ORIGINAL ARTICLE
Screening for osteoporosis after trauma: A new approach using quantitative computed tomography of the skull Amber Taylor, MD, Kenneth Waxman, MD, Seema Izfar, MD, Jonathan Grotts, MA, and Samantha Yim, RN, Ventura, California
The diagnosis of osteoporosis is important in the care of elderly patients at risk of trauma. While pelvis computed tomography (CT) is accurate in the measurement of bone mineral density, axial skull CT has not previously been evaluated for this purpose. This study investigated whether data from axial skull CT scans can screen for osteoporosis. METHODS: Bone density measurements were derived from digital analysis of routine scans of the head and pelvis using quantitative CT. The study took place from October 2010 to November 2011 at a medium-sized community hospital. The first study phase included patients older than 18 years who had both a head and a pelvis CT scan within 30 days. The known diagnostic value for osteoporosis on pelvis CT scans was used to derive a diagnostic value for head CT. The second study phase included adult trauma patients who underwent noncontrast head CT during an initial trauma evaluation. A subgroup analysis was performed during Phase II on patients older than 65 years to identify the incidence of fracture as it is affected by age and bone mineral density. RESULTS: Our data demonstrated that head CT was able to identify osteoporosis with a sensitivity of 0.70, a specificity of 0.81, and an accuracy of 0.76 compared with pelvic CT. Of 261 trauma patients, 54% had bone disease based on axial skull CT criteria. Patients older than 65 years with a positive screen result for osteoporosis on head CT were twice as likely to have a fracture. CONCLUSION: Analysis of data from head CT scans has the potential to provide a useful screen for osteoporosis. Adding this analysis to CT scans performed for elderly trauma patients could result in improved diagnosis and treatment of osteoporosis. (J Trauma Acute Care Surg. 2014;77: 635Y639. Copyright * 2014 by Lippincott Williams & Wilkins) LEVEL OF EVIDENCE: Diagnostic study, level II. KEY WORDS: Osteoporosis; screening; head CT; quantitative CT; geriatric trauma. BACKGROUND:
O
steoporosis is a disease of increasing prevalence. In the year 2000, there were an estimated 9.0 million fractures associated with osteoporosis in the United States. Of these, 1.6 million were fractures of the hip, 1.7 million were of the forearm, and 1.4 million were vertebral. Total disabilityadjusted life years lost because of these osteoporotic fractures were estimated to be 5.8 million.1 Treatment using a pharmacologic agent along with vitamin D and calcium has been shown to be cost-effective in several patient populations: those with low bone density measurements who have fractures related to bone fragility, those with osteoporosis based on World Health Organization criteria, and those with osteopenia and additional risk factors.2 Early identification and treatment of patients who meet these criteria can prevent fractures and save health care dollars. However, large numbers of patients who might benefit from bone density analysis do not receive this screening. Low bone density measurement results in higher incidence of fractures following trauma.3 Elderly patients with a Submitted: April 22, 2014, Revised: June 11, 2014, Accepted: June 23, 2014. From the Santa Barbara Cottage Hospital (A.T., J.G., S.Y.), Santa Barbara; Department of Surgery (K.W.), Ventura County Health System, Ventura, California; and San Antonio Colon and Rectal Clinic (S.I.), San Antonio, Texas. Address for reprints: Kenneth Waxman, MD, Department of Surgery, Ventura County Health System, 3291 Loma Vista Rd Ste 401, Ventura, CA 93003; email: [email protected]. DOI: 10.1097/TA.0000000000000411
history of falls are at higher risk of future falls. Therefore, identification and treatment of osteoporosis in fall patients could result in a decrease in future fractures. At present, screening for osteoporosis is seldom performed during the evaluation of trauma patients who present to emergency departments. Many patients receive computed tomography (CT) of the axial skull during trauma evaluations; additional information is available from this imaging, which can be used to screen for osteoporosis. Bone density measurements from CT scans can be obtained by performing additional analysis of the digital information. Such analysis has been termed quantitative computed tomography (QCT). Analysis of bone density on CT scans performed for injury assessment would have the advantage of providing a diagnosis of osteoporosis without additional testing. QCT from the proximal femur has previously been shown to provide an accurate measure of bone density.4 Although dual-energy x-ray absorptiometry (DXA) is the preferred screening test for osteoporosis as defined by the World Health Organization, QCT might serve as a valuable alternative for trauma patients who receive this test for injury evaluation and for whom DXA has not been previously performed. Studies have shown that accurate bone density measurements can be computed from pelvic and lumbar QCT.5 However, patients presenting with falls most commonly experience head injury and often receive only a CT scan of the axial skull. No study to date has evaluated the accuracy of bone density measurements from the axial skull region using QCT.
J Trauma Acute Care Surg Volume 77, Number 4
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
635
J Trauma Acute Care Surg Volume 77, Number 4
Taylor et al.
If axial skull QCT provided information relative to osteoporosis, this would be of great value. In 2007, 40.7% of patients who presented to emergency departments with head injury underwent a CT examination.6 Many of these individuals are elderly patients who have fallen and are therefore at high risk for complications from osteoporosis. The aim of the present study was to determine whether axial skull CT scans performed during trauma evaluations can provide useful information about bone density. This study was performed by adding evaluation of axial skull bone density measurement onto routine trauma CT scan interpretations. By correlating axial skull QCT measurements to the proven methodology of determining bone density from proximal femur QCT measurements, we wished to determine if axial skull CT scans might also serve to screen for osteoporosis. In the second phase of this study, we tested the validity of this approach by determining whether an at-risk population of trauma patients whose axial skull CT scans measured low bone density were at higher risk for fractures.
PATIENTS AND METHODS Bone density measurements were prospectively collected on patients presenting to our institution, a community hospital, who received axial skull CT scans as part of their injury evaluations. The study period was from October 2010 to November 2011. QCT was performed using the same radiologic technique as would have been performed whether this study was underway, with no additional radiation exposure. A General Electric Lightspeed VCT 64-Slice CT Scanner (General Electric Healthcare, Waukesha, WI) was used. A commercial calibration phantom was added to the CT table to provide calibration for the bone density measurements. Commercially available software was used to analyze the digital data and determine bone density (QCT PRO BIT, Mindways Software, Austin, TX). CT scans and bone density measurements were processed by CT technicians who were trained on the QCT software before initiation of the study. This study was initiated and performed by the authors of this study. No manufacturer provided financial support, and both study design and data analysis were performed solely by the authors. The local institutional review board granted a waiver of consent because the study was determined to be of less than minimal risk to the patients. Bone density measurements obtained from this study were considered experimental and were not released to patients, and no interventions were offered. The
only difference between performing conventional CT scans and those generating bone density measurements was the placement of a calibration phantom underneath patients during the scan, which did not alter radiation exposure. The study included two phases: Phase I was designed to correlate bone density measurements between axial skull CT scans and pelvis CT scans. Inclusion criteria were (1) being 18 years old and (2) having undergone both noncontrast axial skull and pelvis CT scans within 30 days. There were no exclusion criteria. Bone density measurements of the pelvis were taken from the proximal femur. Bone density measurements of the axial skull were taken from the temporal bone. Fifty-four patients met inclusion criteria and were included in this phase of the study. The reference normal for bone density measurements was used to calculate a T-score for each patient’s total hip bone area density measurement (superposition of the femoral neck, trochanter, and intertrochanteric regions). T-score was calculated by subtracting the patient’s bone mineral density (BMD) measurement by the mean of a healthy, young population and then dividing by the SD of the healthy, young population’s BMD (Fig. 1). The BMD values for the healthy normal population were provided by the vendor of the QCT. T-scores from total hip measurements using QCT and DXA have shown high correlation (R = 0.93), although the QCT BMD T-scores were slightly lower than that of the DXA BMD.7 The World Health Organization defines osteoporosis as at least 2.5 SDs below the mean peak bone mass of young healthy individuals as measured by DXA.8 An optimal threshold for axial skull BMD measurements to predict osteoporosis was developed by identifying patients as having osteoporosis if their total hip BMD T-score was less than j2.5. The objectives for Phase I of the study were to define the relationship of BMD between the axial skull and hip and to thereby determine the potential reliability of axial skull measurements to assess for osteoporosis. Phase II was a prospective observational study to assess the feasibility of evaluating BMD in the emergent setting. Patients were included in the study if they were evaluated by the trauma team and received a noncontrast CT scan of the axial skull on initial evaluation. There were no exclusion criteria. Radiology technicians notified the research staff when the bone density phantom was used. The research staff then derived BMD from the QCT software. The incidence of fractures was evaluated through a chart review. Patients who screened positive for osteoporosis were not offered any intervention and were not followed up after the hospital discharge.
Figure 1. Equation for calculating T-score. 636
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 77, Number 4
Taylor et al.
A preplanned subgroup analysis was performed on patients older than 65 years to determine whether fracture rate was affected by age and BMD. The correlations referenced throughout the article were calculated based on Spearman’s correlation coefficient. Youden’s J statistics was used to find the optimal threshold for axial skull BMD using total hip BMD T-score less than j2.5 as the cutoff for osteoporosis. Sensitivity and specificity measurements were used to summarize the efficacy of axial skull BMD to assess for osteoporosis.9 A Student’s t test, Wilcoxon rank-sum test, or Fisher’s exact test were used where appropriate, and a p value of 0.05 was used to indicate statistical significance. Data analysis was performed in the R Statistical Computing Environment (R Core Team 2012).
RESULTS Phase I of the study included 54 patients with both axial skull and pelvis CT scans. There were 30 males (55.5%). The mean (SD) age was 75.0 (14.6) years, with a range from 23 years to 98 years. The majority of study subjects, 43 of 54 patients, were older than 65 years. The mean (SD) hip BMD was 0.68 (0.23) g/cm3, the mean (SD) hip T-score was j2.0 (2.0), and the mean (SD) axial skull BMD was 0.17 (0.08) g/cm3. The correlation was 0.52 between the hip BMD T-score and the axial skull BMD (Fig. 2). There were 27 patients (50%) who had a hip BMD T-score below j2.5, which meets the criterion for a diagnosis of osteoporosis. The corresponding threshold for axial skull BMD to assess for osteoporosis was 0.16 g/cm3, which resulted in a sensitivity of 0.70, a specificity of 0.81, and an overall accuracy of predicting osteoporosis of 0.76. Patients with low BMD on axial skull CT were more likely to be older than 65 years (p = 0.006). The second phase included 261 eligible patients. For Phase II, we used the bone density value of 0.16 g/cm3 as calculated from axial skull CT to identify patients with osteoporosis. The mean (SD) age of these patients was 59.9 (24.7) years, with a mean (SD) axial skull BMD of 0.20 (0.12) g/cm3 for all patients. There were 113 patients (43%) who screened positive for osteoporosis using the criteria from Phase I of the study (Table 1).
TABLE 1. Characteristics of Patients Screened for Osteoporosis by Axial Skull CT No Osteoporosis Osteoporosis (n = 148) (n = 113) Age, mean (SD), y Males, n (%) ISS, median (interquartile range) Axial skull BMD, mean (SD), g/cm3
56.3 (24.9) 86 (58.1) 5 (6) 0.28 (0.09)
64.7 (23.7) 66 (58.4) 5 (8) 0.09 (0.05)
p 0.006 0.961 0.638 G0.001
The correlation between BMD and age separated by sex is presented in Figure 2. There was a negative correlation between age and axial skull BMD in the female population (r = j0.35), but this relationship was not observed in the male population (r = j0.06). The relationship between patients older than 65 years who screened positive for osteoporosis bone density and presence of fractures is shown in Table 2. Of patients older than 65 years, there were twice as many fractures in those with a positive screen result for osteoporosis on axial skull CT scan compared with those who screened negative for osteoporosis (p = 0.031) (Fig. 3).
DISCUSSION Osteoporosis is a disease characterized by low BMD, microarchitectural disruption, and increased skeletal fragility. Osteoporosis predisposes individuals to fractures, often after minor trauma. Such fractures have significant morbidity and mortality especially in the geriatric trauma population.10 DXA is the World Health Organization standard for diagnosis of osteoporosis. However, many at-risk patients do not receive DXA screening because of limited availability in many communities. In addition, DXA screening is frequently not reimbursable.10 DXA is not routinely available in the emergency setting, but injured patients who are at risk for osteoporosis often receive CT scans following trauma. It has been shown that QCT of the spine or hip is an acceptable method for measuring bone density and, in some studies, is a better
TABLE 2. Characteristics of Elderly Patients (Age 9 65 Years) Based on the Presence of Osteoporosis* No Osteoporosis Osteoporosis (n = 64) (n = 63) Age, mean (SD) Males, n (%) ISS, median (interquartile range) Abbreviated Injury Scale (AIS) (Ortho), median (interquartile range) Axial skull BMD, mean (SD), g/cm3 Patients with one fracture, n (%) Patients with two fractures, n (%) Patients with at least one fracture, n (%)
Figure 2. Relationship between axial skull and pelvis BMDs.
p
81.2 (8.3) 27 (42.2) 5 (4.8) 0 (0.8)
83 (8.7) 27 (42.9) 7 (8) 0 (2.8)
0.243 0.939 0.163 0.219
0.28 (0.08)
0.08 (0.06)
G0.001
5 (9) 0 10 (15.6)
9 (14.3) 6 (9.5) 20 (31.7)
0.061 0.003 0.031
*Osteoporosis identified from axial skull BMD.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
637
J Trauma Acute Care Surg Volume 77, Number 4
Taylor et al.
Figure 3. Correlation of axial skull BMD by age for men and women.
predictor of fracture risk compared with DXA.6 Our study suggests that data obtained from the commonly performed axial skull CT scan may also be used to screen for osteoporosis. In our sample, 64.3% of the patients undergoing trauma assessment with axial skull CT were found to screen positive for osteoporosis, with the incidence increasing with age. The mean age of the patients in our study with abnormal bone density measured on axial skull CT was 64.5 years. The incidence of osteoporosis increases with age, and women are more likely than men to sustain fractures.11 In our study population, increased age was negatively correlated with axial skull BMD in women. Fractures are more likely to occur in trauma patients with osteoporosis.12 In our study, a positive screen result for osteoporosis on axial skull CT scan in patients older than 65 years was associated with a higher incidence of fractures. While our study was not designed to determine causality of fractures, the higher incidence of fractures in patients who screen positive for osteoporosis on axial skull CT is consistent with what would be expected from an accurate screening tool for osteoporosis. QCT does not add any radiation exposure to patients. The only costs to add QCT to standard axial skull CT scans may include the purchase of additional software (if an institution does not already possess this software) to analyze the digital data and technician training to make the measurement. The potential benefits of screening at-risk patients for osteoporosis following trauma far outweigh the costs of QCT. There were several limitations of this study. Although our study reached statistical significance, the study size was relatively small. There may have been some selection bias within Phase I of the study because we only included patients who received both axial skull and pelvis CT examinations. These patients may have had characteristics different from those patients who were not getting both examinations. This study was observational; no interventions were performed based on our findings. We did not follow patients to see if osteoporosis was diagnosed. Our sole aim was to determine whether the axial skull CT may be a useful tool to generate data relevant to bone density. Our data show that additional quantitative information may be obtained from routine axial skull CT scans performed during trauma evaluations. We have shown that these data can potentially be used to screen for osteoporosis. A study 638
correlating bone density measurements obtained from axial skull CT with DXA would provide a useful follow-up to the current study. If the usefulness of screening for osteoporosis with axial skull QCT is validated by further studies, we believe this would be a valuable addition to emergency trauma evaluations. Information regarding bone density could be obtained from axial skull QCT and communicated to primary care physicians for treatment, at no additional risk and minimal cost. Routine assessment of bone density in patients presenting after trauma could be integrated into a comprehensive and effective community care model to identify and treat an at-risk population of patients with osteoporosis and potentially decrease their risk of future injuries. AUTHORSHIP S.I. contributed to the study design and data collection. K.W. contributed to the study design, data analysis, and manuscript preparation. A.T. contributed to the data analysis and manuscript preparation. J.G. contributed to the statistical analysis and manuscript preparation. S.Y. contributed to the data collection. ACKNOWLEDGMENTS We thank Ashlie Thompson, CT technician, who generated bone density measurements in real time using the QCT software. We also thank Jennifer Kosek who performed a chart review to identify fractures in Phase II patients. DISCLOSURE This study was supported by the Santa Barbara Cottage Hospital Research Committee.
REFERENCES 1. Johnell O, Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int. 2006;17(12): 1726Y1733. 2. Dawson-Hughes B, Tosteson AN, Melton LJ, Baim S, Favus MJ, Khosla S, Lindsay RL, N. O. F. G. Committee. Implications of absolute fracture risk assessment for osteoporosis practice guidelines in the USA. Osteoporos Int. 2008;19(4):449Y458. 3. Anderson F, Cooper C. The influence of osteoporosis in trauma. Trauma. 1999;1(3):181Y192. 4. Damilakis J, Maris TG, Karantanas AH. An update on the assessment of osteoporosis using radiologic techniques. Eur Radiol. 2007;17(6):1591Y1602. 5. Guglielmi G, Lang TF. Quantitative computed tomography. Semin Musculoskelet Radiol. 2002;6(3):219Y227.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 77, Number 4
6. Larson DB, Johnson LW, Schnell BM, Salisbury SR, Forman HP. National trends in CT use in the emergency department: 1995Y2007. Radiology. 2011;258(1):164Y173. 7. Khoo BC, Brown K, Cann C, Zhu K, Henzell S, Low V, Gustafsson S, Price RI, Prince RL. Comparison of QCT-derived and DXA-derived areal bone mineral density and T scores. Osteoporos Int. 2009;20(9):1539Y1545. 8. Kanis JA. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. WHO Study Group. Osteoporos Int. 1994;4(6):368Y381.
Taylor et al.
9. Youden WJ. Index for rating diagnostic tests. Cancer. 1950;3(1):32Y35. 10. WHO scientific group on the assessment of osteoporosis at primary health care level. Brussels, Belgium: World Health Organization; 2004. Available at: www.who.int/chp/topics/Osteoporosis.pdf. Accessed March 12, 2014. 11. Gerdhem P. Osteoporosis and fragility fractures. Best Pract Res Clin Rheumatol. 2013;27(6):743Y755. 12. Mackey D, Lui L, Cawthon P, Bauer D, Nevitt M, Cauley J, Hillier T, Lewis C, Barrett-Connor E, Cummings S. High-trauma fractures and low bone mineral density in older women and men. JAMA. 2007;298(20):2381Y2388.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
639
Screening for osteoporosis after trauma: A new approach using quantitative computed tomography of the skull Amber Taylor, MD, Kenneth Waxman, MD, Seema Izfar, MD, Jonathan Grotts, MA, and Samantha Yim, RN, Ventura, California
The diagnosis of osteoporosis is important in the care of elderly patients at risk of trauma. While pelvis computed tomography (CT) is accurate in the measurement of bone mineral density, axial skull CT has not previously been evaluated for this purpose. This study investigated whether data from axial skull CT scans can screen for osteoporosis. METHODS: Bone density measurements were derived from digital analysis of routine scans of the head and pelvis using quantitative CT. The study took place from October 2010 to November 2011 at a medium-sized community hospital. The first study phase included patients older than 18 years who had both a head and a pelvis CT scan within 30 days. The known diagnostic value for osteoporosis on pelvis CT scans was used to derive a diagnostic value for head CT. The second study phase included adult trauma patients who underwent noncontrast head CT during an initial trauma evaluation. A subgroup analysis was performed during Phase II on patients older than 65 years to identify the incidence of fracture as it is affected by age and bone mineral density. RESULTS: Our data demonstrated that head CT was able to identify osteoporosis with a sensitivity of 0.70, a specificity of 0.81, and an accuracy of 0.76 compared with pelvic CT. Of 261 trauma patients, 54% had bone disease based on axial skull CT criteria. Patients older than 65 years with a positive screen result for osteoporosis on head CT were twice as likely to have a fracture. CONCLUSION: Analysis of data from head CT scans has the potential to provide a useful screen for osteoporosis. Adding this analysis to CT scans performed for elderly trauma patients could result in improved diagnosis and treatment of osteoporosis. (J Trauma Acute Care Surg. 2014;77: 635Y639. Copyright * 2014 by Lippincott Williams & Wilkins) LEVEL OF EVIDENCE: Diagnostic study, level II. KEY WORDS: Osteoporosis; screening; head CT; quantitative CT; geriatric trauma. BACKGROUND:
O
steoporosis is a disease of increasing prevalence. In the year 2000, there were an estimated 9.0 million fractures associated with osteoporosis in the United States. Of these, 1.6 million were fractures of the hip, 1.7 million were of the forearm, and 1.4 million were vertebral. Total disabilityadjusted life years lost because of these osteoporotic fractures were estimated to be 5.8 million.1 Treatment using a pharmacologic agent along with vitamin D and calcium has been shown to be cost-effective in several patient populations: those with low bone density measurements who have fractures related to bone fragility, those with osteoporosis based on World Health Organization criteria, and those with osteopenia and additional risk factors.2 Early identification and treatment of patients who meet these criteria can prevent fractures and save health care dollars. However, large numbers of patients who might benefit from bone density analysis do not receive this screening. Low bone density measurement results in higher incidence of fractures following trauma.3 Elderly patients with a Submitted: April 22, 2014, Revised: June 11, 2014, Accepted: June 23, 2014. From the Santa Barbara Cottage Hospital (A.T., J.G., S.Y.), Santa Barbara; Department of Surgery (K.W.), Ventura County Health System, Ventura, California; and San Antonio Colon and Rectal Clinic (S.I.), San Antonio, Texas. Address for reprints: Kenneth Waxman, MD, Department of Surgery, Ventura County Health System, 3291 Loma Vista Rd Ste 401, Ventura, CA 93003; email: [email protected]. DOI: 10.1097/TA.0000000000000411
history of falls are at higher risk of future falls. Therefore, identification and treatment of osteoporosis in fall patients could result in a decrease in future fractures. At present, screening for osteoporosis is seldom performed during the evaluation of trauma patients who present to emergency departments. Many patients receive computed tomography (CT) of the axial skull during trauma evaluations; additional information is available from this imaging, which can be used to screen for osteoporosis. Bone density measurements from CT scans can be obtained by performing additional analysis of the digital information. Such analysis has been termed quantitative computed tomography (QCT). Analysis of bone density on CT scans performed for injury assessment would have the advantage of providing a diagnosis of osteoporosis without additional testing. QCT from the proximal femur has previously been shown to provide an accurate measure of bone density.4 Although dual-energy x-ray absorptiometry (DXA) is the preferred screening test for osteoporosis as defined by the World Health Organization, QCT might serve as a valuable alternative for trauma patients who receive this test for injury evaluation and for whom DXA has not been previously performed. Studies have shown that accurate bone density measurements can be computed from pelvic and lumbar QCT.5 However, patients presenting with falls most commonly experience head injury and often receive only a CT scan of the axial skull. No study to date has evaluated the accuracy of bone density measurements from the axial skull region using QCT.
J Trauma Acute Care Surg Volume 77, Number 4
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
635
J Trauma Acute Care Surg Volume 77, Number 4
Taylor et al.
If axial skull QCT provided information relative to osteoporosis, this would be of great value. In 2007, 40.7% of patients who presented to emergency departments with head injury underwent a CT examination.6 Many of these individuals are elderly patients who have fallen and are therefore at high risk for complications from osteoporosis. The aim of the present study was to determine whether axial skull CT scans performed during trauma evaluations can provide useful information about bone density. This study was performed by adding evaluation of axial skull bone density measurement onto routine trauma CT scan interpretations. By correlating axial skull QCT measurements to the proven methodology of determining bone density from proximal femur QCT measurements, we wished to determine if axial skull CT scans might also serve to screen for osteoporosis. In the second phase of this study, we tested the validity of this approach by determining whether an at-risk population of trauma patients whose axial skull CT scans measured low bone density were at higher risk for fractures.
PATIENTS AND METHODS Bone density measurements were prospectively collected on patients presenting to our institution, a community hospital, who received axial skull CT scans as part of their injury evaluations. The study period was from October 2010 to November 2011. QCT was performed using the same radiologic technique as would have been performed whether this study was underway, with no additional radiation exposure. A General Electric Lightspeed VCT 64-Slice CT Scanner (General Electric Healthcare, Waukesha, WI) was used. A commercial calibration phantom was added to the CT table to provide calibration for the bone density measurements. Commercially available software was used to analyze the digital data and determine bone density (QCT PRO BIT, Mindways Software, Austin, TX). CT scans and bone density measurements were processed by CT technicians who were trained on the QCT software before initiation of the study. This study was initiated and performed by the authors of this study. No manufacturer provided financial support, and both study design and data analysis were performed solely by the authors. The local institutional review board granted a waiver of consent because the study was determined to be of less than minimal risk to the patients. Bone density measurements obtained from this study were considered experimental and were not released to patients, and no interventions were offered. The
only difference between performing conventional CT scans and those generating bone density measurements was the placement of a calibration phantom underneath patients during the scan, which did not alter radiation exposure. The study included two phases: Phase I was designed to correlate bone density measurements between axial skull CT scans and pelvis CT scans. Inclusion criteria were (1) being 18 years old and (2) having undergone both noncontrast axial skull and pelvis CT scans within 30 days. There were no exclusion criteria. Bone density measurements of the pelvis were taken from the proximal femur. Bone density measurements of the axial skull were taken from the temporal bone. Fifty-four patients met inclusion criteria and were included in this phase of the study. The reference normal for bone density measurements was used to calculate a T-score for each patient’s total hip bone area density measurement (superposition of the femoral neck, trochanter, and intertrochanteric regions). T-score was calculated by subtracting the patient’s bone mineral density (BMD) measurement by the mean of a healthy, young population and then dividing by the SD of the healthy, young population’s BMD (Fig. 1). The BMD values for the healthy normal population were provided by the vendor of the QCT. T-scores from total hip measurements using QCT and DXA have shown high correlation (R = 0.93), although the QCT BMD T-scores were slightly lower than that of the DXA BMD.7 The World Health Organization defines osteoporosis as at least 2.5 SDs below the mean peak bone mass of young healthy individuals as measured by DXA.8 An optimal threshold for axial skull BMD measurements to predict osteoporosis was developed by identifying patients as having osteoporosis if their total hip BMD T-score was less than j2.5. The objectives for Phase I of the study were to define the relationship of BMD between the axial skull and hip and to thereby determine the potential reliability of axial skull measurements to assess for osteoporosis. Phase II was a prospective observational study to assess the feasibility of evaluating BMD in the emergent setting. Patients were included in the study if they were evaluated by the trauma team and received a noncontrast CT scan of the axial skull on initial evaluation. There were no exclusion criteria. Radiology technicians notified the research staff when the bone density phantom was used. The research staff then derived BMD from the QCT software. The incidence of fractures was evaluated through a chart review. Patients who screened positive for osteoporosis were not offered any intervention and were not followed up after the hospital discharge.
Figure 1. Equation for calculating T-score. 636
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 77, Number 4
Taylor et al.
A preplanned subgroup analysis was performed on patients older than 65 years to determine whether fracture rate was affected by age and BMD. The correlations referenced throughout the article were calculated based on Spearman’s correlation coefficient. Youden’s J statistics was used to find the optimal threshold for axial skull BMD using total hip BMD T-score less than j2.5 as the cutoff for osteoporosis. Sensitivity and specificity measurements were used to summarize the efficacy of axial skull BMD to assess for osteoporosis.9 A Student’s t test, Wilcoxon rank-sum test, or Fisher’s exact test were used where appropriate, and a p value of 0.05 was used to indicate statistical significance. Data analysis was performed in the R Statistical Computing Environment (R Core Team 2012).
RESULTS Phase I of the study included 54 patients with both axial skull and pelvis CT scans. There were 30 males (55.5%). The mean (SD) age was 75.0 (14.6) years, with a range from 23 years to 98 years. The majority of study subjects, 43 of 54 patients, were older than 65 years. The mean (SD) hip BMD was 0.68 (0.23) g/cm3, the mean (SD) hip T-score was j2.0 (2.0), and the mean (SD) axial skull BMD was 0.17 (0.08) g/cm3. The correlation was 0.52 between the hip BMD T-score and the axial skull BMD (Fig. 2). There were 27 patients (50%) who had a hip BMD T-score below j2.5, which meets the criterion for a diagnosis of osteoporosis. The corresponding threshold for axial skull BMD to assess for osteoporosis was 0.16 g/cm3, which resulted in a sensitivity of 0.70, a specificity of 0.81, and an overall accuracy of predicting osteoporosis of 0.76. Patients with low BMD on axial skull CT were more likely to be older than 65 years (p = 0.006). The second phase included 261 eligible patients. For Phase II, we used the bone density value of 0.16 g/cm3 as calculated from axial skull CT to identify patients with osteoporosis. The mean (SD) age of these patients was 59.9 (24.7) years, with a mean (SD) axial skull BMD of 0.20 (0.12) g/cm3 for all patients. There were 113 patients (43%) who screened positive for osteoporosis using the criteria from Phase I of the study (Table 1).
TABLE 1. Characteristics of Patients Screened for Osteoporosis by Axial Skull CT No Osteoporosis Osteoporosis (n = 148) (n = 113) Age, mean (SD), y Males, n (%) ISS, median (interquartile range) Axial skull BMD, mean (SD), g/cm3
56.3 (24.9) 86 (58.1) 5 (6) 0.28 (0.09)
64.7 (23.7) 66 (58.4) 5 (8) 0.09 (0.05)
p 0.006 0.961 0.638 G0.001
The correlation between BMD and age separated by sex is presented in Figure 2. There was a negative correlation between age and axial skull BMD in the female population (r = j0.35), but this relationship was not observed in the male population (r = j0.06). The relationship between patients older than 65 years who screened positive for osteoporosis bone density and presence of fractures is shown in Table 2. Of patients older than 65 years, there were twice as many fractures in those with a positive screen result for osteoporosis on axial skull CT scan compared with those who screened negative for osteoporosis (p = 0.031) (Fig. 3).
DISCUSSION Osteoporosis is a disease characterized by low BMD, microarchitectural disruption, and increased skeletal fragility. Osteoporosis predisposes individuals to fractures, often after minor trauma. Such fractures have significant morbidity and mortality especially in the geriatric trauma population.10 DXA is the World Health Organization standard for diagnosis of osteoporosis. However, many at-risk patients do not receive DXA screening because of limited availability in many communities. In addition, DXA screening is frequently not reimbursable.10 DXA is not routinely available in the emergency setting, but injured patients who are at risk for osteoporosis often receive CT scans following trauma. It has been shown that QCT of the spine or hip is an acceptable method for measuring bone density and, in some studies, is a better
TABLE 2. Characteristics of Elderly Patients (Age 9 65 Years) Based on the Presence of Osteoporosis* No Osteoporosis Osteoporosis (n = 64) (n = 63) Age, mean (SD) Males, n (%) ISS, median (interquartile range) Abbreviated Injury Scale (AIS) (Ortho), median (interquartile range) Axial skull BMD, mean (SD), g/cm3 Patients with one fracture, n (%) Patients with two fractures, n (%) Patients with at least one fracture, n (%)
Figure 2. Relationship between axial skull and pelvis BMDs.
p
81.2 (8.3) 27 (42.2) 5 (4.8) 0 (0.8)
83 (8.7) 27 (42.9) 7 (8) 0 (2.8)
0.243 0.939 0.163 0.219
0.28 (0.08)
0.08 (0.06)
G0.001
5 (9) 0 10 (15.6)
9 (14.3) 6 (9.5) 20 (31.7)
0.061 0.003 0.031
*Osteoporosis identified from axial skull BMD.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
637
J Trauma Acute Care Surg Volume 77, Number 4
Taylor et al.
Figure 3. Correlation of axial skull BMD by age for men and women.
predictor of fracture risk compared with DXA.6 Our study suggests that data obtained from the commonly performed axial skull CT scan may also be used to screen for osteoporosis. In our sample, 64.3% of the patients undergoing trauma assessment with axial skull CT were found to screen positive for osteoporosis, with the incidence increasing with age. The mean age of the patients in our study with abnormal bone density measured on axial skull CT was 64.5 years. The incidence of osteoporosis increases with age, and women are more likely than men to sustain fractures.11 In our study population, increased age was negatively correlated with axial skull BMD in women. Fractures are more likely to occur in trauma patients with osteoporosis.12 In our study, a positive screen result for osteoporosis on axial skull CT scan in patients older than 65 years was associated with a higher incidence of fractures. While our study was not designed to determine causality of fractures, the higher incidence of fractures in patients who screen positive for osteoporosis on axial skull CT is consistent with what would be expected from an accurate screening tool for osteoporosis. QCT does not add any radiation exposure to patients. The only costs to add QCT to standard axial skull CT scans may include the purchase of additional software (if an institution does not already possess this software) to analyze the digital data and technician training to make the measurement. The potential benefits of screening at-risk patients for osteoporosis following trauma far outweigh the costs of QCT. There were several limitations of this study. Although our study reached statistical significance, the study size was relatively small. There may have been some selection bias within Phase I of the study because we only included patients who received both axial skull and pelvis CT examinations. These patients may have had characteristics different from those patients who were not getting both examinations. This study was observational; no interventions were performed based on our findings. We did not follow patients to see if osteoporosis was diagnosed. Our sole aim was to determine whether the axial skull CT may be a useful tool to generate data relevant to bone density. Our data show that additional quantitative information may be obtained from routine axial skull CT scans performed during trauma evaluations. We have shown that these data can potentially be used to screen for osteoporosis. A study 638
correlating bone density measurements obtained from axial skull CT with DXA would provide a useful follow-up to the current study. If the usefulness of screening for osteoporosis with axial skull QCT is validated by further studies, we believe this would be a valuable addition to emergency trauma evaluations. Information regarding bone density could be obtained from axial skull QCT and communicated to primary care physicians for treatment, at no additional risk and minimal cost. Routine assessment of bone density in patients presenting after trauma could be integrated into a comprehensive and effective community care model to identify and treat an at-risk population of patients with osteoporosis and potentially decrease their risk of future injuries. AUTHORSHIP S.I. contributed to the study design and data collection. K.W. contributed to the study design, data analysis, and manuscript preparation. A.T. contributed to the data analysis and manuscript preparation. J.G. contributed to the statistical analysis and manuscript preparation. S.Y. contributed to the data collection. ACKNOWLEDGMENTS We thank Ashlie Thompson, CT technician, who generated bone density measurements in real time using the QCT software. We also thank Jennifer Kosek who performed a chart review to identify fractures in Phase II patients. DISCLOSURE This study was supported by the Santa Barbara Cottage Hospital Research Committee.
REFERENCES 1. Johnell O, Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int. 2006;17(12): 1726Y1733. 2. Dawson-Hughes B, Tosteson AN, Melton LJ, Baim S, Favus MJ, Khosla S, Lindsay RL, N. O. F. G. Committee. Implications of absolute fracture risk assessment for osteoporosis practice guidelines in the USA. Osteoporos Int. 2008;19(4):449Y458. 3. Anderson F, Cooper C. The influence of osteoporosis in trauma. Trauma. 1999;1(3):181Y192. 4. Damilakis J, Maris TG, Karantanas AH. An update on the assessment of osteoporosis using radiologic techniques. Eur Radiol. 2007;17(6):1591Y1602. 5. Guglielmi G, Lang TF. Quantitative computed tomography. Semin Musculoskelet Radiol. 2002;6(3):219Y227.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 77, Number 4
6. Larson DB, Johnson LW, Schnell BM, Salisbury SR, Forman HP. National trends in CT use in the emergency department: 1995Y2007. Radiology. 2011;258(1):164Y173. 7. Khoo BC, Brown K, Cann C, Zhu K, Henzell S, Low V, Gustafsson S, Price RI, Prince RL. Comparison of QCT-derived and DXA-derived areal bone mineral density and T scores. Osteoporos Int. 2009;20(9):1539Y1545. 8. Kanis JA. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. WHO Study Group. Osteoporos Int. 1994;4(6):368Y381.
Taylor et al.
9. Youden WJ. Index for rating diagnostic tests. Cancer. 1950;3(1):32Y35. 10. WHO scientific group on the assessment of osteoporosis at primary health care level. Brussels, Belgium: World Health Organization; 2004. Available at: www.who.int/chp/topics/Osteoporosis.pdf. Accessed March 12, 2014. 11. Gerdhem P. Osteoporosis and fragility fractures. Best Pract Res Clin Rheumatol. 2013;27(6):743Y755. 12. Mackey D, Lui L, Cawthon P, Bauer D, Nevitt M, Cauley J, Hillier T, Lewis C, Barrett-Connor E, Cummings S. High-trauma fractures and low bone mineral density in older women and men. JAMA. 2007;298(20):2381Y2388.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
639
