Back pain, disability, and radiographic vertebral fracture in European women: a prospective study

Transcription

1 Osteoporos Int (2004) 15: DOI /s ORIGINAL ARTICLE Back pain, disability, and radiographic vertebral fracture in European women: a prospective study T.W. O Neill Æ W. Cockerill Æ C. Matthis Æ H.H. Raspe M. Lunt Æ C. Cooper Æ D. Banzer Æ J.B. Cannata M. Naves Æ B. Felsch Æ D. Felsenberg Æ J. Janott O. Johnell Æ J.A. Kanis Æ G. Kragl Æ A. Lopes Vaz G. Lyritis Æ P. Masaryk Æ G. Poor Æ D.M. Reid W. Reisinger Æ C. Scheidt-Nave Æ J.J. Stepan C.J. Todd Æ A.D. Woolf Æ J. Reeve Æ A.J. Silman Received: 27 November 2003 / Accepted: 11 February 2004 / Published online: 12 May 2004 Ó International Osteoporosis Foundation and National Osteoporosis Foundation 2004 Abstract Vertebral fractures are associated with back pain and disability. There are, however, few prospective data looking at back pain and disability following identification of radiographic vertebral fracture. The aim of this analysis was to determine the impact of radiographically identified vertebral fracture on the subsequent occurrence of back pain and disability. Women T.W. O Neill (&) Æ W. Cockerill Æ M. Lunt Æ A.J. Silman ARC Epidemiology Research Unit, University of Manchester, Manchester, M13 9PT, UK Terry@fs1.ser.man.ac.uk C. Matthis Æ H.H. Raspe Institute for Social Medicine, Medical University of Lubeck, Lubeck, Germany C. Cooper MRC Environmental Epidemiology Unit, Southampton General Hospital, Southampton, UK D. Banzer Department of Radiology, Behring Hospital, Berlin, Germany J.B. Cannata Æ M. Naves Asturia General Hospital, Oviedo, Spain B. Felsch Clinic for Internal Medicine, Jena, Germany D. Felsenberg Department of Radiology and Nuclear Medicine, Free University, Berlin, Germany J. Janott Ruhr University, Bochum, Germany O. Johnell Department of Orthopedics, Lund University, Malmo, Sweden J.A. Kanis Centre for Metabolic Bone Disease, Sheffield, UK G. Kragl Medical Academy, Erfurt, Germany A. Lopes Vaz Hospital de San Joao, Oporto, Portugal aged 50 years and over were recruited from population registers in 18 European centers for participation in the European Prospective Osteoporosis Study. Participants completed an interviewer-administered questionnaire which included questions about back pain in the past year and various activities of daily living, and they had lateral spine radiographs performed. Participants in these centers were followed prospectively and had repeat G. Lyritis Laboratory for the Research of Musculoskeletal System, University of Athens, Athens, Greece P. Masaryk Institute of Rheumatic Diseases, Piestany, Slovakia G. Poor National Institute of Rheumatology and Physiotherapy, Budapest, Hungary D.M. Reid Department of Medicine and Therapeutics, University of Aberdeen, Aberdeen, UK W. Reisinger Institute for Diagnostic Radiology, Humboldt University, Berlin, Germany C. Scheidt-Nave Department of General Practice, University of Goettingen, Goettingen, Germany J.J. Stepan Department of Medicine, Charles University, Prague, Czech Republic C.J. Todd School of Nursing, Midwifery and Health Visiting, University of Manchester, Manchester, UK A.D. Woolf Department of Rheumatology, Royal Cornwall Hospital, Truro, UK J. Reeve University Department of Medicine and Institute of Public Health, Cambridge, UK

2 761 spine radiographs performed a mean of 3.7 years later. In addition they completed a questionnaire with the same baseline questions concerning back pain and activities of daily living. The presence of prevalent and incident vertebral fracture was defined using established morphometric criteria. The data were analyzed using logistic regression with back pain or disability (present or absent) at follow-up as the outcome variable with adjustment made for the baseline value of the variable. The study included 2,260 women, mean age 62.2 years. The mean time between baseline and follow-up survey was 5.0 years. Two hundred and forty participants had prevalent fractures at the baseline survey, and 85 developed incident fractures during follow-up. After adjustment for age, center, and the baseline level of disability, compared with those without baseline prevalent fracture, those with a prevalent fracture (odds ratio [OR]=1.4; 95% confidence interval [CI] 1.0 to 2.0) or an incident fracture (OR=1.7; 95% CI, 0.9 to 3.2) were more likely to report disability at follow-up, though the confidence intervals embraced unity. Those with both a prevalent and incident fracture, however, were significantly more likely to report disability at follow-up (OR=3.1; 95% CI, 1.4 to 7.0). After adjustment for age, center, and frequency of back pain at baseline, compared with those without baseline vertebral