Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print September 16, 2004, 10.1148/radiol.2332031640
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2332031640v1
233/2/477    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kleinman, P. L.
Right arrow Articles by Savageau, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kleinman, P. L.
Right arrow Articles by Savageau, J. A.
(Radiology 2004;233:477-485.)
© RSNA, 2004

Pediatric Imaging

Suspected Infant Abuse: Radiographic Skeletal Survey Practices in Pediatric Health Care Facilities1

Patricia L. Kleinman, RT(R), MPH, Paul K. Kleinman, MD and Judith A. Savageau, MPH

1 From the Department of Family Medicine & Community Health, University of Massachusetts Medical School, Worcester, Mass (P.L.K., J.A.S.); and Department of Radiology, Children’s Hospital, Harvard Medical School, 300 Longwood Ave, Boston, MA 02115 (P.K.K.). From the 2003 RSNA scientific assembly. Received October 8, 2003; revision requested December 11; revision received February 5, 2004; accepted March 2. Address correspondence to P.K.K. (e-mail: paul.kleinman@childrens.harvard.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To determine current national radiographic skeletal survey imaging practices, including migration to digital technologies, for evaluation of suspected infant abuse.

MATERIALS AND METHODS: Of 155 children’s health care facilities in the United States in which radiographic skeletal surveys are performed for suspected infant (<1 year old) abuse, 137 (88.4%) agreed to complete a questionnaire. Questions included facility type, imaging department volume, radiographic equipment, and details of skeletal survey imaging practices. Statistical analysis was performed with {chi}2, Fisher exact, Pearson correlation, Spearman rank correlation, and Student t tests.

RESULTS: One hundred seven completed questionnaires were returned. Forty-seven (43.9%) facilities used screen-film imaging; 60 (56.1%) used digital. Of screen-film users, 25 (53.2%) had already migrated or planned to migrate to digital within 1 year. Of screen-film users, 27 (60.0%) reported use of a high-detail imaging system, while 13 (21.7%) digital users employed a high-resolution technique (P < .001). Eighty-four (78.5%) facilities reported more than 10 images in their protocol, and 45 (42.0%) specified more than 15 images. Only one (0.9%) facility obtained fewer than three images. Upper extremities were imaged separately with at least two exposures in 81 (75.7%) facilities. Lower extremities were imaged separately with at least two exposures in 82 (76.6%) facilities. One hundred five (98.1%) facilities reported acquisition of lateral spinal views.

CONCLUSION: Within U.S. pediatric health care facilities, most skeletal surveys in cases of suspected infant abuse include separate frontal views of the appendicular skeleton and frontal and lateral views of the axial skeleton. Imaging protocols and other image quality determinants vary widely, and as U.S. pediatric health care facilities migrate from film-based to digital imaging technology, imaging practices directly applicable to the digital environment are being retained; however, less attention is being paid to technical elements specific to digital imaging that affect high-detail image quality.

© RSNA, 2004

Index terms: Bones, radiography • Infants, injuries • Infants, skeletal system


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Child maltreatment constitutes a major global public health problem. In the United States, there are approximately 3 million reports of child abuse and neglect annually, with most injuries occurring in young children (1). Fractures are common findings in infants and are often critical in the documentation of inflicted injury (2–4). Certain fractures, often subtle, carry a high specificity for the diagnosis of infant abuse, and therefore a high-quality radiographic skeletal examination is essential to visualize these important injuries (2). Studies have shown that with each episode of abuse, there is an increased likelihood of repeated assaults of increasing severity, occasionally resulting in death (5,6). Because fractures may be the only reliable evidence of serious physical trauma, failure to document injuries because of poor radiographic imaging practices may result in the return of a child to a potentially hostile environment.

For more than 50 years, fractures in cases of suspected infant abuse have been studied with a variety of radiographic imaging techniques, ranging from a "babygram"—an x-ray examination of the infant skeleton on one or two 14 x 17-inch films—to a formal skeletal survey, which is a series of radiographic images that encompasses the entire skeleton (7–12). The American College of Radiology (ACR) and the Section on Radiology of the American Academy of Pediatrics recommend that a high-quality radiographic skeletal survey (a) involve the use of a high-resolution imaging system, with technical factors designed to optimize image contrast and spatial resolution, and (b) include an appropriate number of well-collimated views of each anatomic region (11,12). A properly performed radiographic skeletal survey examination constitutes a powerful tool for early detection and characterization of fractures in cases of suspected infant abuse (13,14). To date, there have been no studies in which investigators have systematically evaluated the extent to which this rigorous approach has been implemented in cases of suspected infant abuse.

