The authors have no conflict of interest
Prevalence of Paget's Disease of Bone in Italy†
Version of Record online: 31 MAY 2005
Copyright © 2005 ASBMR
Journal of Bone and Mineral Research
Volume 20, Issue 10, pages 1845–1850, October 2005
How to Cite
Gennari, L., Di Stefano, M., Merlotti, D., Giordano, N., Martini, G., Tamone, C., Zatteri, R., De Lucchi, R., Baldi, C., Vattimo, A., Capoccia, S., Burroni, L., Geraci, S., De Paola, V., Calabrò, A., Avanzati, A., Isaia, G. and Nuti, R. (2005), Prevalence of Paget's Disease of Bone in Italy. J Bone Miner Res, 20: 1845–1850. doi: 10.1359/JBMR.050518
- Issue online: 4 DEC 2009
- Version of Record online: 31 MAY 2005
- Manuscript Accepted: 27 MAY 2005
- Manuscript Revised: 3 MAY 2005
- Manuscript Received: 4 FEB 2005
- Paget's disease of bone;
- alkaline phosphatase;
- bone scintigraphy;
- pelvic X-ray
We examined the prevalence of PDB in Italy from radiological, scintigraphic, and biochemical surveys in two Italian towns. Prevalence rates varied from 0.7% to 2.4%, were higher in males than in females, and slightly differed between the two towns. Unlike previous studies in populations of British descent, no secular trend for a decreasing prevalence emerged.
Introduction: Clinical, radiological, and necropsy data from different countries suggested pronounced geographical variations in the prevalence of Paget's disease of bone (PDB). Despite the impact of the disease on the population, there are limited data on the prevalence of PDB in Italy.
Materials and Methods: The objective of this study was to estimate the prevalence of PDB in the district of Siena (Central Italy) and Turin (Northern Italy) from radiological, biochemical, and scintigraphic surveys. We examined a sample of 1778 consecutive pelvic radiographs performed between 1999 and 2000 at the Hospital Radiology Unit in Siena and 6609 pelvic radiographs performed in 1986–1987, 1992–1993, and 1999–2002 from the Radiology Department of Molinette Hospital in Turin. In Siena, 7906 consecutive99mTC-MDP bone scans performed over a 4-year period (January 2000 to May 2004) were also screened for the presence of PDB, and the prevalence of elevated alkaline phosphatase (ALP) levels (>300 UI/liter) was estimated from 7449 computerized medical records over a 3-year period (January 2000 to February 2003). The finding of PDB on the pelvic radiograph and bone scan was based on standardized radiological criteria.
Results: At the end of the radiological surveys, 16/1778 pelvic PDB cases (8 males and 8 females) were observed in Siena and 41/6609 (27 males and 14 females) in Turin. The crude prevalence of the disease was 0.89% in Siena and 0.62% in Turin. Given that pelvic involvement is commonly described in 60–90% of PDB patients, the estimated overall prevalence of PDB ranged from 1.0% to 1.5% in Siena and from 0.7% to 1.0% in Turin. No decrease in the prevalence of PDB was evident after comparison of prevalence rates from different periods. Biochemical analyses showed 296/7449 subjects with elevated ALP levels and normal liver enzymes, 87 of whom had confirmed diagnosis of PDB. The estimated prevalence of biochemical PDB was 1.5%. The scintigraphic survey showed a PDB prevalence of 194/7906 (2.4%), which was significantly higher than the radiological and biochemical estimates.
Conclusions: Our surveys suggest that PDB in Italy has an estimated prevalence of at least 1%, comparable with that observed in United States and other European countries, but lower than that described in Britain and New Zealand. No secular trend for a decreasing prevalence of PDB was observed.
PAGET's DISEASE OF bone (PDB) is a chronic focal bone remodeling disorder that typically results in enlarged and deformed bones in one or more regions of the skeleton. (1, 2) This disease is most often asymptomatic but can cause a variety of medical complications, resulting in considerable morbidity and reduced quality of life. (3) PDB is most common in white people of European descent, but it also occurs in blacks, whereas it is rare in people of Asian descent. (4) Clinical, radiological, and necropsy data from different European countries has suggested pronounced geographical variations in the prevalence of the disease, ranging from 3–5% in Britain to 0.4% in Scandinavian countries. (5) Moreover, recent studies in different populations suggested a decrease in prevalence and clinical severity of PDB. (6–10) Despite the impact of the disease on the population, there are limited data about the epidemiology and the true prevalence of PDB in Italy. The objective of this study was to estimate the prevalence of PDB in two Italian towns: the district of Siena, from radiological, biochemical, and scintigraphic surveys, and the city of Turin, from pelvic radiological surveys of different decades. Siena is a town of central Italy, located in a rural region that remained fairly isolated until recent times, with a low immigration flow. In contrast, Turin, which is located in northwest Italy, recognized a strong immigration flow from other Italian regions in the 1960s (mainly from southern and northeastern, but also from central Italian regions); therefore, only about 50% of the actual population was born in this district, and the Turin population could be considered sufficiently representative of the whole Italian population.