fracture, those with a prevalent fracture were no more likely to report back pain at follow-up (OR=1.2; 95%CI, 0.8 to 1.7). There was a small increased risk among those with a preexisting fracture who had sustained an incident fracture during follow-up (OR=1.6; 95%CI, 0.6 to 4.1) though the confidence intervals embraced unity. In conclusion, although there was no significant increase in the level of back pain an average of 5 years following identification of radiographic vertebral fracture, women who suffered a further fracture during follow-up experienced substantial levels of disability with impairment in key physical functions of independent living. Keywords Back pain Æ Disability Æ Prospective study Æ Vertebral fracture Introduction Radiographically identified vertebral fractures are associated with back pain and disability, with the strength of these associations increasing with the number and severity of fractures [1, 2, 3, 4, 5, 6, 7]. There are, however, few data from prospective population-based studies which look at change in back pain and disability following identification of radiographic vertebral fracture. Clinic-based studies suggest that functional impairment following clinically apparent vertebral fracture persists though pain decreases [8, 9]. The majority of vertebral fractures do not, however, come to clinical attention. Data from a recent population survey suggest no increase in either back pain or functional impairment following identification of radiographic vertebral fracture during a follow-up of 3.7 years unless a further fracture had occurred [10]. We used data from the European Prospective Osteoporosis Study to determine whether the presence of a radiographically identified vertebral fracture at baseline influences the occurrence of back pain and disability during follow-up. We also looked at the influence on these outcomes of a further fracture occurring during the follow-up period. Methods The participants included in this analysis were recruited during the course of a multicenter population based survey of vertebral osteoporosis the European Prospective Osteoporosis Study [EPOS]. The detailed methods of this study are reported elsewhere [11]. In brief, participants were recruited from population-based registers in 18 centers. Stratified random sampling was used with the aim of recruiting in each center, a target number of 50 participants in each of six, 5-year-age and sex bands, 50 54, 55 59, 60 64, 65 69, 70 74, and 75 years and over. Participants were invited to attend for an interviewer-administered lifestyle questionnaire and lateral radiographs of the thoracic and lumbar spine. The questionnaire included questions about back pain in the past year (response set = yes/no). Also a 12- item, back-specific, activities-of-daily-living (ADL) instrument (response set for each item = can do without difficulty/can do with some difficulty/can t do or only with help; see Appendix ) [12]. Participants were invited to attend for a repeat spinal radiograph an average of 3.7 years after the baseline examination. Both baseline and follow-up spinal radiographs were evaluated morphometrically. Prevalent vertebral fracture was defined morphometrically at baseline according to the McCloskey-Kanis method [13]. A participant was classified as having an incident vertebral fracture if there was evidence of a McCloskey- Kanis fracture in the second radiograph and a reduction of at least 20% or more in at least one vertebral height, with the reduction in height being at least 4 mm [14]. The back pain and ADL questions were readministered by postal questionnaire. In the 11 non-german centers the questionnaire was readministered on a single occasion. In the 7 German centers a questionnaire was administered 6-monthly for a period of 3 years. Participants gave informed consent to their participation in the manner required by their centers research ethics committees. Analysis Level of disability was calculated from the activitiesof-daily-living questionnaire. The trichotomous reply format (can do without difficulty / can do with some difficulty / can t do or only with help) resulted in a score

