas only 20% of fractures in children between the ages of 2 to 3 years were determined to be the result of abuse.⁴

In infancy, a pattern injury like a spiral fracture is often inflicted. A spiral or oblique pattern implies application of a torsional force resulting from manual torsion with abuse, or twisting against a planted foot with a fall. As is the case with accidental injuries, most inflicted fractures of the femoral shaft involve the middle and distal third.⁵ Leventhal et al⁶ performed retrospective chart reviews of children <3 years of age who were examined for a fracture. For femoral fractures, the child's age was the best distinguishing feature of etiology: in children <12 months, 60% were classified as abuse compared with 0% in children older than 23 months. History characteristics were found to be vague in the abuse cases where the second most commonly occurring event was when the caretaker noted a change in the child's behavior, such as a nonmoving extremity: the majority (79%) in this group were classified as abuse. This type of history along with an abnormality in the child, such as swelling without report of any accident, occurred in 52% of the abuse cases. In contrast to nonambulatory children, oblique midshaft femur fractures have been reported to occur as a result of accidental injury in ambulatory children.⁶

The following 2 case reports document the possibility of distal femur fractures in infancy in a manner other than child abuse. Two infants, both <9 months old, with oblique femur fractures, were referred to the Broward County Child Protection Team for possible physical abuse. During the course of the investigation, it was learned that both infants had reportedly played in a product called an Exersaucer (Evenflo, Picqua, OH) before their parents suspecting an injury to their legs. The Exersaucer is a stationary activity center and is a successor to the infant walker. The Exersaucer's seat swivels as the child uses his feet to push off on the saucer-shaped base. The seat can swivel in both directions and the base rocks. The height of the seat can be adjusted to 3 different levels (Fig 1). The child can plant his feet on the base such that a twisting motion is generated with the seat and the base moving in opposite directions.

View larger version (42K): [in this window] [in a new window]   Fig. 1.   Exersaucer (Evenflo, Picqua, OH).

	CASE REPORTS

Case 1

C. S., a 7-month-old female, presented to her pediatrician's office on September 5, 1996, with an approximately 48-hour history of irritability and decreased range of motion of the right lower extremity. She normally had been able to stand, move, and kick with both lower extremities. The pediatrician noted swelling and discomfort in the right lower extremity, and the child was referred for an orthopedic consult. The child seemed comfortable when the right lower extremity was not moved. No bruises were noted on physical examination. Femur films revealed an acute oblique distal femur fracture involving the physeal plate. This was a Salter-Harris type II fracture with a large posterior metaphyseal fragment with an anterior extension into the physis (Fig 2). A bone survey was negative for any other fractures. On initial interview, the parents denied any history of trauma. They clearly noted a lack of movement of the right leg on the night of September 3. When asked about other possibilities, they recalled that the child had been using an Exersaucer. They brought it to the hospital and stated that they had not adjusted the height of the seat to lengthen it in conjunction with the child's growth. They demonstrated that the child could thus plant her feet on the base of the product and twist. The child underwent closed reduction and application of a spica cast. Femur films done through a fiberglass cast on 9/9 revealed an oblique fracture of the distal metaphysis of the femur with apparent extension into the physeal plate on the lateral view. There was minimal posterior displacement of the distal fragment of approximately 2 mm. Some early periosteal reaction is suggested (Fig 3). The mechanism of this injury was felt to possibly be consistent with the use of the Exersaucer as reported by the parents. No family concerns were identified during the Child Protection Team Family Assessment, which is a detailed psychosocial interview of the family members. Child Protective Services closed the case without any indicators of abuse.

View larger version (73K): [in this window] [in a new window]   Fig. 2.   Right femur, September 5, 1996.

View larger version (119K): [in this window] [in a new window]   Fig. 3.   Right femur, September 9, 1996.

Case 2

E. M., a 4.5-month-old female, was reportedly playing in her Exersaucer the day before her presentation to the emergency department of a community hospital with the chief complaint of not moving her leg for 1 day. The mother noted the child was not moving her left leg when she put her to bed, and the child cried when her left leg was touched. There was no history of injury given. No bruises were noted on the nursing admission assessment. Femur films done in the emergency department revealed a minimally displaced oblique fracture extending from the physis into the posterior metaphysis (Fig 4). A bone survey did not reveal any other fractures. The child underwent traction and placement of a long leg cast. Healing films (Fig 5) revealed a left oblique distal femur fracture with posteromedial callus formation. The mechanism of this injury was felt to possibly be consistent with the described use of the Exersaucer. Child Protective Services closed the case without evidence of abuse.

View larger version (66K): [in this window] [in a new window]   Fig. 4.   Left femur, March 1, 1997.

View larger version (90K): [in this window] [in a new window]   Fig. 5.   Left femur, March 11, 1997.

The 2 above incidents were reported to the Consumer Product Safety Commission. These reports are currently under investigation.

	DISCUSSION

Top Abstract Introduction Discussion Conclusion References

A pediatric radiologist and pediatric orthopedist independently reviewed both children's films. They concluded that these fractures are similar in geometry and reflect a similar causative mechanism. Both show an oblique fracture line beginning at the anterolateral aspect of the growth plate ending in the posteromedial aspect of the femoral metaphysis. A mechanism of injury to these children might be as follows: The child's feet are planted on the base of the Exersaucer. As the child rotates, the force generated is transmitted to the weakest part of the child's leg, the anterolateral aspect of the knee or growth plate, where the fracture begins. The twisting motion from the child's rotation causes the oblique fracture. As the child rotates outward in a partially or fully weight-bearing position with the foot of the involved extremity fixed in place and the knee fully extended or even slightly flexed, a varus force is created: the fracture begins proximal to the growth plate, travels laterally and proximally into the metaphysis behind the knee where the fracture ends, leaving the periosteum separated from the bone. The fracture ends medially because of the rotational component. As the bone heals, new bone forms under the periosteum forming a callus posterior and medial to the knee. Healing films of both children confirm that this callus is located at the posteromedial aspect of the knee.

This type of fracture might also be sustained by a child swiveling in the seat in 1 direction then stopping suddenly by planting a foot and hyperextending the knee. The momentum of the body against the planted, hyperextended, and fixed knee would generate torsional forces into the femur or tibia. Depending on the point of maximal stress concentration, a fracture may result. A likely point of stress concentration would be above the knee: the middle of a long lever arm fixed at the base of the toy by the foot and by the hip locked in the seat. In full or even hyperextension, and with external rotation of the body over the fixed lower extremity, the forces generated by this sudden stop would be anterolateral directed posteromedial. The physeal plate would be the weak link resulting in the initial fracture line starting anterolateral in the physis extending obliquely proximal into the metaphyseal bone resulting from the torsional load. A Salter-Harris II fracture would be the result just as these 2 cases demonstrate.

A child's bone is inherently weakest at the physis or growth plate. Injuries predictably involve this area, so this type of fracture is quite typical in growing bones, particularly when angulatory and rotational forces are applied.⁷ Furthermore, metaphyseal and cortical bone is more porous and not nearly so organized into lamellar structures in infants making failure in compression, particularly when combined with rotation, more likely.⁸ Lastly, because these fractures involve the growth plate, growth arrest is a possibility.

This oblique fracture is in contrast to the classic metaphyseal lesion (CML) described by Paul Kleinman.⁹ This lesion was described in 11 infants ages 1 to 5 months old who had died as a result of inflicted injuries that were examined both histologically and with detailed radiographic studies. In the same paper, the authors describe another radiographic pattern of the healing CML, a relatively discrete medial metaphyseal fragment corresponding to the corner fracture pattern.¹⁰ Dr Kleinman described the classic CML as always involving the posteromedial aspect of the femur, with anterior and lateral extension occurring in more diffuse lesions. The projection of the fracture fragment radiographically will result in either a "corner fracture" pattern or a "bucket handle" fracture. Only a small part of the metaphysis is visible by radiograph.

In both of our case reports, the fracture originates in the physis and extends obliquely into the metaphysis. The CML lesion, on the other hand, originates in the metaphysis, parallels the growth plate, and is entirely through the most distal portion of the metaphysis. It does not extend into the physis. The different mechanisms of injury explain the difference in the pattern of these 2 types of fractures. Classically, the CML lesion is thought to result from primarily a traction injury. In contrast, our 2 children sustained weight-bearing, torsional injuries. Each of these mechanisms creates the respective fracture anatomy.

Both parents report the use of the Exersaucer before the onset of their children's symptoms. The parents of E. M. spontaneously mentioned the toy when they brought the child to the emergency department. The parents of C. S. mentioned the toy when told by one of the authors (P. G.) that a possible twisting motion to the leg caused the fracture. The direction and type of both of these fractures fits the forces generated when a child is using the Exersaucer. One would not have expected the parents to directly link the toy to the injury as neither parent reported a distinct episode of sudden onset of crying while the child was using the toy.

A literature search was performed via the National Center for Biotechnology Information PubMed. No matches were found for the search words Exersaucer and femur fracture. We believe this is the first published possible association between this type of product and the described fractures.

	CONCLUSION

Top Abstract Introduction Discussion Conclusion References

These 2 identical femur fractures sustained by nonambulating infants serve as a reminder that a good history must always be obtained in cases of pediatric trauma. These fractures were initially believed to possibly be the result of physical abuse. Additional questioning revealed a possible accidental mechanism that could be consistent with the injuries sustained by these children. Thus, although the majority of femur fractures in this age group are the result of abuse, these 2 cases demonstrate that there are exceptions. As these are just 2 incidents, a conclusion as to the safety of this product cannot be drawn without additional investigation.

	ACKNOWLEDGMENTS

We thank Dr Robert W. Block for his critical review of the manuscript.

We greatly appreciate the thoughtful comments of Dr Betty Spivack and Dr Randall Alexander.

Penny Grant, MD
Department of Pediatrics
University of Oklahoma
Tulsa, OK 74129

Maria B. Mata, MD
Department of Radiology
Broward General Medical Center
Fort Lauderdale, FL 33316

Michael Tidwell, MD
Department of Orthopedics
Miami Children's Hospital
Miami, FL 33155

	FOOTNOTES

Received for publication May 12, 2000; accepted Feb 1, 2001.

Address correspondence to Penny Grant, MD, Department of Pediatrics, University of Oklahoma: Tulsa, Child Abuse Network, 2829 South Sheridan Rd, Tulsa, OK 74129. E-mail: penny-grant{at}ouhsc.edu

	ABBREVIATIONS

CML, classic metaphyseal lesion.

	REFERENCES

Top Abstract Introduction Discussion Conclusion References

1. Reece RM. Child Abuse: Medical Diagnosis and Management. Malvern, PA: Lea & Febiger; 1994:32
2. Beals RK, Tufts E Fractured femur in infancy: the role of child abuse. J Pediatr Orthop 1983; 5:583-586
3. Gross RH, Stranger M Causative factors responsible for femoral fracture. J Pediatr Orthop 1983; 3:341-343 [Medline]
4. Reece RM. Child Abuse: Medical Diagnosis and Management. Malvern, PA: Lea & Febiger; 1994:36
5. Kleinman P. The Diagnostic Imaging of Child Abuse. 2nd ed. St Louis, MO: Mosby; 1998:34
6. Leventhal JM, Thomas SA, Rosenfield NS, Markowitz RI Fractures in young children: distinguishing child abuse from unintentional injuries. Am J Dis Child 1993; 147:87-92 [Abstract]
7. Rang M. Children's Fractures. 2nd ed. Philadelphia, PA: JB Lippincott Company; 1983:7
8. Bright RW, Elmore SM Physical properties of epiphyseal plate cartilage. Surg Forum 1968; 19:463 [Medline]
9. Kleinman PK, Marks SC A regional approach to the classic metaphyseal lesion in abused infants: the distal femur. AJR Am J Roentgenol 1998; 170:44