As pediatric health care imaging facilities migrate from film-based to digital technologies, it is unknown to what extent departments currently rely on digital imaging for their skeletal survey examinations. Although certain elements of the ACR standard for film-based skeletal surveys are applicable to digital radiography, fundamental differences in image acquisition, processing, and display of information require an imaging approach that is tailored to digital technology. Thus, the purpose of our study was to determine current national radiographic skeletal survey imaging practices, including migration to digital technologies, in the evaluation of suspected infant abuse.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Study Participants
This study was approved by the institutional review boards of the two medical facilities at which this research was conducted. Informed consent was obtained from all participants in the final study group. The subject population was selected from the National Association of Children’s Hospitals and Related Institutions (NACHRI) membership pool (15), which was accessible through the Internet as of August 30, 2002. NACHRI is the largest national organization representing children’s hospitals. It is a voluntary nonprofit membership organization of children’s hospitals, large pediatric units of medical centers, and related health systems.

There were 176 NACHRI institutional members, associate members, and supporters within the United States at the commencement of this study. The subject population of health care facilities was grouped by NACHRI into three categories: (a) institutional membership (self-governing, nonprofit, requiring less than a 30-day hospital stay); type IA, children’s hospitals (n = 45, 25.6%); type IB, children’s specialty hospitals (n = 35, 19.9%); type IC, children’s hospitals within a hospital (n = 57, 32.4%); (b) associate membership (n = 17, 9.6%); and (c) supporter (n = 22, 12.5%).

Health care facilities were included in this study if they performed skeletal survey examinations for suspected infant abuse. Of the 176 facilities within the subject population, 21 (11.9%) were excluded because they did not perform skeletal survey examinations (type IB, n = 16 [9.1%]; supporters, n = 5 [2.8%]). Of the 155 (88.1%) eligible participants, 18 (11.6%) were unable to be contacted (type IA, n = 1 [0.7%]; type IB, n = 7 [4.5%]; type IC, n = 5 [3.2%]; associate, n = 2 [1.3%]; supporter, n = 3 [1.9%]). Questionnaires were mailed to 137 consenting participants.

Data Collection
This cross-sectional study involved collection of data regarding current radiographic skeletal survey imaging practices performed in infants (<1 year old) between August 30, 2002, and May 16, 2003. Skeletal survey imaging data were gathered by using a 10-page data collection tool (questionnaire). A radiologist at each facility was contacted initially (P.L.K., P.K.K.) by means of e-mail, fax, or telephone if necessary and asked if he or she would be willing to participate in this study. Radiologists who agreed to participate were asked to supply the name, title (radiologist or other radiology personnel), and e-mail address of the person "most knowledgeable" with regard to skeletal survey imaging protocols in their facilities. A questionnaire was mailed to each contact person. Follow-up on surveys not returned in a timely manner was conducted (P.L.K., P.K.K.) by means of e-mail, telephone, mail, or fax. A determination was made (P.L.K., P.K.K.) if either the referring or contact radiologist at each facility was a current member of the Society for Pediatric Radiology on the basis of the membership directory accessible through the Internet (www.pedrad.org/) as of May 2003.

Data Collection Tool (Questionnaire)
The data collection tool was used to assess the following variables: (a) characteristics of pediatric health care facilities (type of facility, NACHRI classification, number of pediatric imaging studies and number of skeletal surveys performed annually); (b) services provided (hours skeletal survey examinations were performed and reviewed by radiologists and who interpreted the examinations); (c) skeletal survey imaging practices (primary imaging examination, type of imaging system used for skeletal surveys [screen-film or digital], characteristics of imaging system [general purpose vs high resolution], optimization of imaging technologies, radiographic equipment specifications [focal spot size, peak voltage range, and grid use], skeletal survey imaging protocols [number and type of images acquired], and migration from screen-film to digital imaging technologies). To clarify questions regarding the type of screen-film imaging system used, a follow-up e-mail was sent to 17 facilities.

Statistical Analysis
Descriptive data were tabulated and analyzed with Epi Info 6, version 6.04d, which is a word-processing database and statistical analysis program obtained from the Web site of the Center for Disease Control (www.cdc.gov/epiinfo/Epi6/EI6dnjp.htm). Data were also analyzed with Statistix for Windows (Analytical Software, Tallahassee, Fla), which is a data analysis program. Excel 2000, version 9.0 (Microsoft, Redmond, Wash), was used to chart data.

Categoric (nominal and ordinal) and continuous variables were analyzed by using uni- and bivariate statistics. Univariate analysis included the reporting of frequencies, percentages, sums, means, standard deviations, medians, modes, ranges, minimums, maximums, and 25th and 75th percentiles for all variables when appropriate. Bivariate analysis included {chi}2 testing to evaluate differences between (a) screen-film and digital imaging users with regard to type of facility, high-resolution imaging use, focal spot size selection, peak voltage range, and grid use; (b) high-resolution use and type of facility; and (c) services provided (hours skeletal surveys were performed and reviewed and who interpreted them) and type of facility.