3 762 between 0% (12 unable to do ) and 100% (12 can do ). A functional capacity lower than or equal to 70% was defined as indicative of disability [12]. The responses from the final follow-up of the 7 German centers were combined with the follow-up of the 11 non-german centers. The German follow-up included a question about back pain in the past 6 months, while for the non-german centers it was back pain in the past year. To create an equivalent response, Germans who reported back pain in the past 6 months at either of the last two 6-month follow-ups were classified as reporting back pain in the past year. Logistic regression was used to explore the association between the outcome variables, back pain in the past year and disability, and vertebral fracture, with the results expressed as odds ratios and 95% confidence intervals. The vertebral fracture status of each subject was characterized by a single explanatory variable with four categories: (1) no fracture, (2) prevalent fracture at baseline only, (3) incident fracture only, and (4) both baseline prevalent and incident fractures. Statistical adjustments were made for age, center, and the baseline value of the outcome variable. Because of the small number of men with incident vertebral fractures in this study, data are presented for women only. Results Participants In total, 2,260 women, mean age 62.2 years (SD=7.7), responded to the follow-up questionnaire and had paired radiographs available for analysis. Of these, 240 had evidence of one or more vertebral fractures at baseline. Fifty-three (2.4%) women without a prior fracture had evidence of a new (incident fracture) on the repeat spinal radiograph. Two hundred and eight (9.2%) had evidence of a baseline fracture only, while 32 (1.4%) had evidence of both prevalent and incident fractures. The mean time between the baseline and follow-up survey was 5.0 years (SD=1.5), while the mean time between the repeat radiograph and follow-up survey was 1.3 years (SD=1.4). Table 1 Frequency of back pain in past year and disability by baseline vertebral fracture status No fracture (n=2020) Back pain in past year a (%) Baseline Follow-up 69.9 d 72.5 d Disability b (%) Baseline c Follow-up 29.6 d 42.0 c,d a Yes vs no b 70% vs >70% functional capacity c Prevalent fracture vs no prevalent fracture, p<0.05 d Baseline vs follow-up, p<0.05 Prevalent fracture (n=240) Long-term impact of prevalent vertebral fracture The associations between back pain and disability at follow-up and vertebral fracture are shown in Table 2. After adjustment for age, center, and the baseline level of function, compared with those without baseline prevalent fracture, those with a prevalent fracture (OR=1.4; 95% CI, 1.0 to 2.0) or incident fracture (OR=1.7; 95% CI, 0.9 to 3.2) were slightly more likely to report disability at follow-up, though the confidence intervals embraced unity. The strongest association was for those with a preexisting fracture who subsequently sustained an incident fracture (OR=3.1; 95% CI, 1.4 to 7.0). The strength of the association remained unaltered after adjustment for back pain in the past year at both follow-up and baseline (OR=3.3; 95% CI, 1.4 to 8.1). After adjustment for age, center, and back pain at baseline, compared with those without baseline vertebral fracture, those with a prevalent fracture were no more likely to report back pain at follow-up (OR=1.2; 95% CI, 0.8 to 1.7). There was a small increased risk among those with a preexisting fracture who had sustained an incident fracture during follow-up (OR=1.6; 95% CI, 0.6 to 4.1) although again the confidence intervals embraced unity. The association between individual activities of daily living and vertebral fracture are shown in Table 3. In this analysis, impairment in an individual activity is taken as those with any difficulty performing it. As with Frequency of self-reported back pain and disability The frequency of back pain and disability (as defined by the functional ability questions) at both baseline and follow-up for women with and without a prevalent vertebral fracture at baseline is shown in Table 1. Compared with the baseline survey, a significantly greater proportion of participants reported back pain and disability at follow-up. This was true for both those with and without fracture. Compared with those without a prevalent fracture, a significantly greater proportion of those with prevalent vertebral fracture reported disability at both baseline and follow-up (Table 1). Table 2 Association between vertebral fracture, back pain, and disability. OR = odds ratio, CI = confidence interval Fracture status Back pain past year a OR (95% CI) c Disability b OR (95% CI) c None (n=2,020) Referent Referent Prevalent only (n=208) 1.2 ( ) 1.4 ( ) Incident only (n=53) 0.9 ( ) 1.7 ( ) Prevalent and incident (n=32) 1.6 ( ) 3.1 ( ) a Yes vs no b 70% vs >70% functional capacity c Adjustments made for age, center, and baseline value of outcome variable

4 763 Table 3 Association between individual activities and vertebral fracture. OR = odds ratio, CI = confidence interval Activity Fracture status Prevalent OR (95% CI) a Incident OR (95% CI) a Prevalent and incident OR (95% CI) a Reach a book from a high shelf? 1.6 ( ) 1.4 ( ) 2.5 ( ) Lift a heavy object of at least 10 kilos? 1.7 ( ) 1.5 ( ) 3.6 ( ) Wash and dry yourself all over? 1.1 ( ) 0.9 ( ) 2.4 ( ) Bend forward and pick up a light object from the floor? 1.3 ( ) 1.2 ( ) 2.3 ( ) Wash your hair over a basin? 1.4 ( ) 1.5 ( ) 2.1 ( ) Sit for 1 hour on a hard chair? 1.4 ( ) 0.9 ( ) 2.7 ( ) Stand continuously for 30 minutes in a queue? 1.4 ( ) 1.2 ( ) 2.0 ( ) Raise yourself in bed from a lying position? 1.3 ( ) 1.0 ( ) 1.7 ( ) Take your socks off your feet? 1.3 ( ) 1.3 ( ) 1.7 ( ) Bend from seated, and pick up a small object? 1.3 ( ) 0.7 ( ) 1.9 ( ) Lift box containing 6 litre bottles onto a table? 1.7 ( ) 1.3 ( ) 1.5 ( ) Run 100 metres fast without stopping? 1.4 ( ) 1.1 ( ) 2.4 ( ) a Adjustments made for age, center, and baseline value of outcome variable the results using the summary disability variable, the strength of the associations were more marked for those women with an incident and a prevalent fracture than for either incident or prevalent fractures alone. We looked also at the influence of number of fractures at baseline on subsequent outcome. Because of the effect of incident fractures on outcome we restricted this analysis to those without incident fractures. After adjusting for baseline level of disability, age, and center, compared with those without prevalent fracture there was a small increase in the risk of disability at follow-up among those with a single fracture (OR=1.3; 95% CI, 0.8 to 1.9) and a slightly higher risk among those with two or more fractures (OR=1.7; 95% CI, 0.9 to 3.1), though the confidence intervals around both estimates embraced unity. After adjusting for baseline level of back pain, age, and center, there was no increase in the risk of reporting back pain at follow-up with increasing number of baseline fractures. Discussion In this population survey there was no increase in reported back pain or disability an average of 5 years following identification of radiographic vertebral fracture unless a further fracture occurred, in which case there was a significant increase in self-reported disability. The result emphasizes the importance of further fractures for the development of disability with osteoporosis. Our study was prospective with information concerning the outcome variables back pain and disability obtained at both baseline and follow-up. There are, however, several limitations to be considered in interpreting the results. The follow-up questionnaire was completed a mean of 1.3 years following the repeat radiograph. It is possible that during this time a small number of new vertebral fractures may have occurred in the referent group (those without fracture). The effect of any such misclassification of fracture status, however, would be to tend to reduce the chance of finding significant associations with the outcome variables. Back pain is common among middle aged and elderly women, and for the majority of individuals it is due to other causes including disk degeneration and osteoarthritis of the facet joints. The effect of the relatively high background frequency in the general population would be to mask the impact linked with vertebral fracture in this setting. Back pain and disability related to vertebral fracture may be transient. Our study which looked at these outcomes on two occasions may have missed transient episodes particularly among those with recent fractures and is likely therefore to have underestimated the health burden related to these events. It was not possible to date the onset of the vertebral fracture identified at the baseline survey, which may have occurred many years previously. Our results should not be therefore interpreted as changes in back pain and disability following the actual fracture event, but rather to changes following their identification on screening radiographs. Finally, our results refer to European women only and may not apply to other ethnic groups or to men. In a clinic-based study, perception of average pain was higher, and there was a lower degree of well-being in women with recent vertebral fractures (<2 years) compared with those with older fractures (>2 years); however, limitations in everyday life were similar in the two groups, suggesting that these persisted following fracture [8]. In a separate clinic-based study, 100 women with vertebral fracture were asked to estimate the date of onset of their fracture. Both quality of life (as assessed using SF-36) and function (assessed using the modified Barthel index and the timed get-up-and-go test) were impaired in those with fracture compared with a group of controls without fracture. Among those with fracture there was no association between quality of life or functional ability and time since fracture (mean 5.1 years), suggesting that these adverse outcomes persisted [9].