MATERIALS AND METHODS
Radiographic survey in Turin
The radiographic sample consisted of a total amount of 6609 pelvic radiographs. We evaluated all the subjects ≥60 years of age who had a pelvic X-ray performed in the years 1986–1987, 1992–1993, and 1999–2002 at the Radiology Department of Molinette Hospital in Turin. The selected X-ray films, taken from stored films within the Radiology Department, showed the entire pelvis and sacrum, the lumbar vertebrae, and the femoral heads, sites that are commonly affected by PDB. Radiographs were sequentially mounted on multiview X-ray view boxes and reviewed for the presence of PDB. Readings of the radiographs were performed simultaneously by two readers (RZ and RDL) over a 6-day period. All positive X-rays were reviewed a second time by both readers before removal of the films from the view boxes. A further review was conducted at a later date by the same readers. Discrepancies were discussed at the end of each reading session, and an adjudicated reading was recorded. The finding of PDB on the pelvic radiograph was based on the following standardized radiographic criteria(11): (1) expansion of bone size; (2) thickened, disorganized trabeculae, (3) thickened, expanded cortex; (4) osteosclerosis; and (5) deformity.
The birthplace of all PDB patients living in Turin was recorded; patients from other Italian regions moved in Turin mainly in the 1960s, and all of them were in Turin before starting the survey in 1986. Between- and within-observer repeatability of radiographic grading was assessed using the κ statistic.
Radiographic survey in Siena
The radiographic sample consisted of 1778 consecutive pelvic radiographs performed between 1999 and 2000 at the Hospital Radiology Unit of the University of Siena. The selected X-ray films met the same criteria of that performed in Turin, showing the entire pelvis and sacrum, the lumbar vertebrae, and the femoral heads. Patients with previously known PDB were excluded from analysis. Radiographs were sequentially mounted on multiview X-ray view boxes and reviewed for the presence of PDB simultaneously by two readers (CB and SC) over a 6-day period. All positive X-rays were reviewed a second time, and a further review was conducted at a later date by the same readers. Discrepancies were discussed at the end of each reading session, and an adjudicated reading was recorded. The diagnosis of PDB on the pelvic radiograph was based on the same standardized radiographic criteria used in Turin. (11) Between- and within-observer repeatability of radiographic grading was assessed using the κ statistic. In addition, the presence of vertebral deformities and of osteoarthritis at the lumbar vertebrae, at the hip, and at the sacroiliac joints was recorded for each patient.
To estimate the skeletal distribution of pagetic lesions in the Italian population, we analyzed clinical and radiological data from 120 consecutive PDB cases in Siena that attended our outpatient clinic. Their mean age was 69 ± 12 years, with an age range of 46–89 years.
Biochemical survey in Siena
The prevalence of elevated plasma alkaline phosphatase activity (ALP; normal range, 106–298 UI/liter) in subjects ≥40 years of age was estimated from consecutive computerized records of our laboratory over a 3-year period (January 2000 to February 2003). Samples were considered only if measurements of liver enzymes aspartate aminotransferase, alanine aminotransferase, and γ-glutamyltranspeptidase had been simultaneously assessed. Biochemical diagnosis of PDB was made if ALP was >300 UI/liter, and liver enzymes were in the normal range. A total of 7449 samples were valuable. A group (n = 1180) of these patients subsequently underwent biochemical screening of markers of bone turnover, namely serum bone ALP (BALP; Alkphase-B; Metra Biosystem, Mountain View, CA, USA), as an index of bone formation, and serum C-terminal telopeptides of type I collagen (Serum CrossLaps; Osteometer, Herlev, Denmark), as an index of bone resorption.
Scintigraphic survey in Siena
The evaluation of scintigraphic prevalence of PDB in Siena was performed by the analysis of 7906 consecutive99mTC-MDP bone scans performed over a 4-year period (January 2000 to May 2004) at the Nuclear Medicine Unit of the Department of Radiology of our university. Readings of the bone scans were performed simultaneously by two physicians (AV and LB). Discrepancies were discussed at the end of each reading session, and an adjudicated reading was recorded. No subject was included more than once. The diagnosis of PDB on the bone scan was based on the following standardized criteria(12, 13): (1) areas of intensively increased uptake, often showing a “V” or “flame-shaped” leading edge; (2) ring of increased activity in the margins of the skull; (3) deformity and expansion of bone size; and (4) tracer accumulation throughout one or more vertebrae, affecting the body and posterior elements, including the spinous and transverse processes. A repeatability study was performed on 30 pagetic and 30 nonpagetic bone scans on two occasions separated by 1 month to assess the within- and between-observer variation. Repeatability was calculated using the κ statistic.
Radiological prevalence of PDB in Turin
Overall, 41 (27 males and 14 females) of the 6609 pelvic radiographs showed PDB. Their mean age was 76 ± 8.8 years, with an age range of 60–94 years. The distribution and localization of PDB on the radiographs is summarized in Table 1. Pelvic PDB was present in 33 (80.5%) radiographs, and unilateral involvement of the pelvis was present in 21 (51.2%) patients. Of the eight patients without pelvic involvement, three showed lumbar vertebrae PDB and five had PDB of the proximal right (three patients) or left (two patients) femur. All radiographic lesions were predominantly sclerotic in nature. Concordance between observers was 0.99%, for a κ statistic of 0.93. The within-observer repeatability estimated on 30 affected and 30 nonaffected radiographs was 0.96%, with a κ value of 0.93%.
The overall prevalence of pelvic PDB was 41 of 6609 (0.62%). Given that, in previous studies, the pelvic involvement is commonly described in 60–90% of PDB patients, the estimated overall prevalence of PDB in Turin population ranged from 0.69% to 1.03%.
We divided subjects according to age and date of the pelvic X-ray examination; the results are summarized in Table 2. The crude prevalence of the disease differed according to year of examination, being 10 of 2305 (0.43%), 7/951 (0.74%) males, and 3/1354 (0.22%) females for 1986–1987; 8 of 1101 (0.73%), 5/409 (1.22%) males, and 3/692 (0.43%) females for 1992–1993; 12 of 1984 (0.60%), 8/798 (1%) males, and 4/1186 (0.34%) females for 1999–2000; and finally, 11 of 1219 (0.90%), 7/500 (1.4%) males, and 4/719 (0.56%) females for 2001–2002. Prevalence rates increased with age, more steeply in men, because percentage values were higher in patients ≥75 years of age than in patients 60–74 years of age.
Radiological prevalence of PDB in Siena
Sixteen (eight males and eight females) of the 1778 radiographs showed PDB. Their mean age was 76 ± 8.6 years, with an age range of 57–88 years. The distribution and localization of PDB on the radiographs is summarized in Table 3. Pelvic PDB was present in 14 (87.5%) radiographs and unilateral involvement of the pelvis was present in 9 (64.3%) patients. Within- and between-observer repeatability was good (0.97% and 0.98%, with κ values 0.95 and 0.87, respectively). Of the two patients without pelvic involvement, one had PDB in the third lumbar vertebra and the other had PDB of the proximal right femur. All radiographic lesions were predominantly sclerotic in nature. Seven PDB patients (43.7%) showed radiological evidence of osteoarthritis at the lumbar vertebrae, and six patients (37.5%) had osteoarthritis of the hip. Among the pagetic subjects with hip osteoarthritis, three had bilateral hip involvement and three had unilateral hip involvement. The sacroiliac joints were fused in 1 (6.2%) of the 16 PDB patients. The cumulative radiological prevalence of osteoarthritis in our pagetic patients was 11 of 16 (68.7%). Three (18.7%) PDB cases (two females and one male) had vertebral fractures.
The crude prevalence of the disease was 16 of 1778 (0.89%). Given that in previous studies the pelvic involvement is commonly described in 60–90% of PDB patients, the estimated overall prevalence of PDB in Siena ranged from 0.98% to 1.48%. The analysis of the 120 consecutive PDB patients who referred to our Metabolic Bone Disease Unit showed pelvic involvement in 82 (68%), proximal femur involvement in 24 (20%), and lumbar spine involvement in 19 (16%). The number of PDB outpatient subjects with the involvement of at least one of these sites was 94 (78%). Accordingly, the estimated overall prevalence of PDB in our population was 1.15%.
The distribution of PDB by decade is shown in Fig. 1. Prevalence rates increased with age, peaking in the seventh and eighth decades. Only one subject <60 years of age had radiographic evidence of PDB.
Biochemical prevalence of PDB in Siena
Results from the biochemical screening are summarized in Table 4. Of interest, 296 (183 females and 113 males) of the 7449 subjects showed elevated ALP levels (>300 UI/liter) and normal liver enzymes, with a prevalence of “biochemical hyperphosphatasia” of 3.97%. Moreover, 236 of the 296 subjects with elevated ALP levels and 944 subjects with normal ALP levels had a biochemical analysis of bone turnover markers and eventually scintigraphic and radiographic screening for PDB. Interestingly, 87 subjects (36.8%) with elevated ALP also showed increased BALP and radiological or scintigraphic diagnosis of PDB. There was a slight male predominance, with a male to female ratio of 1.3:1; 63 cases were polyostotic and 24 monostotic. In addition, we also diagnosed PDB in 12 of the 944 subjects (1.3%) with normal ALP levels that were further studied. In those subjects, the disease was diagnosed mostly in one bone, with only four subjects having more than one localization. Based on the observed 3.97% prevalence of elevated ALP levels and the verified diagnosis of PDB in 36.8% of these subjects, the estimated prevalence of PDB was 1.46%. When prevalence was adjusted to take into account that about 20% of patients with PDB have ALP levels within the normal range, the overall prevalence of PDB was 1.75%. Moreover, prevalence rates were higher in males (1.88%) than in females (1.14%) and increased steadily from the 40–49 year age group through the 80–89 year age group (Table 4; Fig. 1).
Scintigraphic prevalence of PDB in Siena
The analysis of the 7906 consecutive99mTC-MDP bone scans performed over a 4-year period showed the presence of 194 PDB cases. Their mean age was 71 ± 11 years, with an age range between 43 and 87 years. Concordance between observers was 0.99%, for a κ statistic of 0.94. The within-observer repeatability estimated on 50 affected and 50 nonaffected scans was 0.94%, with a κ value of 0.89%. Altogether, there were 116 PDB men and 78 PDB women, with a male to female ratio of 1.5:1. A predominance of polyostotic (n = 116) over monostotic (n = 78) PDB cases was observed. Sites of involvement, in order of frequency, included pelvis (65%), vertebrae (31%), femur (30%), tibias (28%), skull (26%), humerus (9%), ribs (7%), calcis (4%), ulna (2%), scapula (1%), hands (3%), clavicle (2%), mandible (1%), and sternum (1%). Overall, 71% of cases showed PDB in locations that would have been determined on a pelvis film, similar to that observed in the 120 consecutive PDB subjects. The estimated prevalence of “scintigraphic PDB” was 2.45%, which is higher than the estimates from both the biochemical and the radiographic surveys. An age-related increase in prevalence rates was also observed, with a peak from the sixth to eighth decades (Fig. 1).
Estimated number of subjects with PDB in Italy
To obtain an estimate of the number of PDB subjects in our country, we considered prevalence rates from radiographic, biochemical, and scintigraphic surveys, multiplied by the number of subjects according to census data in Italy, Siena and Turin (http://demo.istat.it/pop2002/index.html). Results are summarized on Table 5. The resulting sums suggest that there may be currently from 124,788 to 338,576 people (considering the radiographic and scintigraphic prevalence rates, respectively) potentially affected with PDB in Italy.
Epidemiological studies on PDB have shown a sharp geographical variation in prevalence rates within Europe. The disorder seemed more prevalent in Britain and France than in other countries. (4, 5, 7) Estimates of the prevalence of PDB have been derived from different approaches, such as questionnaires, autopsies, or radiographic surveys of subjects investigated in hospitals. (4–11, 14) In this study, by the analysis of radiological, biochemical, and scintigraphic data from subjects studied in our hospitals, we were able to estimate the prevalence of PDB to be within 1.0-2.4% in the general population from the city and the surroundings of Siena and within 0.7-1.0% in the general population from the city and the surroundings of Turin. The results of our study provide information on the current burden of PDB in Italy and confirm previous literature data about increasing PDB prevalence with age. To our knowledge, there have been no similar studies on the prevalence of PDB in Italy. In a previous postal questionnaire and radiological survey on the European distribution of PDB, the prevalence of the disease in two Italian towns, Milan and Palermo, was reported to be 1.0% and 0.5%, respectively. (5) The sample size, however, was too low to obtain a clear picture of the prevalence of PDB in our country.