The Fisher exact test was used when one or more of the cells in the {chi}2 table had a value of less than five. Pearson correlation (r) was used to measure the strength of the linear correlation between the number of pediatric imaging examinations and the number of skeletal surveys performed annually. Spearman rank correlation (rs) was used to assess nonparametric measurements of correlation between the number of skeletal surveys performed annually and the number of images acquired in the skeletal survey. The Student t test was used to estimate differences between screen-film and digital imaging facilities with regard to the mean number of pediatric imaging examinations performed annually, skeletal surveys performed annually, and images acquired in the skeletal survey examination. For all statistical tests performed, P < .05 was considered to indicate a significant difference.

For the purpose of analysis, when facilities reported use of a grid for imaging of any anatomic region, they were considered grid users. Facilities that reported use of a high-detail imaging system/technique for exposures of the extremities only were considered high-detail system/technique users. With regard to hours that skeletal surveys were performed and reviewed by a radiologist, services may have been provided by facilities in only a portion of the time period designated in the survey questions.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
General Data
One hundred seven completed questionnaires were returned (completion rate, 78.1% [107 of 137 consenting participants]; response rate, 69.0% [107 of 155 eligible participants]). Ninety-seven (90.6%) referring or contact radiologists for this study were members of the Society for Pediatric Radiology. Ninety-nine (92.5%) radiologists and eight (7.5%) other radiology personnel (technologist, supervisor, or director in radiology) reported that they completed the questionnaire.

Characteristics and skeletal survey practices of screen-film and digital imaging facilities surveyed are summarized in Table 1. There was a significant moderately positive linear correlation (r = 0.513, P < .001) between the number of pediatric imaging examinations and the number of skeletal surveys performed annually, as reported by the facilities (Fig 1). There was no significant linear correlation (rs = 0.065) between the number of skeletal surveys performed annually and the number of skeletal survey images acquired.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Facility Characteristics and Skeletal Survey Practices

 


View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Plot shows number of pediatric imaging examinations versus skeletal survey examinations performed annually at each facility.

 
The primary (first) imaging examination of choice in cases of suspected infant abuse was the radiographic skeletal survey, as reported by 102 (95.3%) facilities. Five (4.7%) facilities reported routine performance of both a radiographic skeletal survey and a nuclear medicine bone scan.

Screen-Film versus Digital Imaging Practices
Forty-seven (43.9%) facilities used screen-film imaging, and 60 (56.1%) facilities used digital (computed radiography, 56 facilities [52.3%]; direct digital radiography, four facilities [3.7%]). There was a borderline significantly increased frequency of use ({chi}2 = 3.63, P = .057) of digital imaging technology among children’s hospitals within a hospital when compared with freestanding children’s hospitals (Fig 2). There was a significantly decreased frequency of use ({chi}2 = 4.97, P = .026) of digital imaging technology among freestanding children’s hospitals when compared with all other facilities. There was no significant difference between the mean number of pediatric imaging examinations performed annually (t = 0.73, P = .469) or the mean number of skeletal surveys performed annually (t = 0.62, P = .535) in screen-film imaging facilities when compared with those in digital imaging facilities.



View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Bar graph shows comparison of screen-film and digital imaging facilities in which pediatric skeletal survey examinations are performed.

 
The total number of images in the skeletal survey protocol ranged from two to more than 20 (Fig 3). Eighty-four (78.5%) facilities reported acquisition of more than 10 images, and 45 (42.0%) facilities specified acquisition of more than 15 images. Twelve (11.2%) facilities reported acquisition of more than 20 images. Only one (0.9%) facility acquired fewer than three images, constituting a babygram. We found no significant difference in the number of images in the skeletal surveys performed at screen-film facilities when compared with that at digital facilities.



View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Bar graph shows the number of images that constitute a pediatric skeletal survey examination.

 
Of screen-film users, 27 (60.0%) used a high-detail imaging system, compared with 13 (21.7%) digital users who used a high-resolution imaging technique ({chi}2 = 16.02, P < .001) (Table 1). There was no significant difference ({chi}2 = 1.59, P = .208) between the use of high-resolution imaging as reported by freestanding children’s hospitals when compared with children’s hospitals within a hospital.

Screen-Film Imaging Facilities
Of the 27 screen-film facilities that reported use of a high-detail imaging system, nine (34.6%) used a single-screen/single-emulsion system, 15 (57.7%) used a double-screen/double-emulsion system, one (3.8%) used a double-screen/single-emulsion system, one (3.8%) used a single-screen/double-emulsion system, and one did not respond. Twenty (43.5%) facilities reported use of film manufactured by Agfa, Greenville, SC; four (8.7%) used Fuji, Stamford, Conn; 21 (45.6%) used Kodak, Rochester, NY; one (2.2%) reported other; and one did not respond. Sixteen (34.8%) facilities reported use of Agfa cassettes, five (10.9%) used Fuji, 18 (39.1%) used Kodak, seven (15.2%) reported other, and one did not respond.

