13
Journal of Back and Musculoskeletal Rehabilitation 28 (2015) 13–18 DOI 10.3233/BMR-140479 IOS Press
Association between bone scintigraphy features of spinal degeneration and anthropometric and demographic variables Sofia Chatziioannoua, Maria Kallergib,∗ , Pinelopi Karampinaa, Polixeni Zotoua, Sotiris Bakalisa , Vassiliki Lyraa , George Lamprakopoulosa, Ioannis Armeniakosa and Spiros Pneumaticosc a
National and Kapodistrian University of Athens, 2nd Department of Radiology, Nuclear Medicine Section, “Attikon” University General Hospital, Athens, Greece b Department of Biomedical Engineering, Technological Educational Institute of Athens, Athens, Greece c National and Kapodistrian University of Athens, 3rd Department of Orthopaedic Surgery, “KAT” General Hospital, Athens, Greece
Abstract. BACKGROUND AND OBJECTIVES: Bone scintigraphy is a molecular imaging technique routinely used for the evaluation of benign and malignant bone abnormalities. This study aimed at evaluating spinal degenerative changes detected by bone scintigraphy and determining associations between image features and patients’ anthropometric and demographic variables. MATERIAL AND METHOD: In a cross-sectional study, 64 men and 52 women underwent bone scintigraphy. Experts identified all image regions suggesting degenerative joint disease (DJD) and classified region intensity on a 3-point scale. Image characteristics were correlated to the patients’ body mass index (BMI), age, weight, height, activity level, and sex. Data analysis included descriptive statistics and association coefficients. RESULTS: DJD was found in 53 patients (46%). In men, there was weak but statistically significant correlation between DJD and activity level, and DJD and age, but not BMI or weight. In women, only a weak, not statistically significant, linear correlation was found between DJD and BMI, and DJD and weight. CONCLUSION: Molecular imaging with bone scintigraphy showed that spinal degenerations are associated with different anthropometric and demographic features in men and women. Interestingly, no association was found between DJD and increased body weight in men while a weak association may exist in women. The results prompt for additional studies to better determine the risk factors for DJD and low back pain in male and female patients. LEVEL OF EVIDENCE: Diagnostic study, Level II (retrospective study). Keywords: Bone scintigraphy, molecular imaging, spinal degeneration, weight, low back pain, body mass index, musculoskeletal disorder
1. Introduction Degenerative bone disease is a complex, multifactorial condition. Hereditary and environmental factors such as age, weight, height, gender, occupation, smoking, exposure to vibration, and genetic factors ∗ Corresponding author: Maria Kallergi, Ph.D., Department of Biomedical Engineering, Technological Educational Institute of Athens, 28 Ag.Spiridona St., Athens 12243, Greece. Tel.: +30 210 5385531; E-mail: [email protected].
have all been implicated in its causation [1,2]. Occupational conditions are known to induce degenerative joint changes as a result of increased load to the spine, knees and hips, due to monotonous and repetitive body position that leads to injury of the nervous and circulatory system, leading to damage and pain [3–5]. Genetic influence also seems to play a significant role on an earlier onset of lumbar spine degeneration [6–9]. Many studies have demonstrated a strong correlation between obesity and knee osteoarthritis [10–17]. The latter association is more often seen in older women with peripheral body girth pattern and coexisting fac-
c 2015 – IOS Press and the authors. All rights reserved ISSN 1053-8127/15/$35.00
14
S. Chatziioannou et al. / Association between bone scintigraphy features of spinal degeneration
tors such as knee injury and high physical activity levels. Body weight is considered one of the most significant risk factors for the development of degenerative bone disease [13]. Physicians urge overweight or obese patients with degenerative bone disease, who present with back or peripheral joint pain, to lose weight. It has been suggested that obesity is responsible for the development of osteoarthritis as a result of mechanical or metabolic effect upon the bones [14]. However, many of the referenced studies have not found a significant relationship between obesity and hip osteoarthritis and, similarly, the association between body weight and DJD of the spine has not been established yet. Radiography and computed tomography (CT) are the most common diagnostic tools for the evaluation of joint and spinal degeneration [18]. Consequently, there are reports of several observational studies that aim at exploring the association between abnormal lumbar radiographic findings and low back pain. A systematic review of such studies concluded that “there was no firm evidence for the presence or absence of a causal relationship between radiographic findings and nonspecific low back pain.” [18]. So, the question rises whether this non-specific result may be due to the limitation of the imaging technique, i.e., radiography. In this study, we attempt a different approach to the evaluation of DJD of the spine by using molecular instead of anatomical imaging. Radionuclide bone scintigraphy is one of the most widely used molecular imaging modalities for the evaluation of benign and malignant bone abnormalities [19–21]. This non-invasive technique is widely available, of relatively low cost, and practically sideeffect free. It can detect bone areas with increased osteoblastic activity or synovial changes caused by hyperaemia or inflammation. It can also depict degenerative changes, particularly those with a high degree of remodelling [21]. The purpose of this work was to use bone scintigraphy to evaluate, in a cross-sectional study, the possible association between anthropometric and demographic factors and DJD.
Institution’s Review Board. The majority of patients were referred for staging of new neoplastic disease, primarily breast or lung cancer. Twenty three patients were excluded because they had known metastatic or metabolic bone disease or presented one or more lesions on the bone scintigraphy consistent or suspicious for metastatic disease. The final study population included 116 adult patients, 64 males and 52 females, with ages ranging from 27 to 92 years (average age of 68 years). All patients were asked to complete a questionnaire and provide the following information: Age, profession, reason for referral, history of orthopaedic interventions, surgeries, pain presence or absence, pain location, and other major medical history. Weight and height were measured in the clinic. Based on the questionnaire responses regarding history of orthopaedic interventions and pain symptoms, only 10% (12/116) of the patients had prior knee or hip surgery, 48% (56/116) had no pain symptoms, 18% (22/116) complained of spinal pain (cervical, thoracic, or lumbar), 29% (34/116) had joint pain (knee, elbow, or hip pain), and 4% (5/116) complained of diffused pain. 2.2. Body weight and body mass index calculation Based on the collected information, the average patient weight was 74.1 kg with a standard deviation of 13.9 kg. The average patient height was 1.7 m with a standard deviation of 0.1 m. From the weight and height data, the Body Mass Index (BMI) was calculated by the formula: BM I = HW2 (kg) (m2 ) , where W is the weight in kg and H is the height in m. The average BMI was 26.7 kg/m2 with a standard deviation of 4.5 kg/m2 . 2.3. Activity level
2. Material and method
Based on their profession, the patients were placed in one of six categories representing the level of their muscular effort or activity. Table 1 shows the correspondence between level of activity during an 8-hour workday and profession as established by a panel of experts who reviewed all demographic information. It also shows the number of patients categorized at each level.
2.1. Patients
2.4. Bone scintigraphy
One hundred and thirty nine (139) consecutive patients referred to our Department for bone scintigraphy were enrolled in our study, which was approved by our
Patients received 740 MBq (20mCi) of 99m Tc MDP (Sodium medronate – Amersham Medronate II Agent) intravenously. Whole-body anterior and posterior im-
S. Chatziioannou et al. / Association between bone scintigraphy features of spinal degeneration
15
Table 1 Categorization of the patients according to their activity level Level of activity 0 1 2 3 4 5
Profession Minimum activity (0% standing); working in a sitting job, e.g., office employee, taxi driver, retiree Low activity (< 25% standing); working primarily in a sitting job with infrequent movements and strain, e.g., businessmen, most public sector employees, librarian Low-to-average activity (25%–50% standing), e.g., sales people, people running errands, couriers, private sector employees Average activity (50% standing) (e.g., housewives) Significant activity (50%–75% standing), e.g., farmers and environmental services personnel Maximum activity (almost 100% standing); working standing up or constantly moving with short break intervals (e.g., construction workers, heavy industry workers)
ages were performed at 2–4 hours post injection. Whole body imaging was performed with a single head gamma camera (Millennium MPR/MPS GE Medical systems) with a table motion speed of 11 cm/min, 256 × 1024 matrix, energy peaked at 140 keV, window 20%, and a Low-Energy-High-Resolution collimator. Additional spot views were obtained when necessary. The scans were reviewed by an experienced nuclear medicine physician, who identified all areas of increased uptake of the radiotracer in the spine as well as in the peripheral joints (e.g., knees, shoulders, hips) and graded the intensity of each finding from 1– 3 (1:mild, 2:moderate, 3:severe). If no abnormal uptake was observed, the scan was assigned zero intensity. The expert was blinded to the patients’ clinical history and symptoms. 2.5. Data analysis Data were characterized for location and variability with descriptive statistics, e.g., means, and standard deviation [22]. In addition, linear and nonlinear relationships between pairs of variables were investigated by generating scatter plots and estimating the corresponding correlation coefficients [23,24]. Correlation analysis was done with the following methods: 1) Pearson’s linear correlation coefficient was calculated with the null hypothesis being that the correlation coefficient between pairs of datasets is 0. 2) Spearman’s nonlinear association coefficient was calculated with the null hypothesis being that the correlation coefficient between pairs of datasets is 0. 3) Somers’ DXY ordinal measure of association was estimated to evaluate how well the knowledge of the maximum intensity of scintigraphy findings could predict the patients’ pain symptoms.
Number of patients 16 24 8 26 26 16
Correlation results were further analyzed in terms of statistical significance by: [22] 1) Estimating corresponding P values given the calculated coefficients and the sample size. (Null hypothesis was tested at the 0.05 and 0.01 levels.) 2) Estimating confidence intervals (CI) for a Pvalue of 0.05. For the calculation of a CI, the Pearson’s correlation coefficient was first transformed using Fisher’s z-transformation. Then, the standard error was estimated followed by the 95% CI of the transformed correlation coefficient. The latter was used to estimate the 95% CI of the correlation coefficient by applying the inverse Fisher’s transform on its lower and upper limits. 3) Applying a one-tailed t-test to determine the significance of the Somers’ D value. For the analysis, the data were reorganized, reordered, and ranked in two groups: Group A for spine data only and Group B for all joint data. Spine data included findings in the cervical, thoracic and lumbar spine. All joint data included all findings independent of location, e.g., spine, knees, shoulders, hip. We investigated the correlation of the maximum intensity and the number of findings in each data group to the patients’ five characteristics, i.e., BMI, weight, height, activity, and age. This was done for all patients (116) as well as the male (64) and female (52) patients separately. Hence, 60 correlation coefficients were calculated for each data group while 95% CIs were estimated for the pairs where statistically significant correlation was found.
3. Results Overall, 53 patients (45.7%) had findings in the bone scintigraphy consistent with DJD. A total of 229 ar-
16
S. Chatziioannou et al. / Association between bone scintigraphy features of spinal degeneration
Table 2 Linear correlation coefficients between the maximum intensity of the findings in Group A (spine only) and Group B (all) data and the patients’ characteristics, i.e., BMI, weight, height, activity level, and age in years. Correlation was studied for all patients (116), female patients only (52), and male patients only (64). The cells highlighted in gray represent statistically significant linear correlations and the corresponding 95% CIs are listed in brackets Group
#
Pair of data correlated
A (spine)
1 2 3 4 5
Max intensity vs. BMI Max intensity vs. Weight Max intensity vs. Height Max intensity vs. Activity level Max intensity vs. Age
All patients (116) 0.09 0.10 0.04 0.07 0.22 [0.04,0.39]
B (all)
6 7 8 9 10
Max intensity vs. BMI Max intensity vs. Weight Max intensity vs. Height Max intensity vs. Activity level Max intensity vs. Age
0.11 0.14 0.08 0.08 0.20 [0.02,0.37]
Linear correlation coefficients Female patients Male patients (52) (64) 0.24 −0.04 0.24 −0.03 0.00 −0.03 0.05 0.10 0.14 0.26 [0.01,0.48] 0.24 0.20 −0.09 0.07 0.14
−0.01 0.04 0.06 0.11 0.28 [0.03,0.49]
Table 3 Linear correlation coefficients between the number (Nb) of findings in Group A (spine only) and Group B (all) data and the patients’ characteristics, i.e., BMI, weight, height, activity level, and age in years. Correlation was studied for all patients (116), female patients only (52), and male patients only (64). The cells highlighted in gray represent statistically significant linear correlations and the corresponding 95% CIs are listed in brackets Group
#
Pair of data correlated
A (spine)
1 2 3 4 5
Nb of findings vs. BMI Nb of findings vs. Weight Nb of findings vs. Height Nb of findings vs. Activity level Nb of findings vs. Age
All patients (116) 0.12 0.12 0.03 0.33 [0.15,0.48] 0.25 [0.07,0.41]
B (all)
6 7 8 9 10
Nb of findings vs. BMI Nb of findings vs. Weight Nb of findings vs. Height Nb of findings vs. Activity level Nb of findings vs. Age
0.13 0.15 0.06 0.33 [0.15,0.48] 0.25 [0.07,0.41]
eas of increased uptake were identified, the majority of which (78%) were in the spine and particularly the thoracic and lumbar spine, which, combined, accounted for 74% of the findings. Table 2 presents the Pearson’s correlation coefficients between maximum intensity of a finding and patients’ characteristics for both Groups A and B as a function of patient population. The gray highlighted cells show the pairs that yielded statistically significant linear correlation. In general, no strong correlations were identified, namely all were below 0.5. However, the sample size was sufficiently large to suggest that there is statistically significant correlation between the maximum intensity of a finding and the patients’ age. A detailed review of the data further suggested that it is the male patients and the findings for DJD of the spine in particular that cause this correlation. The correlation results between the number of findings in the spine (Group A data) and all joints (Group
Linear correlation coefficients Female patients Male patients (52) (64) 0.20 0.06 0.21 0.00 0.03 −0.14 0.06 0.50 [0.28,0.67] 0.03 0.36 [0.13,0.56] 0.23 0.22 −0.02 0.09 0.07
0.07 0.03 −0.11 0.48 [0.26,0.65] 0.38 [0.15,0.58]
B data) and the patients’ characteristics are presented in Table 3. The gray highlighted cells show the pairs that yielded statistically significant linear correlation. Significant correlations were again identified between the number of findings and the patients’ age. This time, however, a second significant correlation was observed between the number of findings and the activity level of the patients. As previously, it is the male patient population that primarily accounts for this correlation outcome. The Somers’ DXY coefficient was estimated to evaluate the association between maximum intensity of scintigraphy findings and pain. For the analysis, the ordered independent variable was the maximum intensity (ranging from 0 to 3) and the binary dependent variable was pain (coded 0 for no pain and 1 for pain). Considering maximum intensity as the predictor or independent variable X and pain as the outcome variable Y, Somers’ DXY was estimated to be 0.46 suggesting
S. Chatziioannou et al. / Association between bone scintigraphy features of spinal degeneration
that maximum intensity could predict pain in 46% of the cases; note that Somers’ coefficient’s values range from −1 to +1. This association is statistically significant based on a one-tailed t-test.
4. Discussion and conclusion In the present study, bone scintigraphy was used to identify active DJD as opposed to traditional anatomic imaging with radiography or computed tomography (CT), the findings of which are often due to old and inactive disease. Bone scintigraphy has the potential to detect bone changes earlier than radiography [20] and demonstrates the metabolically active osseous or joint changes, which are most likely responsible for the development of pain [19]. The results of the study led to the following conclusions: i) Statistically significant correlation exists between age and DJD of the spine or, generally, age and joint disease, as indicated by both the maximum intensity and the number of findings analysis in Tables 2 and 3. ii) Statistically significant correlation exists between activity level and DJD of the spine or, generally, activity level and joint disease, as indicated by the number of findings analysis in Table 3. iii) Weak, not statistically significant, correlations were observed between BMI and DJD of the spine or BMI and joint disease. The stronger correlations were observed for the female patients with coefficients in the range of 0.20–0.24 and 95% CIs in the interval [−0.07, 0.48]. iv) Weak, not statistically significant, correlations were observed between weight and DJD of the spine or weight and joint disease. Again, the stronger correlations were observed for the female patients with coefficients in the range of 0.20–0.24 and 95% CIs in the interval [−0.07, 0.47]. v) Correlation results differed between men and women suggesting that the two groups should be evaluated separately. Back problems, and possibly other skeletal problems, in men were correlated with the type of their activity, increasing linearly with age both in terms of number and intensity. Weight or BMI did not seem to be a risk factor for degenerative spine disease in men. In contrast, female patients did not present
17
any statistically significant correlations at the P = 0.01 or P = 0.05 levels. The results for the female patients, however, suggested that a linear correlation may exist between BMI and maximum intensity or between BMI and number of findings either in the spine or all joint locations. A similar outcome was observed for the weight. vi) Statistically significant association was found between maximum intensity of scintigraphy findings and pain. Results suggested that pain may be possible to predict in approximately 50% of the patients based on the intensity of the scintigraphy findings. However, the current study data did not allow the definition of an intensity “threshold” that could be used to predict painful versus non-painful degenerative disease. The studied population was sufficiently large in size to allow us to reach statistically significant conclusions in some associations despite the relatively small correlation coefficients (see Table 3). The unevenness of the CIs with the lower limit being very close to zero in some comparisons (see 3rd and 4th conclusions above) suggests that a larger patient population may have led to statistically significant correlations between, for example, weight or BMI and DJD of the spine in women. The strength of the association, however, would still be relatively weak. Analysis of the data for nonlinear associations was negative, namely there were no nonlinear relationships between the various pairs of data with the Spearman coefficient remaining below 0.13 in all comparisons. This study included a group of patients that underwent bone scintigraphy in a tertiary care center for a wide range of indications. Consequently, the study population does not represent well the general population and one could argue that it introduces a selection bias that could be in favor or against the presence of DJD. Degenerative bone disease, however, affects all patients independent of whether they have other underlying pathology or not, and so the results of the study, generally or conditionally applicable, hold. In conclusion, molecular imaging findings suggest that level of activity and age are strong risk factors for the development of DJD in men in agreement with a recent review [25]. The age/DJD correlation in this group may be due to a combination of normal aging changes in the body and changes caused by the years the patients were engaged to a certain activity. The distinction between the two causes was not possible in this study. It should be noted, however, that this dependence was clearly not observed for the female patients.
18
S. Chatziioannou et al. / Association between bone scintigraphy features of spinal degeneration
Weight and BMI seem to be stronger DJD factors in women than men. The latter group showed negligible correlation while female patients showed linear correlation coefficients that were approaching statistical significance. Hence, sex appeared to be a critical factor in identifying the causes of DJD of the spine. Certainly, bone scintigraphy is not used routinely in every day practice and is not the method of choice for the identification of degenerative bone changes. Other imaging modalities, such as plain radiographs and CT, are more widely used for the demonstration of degenerative changes and studies based on those modalities lead to different associations. Kalichman et al, for example, observed significantly higher prevalence of facet joint osteoarthritis in the CT scans of obese patients [26]. Bone scintigraphy, however, as a molecular imaging technique, has demonstrated the ability to identify the active and, consequently, painful degenerative changes and, thus, the changes that lead to symptomatology that may require further treatment [20,21]. Therefore, bone scintigraphy may better delineate the presence of painful and clinically relevant degenerative spine changes. Molecular findings seem also to suggest that male and female patients merit a different evaluation for spinal degeneration and low back pain, an aspect worth pursuing further to avoid misinterpretations and erroneous generalizations.
References [1]
[2]
[3]
[4]
[5] [6]
[7] [8]
Mody GM, Brooks PM. Improving musculoskeletal health: global issues. Best Pract Res Clin Rheumatol 2012; 26(2): 237-249. Haldeman S, Kopansky-Giles D, Hurwitz EL, Hoy D, Mark Erwin W, Dagenais S, Kawchuk G, Strömqvist B, and Walsh N. Advancements in the management of spine disorders. Best Pract Res Clin Rheumatol 2012; 26:263-280. Vignon E, Valat JP, Rossignol M, Avouac B, Rozenberg S, Thoumie P, Avouac J, Nordin M, Hilliguin P. Osteoarthritis of the knee and hip and activity: A systematic international review and synthesis (OASIS). Joint Bone Spine 2006; 73:442455. Lundberg U. Stress responses in low-status jobs and their relationship to health risks: musculoskeletal disorders. Ann N Y Acad Sci 1999; 896:162-172. Seidel H. Selected health risks caused by long-term, wholebody vibration. Am J Ind Med 1993; 23:589-604. Zhang Y, Sun Z, Liu J, Guo X. Advances in susceptibility genetics of intervertebral degenerative disc disease.Int J Biol Sci 2008; 4:283-290. Frymoyer J. Lumbar disk disease: epidemiology. Instr Course Lect 1992; 41:217-223. Battié MC, Videman T, Levalahti E, Gill K, Kaprio J.
[9] [10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23] [24]
[25] [26]
Heritability of low back pain and the role of disc degeneration. Pain 2007; 131:272-280. Ala-Kokko L. Genetic risk factors for lumbar disc disease. Ann Med 2002; 34:42-47. Felson DT, Zhang Y, Hannan M, Naimark A, Weissman B, Aliabadi P, Levy D. Risk factors for incident radiographic knee osteoarthritis in the elderly: The Framingham Study. Arthritis Rheum 1997; 40:728-733. Felson DT, Anderson JJ, Naimark A, Walker AM, Meenan RF. Obesity and knee osteoarthritis. The Framingham Study. Ann Intern Med 1988; 109:18-24. Gelber AC, Hochberg MC, Mead LA, Wang NY, Wigley FM, Klag MJ. Body mass index in young men and the risk of subsequent knee and hip osteoarthritis. Am J Med 1999; 107:542548. Like M, Solovieva S, Lamminen A, Luoma K, Leino-Arjas P, Luukkonen R, and Riihimaki H. Disc degeneration of the lumbar spine in relation to overweight. Int J Obes (Lond). 2005; 29(8):903-908. Sandmark H, Hogstedt C, Lewold S, Vingård E. Osteoarthrosis of the knee in men and women in association with overweight, smoking, and hormone therapy. Ann Rheum Dis 1999; 58:151–155. Coggon D, Reading I, Croft P, McLaren M, Barrett D, Cooper C. Knee osteoarthritis and obesity. Int J Obes Relat Metab Disord. 2001; 25:622-627. Grotle M, Hagen KB, Natvig B, Dahl FA, Kvien TK. Obesity and osteoarthritis in knee, hip and/or hand: an epidemiological study in the general population with 10 years follow-up. BMC Musculoskelet Disord. 2008; 9:132. Davis MA, Neuhaus JM, Ettinger WH, Mueller WH. Body fat distribution and osteoarthritis. Am J Epidemiol. 1990; 132:701-707. Van Tulder MW, Assendelft WJ, Koes BW, Bouter LM. Spinal radiographic findings and nonspecific low back pain: A systematic review of observational studies. Spine. 1997; 22(4): 427-434. Horger M, Bares R. The role of single-photon emission computed tomography/computed tomography in benign and malignant bone disease. Semin Nucl Med. 2006; 36:286-294. Pneumaticos SG, Chatziioannou SN, Hipp JA, Moore WH, Esses SI. Low back pain: Prediction of short-term outcome of facet joint injection with bone scintigraphy. Radiology 2006; 238:693-698. Hutton CW, Higgs ER, Jackson PC, Watt I. 99mTc HMDP bone scanning in generalised nodal osteoarthritis. II. The four hour bone scan image predicts radiographic change. Ann Rheum Dis. 1986; 45:622-626. Whitley E, Ball J. Statistics review 1: presenting and summarising data. Crit Care. 2002; 6(1):66-71. Bewick V, Cheek L, Ball J. Statistics review 7: Correlation and regression. Crit Care. 2003; 7:451-459. Wessa P. Somers Dxy Rank Correlation (v1.0.2) in Free Statistics Software (v1.1.23-r7). Office for Research Development and Education, 2006 [updated 2014 Jan; cited 2014 Feb 20]. Available from: http://www.wessa.net/rwasp_ somers.wasp/. Boqduk N. Degenerative joint disease of the spine. Radiol Clin North Am 2012; 50(4):613-628. Kalichman L, Guermazi A, Li L, Hunter DJ. Association between age, sex, BMI and CT-evaluated spinal degeneration features. J Back Musculoskelet Rehabil. 2009; 22:189-195.
Copyright of Journal of Back & Musculoskeletal Rehabilitation is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Journal of Back and Musculoskeletal Rehabilitation 28 (2015) 13–18 DOI 10.3233/BMR-140479 IOS Press
Association between bone scintigraphy features of spinal degeneration and anthropometric and demographic variables Sofia Chatziioannoua, Maria Kallergib,∗ , Pinelopi Karampinaa, Polixeni Zotoua, Sotiris Bakalisa , Vassiliki Lyraa , George Lamprakopoulosa, Ioannis Armeniakosa and Spiros Pneumaticosc a
National and Kapodistrian University of Athens, 2nd Department of Radiology, Nuclear Medicine Section, “Attikon” University General Hospital, Athens, Greece b Department of Biomedical Engineering, Technological Educational Institute of Athens, Athens, Greece c National and Kapodistrian University of Athens, 3rd Department of Orthopaedic Surgery, “KAT” General Hospital, Athens, Greece
Abstract. BACKGROUND AND OBJECTIVES: Bone scintigraphy is a molecular imaging technique routinely used for the evaluation of benign and malignant bone abnormalities. This study aimed at evaluating spinal degenerative changes detected by bone scintigraphy and determining associations between image features and patients’ anthropometric and demographic variables. MATERIAL AND METHOD: In a cross-sectional study, 64 men and 52 women underwent bone scintigraphy. Experts identified all image regions suggesting degenerative joint disease (DJD) and classified region intensity on a 3-point scale. Image characteristics were correlated to the patients’ body mass index (BMI), age, weight, height, activity level, and sex. Data analysis included descriptive statistics and association coefficients. RESULTS: DJD was found in 53 patients (46%). In men, there was weak but statistically significant correlation between DJD and activity level, and DJD and age, but not BMI or weight. In women, only a weak, not statistically significant, linear correlation was found between DJD and BMI, and DJD and weight. CONCLUSION: Molecular imaging with bone scintigraphy showed that spinal degenerations are associated with different anthropometric and demographic features in men and women. Interestingly, no association was found between DJD and increased body weight in men while a weak association may exist in women. The results prompt for additional studies to better determine the risk factors for DJD and low back pain in male and female patients. LEVEL OF EVIDENCE: Diagnostic study, Level II (retrospective study). Keywords: Bone scintigraphy, molecular imaging, spinal degeneration, weight, low back pain, body mass index, musculoskeletal disorder
1. Introduction Degenerative bone disease is a complex, multifactorial condition. Hereditary and environmental factors such as age, weight, height, gender, occupation, smoking, exposure to vibration, and genetic factors ∗ Corresponding author: Maria Kallergi, Ph.D., Department of Biomedical Engineering, Technological Educational Institute of Athens, 28 Ag.Spiridona St., Athens 12243, Greece. Tel.: +30 210 5385531; E-mail: [email protected].
have all been implicated in its causation [1,2]. Occupational conditions are known to induce degenerative joint changes as a result of increased load to the spine, knees and hips, due to monotonous and repetitive body position that leads to injury of the nervous and circulatory system, leading to damage and pain [3–5]. Genetic influence also seems to play a significant role on an earlier onset of lumbar spine degeneration [6–9]. Many studies have demonstrated a strong correlation between obesity and knee osteoarthritis [10–17]. The latter association is more often seen in older women with peripheral body girth pattern and coexisting fac-
c 2015 – IOS Press and the authors. All rights reserved ISSN 1053-8127/15/$35.00
14
S. Chatziioannou et al. / Association between bone scintigraphy features of spinal degeneration
tors such as knee injury and high physical activity levels. Body weight is considered one of the most significant risk factors for the development of degenerative bone disease [13]. Physicians urge overweight or obese patients with degenerative bone disease, who present with back or peripheral joint pain, to lose weight. It has been suggested that obesity is responsible for the development of osteoarthritis as a result of mechanical or metabolic effect upon the bones [14]. However, many of the referenced studies have not found a significant relationship between obesity and hip osteoarthritis and, similarly, the association between body weight and DJD of the spine has not been established yet. Radiography and computed tomography (CT) are the most common diagnostic tools for the evaluation of joint and spinal degeneration [18]. Consequently, there are reports of several observational studies that aim at exploring the association between abnormal lumbar radiographic findings and low back pain. A systematic review of such studies concluded that “there was no firm evidence for the presence or absence of a causal relationship between radiographic findings and nonspecific low back pain.” [18]. So, the question rises whether this non-specific result may be due to the limitation of the imaging technique, i.e., radiography. In this study, we attempt a different approach to the evaluation of DJD of the spine by using molecular instead of anatomical imaging. Radionuclide bone scintigraphy is one of the most widely used molecular imaging modalities for the evaluation of benign and malignant bone abnormalities [19–21]. This non-invasive technique is widely available, of relatively low cost, and practically sideeffect free. It can detect bone areas with increased osteoblastic activity or synovial changes caused by hyperaemia or inflammation. It can also depict degenerative changes, particularly those with a high degree of remodelling [21]. The purpose of this work was to use bone scintigraphy to evaluate, in a cross-sectional study, the possible association between anthropometric and demographic factors and DJD.
Institution’s Review Board. The majority of patients were referred for staging of new neoplastic disease, primarily breast or lung cancer. Twenty three patients were excluded because they had known metastatic or metabolic bone disease or presented one or more lesions on the bone scintigraphy consistent or suspicious for metastatic disease. The final study population included 116 adult patients, 64 males and 52 females, with ages ranging from 27 to 92 years (average age of 68 years). All patients were asked to complete a questionnaire and provide the following information: Age, profession, reason for referral, history of orthopaedic interventions, surgeries, pain presence or absence, pain location, and other major medical history. Weight and height were measured in the clinic. Based on the questionnaire responses regarding history of orthopaedic interventions and pain symptoms, only 10% (12/116) of the patients had prior knee or hip surgery, 48% (56/116) had no pain symptoms, 18% (22/116) complained of spinal pain (cervical, thoracic, or lumbar), 29% (34/116) had joint pain (knee, elbow, or hip pain), and 4% (5/116) complained of diffused pain. 2.2. Body weight and body mass index calculation Based on the collected information, the average patient weight was 74.1 kg with a standard deviation of 13.9 kg. The average patient height was 1.7 m with a standard deviation of 0.1 m. From the weight and height data, the Body Mass Index (BMI) was calculated by the formula: BM I = HW2 (kg) (m2 ) , where W is the weight in kg and H is the height in m. The average BMI was 26.7 kg/m2 with a standard deviation of 4.5 kg/m2 . 2.3. Activity level
2. Material and method
Based on their profession, the patients were placed in one of six categories representing the level of their muscular effort or activity. Table 1 shows the correspondence between level of activity during an 8-hour workday and profession as established by a panel of experts who reviewed all demographic information. It also shows the number of patients categorized at each level.
2.1. Patients
2.4. Bone scintigraphy
One hundred and thirty nine (139) consecutive patients referred to our Department for bone scintigraphy were enrolled in our study, which was approved by our
Patients received 740 MBq (20mCi) of 99m Tc MDP (Sodium medronate – Amersham Medronate II Agent) intravenously. Whole-body anterior and posterior im-
S. Chatziioannou et al. / Association between bone scintigraphy features of spinal degeneration
15
Table 1 Categorization of the patients according to their activity level Level of activity 0 1 2 3 4 5
Profession Minimum activity (0% standing); working in a sitting job, e.g., office employee, taxi driver, retiree Low activity (< 25% standing); working primarily in a sitting job with infrequent movements and strain, e.g., businessmen, most public sector employees, librarian Low-to-average activity (25%–50% standing), e.g., sales people, people running errands, couriers, private sector employees Average activity (50% standing) (e.g., housewives) Significant activity (50%–75% standing), e.g., farmers and environmental services personnel Maximum activity (almost 100% standing); working standing up or constantly moving with short break intervals (e.g., construction workers, heavy industry workers)
ages were performed at 2–4 hours post injection. Whole body imaging was performed with a single head gamma camera (Millennium MPR/MPS GE Medical systems) with a table motion speed of 11 cm/min, 256 × 1024 matrix, energy peaked at 140 keV, window 20%, and a Low-Energy-High-Resolution collimator. Additional spot views were obtained when necessary. The scans were reviewed by an experienced nuclear medicine physician, who identified all areas of increased uptake of the radiotracer in the spine as well as in the peripheral joints (e.g., knees, shoulders, hips) and graded the intensity of each finding from 1– 3 (1:mild, 2:moderate, 3:severe). If no abnormal uptake was observed, the scan was assigned zero intensity. The expert was blinded to the patients’ clinical history and symptoms. 2.5. Data analysis Data were characterized for location and variability with descriptive statistics, e.g., means, and standard deviation [22]. In addition, linear and nonlinear relationships between pairs of variables were investigated by generating scatter plots and estimating the corresponding correlation coefficients [23,24]. Correlation analysis was done with the following methods: 1) Pearson’s linear correlation coefficient was calculated with the null hypothesis being that the correlation coefficient between pairs of datasets is 0. 2) Spearman’s nonlinear association coefficient was calculated with the null hypothesis being that the correlation coefficient between pairs of datasets is 0. 3) Somers’ DXY ordinal measure of association was estimated to evaluate how well the knowledge of the maximum intensity of scintigraphy findings could predict the patients’ pain symptoms.
Number of patients 16 24 8 26 26 16
Correlation results were further analyzed in terms of statistical significance by: [22] 1) Estimating corresponding P values given the calculated coefficients and the sample size. (Null hypothesis was tested at the 0.05 and 0.01 levels.) 2) Estimating confidence intervals (CI) for a Pvalue of 0.05. For the calculation of a CI, the Pearson’s correlation coefficient was first transformed using Fisher’s z-transformation. Then, the standard error was estimated followed by the 95% CI of the transformed correlation coefficient. The latter was used to estimate the 95% CI of the correlation coefficient by applying the inverse Fisher’s transform on its lower and upper limits. 3) Applying a one-tailed t-test to determine the significance of the Somers’ D value. For the analysis, the data were reorganized, reordered, and ranked in two groups: Group A for spine data only and Group B for all joint data. Spine data included findings in the cervical, thoracic and lumbar spine. All joint data included all findings independent of location, e.g., spine, knees, shoulders, hip. We investigated the correlation of the maximum intensity and the number of findings in each data group to the patients’ five characteristics, i.e., BMI, weight, height, activity, and age. This was done for all patients (116) as well as the male (64) and female (52) patients separately. Hence, 60 correlation coefficients were calculated for each data group while 95% CIs were estimated for the pairs where statistically significant correlation was found.
3. Results Overall, 53 patients (45.7%) had findings in the bone scintigraphy consistent with DJD. A total of 229 ar-
16
S. Chatziioannou et al. / Association between bone scintigraphy features of spinal degeneration
Table 2 Linear correlation coefficients between the maximum intensity of the findings in Group A (spine only) and Group B (all) data and the patients’ characteristics, i.e., BMI, weight, height, activity level, and age in years. Correlation was studied for all patients (116), female patients only (52), and male patients only (64). The cells highlighted in gray represent statistically significant linear correlations and the corresponding 95% CIs are listed in brackets Group
#
Pair of data correlated
A (spine)
1 2 3 4 5
Max intensity vs. BMI Max intensity vs. Weight Max intensity vs. Height Max intensity vs. Activity level Max intensity vs. Age
All patients (116) 0.09 0.10 0.04 0.07 0.22 [0.04,0.39]
B (all)
6 7 8 9 10
Max intensity vs. BMI Max intensity vs. Weight Max intensity vs. Height Max intensity vs. Activity level Max intensity vs. Age
0.11 0.14 0.08 0.08 0.20 [0.02,0.37]
Linear correlation coefficients Female patients Male patients (52) (64) 0.24 −0.04 0.24 −0.03 0.00 −0.03 0.05 0.10 0.14 0.26 [0.01,0.48] 0.24 0.20 −0.09 0.07 0.14
−0.01 0.04 0.06 0.11 0.28 [0.03,0.49]
Table 3 Linear correlation coefficients between the number (Nb) of findings in Group A (spine only) and Group B (all) data and the patients’ characteristics, i.e., BMI, weight, height, activity level, and age in years. Correlation was studied for all patients (116), female patients only (52), and male patients only (64). The cells highlighted in gray represent statistically significant linear correlations and the corresponding 95% CIs are listed in brackets Group
#
Pair of data correlated
A (spine)
1 2 3 4 5
Nb of findings vs. BMI Nb of findings vs. Weight Nb of findings vs. Height Nb of findings vs. Activity level Nb of findings vs. Age
All patients (116) 0.12 0.12 0.03 0.33 [0.15,0.48] 0.25 [0.07,0.41]
B (all)
6 7 8 9 10
Nb of findings vs. BMI Nb of findings vs. Weight Nb of findings vs. Height Nb of findings vs. Activity level Nb of findings vs. Age
0.13 0.15 0.06 0.33 [0.15,0.48] 0.25 [0.07,0.41]
eas of increased uptake were identified, the majority of which (78%) were in the spine and particularly the thoracic and lumbar spine, which, combined, accounted for 74% of the findings. Table 2 presents the Pearson’s correlation coefficients between maximum intensity of a finding and patients’ characteristics for both Groups A and B as a function of patient population. The gray highlighted cells show the pairs that yielded statistically significant linear correlation. In general, no strong correlations were identified, namely all were below 0.5. However, the sample size was sufficiently large to suggest that there is statistically significant correlation between the maximum intensity of a finding and the patients’ age. A detailed review of the data further suggested that it is the male patients and the findings for DJD of the spine in particular that cause this correlation. The correlation results between the number of findings in the spine (Group A data) and all joints (Group
Linear correlation coefficients Female patients Male patients (52) (64) 0.20 0.06 0.21 0.00 0.03 −0.14 0.06 0.50 [0.28,0.67] 0.03 0.36 [0.13,0.56] 0.23 0.22 −0.02 0.09 0.07
0.07 0.03 −0.11 0.48 [0.26,0.65] 0.38 [0.15,0.58]
B data) and the patients’ characteristics are presented in Table 3. The gray highlighted cells show the pairs that yielded statistically significant linear correlation. Significant correlations were again identified between the number of findings and the patients’ age. This time, however, a second significant correlation was observed between the number of findings and the activity level of the patients. As previously, it is the male patient population that primarily accounts for this correlation outcome. The Somers’ DXY coefficient was estimated to evaluate the association between maximum intensity of scintigraphy findings and pain. For the analysis, the ordered independent variable was the maximum intensity (ranging from 0 to 3) and the binary dependent variable was pain (coded 0 for no pain and 1 for pain). Considering maximum intensity as the predictor or independent variable X and pain as the outcome variable Y, Somers’ DXY was estimated to be 0.46 suggesting
S. Chatziioannou et al. / Association between bone scintigraphy features of spinal degeneration
that maximum intensity could predict pain in 46% of the cases; note that Somers’ coefficient’s values range from −1 to +1. This association is statistically significant based on a one-tailed t-test.
4. Discussion and conclusion In the present study, bone scintigraphy was used to identify active DJD as opposed to traditional anatomic imaging with radiography or computed tomography (CT), the findings of which are often due to old and inactive disease. Bone scintigraphy has the potential to detect bone changes earlier than radiography [20] and demonstrates the metabolically active osseous or joint changes, which are most likely responsible for the development of pain [19]. The results of the study led to the following conclusions: i) Statistically significant correlation exists between age and DJD of the spine or, generally, age and joint disease, as indicated by both the maximum intensity and the number of findings analysis in Tables 2 and 3. ii) Statistically significant correlation exists between activity level and DJD of the spine or, generally, activity level and joint disease, as indicated by the number of findings analysis in Table 3. iii) Weak, not statistically significant, correlations were observed between BMI and DJD of the spine or BMI and joint disease. The stronger correlations were observed for the female patients with coefficients in the range of 0.20–0.24 and 95% CIs in the interval [−0.07, 0.48]. iv) Weak, not statistically significant, correlations were observed between weight and DJD of the spine or weight and joint disease. Again, the stronger correlations were observed for the female patients with coefficients in the range of 0.20–0.24 and 95% CIs in the interval [−0.07, 0.47]. v) Correlation results differed between men and women suggesting that the two groups should be evaluated separately. Back problems, and possibly other skeletal problems, in men were correlated with the type of their activity, increasing linearly with age both in terms of number and intensity. Weight or BMI did not seem to be a risk factor for degenerative spine disease in men. In contrast, female patients did not present
17
any statistically significant correlations at the P = 0.01 or P = 0.05 levels. The results for the female patients, however, suggested that a linear correlation may exist between BMI and maximum intensity or between BMI and number of findings either in the spine or all joint locations. A similar outcome was observed for the weight. vi) Statistically significant association was found between maximum intensity of scintigraphy findings and pain. Results suggested that pain may be possible to predict in approximately 50% of the patients based on the intensity of the scintigraphy findings. However, the current study data did not allow the definition of an intensity “threshold” that could be used to predict painful versus non-painful degenerative disease. The studied population was sufficiently large in size to allow us to reach statistically significant conclusions in some associations despite the relatively small correlation coefficients (see Table 3). The unevenness of the CIs with the lower limit being very close to zero in some comparisons (see 3rd and 4th conclusions above) suggests that a larger patient population may have led to statistically significant correlations between, for example, weight or BMI and DJD of the spine in women. The strength of the association, however, would still be relatively weak. Analysis of the data for nonlinear associations was negative, namely there were no nonlinear relationships between the various pairs of data with the Spearman coefficient remaining below 0.13 in all comparisons. This study included a group of patients that underwent bone scintigraphy in a tertiary care center for a wide range of indications. Consequently, the study population does not represent well the general population and one could argue that it introduces a selection bias that could be in favor or against the presence of DJD. Degenerative bone disease, however, affects all patients independent of whether they have other underlying pathology or not, and so the results of the study, generally or conditionally applicable, hold. In conclusion, molecular imaging findings suggest that level of activity and age are strong risk factors for the development of DJD in men in agreement with a recent review [25]. The age/DJD correlation in this group may be due to a combination of normal aging changes in the body and changes caused by the years the patients were engaged to a certain activity. The distinction between the two causes was not possible in this study. It should be noted, however, that this dependence was clearly not observed for the female patients.
18
S. Chatziioannou et al. / Association between bone scintigraphy features of spinal degeneration
Weight and BMI seem to be stronger DJD factors in women than men. The latter group showed negligible correlation while female patients showed linear correlation coefficients that were approaching statistical significance. Hence, sex appeared to be a critical factor in identifying the causes of DJD of the spine. Certainly, bone scintigraphy is not used routinely in every day practice and is not the method of choice for the identification of degenerative bone changes. Other imaging modalities, such as plain radiographs and CT, are more widely used for the demonstration of degenerative changes and studies based on those modalities lead to different associations. Kalichman et al, for example, observed significantly higher prevalence of facet joint osteoarthritis in the CT scans of obese patients [26]. Bone scintigraphy, however, as a molecular imaging technique, has demonstrated the ability to identify the active and, consequently, painful degenerative changes and, thus, the changes that lead to symptomatology that may require further treatment [20,21]. Therefore, bone scintigraphy may better delineate the presence of painful and clinically relevant degenerative spine changes. Molecular findings seem also to suggest that male and female patients merit a different evaluation for spinal degeneration and low back pain, an aspect worth pursuing further to avoid misinterpretations and erroneous generalizations.
References [1]
[2]
[3]
[4]
[5] [6]
[7] [8]
Mody GM, Brooks PM. Improving musculoskeletal health: global issues. Best Pract Res Clin Rheumatol 2012; 26(2): 237-249. Haldeman S, Kopansky-Giles D, Hurwitz EL, Hoy D, Mark Erwin W, Dagenais S, Kawchuk G, Strömqvist B, and Walsh N. Advancements in the management of spine disorders. Best Pract Res Clin Rheumatol 2012; 26:263-280. Vignon E, Valat JP, Rossignol M, Avouac B, Rozenberg S, Thoumie P, Avouac J, Nordin M, Hilliguin P. Osteoarthritis of the knee and hip and activity: A systematic international review and synthesis (OASIS). Joint Bone Spine 2006; 73:442455. Lundberg U. Stress responses in low-status jobs and their relationship to health risks: musculoskeletal disorders. Ann N Y Acad Sci 1999; 896:162-172. Seidel H. Selected health risks caused by long-term, wholebody vibration. Am J Ind Med 1993; 23:589-604. Zhang Y, Sun Z, Liu J, Guo X. Advances in susceptibility genetics of intervertebral degenerative disc disease.Int J Biol Sci 2008; 4:283-290. Frymoyer J. Lumbar disk disease: epidemiology. Instr Course Lect 1992; 41:217-223. Battié MC, Videman T, Levalahti E, Gill K, Kaprio J.
[9] [10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23] [24]
[25] [26]
Heritability of low back pain and the role of disc degeneration. Pain 2007; 131:272-280. Ala-Kokko L. Genetic risk factors for lumbar disc disease. Ann Med 2002; 34:42-47. Felson DT, Zhang Y, Hannan M, Naimark A, Weissman B, Aliabadi P, Levy D. Risk factors for incident radiographic knee osteoarthritis in the elderly: The Framingham Study. Arthritis Rheum 1997; 40:728-733. Felson DT, Anderson JJ, Naimark A, Walker AM, Meenan RF. Obesity and knee osteoarthritis. The Framingham Study. Ann Intern Med 1988; 109:18-24. Gelber AC, Hochberg MC, Mead LA, Wang NY, Wigley FM, Klag MJ. Body mass index in young men and the risk of subsequent knee and hip osteoarthritis. Am J Med 1999; 107:542548. Like M, Solovieva S, Lamminen A, Luoma K, Leino-Arjas P, Luukkonen R, and Riihimaki H. Disc degeneration of the lumbar spine in relation to overweight. Int J Obes (Lond). 2005; 29(8):903-908. Sandmark H, Hogstedt C, Lewold S, Vingård E. Osteoarthrosis of the knee in men and women in association with overweight, smoking, and hormone therapy. Ann Rheum Dis 1999; 58:151–155. Coggon D, Reading I, Croft P, McLaren M, Barrett D, Cooper C. Knee osteoarthritis and obesity. Int J Obes Relat Metab Disord. 2001; 25:622-627. Grotle M, Hagen KB, Natvig B, Dahl FA, Kvien TK. Obesity and osteoarthritis in knee, hip and/or hand: an epidemiological study in the general population with 10 years follow-up. BMC Musculoskelet Disord. 2008; 9:132. Davis MA, Neuhaus JM, Ettinger WH, Mueller WH. Body fat distribution and osteoarthritis. Am J Epidemiol. 1990; 132:701-707. Van Tulder MW, Assendelft WJ, Koes BW, Bouter LM. Spinal radiographic findings and nonspecific low back pain: A systematic review of observational studies. Spine. 1997; 22(4): 427-434. Horger M, Bares R. The role of single-photon emission computed tomography/computed tomography in benign and malignant bone disease. Semin Nucl Med. 2006; 36:286-294. Pneumaticos SG, Chatziioannou SN, Hipp JA, Moore WH, Esses SI. Low back pain: Prediction of short-term outcome of facet joint injection with bone scintigraphy. Radiology 2006; 238:693-698. Hutton CW, Higgs ER, Jackson PC, Watt I. 99mTc HMDP bone scanning in generalised nodal osteoarthritis. II. The four hour bone scan image predicts radiographic change. Ann Rheum Dis. 1986; 45:622-626. Whitley E, Ball J. Statistics review 1: presenting and summarising data. Crit Care. 2002; 6(1):66-71. Bewick V, Cheek L, Ball J. Statistics review 7: Correlation and regression. Crit Care. 2003; 7:451-459. Wessa P. Somers Dxy Rank Correlation (v1.0.2) in Free Statistics Software (v1.1.23-r7). Office for Research Development and Education, 2006 [updated 2014 Jan; cited 2014 Feb 20]. Available from: http://www.wessa.net/rwasp_ somers.wasp/. Boqduk N. Degenerative joint disease of the spine. Radiol Clin North Am 2012; 50(4):613-628. Kalichman L, Guermazi A, Li L, Hunter DJ. Association between age, sex, BMI and CT-evaluated spinal degeneration features. J Back Musculoskelet Rehabil. 2009; 22:189-195.
Copyright of Journal of Back & Musculoskeletal Rehabilitation is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