Prevalence rates observed in this study differed slightly between Siena and Turin. In addition, PDB prevalence in Siena is similar to that previously reported in Milan, northern Italy, whereas it seems much higher than that observed in Sicily. These findings suggest the possibility of regional variation in the prevalence of PDB in Italy. Indeed, localized areas of high prevalence of PDB have been described in Lancashire, UK, in the northeastern United States, and in the Sierra de La Cabrera district in Spain. (7, 14–17) In contrast no apparent geographical variation has been observed in the distribution of PDB in New Zealand. (9) Further epidemiological studies are needed to verify the hypothesis of geographical variations and of a regional clustering of PDB in Italy.
Results from this study indicate an important prevalence of PDB in our country. According to census data, the estimated number of PDB patients in Turin and in Siena is 4476 and 1334, respectively. Moreover, even though an extrapolation of our results to the entire Italian population may not be accurate, it can be estimated that ∼150,000-300,000 subjects may be affected with PDB in our country. Results from these estimates confirm PDB to be the most common bone remodeling disorder in elderly people in Italy, excluding osteoporosis. It is likely that most of these subjects still require to be diagnosed. Moreover, even though the majority of these subjects will remain asymptomatic, a consistent proportion will certainly require treatment with inhibitors of bone resorption. Our prevalence estimates could be useful in predicting the number of patients in whom treatment with these agents might be required in future.
Importantly, as in other previous studies in different populations, prevalence rates increased significantly with age, peaking in those over age 75 with an age-specific-adjusted prevalence rate of about 3%. There are several potential interpretations of these data. The most likely is that PDB may not develop until an older age and that there may be a lag time in the development of the disease. However, the possibility of a decreasing incidence of PDB in Italian population, as previously observed in other populations, cannot be ruled out. The decrease was particularly evident in populations at higher incidence of PDB, such as in the United Kingdom, New Zealand, and Australia. (6, 7, 9) In contrast, a recent population study(s) in a Dutch sample from the Rotterdam Study indicated an overall estimated prevalence of PDB in The Netherlands of 3.6%, which is significantly higher than the 0.6% reported in 1982 from Detheridge et al. (18) Results from this study allow us to obtain some indications about possible differences in the prevalence of PDB in Italy. If we compare the Italian data from the 1982 European study of Eekhoff et al. to estimates obtained in this study we cannot observe any significant decrease in prevalence rates. This is consistent with the longitudinal results from the radiological survey in Turin, showing no significant reduction in PDB prevalence among the four selected periods: 1986–1987 (0.43%), 1992–1993 (0.73%), 1999–2000 (0.60%), and 2001–2002 (0.90%). Moreover, the lack of significant age-related differences in prevalence rates among subjects older than 60 years in the bone scan survey also support this hypothesis. In fact, PDB can be detected scintigraphically at an early stage, and new scintigraphic lesions virtually never appear over the course of the disease. Possibly, the prevalence of PDB is particularly reduced in those areas where earlier it had been highest, such as Lancashire, UK. (7) The reason for these modifications in prevalence estimates over time is actually unknown and may be caused by either immigration flows (changing the ethnic mix and the genetic background of the population) or decreased exposure to an environmental trigger (i.e., viral infection) in the high prevalence areas.
There was a slight predominance of men in our scintigraphic and biochemical surveys, as well as in the two radiological surveys. Taken together, these data are consistent with the sex- and age-specific differences previously reported for PDB prevalence in different studies. (19) Interestingly, both the radiological and the scintigraphic surveys evidenced a tendency for the right side of the body to be affected more than the left, in keeping with previous observations. (20) The reasons explaining this tendency are actually unknown and deserve further study. Moreover, results from the radiological survey in Siena also confirmed osteoarthritis to be a frequent complication of PDB. (21)
We found very good agreement in prevalence rates assessed by biochemical and radiological methods, especially in the younger age groups, suggesting that biochemical screening may offer an alternative and valid method of assessing PDB prevalence, particularly because very large samples can be screened in a shorter time. One in three subjects with raised serum ALP levels had confirmed PDB diagnosis, accounting for a 45-fold higher risk of PDB in subjects with higher ALP activity. These estimates are higher than those previously reported in a recent population-based study from the Rotterdam cohort, which showed an 11-fold increased risk for the disease in subjects with elevated ALP, (18) whereas they are in keeping with findings from patients followed in bone clinics, 90% of whom have raised serum ALP activity. (19) Indeed, even tough comparisons between these observations are difficult; differences among studies are mainly caused by different methodological approaches (hospital-based and population-based). In fact, it is likely that only a small proportion of cases with PDB come to clinical attention, most likely those with the more severe disease. Consistently, we observed in our biochemical survey an increased proportion of polyostotic than monostotic PDB cases. Similarly, a disease-based selection bias can be responsible of the much higher prevalence of PDB observed in the scintigraphic survey, because the hospital of Siena has a recognized experience for the diagnosis and treatment of PDB for many years. Moreover, bone scans are a sensitive way to evaluate potential sites of PDB but are clearly not diagnostic, and we do not have informations about the X-ray confirmation of the diagnosis in these subjects. Finally, radiological surveys also have some limitations. We examined radiographs from the pelvis and surrounding tissue, and because PDB may affect any part of the skeleton, it is likely that not all cases of PDB were detected among the analyzed subjects. Moreover, the indications for abdominal radiography have probably changed over the past years because of the increasing use of ultrasounds, NMR, and CT techniques.
In conclusion, results from our surveys suggest that PDB in Italy has an estimated prevalence of at least 1%, comparable with that observed in United States and other European countries, but lower than that described in high prevalence areas such as Britain and New Zealand. No secular trend for a decreasing prevalence of PDB has emerged from these surveys.
- 11877 On a form of chronic inflammation of bones (osteitis deformans). Med Chir Trans 60:37–63.
- 21991 Paget's disease of bone. In: CoeFL, FavusMJ (eds.) Disorders of Bone and Mineral Metabolism. Raven Press, New York, NY, USA, pp. 1027–1064.
- 31996 Paget's disease of bone and quality of life. J Bone Miner Res 11:1897–1904., , , ,
- 41981 The epidemiology of Paget's disease. Metab Bone Dis Relat Res 3:231–233.
- 51982 European distribution of Paget's disease of bone. BMJ 285:1005–1008., ,
- 61997 Evidence for secular change in Paget's disease. Bone 20:69–71., , , , ,
- 71999 The epidemiology Paget's disease in Britain: Is the prevalence decreasing? J Bone Miner Res 14:192–197., , , , ,
- 82000 Long-term trends in the incidence of Paget's disease of bone. Bone 27:423–427., , ,
- 92002 Paget's disease in New Zealand: Evidence for declining prevalence. Bone 31:616–619., , , ,
- 102002 Is clinical expressiveness of Paget's disease of bone decreasing? Bone 30:399–403., , , ,
- 111981 Paget's disease. In: ResnickD, NiwayamaG (eds.) Diagnosis of Bone and Joint Disorders, vol. 2. W. B. Saunders, Philadelphia, PA, USA, pp. 1721–1754.,
- 121997 Bone scintigraphy in metabolic bone disease. Semin Nucl Med 27:291–305.,
- 13Scintigraphy in metabolic bone disease. In: FavusMJ (ed.) Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism, 5th ed. Lippincott Williams and Wilkins, Philadelphia, PA, USA, pp. 189–194.,
- 142000 Prevalence of pelvic Paget's disease of bone in the United States. J Bone Miner Res 15:461–465., , ,
- 151980 Paget's disease of bone: The Lancashire focus. BMJ 280:1105–1107., , ,
- 162003 Identification of possible areas of high prevalence of Paget's disease of bone in Spain. Clin Exp Rheumatol 21:635–638., , , , ,
- 171990 Features of Paget's disease of bone in a new high-prevalence focus. Med Clin (Barc) 95:169–174., , , ,
- 182004 Paget's disease of bone in The Netherlands: A population-based radiological and biochemical survey-The Rotterdam Study. J Bone Miner Res 19:566–570., , , , , ,
- 191998 Pathophysiology and Treatment of Paget's Disease of Bone, 2nd ed, Martin Dunitz, London, UK.
- 201981 Paget's disease of bone: The anatomical distribution. Metab Bone Dis Relat Res 3:239–241.
- 212002 Incidence and natural history of Paget's disease of bone in England and Wales. J Bone Miner Res 17:465–471., , , , ,