Of facilities in which screen-film imaging is used for skeletal surveys, two (4.2%) reported having already migrated to digital radiography, nine (19.1%) are currently migrating, 14 (29.8%) plan to migrate within 1 year, 13 (27.7%) plan to migrate within 5 years, and nine (19.1%) reported no current plan to migrate.

Digital Imaging Facilities
Of digital users, 44 (73.3%) reported interpretation of skeletal survey images from a monitor, and 16 (26.7%) from printed images. Fifteen (25.4%) facilities used the manufacturers’ preset exposure factors after installation of their imaging equipment, compared with 44 (74.6%) facilities that reprogrammed the exposure factors to their specifications. One facility did not respond. Twenty-two (37.3%) digital facilities reported Agfa as the manufacturer of their computed radiography or direct digital radiography imaging equipment; 22 (37.3%) reported Fuji; 11 (18.6%) reported Kodak; two (3.4%) reported Siemens, Malvern, Pa; two (3.4%) reported other; and one did not respond.

Technical Factors
Table 2 shows data regarding focal spot size, peak voltage range, and grid use. There was no significant difference between the use of small and variable focal spot sizes when comparing screen-film and digital facilities. There was a significantly increased frequency of use ({chi}2 = 8.54, P = .003) in the 50-kVp range among screen-film facilities when compared with digital facilities. There was no significant difference between screen-film and digital facilities with regard to grid use.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Focal Spot Size, Peak Voltage Range, and Grid Use

 
Imaging Protocols
Upper extremities were imaged separately with at least two exposures in 81 (75.7%) of 107 facilities (Fig 4). Seventy-four (69.2%) facilities imaged each hand separately. Thirty-nine (36.4%) facilities imaged each upper extremity with three separate exposures. Lateral views of the upper extremities were acquired at 11 (10.3%) facilities. Facilities reported an average of 4.2 exposures ± 1.6 (standard deviation) (range, 2.0–6.0 exposures) of both upper extremities combined. Each lower extremity was imaged separately with at least two exposures in 82 (76.6%) facilities (Fig 5). The foot was imaged separately from the lower leg at 80 (74.8%) facilities. Forty-seven (43.9%) facilities imaged each lower extremity with three separate exposures. Lateral views of the lower extremities were reported at 18 (16.8%) facilities. An average of 4.2 exposures ± 1.9 (range, 1.0–6.0 exposures) were acquired of both lower extremities combined. The thorax and pelvis were imaged separately at 75 (70.1%) facilities.



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Bar graph shows comparison of upper-extremity imaging protocols. Of note, the number of images signifies the total number obtained in each upper extremity separately. One refers to the entire upper extremity (shoulder to hand) on one image. Two (A) refers to the humerus and forearm together on one image and the hand on a separate image. Two (B) refers to the humerus on one image and the forearm and hand together on one image. Three refers to the humerus, forearm, and hand each on a separate image.

 


View larger version (18K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Bar graph shows comparison of lower-extremity imaging protocols. Of note, the number of images signifies the total number obtained in both lower extremities combined. One refers to both lower extremities (hips to feet) together on one image. Two refers to each lower extremity (hip to foot) exposed separately. Four (A) refers to femur and lower leg together on one image and the foot on a separate image. Four (B) refers to the femur on one image and the lower leg and foot together on a separate image. Six refers to each lower extremity exposed separately with three exposures.

 
One hundred five (98.1%) facilities reported acquisition of lateral views of the spine. These data are shown in Figure 6. Fifty-two (48.6%) facilities reported acquisition of a separate view of the cervical spine, and 21 (19.6%) did not image the cervical spine in their lateral spine imaging protocol. One hundred three (96.3%) facilities reported imaging the skull with at least an anteroposterior and lateral view, and 15 (14.0%) included three or more skull images in their protocol.



View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Bar graph shows comparison of lateral spine imaging protocols. One (A) refers to cervical, thoracic, and lumbar spine imaged together. One (B) refers to thoracic and lumbar spine imaged together with no cervical spine image. Two (A) refers to the cervical spine imaged separately and the thoracic and lumbar spine imaged together. Two (B) refers to thoracic spine and lumbar spine each imaged separately with no cervical spine image. Three refers to cervical, thoracic, and lumbar spine each imaged separately.

 
Imaging Services
Table 3 summarizes the imaging services provided according to type of imaging facility. There was a significantly increased relative frequency ({chi}2 = 11.87, P < .001) of pediatric radiologists who "always" interpreted skeletal survey examinations in freestanding children’s hospitals when compared with children’s hospitals within a hospital. At least 84% of each type of facility reported acquisition of skeletal surveys during daytime and nighttime hours.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Services Provided according to Type of Imaging Facility (n = 107)