5 764 The majority of fractures do not come to clinical attention, and there are few data from population studies concerning the longer term impact of radiographically identified vertebral fractures which include those who do and do not come to clinical attention. In the Study of Osteoporotic Fractures (SOF), during a mean follow-up of 3.7 years, prevalent radiographic vertebral fractures were not associated with a subsequent increase in back pain (OR=1.1) or with back disability (OR=1.2) during follow-up unless a subsequent incident fracture had occurred [10]. Among women who participated in the Hawaii osteoporosis study, the presence of a recent fracture (<4.3 years ago), though not older fractures, predicted current impairment of function and back pain [5, 6]. Prevalent fractures at the baseline examination (about 10 years previously) were not linked with impaired function or back pain at follow-up. Our results confirm the findings from SOF suggesting no increase in back pain or disability following identification of a prevalent vertebral fracture if no new fractures occur during follow-up. They confirm also the finding that the risk of disability is increased if a further fracture occurs during the follow-up period [10]. In contrast, however, to the data from SOF, we observed no significant association between recent (incident only) fractures and back pain. Unlike SOF which asked about the frequency and severity of back pain on an annual basis, we asked only about the occurrence of back pain and, in the majority of centers, at baseline and followup. The discrepancy suggests the possibility that to identify increased back pain following a recent fracture a more sensitive question and/or more frequent surveillance may be required. Our results confirm the significant impact of radiographic vertebral fracture on outcome. Many women, however, with radiographic vertebral fracture do not come to clinical attention [15]. The proportion who do come to attention may be lower still in Europe than in the United States, judging from the contrast between the rates reported for radiographic and clinical fractures, respectively, in Britain [16, 17]. Such women would clearly benefit from treatment to reduce their risk of further fracture and consequent attendant morbidity. Population-wide strategies to identify these women are urgently required. In summary, we observed no significant increase in the frequency of back pain or disability at follow-up an average of 5 years after identification of radiographic vertebral fracture unless a further fracture occurs, in which case the risk of disability is significantly increased. Acknowledgements The study was financially supported by a European Union concerted action grant under Biomed-1 (BMH1CT920182), and also EU grants C1PDCT925102, ERBC1PDCT & The central coordination was also supported by the World Health Organization, the European Foundation for Osteoporosis and Bone Disease, the Medical Research Council (G ), and the UK Arthritis Research Campaign. The EU s PECO program linked to Biomed-1 funded in part the participation of the Budapest, Prague, and Piestany centers. The central X-ray evaluation was generously sponsored by the Bundesministerium fur Forschung and Technologie, Germany. Individual participating centers acknowledge the receipt of locally acquired support for their data collection. We would like to thank the following individuals: Czech Republic: Prague, M. Linduskova; Germany: Berlin Steglitz, I. Keller-Janker, B. Rothenburg; Berlin Postdam, C. Popovici;Bochum, M. Bohle, S. Hering, A. Pfeiffer, A. Weber, V. WieBe, H. Seelbach; Erfurt, M. Angrick, C. Dodenhof; Heidelberg, G. Leidig-Bruckner, B. Limberg; Jena, G. Lehmann, I. Marzoll; Lubeck, A. Raspe, E. Taubert; Greece: Athens, M. Katsiri, P. Papangelopoulou, G. Petta, P. Raptou; Poland: Szczecin, R. Celibala, E. Gromniak, A. Krzysztalowski, K. Napierata, J. Ogonowski; Portugal: Oporto, I. Brito, J. Brito, C. Maia, C. Vaz; Slovakia: Piestany, E. Brisudova, T. Hornakova, E. Martancikova, J. Tomkuljakova; Spain: Oviedo, C. Gomez Alonso, J.B. Diaz Lopez, A. Rodriguez Rebollar; Sweden: Malmo, A. Rafsted; United Kingdom: Aberdeen, R. Smith; Cambridge, A. Martin; Harrow, A. Nicholls, C. Oxbrough, L. Peter, O. Waldron, J. Walton, K. Walton; Truro, A. Deodhar, J. Parsons Appendix: ADL Questionnaire Can you reach for example a book from a high shelf or cupboard? Can you lift a heavy object of at least 10 kilo (e.g., a full suitcase) and carry it for 10 metres? Can you wash and dry yourself all over? Can you bend forward to pick up a small lightweight object from the floor? Can you wash your hair over a washbasin? Can you sit for 1 hour on a hard chair? Can you stand continuously for 30 minutes (for example in a queue)? Can you raise yourself in bed from a lying position? Can you take your socks or similar garments on and off your feet? Can you bend down from a seated position and pick up a small object at the side of your chair? Can you lift a box containing 6-litre bottles of liquid onto a table? Can you run 100 metres fast without stopping in order that you can catch a bus? Response set for each item Can do without difficulty / Can do with some difficulty / Can t do, or only with help References 1. Burger H, Van Daele PLA, Grashuis K, Hofman A, Grobbee DE, Schutte HE et al (1997) Vertebral deformities and functional impairment in men and women. J Bone Miner Res 12: Ettinger B, Black DM, Nevitt MC, Rundle AC, Cauley JA, Cummings SR et al (1992) Contribution of vertebral deformities to chronic back pain and disability. J Bone Miner Res 7(4): Leidig G, Minne HW, Sauer P, Wuster C, Wuster J, Lojen M et al (1990) A study of complaints and their relation to vertebral destruction in patients with osteoporosis. Bone Miner 8:

6 Lyles KW, Gold DT, Shipp KM, Pieper CF, Martinez S, Mulhausen PL (1993) Association of osteoporotic vertebral compression fractures with impaired functional status. Am J Med 94: Huang C, Ross PD, Wasnich RD (1996) Vertebral fracture and other predictors of physical impairment and health care utilization. Arch Intern Med 156: Huang C, Ross PD, Wasnich RD (1996) Vertebral fractures and other predictors of back pain among older women. J Bone Miner Res 11: Matthis C, Weber U, O Neill TW, Raspe H and the European Vertebral Osteoporosis Study Group (1998) Health impact associated with vertebral deformities: results from the European Vertebral Osteoporosis Study (EVOS). Osteoporos Int 8: Begerow B, Pfeifer M, Pospeschill M, Scholz M, Schlotthauer T, Lazarescu A et al (1999) Time since vertebral fracture: an important variable concerning quality of life in patients with postmenopausal osteoporosis. Osteoporos Int 10: Hall SE, Criddle RA, Comito TL, Prince RL (1999) A casecontrol study of quality of life and functional impairment in women with long-standing vertebral osteoporotic fracture. Osteoporos Int 9: Nevitt MC, Ettinger B, Black DM, Stone K, Jamal SA, Ensrud K et al (1998) The association of radiographically detected vertebral fractures with back pain and function: a prospective study. Ann Intern Med 128: O Neill TW, Felsenberg D, Varlow J, Cooper C, Kanis JA, Silman AJ, and the European Vertebral Osteoporosis Study Group (1996) The prevalence of vertebral deformity in European men and women: the European Vertebral Osteoporosis Study. J Bone Miner Res 11: Kohlmann T, Raspe H (1996) Der Funktionsfragebogen Hannover zur alltagsnahen Diagnostik der Funktionsbeeintrachigung durch Ruckenschmerzen (FFbH-R). Rehabilitation 35:I VIII 13. McCloskey EV, Spector TD, Eyres KS, Fern ED, O Rourke N, Vasikaran S, Kanis JA (1993) The assessment of vertebral deformity: a method for use in population studies and clinical trials. Osteoporos Int 3: Lunt M, Ismail AA, Felsenberg D, Cooper C, Kanis JA, Reeve J et al (2002) Defining incident vertebral deformities in population studies: a comparison of mophometric criteria. Osteoporos Int 13: Cooper C, Atkinson EJ, O Fallon WM, Melton LJ III (1992) Incidence of clinically diagnosed vertebral fractures: a population based study in Rochester, Minnesota, J Bone Miner Res 7: van Staa TP, Dennison EM, Leufkens HGM, Cooper C (2001) Epidemiology of fractures in England and Wales. Bone 29: The European Prospective Osteoporosis Study Group (2002) Incidence of vertebral fracture in Europe: results from the European Prospective Osteoporosis Study (EPOS). J Bone Miner Res 17: