L. Sinigaglia

Determinants of adherence to osteoporosis treatment in clinical practice

IntroductionPoor adherence to prescribed treatments is widespread in clinical practice and this can lead to potentially life-threatening events. This problem is apparently very common for osteoporosis treatment but the causes of discontinuation and low compliance are complex and poorly defined.MethodsGlobal adherence to osteoporosis treatment was specifically addressed in a nation-wide survey carried out in 9851 postmenopausal women referred to 141 Italian centres for osteoporosis management for a follow-up assessment, at least one year after having been prescribed a treatment with one of the following drugs: calcium±vitamin D supplements alone (CaVitD), hormone replacement therapy (HRT), raloxifene 60 mg (RLX), intramuscular clodronate 100 mg/7-14 days (CLOD), risedronate 5 mg/day (RIS) and alendronate 10mg/daily (ALN10) or 70 mg once weekly (ALN OW).ResultsOverall 19.1% of the patients discontinued the prescribed drug before attending the bone mass re-evaluations, more than half of them within the first 6 months. The discontinuation rate was significantly different between the treatments. The medications most frequently interrupted within one year were CLOD (28.7%; p<0.01 versus any other treatment), while by far the least interrupted was ALN-OW (6.9%; p<0.001 versus any other treatment). The most frequent reasons for discontinuation were drug related side effects, insufficient motivation to treatment and fear of side effects. The prevalence of the reasons for discontinuation were different among treatments: safety concerns were very common for HRT, lack of motivation was the most common cause for CaVitD and CLOD, and drug related side effects for RIS, ALN and RLX. Persistence to treatment was significantly higher in patients with previous vertebral fractures, densitometric osteoporosis, on corticosteroid or anti-inflammatory treatments. A significantly increased risk of treatment interruption was found among patients on benzodiazepine or gastro-protective agents and in patients in whom a bone measurement was not readily available. The highest compliance to recommended dosing was observed with ALN OW and HRT (p<0.001 versus any other) and the lowest for CaVitD (p<0.01 versus any other). Poor treatment compliance (<50% drug taken) was significantly related to benzodiazepine and gastroprotective use, while a significantly better compliance was associated with recognized risk factors for osteoporosis: early menopause, low bone mass values values, previous vertebral fractures. The poorest adherence was observed when treatments were prescribed by General practitioners (GPs), and orthopaedic surgeons (p<0.01 versus global mean).ConclusionsThe results of this large survey of Italian osteoporotic women indicates that the most important determinant of both persistence and compliance to treatment is the type of drug prescribed with a definite advantage of ALN-OW. Treatment compliance is particularly poor for CaVitD and this emphasizes the need for new ways to supplement at least vitamin D. The main reasons for discontinuation are side effects and lack of motivation while the best treatment adherence was observed in patients with severe and well documented osteoporosis.

Prevalence of Osteoporosis by Educational Level in a Cohort of Postmenopausal Women

Abstract: To evaluate whether the prevalence of osteoporosis and related risk factors might be influenced by the level of education, as has been demonstrated for many other chronic diseases, 6160 postmenopausal women at their first densitometric referral were interviewed about reproductive variables, past and current use of estrogens, prevalence of chronic diseases, and lifestyle factors such as calcium intake, physical activity, smoking and overweight. This sample was stratified by years of formal education. Densitometric evaluation was performed by dual-energy X-ray absorptiometry. Age at menarche, past exposure to oral contraceptives, use of hormone replacement therapy, prevalence of chronic diseases, physical activity, overweight and smoking showed significant trends according to the years of education. The prevalence of osteoporosis showed an inverse relationship with level of education, ranging from 18.3% for the most educated to 27.8% for the least educated women. Multiple logistic regression analysis demonstrated a predictive role toward osteoporosis by age, age at menarche and menopause, hormone replacement therapy, calcium intake, physical activity and body mass index. Using the lowest educational level as reference category, increases in educational status were associated with a significantly reduced risk for osteoporosis (OR = 0.76, 95% CI 0.65–0.90 for 6–8 years of schooling; OR = 0.68, 95% CI 0.57–0.82 for 9 years or more). This study shows differences in the prevalence of osteoporosis among educational classes and the protective role played by increases in formal education. If these results are confirmed in other population studies, public health intervention programs will have to consider the socioeconomic and cultural background of the population strata that run a greater risk of osteoporosis.

Drug-induced lupus following treatment with infliximab in rheumatoid arthritis.

After introduction of infliximab for the treatment of rheumatoid arthritis (RA), there have been many reports of patients developing asymptomatic higher rate of antinuclear antibodies and antidsDNA antibodies than in non-infliximab-treated patients. However, only five clinical drug-induced lupus (DIL) cases have been documented following treatment with infliximab, in RA and in Crohns diseases. We report a case of a 69-year-old female with a 5 year history of RA, who was successfully treated with low-dose methotrexate (MTX) and infliximab (initially 3 mg/kg and from the fourth infusion 5 mg/kg) for 23 weeks. Before the sixth infusion, she was diagnosed with DIL by both clinical features (fever> 38°C, recurrence of active synovitis, myalgia, erythematous rash and general malaise) and laboratory findings (antinuclear antibodies 1:160, anti-double-stranded DNA positive by ELISA assay, decreased serum complement C3 and C4, hypergammaglobulinaemia, increased erythrocyte sedimentation rate). After discontinuation of treatment and therapy with oral prednisone, lupus resolved within 8 weeks.

Quantitative heel ultrasound in a population-based study in Italy and its relationship with fracture history: the ESOPO study

We assessed the clinical usefulness of quantitative ultrasound (QUS) in defining the prevalence rates of osteoporosis and osteopenia and their association with fractures of the forearm, vertebrae, and hip. The ESOPO study was conducted in 2001 and assessed a random sample of 11,011 women and 4,981 men, in 83 centers spread all over Italy. A large array of risk factors was investigated, and self-reported history of fractures was collected in a questionnaire. After the patient had undergone interview and a brief physical examination, QUS of the heel was performed, using the Achilles apparatus (GE-Lunar, Madison, USA). The prevalence rate of osteoporosis in women 40–79 years old was approximately 18.5%, while the rate of osteopenia was about 44.7%; in men 60–79 years of age the rates were 10% and 36%, respectively. A strong association with fractures was found for osteoporosis and osteopenia in both men and women, independently of all traditional risk factors, including age. These results confirm the suitability of US measurements as a tool for detecting individuals at risk of fractures.

Effect of Age, Weight and Lifestyle Factors on Calcaneal Quantitative Ultrasound in Premenopausal Women: The ESOPO Study

The objective of this study was to identify the determinants of bone mass as measured by quantitative ultrasound (QUS) in premenopausal women. The study population is part of the “Epidemiological Study On the Prevalence of Osteoporosis” (ESOPO) on risk of the general population of Italy. We report the data on 2727 premenopausal women aged 40–50 years who are still regularly menstruating. Bone stiffness (called simplicity stiffness), which is derived from the values of broadband ultrasound attenuation (BUA) and speed of sound (SoS), was measured by a heel QUS device (Achilles Apparatus, Lunar, Co. USA). The most commonly recognized determinants of bone mass were modelled with stiffness by multiple regression analysis or analysis of variance (ANOVA). Bone stiffness was negatively related to age and number of cigarettes and positively to body weight, body weight at 25 years, height and estimated daily calcium intake. By multiple regression analysis, independent, positive, predictors of bone stiffness were age, weight at 25 years and daily calcium intake. Bone stiffness adjusted for age and body weight at 25 years was positively associated with outdoor activity score and negatively with number of pregnancies, chronic use of any drug, smoking and subjective health status. Bone stiffness was also somewhat (p < 0.015) negatively related to history of prolonged bedrest and thyroxin use. In conclusion, our results indicate that risk factors usually associated in other studies with DXA-BMD in elderly women are also associated with calcaneal bone stiffness, as measured by QUS in premenopausal women. These findings should help to identify premenopausal women at risk and to design an early strategy for osteoporosis prevention based on eliminating modifiable risks.

Unbalanced Diet to Lower Serum Cholesterol Level is a Risk Factor for Postmenopausal Osteoporosis and Distal Forearm Fracture

Abstract: The purpose of this study was to assess whether dietary changes aimed at reducing serum cholesterol can increase the risk of osteoporosis (OP) and fracture. The study group consisted of 311 postmenopausal women with high serum cholesterol levels and following a diet low in dairy products (calcium intake estimated at less than 300 mg/day) for 27.3 ± 29.1 months. This sample was compared with a case–control group of 622 healthy postmenopausal women paired for age and age at menopause and with a calcium intake estimated at more than 1 g/day. Bone mineral density was measured at the lumbar spine by dual-energy X-ray absorptiometry. Prevalence of OP was significantly higher in women with a low dairy calcium intake (42.1% vs 22.3%; p<0.0001), as was the number of Colles” fractures occurring after menopause (4.5% vs 1.6%; p = 0.008). Multiple logistic regression analyses demonstrated that a diet low in dairy calcium was a risk factor for OP (OR = 2.52, 95% CI 1.84–3.45) and Colles” fracture (OR = 2.72, 95% CI 1.18–6.26). In the low dairy calcium group, diet duration significantly influenced the risk of OP (OR = 1.13, 95% CI 1.01–1.25 for 1 year of diet). No differences in further risk factors for coronary heart disease were found between the groups, but the proportion of women physically active was lower in the women with high serum cholesterol levels. A diet that severely limits calcium intake from dairy products in an attempt to correct raised serum cholesterol levels is a risk factor for postmenopausal OP and Colles” fracture. Dietary intervention methods to lower serum cholesterol in postmenopausal women should maintain an adequate calcium intake by providing calcium from low-fat dairy products or calcium supplements.

Transient osteoporosis of the hip: a densitometric study.

SummaryThree new cases of transient osteoporosis of the hip are reported. Diagnosis was achieved by plain radiographs, bone scintiscan, magnetic resonance imaging and X-ray absorptiometry of proximal femurs. The densitometry showed at the Wards triangle a mean reduction of bone mineral density in the affected side of 36%. All subjects were treated with i.v. clodronate for ten consecutive days with a complete recovery of femoral density within 4 months. X-ray absorptiometry allows a quantification of the demineralization process and can be useful in the long term evaluation of this entity.

Prevalence of osteoporosis and fractures in a migrant population from southern to northern Italy: a cross-sectional, comparative study

The purpose of this study was to assess whether a migrant population from southern regions of Italy showed differences in osteoporosis (OP) and fracture prevalence in comparison to a population born and living in Milan, a city in northern Italy. The study group consisted of 1764 postmenopausal women born in southern Italy who migrated after age 20 to Milan, where they had lived for at least 15 years. This group was compared with a sample of 4018 postmenopausal women born, raised and living in Milan. Bone mineral density (BMD) was measured at lumbar spine by dual-energy X-ray absorptiometry. Univariate analysis showed a significantly higher BMI, a greater parity, a lower postmenopausal estrogen (HRT) usage and a lower calcium intake in the Southern group (SG). Densitometric values showed a mean lumbar BMD significantly lower in SG (P<0.001), resulting in a greater prevalence of OP (30.5% versus 24.9%; P<0.001). Consistent with this result, the number of women reporting low energy fractures after menopause was higher in SG (4.8% versus 3.6%; P=0.01). Multiple logistic regression analyses showed that OP was predicted by well-recognized variables, such as age, age at menopause, BMI, HRT, calcium intake together to belonging to SG, which independently increased the risk for OP by 42%. Age, age at menopause, BMI and belonging to SG significantly influenced the fracture risk, but the last variable was no longer associated with an increased fracture risk when BMD was entered in the model. Despite the results of previous epidemiological studies and protective anthropometric measures, a higher OP and fracture prevalence was found in the migrant group from southern Italy to Milan in comparison with the native sample. These differences are not solely explained by lifestyle variables. Our data suggest that changing environmental exposures can influence bone mass and fracture risk.

Is the metatarsal fracture in postmenopausal women an osteoporotic fracture? A cross-sectional study on 113 cases.

To evaluate whether metatarsal fracture in postmenopausal women can be related to osteoporosis, a sample of 113 postmenopausal women with metatarsal fracture due to minor trauma were recruited. Demographic and clinical data were compared with a control group of 339 healthy age-matched women and with a sample of 224 women with wrist fracture. In all women, bone mineral density (BMD) was measured at the lumbar spine by dual-energy X-ray absorptiometry. The average age of the metatarsal fracture group was slightly lower than that of the wrist fracture group (56.9 vs 58.4 years). Women with metatarsal and wrist fracture had a significantly higher age at menarche, lower age at menopause and lower body mass index when compared with controls. In both fracture groups BMD was significantly lower compared with controls. In stepwise logistic regression models, factors associated with metatarsal fracture risk were age at menopause (odds ratio [OR] 0.86; 95% confidence interval [CI] 0.81–0.92) and BMD (OR per — 1 SD 2.44; CI 1.92–3.11). Factors associated with wrist fracture risk included age at menopause (OR 0.89; CI 0.84–0.93) and BMD (OR per — 1 SD 2.65; CI 2.17–3.24). The similarities existing in risk factors and their estimates between a well-recognized osteoporotic fracture such as wrist fracture and metatarsal fracture, support the hypothesis that the latter can be included among osteoporotic fractures.

FRI0172 Drug survival of the first course of anti-TNF agents in patients with rheumatoid arthritis and spondyloarthropathies. results from the monitornet database

Background Pharmaco-epidemiological studies on TNF-inhibitors (TNF-I) in rheumatoid arthritis (RA) and spondyloarthritis (SpA) are providing useful data about effectiveness and safety in clinical practice. Objectives To compare drug survival in RA and SpA according to TNF-I drug using data from a multicentre Italian cohort study. Methods The patients were selected from the Monitornet database, a prospective cohort study to monitor the long-term safety of biologic therapy involving 27 rheumatology centres across Italy, supported by the Italian regulatory agency AIFA. For the purpose of these analyses we included RA or SpA patients, who started a first course infliximab (INF), etanercept (ETA) or adalimumab (ADA). Drug survival was primarily defined from start until first discontinuation (overall drug survival) and secondarily due to ineffectiveness or adverse events (AEs). A first set of analyses assessed the relationship between diagnosis and drug survival using RA as reference category, adjusting for age, gender and comorbidities. A second set of analyses stratified by diagnosis explored the relationship between TNF-I and drug survival using INF as reference category, adjusting for age, gender, comorbidities, disease duration, previous DMARDs, calendar, baseline CRP, disease activity (DAS28 or BASDAI) and severity (HAQ score or BASFI). Drug survival was analyzed using Cox proportional hazards models. Results are presented as hazard ratios (HR) and 95% confidence intervals (CI). Results 1992 RA patients (79.8% women, mean age 55.5 yrs (SD 12.3), mean disease duration 9.6 yrs (SD 8.0), mean baseline DA28 4.4 (SD 1.9), baseline HAQ 1.4 (SD 0.7), median number of previous DMARDs 2 (IQR 1-3) and 993 SpA patients (55.2% men, mean age 49.6 yrs (SD 12.5), mean disease duration 6.9 yrs (7.5), mean baseline BASDAI 4.7 (SD 2.4), mean baseline BASFI 4.6 (SD 2.5) were included in the analyses. 44.2% RA patients started ETA, 21.4% INF, 34.4% ADA, while 45.8% SpA patients ETA, 22.3% INF, 31.9% ADA. Using RA as reference, SpA patients showed a lower drug discontinuation (HR 0.92 [95%CI 0.87, 0.99]), due to a lower withdrawal for inefficacy (HR 0.90 [95%CI 0.82, 0.99]). Both in RA and in SpA, compared to INF, ETA showed a slightly better survival on treatment, also due to lower discontinuation for inefficacy (see table). Diagnosis TNF-I Overall drug survival Drug survival (inefficacy) Drug survival (AEs) HR [95%CI] HR [95%CI] HR [95%CI] RA INF Reference Reference Reference ETA 0.73 (0.56, 0.94) 0.69 (0.51, 0.94) 0.80 (0.48, 1.33) ADA 0.81 (0.54, 1.20) 0.83 (0.61, 1.12) 0.88 (0.54, 1.43) SpA INF Reference Reference Reference ETA 0.54 (0.31, 0.93) 0.48 (0.27, 0.84) 0.57 (0.33, 0.99) ADA 0.65 (0.35, 1.21) 0.76 (0.43, 1.35) 0.46 (0.19, 1.10) Conclusions These data support a better survival on treatment for SpA over RA and for ETA as compared with INF, mainly due to lower discontinuation for ineffectiveness rather than for AEs. Disclosure of Interest None Declared