 
During weekday and weekend nights, approximately between the hours of 5 PM and 8 AM, staff or resident radiologists were more likely to review skeletal surveys acquired at children’s hospitals within a hospital when compared with freestanding children’s hospitals ({chi}2 = 5.33 and P = .021 vs {chi}2 = 3.87 and P = .049, respectively). In freestanding children’s hospitals, as well as children’s hospitals within a hospital, staff or resident radiologists reviewed skeletal surveys at a significantly higher frequency when performed weekdays between 8 AM and 5 PM than when performed at night ({chi}2 = 25.07 and P < .001 vs {chi}2 = 12.36 and P < .001, respectively). Additionally, skeletal surveys at freestanding children’s hospitals were less likely to be reviewed on weekend days when compared with weekdays between 8 AM and 5 PM (two-tailed Fisher exact test, P = .001). Community hospitals were not included in the previous analyses because of their small sample size (n = 4).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
Our findings indicate that the traditional film-based imaging technology used for skeletal surveys in cases of suspected infant abuse has been substantially replaced by digital imaging technology. Of 7893 skeletal survey examinations performed annually at the facilities in our study, 52.3% were performed with digital imaging technology, while 47.7% were performed with screen-film imaging technology. Furthermore, approximately 50% of current film-based imaging users will become digital facilities within 1 year.

Thus, on the basis of the results of the present study, it is estimated that within 1 year, 79% of pediatric health care imaging facilities in the United States will have migrated to digital technology. Despite this rapid migration to a new imaging modality, there are no data available specifically with regard to the suitability of digital radiography for skeletal surveys in cases of suspected infant abuse.

We found that in contrast to screen-film imaging users, who employed a high-detail imaging system in 60.0% of facilities, only 21.7% of digital facilities used a high-resolution imaging technique. To obtain superior skeletal detail in cases of suspected infant abuse, the ACR recommends the use of a high-resolution screen-film imaging system rather than a general pediatric imaging system.

Results of laboratory, radiologic-histopathologic, and clinical studies support the positive relationship between spatial resolution and diagnostic performance of the imaging system in the detection of osseous injuries in abused children (16–18). Outcome data are scarce in the complex medicolegal child abuse arena, but one study (4) suggests that the detection of inflicted skeletal injuries contributes to investigation and prosecution in fatal infant abuse.

The ACR standard specifies a limiting spatial resolution of at least 10 line pairs per millimeter, a performance that is achievable with high-detail screen-film imaging systems that operate at a system speed no higher than 200 (11). Although a single-screen/single-emulsion system of the class used for mammography provides the greatest spatial resolution, slow-speed double-screen/double-emulsion systems provide comparable resolution (16). All current digital radiographic imaging systems designed for general use operate at a spatial resolution well below that of high-detail screen-film imaging systems. Although both computed radiography and direct digital mammographic imaging systems provide high spatial resolution, these technologies are not readily available for general diagnostic imaging applications. Consequently, current digital imaging systems perform at levels of spatial resolution well below their high-detail screen-film imaging counterparts.

Computed radiography systems can be optimized for higher detail by acquiring images at an increased radiation exposure and by adjusting scan speed and processing algorithms (19). It is notable that 74.6% of digital facilities in our study reported reprogramming of exposure factors to their individual specifications for pediatric imaging examinations. Although the performance of digital radiography has been shown to be comparable to screen-film imaging in the demanding field of mammography (20), and digital imaging has clear advantages over film, such as a wide dynamic range and increased contrast capabilities, less attention has been given to the suitability of current general-use digital systems for their application in the rigorous area of imaging in cases of suspected infant abuse.

Other determinants of image quality assessed in the present study were focal spot size and peak voltage range. High skeletal detail is achieved with the use of a small focal spot, and this practice was reported by 69.8% of facilities. It is encouraging that only 7.5% of all facilities in our study used a large focal spot, which results in decreased image sharpness. Modern radiographic equipment may include a variable focal spot that is selected automatically when an exposure setting is chosen. In total, 22.6% of respondents reported the use of a variable focal spot, and it is therefore not possible to discern the effective focal spot size used at these facilities. Facilities that use a variable focal spot system should determine the effective focal spot size when performing skeletal survey examinations for optimization of image quality.

High-contrast images are essential for visualization of subtle skeletal injury and are achievable with the use of a low–peak voltage technique. A total of 80.8% of facilities reported use of less than 70 kVp for their skeletal surveys. Furthermore, only two facilities reported use of 80 kVp or more. It is interesting that the use of the 50-kVp range for skeletal surveys was used significantly less often in digital facilities than in screen-film facilities (P = .003). A possible explanation for this finding is that digital users believe that the higher-contrast capabilities of digital systems permit the acquisition of diagnostic images at a higher peak voltage setting. The possibility that digital skeletal survey imaging may be feasible at higher peak voltage settings than those used with screen-film imaging systems offers the attractive prospect of dose reduction and deserves further study.

The ACR standard recommends that an antiscatter grid not be used for skeletal survey examinations in infants (11). We found that 39.0% of facilities used a grid for skeletal surveys, and there was no significant difference in the frequency of grid use at screen-film facilities when compared with digital facilities. A substantial reduction in radiation exposure could be achieved easily by means of the elimination of grid use in these pediatric health care imaging departments.

In the present study, we explored in detail the imaging protocols used at participating facilities and compared them with the ACR standard (Table 4) (11). Each anatomic region should be radiographed separately to ensure uniform density on each image and to minimize loss of sharpness. We found that the use of the babygram was essentially nonexistent in our study population. Only one facility reported acquisition of fewer than three images in their skeletal survey imaging protocol. A total of 78.5% of facilities reported acquisition of more than 10 images, and 42.0% reported acquisition of more than 15 images, a practice which approximates that recommended by the ACR.


View this table:
[in this window]
[in a new window]

  		TABLE 4. ACR Skeletal Survey Table

 
It is notable that 11.2% of facilities reported acquisition of more than 20 images in their skeletal survey imaging protocol, a figure that exceeds ACR recommendations. Whether a facility was screen-film or digital had no apparent effect on the number of images acquired in the skeletal survey protocol. This finding suggests that in the transition from screen-film to digital imaging technology, facilities have carried over their imaging protocols established previously.

Recognition of extremity injuries remains a primary focus in the diagnosis of infant abuse (2,21). The classic metaphyseal lesion, a high-specificity indicator of inflicted injury in infants, is notoriously difficult to identify and requires high-quality images for accurate diagnosis (2,22,23). Accordingly, the ACR recommends that each upper and lower extremity be imaged with three separate exposures, totaling 12 images, to identify these lesions, in addition to other less-specific osseous injuries (11).

Most injuries of the appendicular skeleton involve the lower extremities. In infant fatalities, 73% of classic metaphyseal lesions involve the femur, tibia, or fibula (22). We found that the lower extremities were imaged with an average of 4.2 exposures, and 43.9% of facilities reported acquisition of three separate images of each lower extremity in accordance with the ACR standard (11). A total of 74.8% facilities reported imaging each foot separately, a practice recommended to identify subtle but important fractures of the metatarsals and phalanges (24).

Both upper extremities were imaged with an average of 4.2 exposures, and 36.4% of facilities followed the ACR standard by imaging each upper extremity with three separate exposures. A total of 69.2% of facilities reported imaging of each hand separately. The fact that a substantial majority of respondents reported acquisition of at least two images of each extremity represents an important improvement over practices advocated in the early child abuse literature (3,7–9). It is of concern, however, that 17.8% of respondents continue to expose both lower extremities, in their entirety, on one image, and 24.3% expose each upper extremity on one image.

Rib fractures are frequently found in child abuse and, in some series, constitute the most common skeletal injuries in abused infants (22,25,26). The ACR recommends a single exposure collimated to the thorax (11). In our study, we found that 29.9% of facilities chose to combine the thorax with the pelvis on a single exposure. This practice is likely to produce an image with unequal density, as well as loss of image sharpness due to a larger exposure field and geometric distortion.

One of the most encouraging findings in the present study is that nearly all facilities (98.1%) imaged the spine in the lateral projection. Although spinal injuries are relatively infrequent in child abuse, in contrast to most other skeletal fractures, they may result in severe long-term disability and, on occasion, death (27). Modest compression deformities of the vertebral bodies are usually asymptomatic, and even major spinal fractures and dislocations may not be suspected clinically (28).

Only 18.6% of facilities reported acquisition of separate lateral views of the lumbosacral spine and the thorax. Half of this group also obtained a separate lateral view of the cervical spine in accordance with ACR recommendations (11). A total of 19.6% of facilities did not image the cervical spine in their lateral spine imaging protocol; however, it is not possible to determine from our data if these facilities chose to include the cervical spine in the lateral view of the skull. It is interesting that the most commonly reported approach to imaging the lateral spine (39.3%) included a single view of the thoracolumbar region and a single view of the cervical region. This finding suggests that these facilities deem a single lateral view of the thoracolumbar spine adequate for assessment of injury in these regions. A total of 70.1% of facilities obtained separate anteroposterior views to encompass the thoracic and lumbar spine; the thoracic spine and upper lumbar spine were included on the anteroposterior thorax view, and the middle and lower lumbar spine were included on the anteroposterior view of the pelvis, in keeping with the ACR standard (11). When the pelvis and thorax are imaged together with a single exposure, subtle pelvic fractures are less likely to be identified (29,30).

In this study, we explored differences in services provided among freestanding children’s hospitals, children’s hospitals within a hospital, and community hospitals in which skeletal surveys are performed in cases of suspected infant abuse. Skeletal surveys were performed and reviewed by staff or resident radiologists at all facilities on weekdays; however, significant variations were observed at night and on weekends. Eighty-four percent of facilities reported acquisition of skeletal surveys during the day and at night, 7 days a week. At freestanding children’s hospitals, radiologist review of skeletal surveys decreased to less than 60% at night, a practice that was significantly less than at children’s hospitals within a hospital. An explanation for this difference may relate to a greater likelihood of a resident or staff radiologist being in-house when a children’s hospital is located within a large general academic medical center rather than in a freestanding facility. On the other hand, a skeletal survey was more likely to "always" be interpreted by a pediatric radiologist at a freestanding children’s hospital when compared with a children’s hospital within a hospital. All community hospitals reported performance and review of skeletal surveys during the day and night, 7 days a week; however, interpretation was at most "occasionally" performed by a pediatric radiologist.

We found that freestanding children’s hospitals have been substantially slower to migrate from screen-film to digital imaging technology than other facilities. A total of 45.3% of freestanding children’s hospitals have migrated to digital imaging technology for skeletal survey examinations, compared with 66.7% of all other facilities combined. This finding may relate to differences in financial resources between freestanding children’s hospitals and larger general hospitals in which children’s health care facilities are located. Other explanations may include image quality and radiation dose considerations.

This study is limited because all participants were drawn from NACHRI, an organization of children’s hospitals and related health care facilities (15). The requirements for membership raise the potential of introducing selection bias into our study methods. This study leaves unanswered questions regarding skeletal survey imaging practices at health care facilities that are not members of NACHRI—in particular, general hospitals with small pediatric populations and outpatient imaging facilities.

In conclusion, the results of this study suggest that within pediatric health care facilities in the United States, most skeletal survey examinations performed in cases of suspected infant abuse include separate frontal views of the appendicular skeleton and frontal and lateral views of the axial skeleton. Our findings show that within our study population of pediatric health care facilities, the babygram is virtually extinct; however, imaging protocols and other technical factors used to determine image quality vary widely. Pediatric health care facilities are rapidly migrating from film-based to digital imaging technology, and although skeletal survey imaging protocols are comparable in screen-film and digital imaging departments, the practice of high-detail imaging is significantly greater at screen-film facilities. Our results suggest that as pediatric health care facilities transition from screen-film technology, they are retaining imaging practices that are directly applicable to the digital environment; however, it appears that less attention is being paid to technical elements specific to digital imaging that affect image quality. Laboratory and clinical studies should be encouraged to compare the diagnostic performance of current digital imaging technologies with traditional high-detail screen-film imaging systems to establish minimum requirements for high-detail digital pediatric skeletal survey examinations in cases of suspected infant abuse.


   ACKNOWLEDGMENTS
 
The authors acknowledge Jacalyn Coghlin-Strom, MD, MPH, for her advice and constructive criticism in the design and implementation of this research.


   FOOTNOTES
 
Abbreviations: ACR = American College of Radiology, NACHRI = National Association of Children’s Hospitals and Related Institutions

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, P.L.K., P.K.K.; study concepts and design, P.L.K., P.K.K., J.A.S.; literature research, P.L.K., P.K.K.; data acquisition, P.L.K., P.K.K.; data analysis/interpretation, P.L.K., P.K.K., J.A.S.; statistical analysis, P.L.K., P.K.K., J.A.S.; manuscript preparation, P.L.K., P.K.K.; manuscript definition of intellectual content, editing, revision/review, and final version approval, P.L.K., P.K.K., J.A.S.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 

  1. U.S. Department of Health and Human Services. Administration on children, youth and families, child maltreatment 2001 Washington, DC: U.S. Government Printing Office, 2003.
  2. Kleinman PK. Skeletal trauma: general considerations. In: Kleinman PK, eds. Diagnostic imaging of child abuse. Baltimore, Md: Williams & Wilkins, 1987; 5-28.
  3. Merten DF, Radkowski MA, Leonidas JC. The abused child: a radiological reappraisal. Radiology 1983; 146:377-381.[Abstract]
  4. Kleinman PK, Blackbourne BD, Marks SC, Karellas A, Belanger PL. Radiologic contributions to the investigation and prosecution of cases of fatal infant abuse. N Engl J Med 1989; 320:507-511.[Abstract]
  5. Alexander R, Crabbe L, Sato Y, Smith W, Bennett T. Serial abuse in children who are shaken. Am J Dis Child 1990; 144:58-60.[Abstract]
  6. Jenny C, Hymel KP, Ritzen A, Reinert SE, Hay TC. Analysis of missed cases of abusive head trauma. JAMA 1999; 281:621-626.[Abstract/Free Full Text]
  7. Caffey J. Multiple fractures in the long bones of infants suffering from chronic subdural hematoma. Am J Roentgenol Radium Ther Nucl Med 1946; 56:163-173.
  8. Cameron JM, Rae L. The radiological diagnosis. In: Cameron JM, Rae L, eds. Atlas of the battered child syndrome. London, England: Churchill Livingstone, 1975; 20-50.
  9. Silverman FN. Radiology and special diagnostic procedures. In: Helfer RE, Kempe RS, eds. The battered child. 3rd ed. Chicago, Ill: University of Chicago Press, 1980; 215-240.
  10. American Academy of Pediatrics. Section on radiology: diagnostic imaging of child abuse. Pediatrics 1991; 87:262-264.[Abstract/Free Full Text]
  11. American College of Radiology: ACR standard for skeletal surveys in children. Res 22. Reston, Va: American College of Radiology, 1997; 51-54.
  12. American Academy of Pediatrics. Section on radiology: diagnostic imaging of child abuse. Pediatrics 2000; 105:1345-1348.[Abstract/Free Full Text]
  13. Belfer RA, Klein BL, Orr L. Use of the skeletal survey in the evaluation of child maltreatment. Am J Emerg Med 2001; 19:122-124.[CrossRef][Medline]
  14. Mandelstam SA, Cook D, Fitzgerald M, Ditchfield MR. Complementary use of radiological skeletal survey and bone scintigraphy in detection of bony injuries in suspected child abuse. Arch Dis Child 2003; 88:387-390.[Abstract/Free Full Text]
  15. The National Association of Children’s Hospitals and Related Institutions. NACHRI’s public site. Available at: www.childrenshospitals.net/ Accessed August 30, 2002.
  16. Kleinman PK, O’Connor B, Nimkin K, et al. Detection of rib fractures in an abused infant using digital radiography: a laboratory study. Pediatr Radiol 2002; 32:896-901.[CrossRef][Medline]
  17. Murphey MD, Bramble JM, Cook LT, Martin NL, Dwyer SJ, 3rd. Nondisplaced fractures: spatial resolution requirements for detection with digital skeletal imaging. Radiology 1990; 174:865-870.[Abstract]
  18. Youmans DC, Don S, Hildebolt C, Shackelford GD, Luker GD, McAlister WH. Skeletal surveys for child abuse: comparison of interpretation using digitized images and screen-film radiographs. AJR Am J Roentgenol 1998; 171:1415-1419.[Abstract/Free Full Text]
  19. James JJ, Davies AG, Cowen AR, O’Connor PJ. Developments in digital radiography: an equipment update. Eur Radiol 2001; 11:2616-2626.[CrossRef][Medline]
  20. Lewin JM, D’Orsi CJ, Hendrick RE, et al. Clinical comparison of full-field digital mammography and screen-film mammography for detection of breast cancer. AJR Am J Roentgenol 2002; 179:671-677.[Abstract/Free Full Text]
  21. King J, Diefendorf D, Apthorp J, Negrete VF, Carlson M. Analysis of 429 fractures in 189 battered children. J Pediatr Orthop 1988; 8:585-589.[Medline]
  22. Kleinman PK, Marks SC, Jr, Richmond RM, Blackbourne BD. Inflicted skeletal injury: a postmortem radiologic-histopathologic study in 31 infants. AJR Am J Roentgenol 1995; 165:647-650.[Abstract/Free Full Text]
  23. Caffey J. Some traumatic lesions in growing bones other than fractures and dislocations: clinical and radiological features. Br J Radiol 1957; 30:225-238.
  24. Nimkin K, Spevak MR, Kleinman PK. Fractures of the hands and feet in child abuse: imaging and pathologic features. Radiology 1997; 203:233-236.[Abstract]
  25. Worlock P, Stower M, Barbor P. Patterns of fractures in accidental and non-accidental injury in children: a comparative study. BMJ 1986; 293:100-102.
  26. Kleinman PK, Marks SC, Jr, Nimkin K, Rayder SM, Kessler SC. Rib fractures in 31 abused infants: postmortem radiologic-histopathologic study. Radiology 1996; 200:807-810.[Abstract]
  27. Kleinman PK. Spinal trauma. In: Kleinman PK, eds. Diagnostic imaging of child abuse. 2nd ed. St Louis, Mo: Mosby, 1998; 149-167.
  28. Levin TL, Berdon WE, Cassell I, Blitman NM. Thoracolumbar fracture with listhesis: an uncommon manifestation of child abuse. Pediatr Radiol 2003; 33:305-310.[Medline]
  29. Kleinman PK. The lower extremity. In: Kleinman PK, eds. Diagnostic imaging of child abuse. 2nd ed. St Louis, Mo: Mosby, 1998; 26-71.
  30. Starling SP, Heller RM, Jenny C. Pelvic fractures in infants as a sign of physical abuse. Child Abuse Negl 2002; 26:475-480.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 RadiologyHome page
P. L. Kleinman, D. Zurakowski, K. J. Strauss, R. H. Cleveland, J. M. Perez-Rosello, D. P. Nichols, K. H. Zou, and P. K. Kleinman
Detection of Simulated Inflicted Metaphyseal Fractures in a Fetal Pig Model: Image Optimization and Dose Reduction with Computed Radiography
Radiology, May 1, 2008; 247(2): 381 - 390.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2332031640v1
233/2/477    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kleinman, P. L.
Right arrow Articles by Savageau, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kleinman, P. L.
Right arrow Articles by Savageau, J. A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE