Prof. Dr. Stefan Cristea
Ortopedie si traumatologie
RISK ASSESSMENT OF NEW FRACTURE FOLLOWING FRAGILITY HIP FRACTURE
RISK ASSESSMENT OF NEW FRACTURE FOLLOWING FRAGILITY HIP FRACTURE
CR. BUDICĂ, ŞT. CRISTEA, GHE. PANAIT
Saint Pantelimon Emergency Hospital, Orthopaedics Department
Background: Osteoporosis, a disease with serious socio-economical implication, characterized by bone mass reduction and modifications of internal bone structure, leads to increased bone fragility and to an increased fracture risk. Hip fracture is the major complication of osteoporosis and this appears as a result of a low energy trauma – usually a fall, the bone resistance is over passed by the force applied to the bone.
Aim & Objectives: Our aim was to evaluate the risk of subsequent hip fractures. We have evaluated the incidence of a new fragility fracture; the risk factors associated to a subsequent fracture; the mortality risk and we have also described the modifications of internal bone structure depending on the compliance with osteoporosis treatment. We have consider a subsequent fracture any type of fracture produced after the hip fracture, no matter if is a hip fracture or not (spine, wrist, shoulder, knee). To obtain these results we developed statistical models based on: survival analysis by Kaplan-Meier, life table, COX proportional hazard regression, regression binary logistic.
Study Limitations: Even if the period of time in this study is relatively long (6 years), the number of patients that could be followed for a long period is quite small because some missed their follow-ups, and others couldn’t be reached and so some indicators were excluded from the study.
Results/Findings: The cohort included 300 patients over 60 years with fragility hip fractures admitted and surgically treated at “Saint Pantelimon” Emergency Hospital / Orthopedics Department during 2002-2008. The exclusion standards used were: the patients treated for osteoporosis, patients with cortical-shaft index over 0.5, patients with severe neuropsychical diseases, the patients deceased during hospitalization and also the patients with a T-score >-2.5 (DXA method) or the patients with no diagnosis test for osteoporosis. During the study we used face to face questionnaires, the response rate was 75% (the final group having 300 patients). The patients kept in the study were followed postoperative at 1 month, 3 and 6 months, 12 months, then yearly. Data required for the survival analysis were obtained over the phone from the patient’s family.
Conclusion: The cumulative incidence of subsequent fractures was 7,4% at one year, 16 % at 2 years from the first hip fracture, the median value of refracture being 49 months (CI 95% 32,2-65,8). Men have a greater risk of a new fracture, even though there is no statically significant difference regarding the risk of fracture by gender. The risk of new fracture is higher in younger patients (60-69 years) than the patients over 80 years. The alendronate and vitamin D and Calcium supplements treatment diminishes the risk of a subsequent fracture with approximately 7%. Hip fractures are associated with high mortality. The cumulative incidence of mortality is 2% at one month, 5,4% at 3 months, 17,5% at 6 months, 27,4% at one year and 36,8% at 2 year after the fracture; the median survival time is 35 months. The patients with recurrent fractures have a greater mortality risk than those who didn’t suffer a new fracture. The cumulative incidence of deaths in patients who refractured is 31% at one year. Prevention of the first hip fracture is ideal, but at least after the first fracture the awareness of the physicians should be heightened to prevent future fracture by modifying the risks, which must include optimal management of other medical diagnoses as well as fall prevention.
Keywords: hip fracture, predictors, mortality, fracture risk, osteoporosis, alendronate treatment, accidental falls.
L’évaluation du risque d’une nouvelle fracture après une fracture de fragilité de la hanche
Introduction: L’ostéoporose, une maladie avec des graves implications socio-économiques, caractérisé par la réduction de la masse osseuse et changements de la structure interne osseuse, conduit à une fragilité osseuse augmentée et un risque élevé de fracture. La fracture de la hanche est la majeure complication de l’ostéoporose et elle vient de se produire après un traumatisme à faible énergie – habituellement une chute accidentelle, la résistance osseuse est surmontée par la force appliqué aux os.
Objectifs: Notre objectif a été d’évaluer le risque de fractures ultérieures de la hanche. Nous avons évalué l’incidence d’une nouvelle fracture de la hanche; les facteurs de risque associés a une fracture ultérieure; le risque de la mortalité et nous avons aussi décrit les changements de la structure interne osseuse dépendent de la conformité avec le traitement de l’ostéoporose. Nous avons considéré une fracture ultérieure n’importe quoi type de fracture qui se produit après une fracture de la hanche, peu importe s’il est ou non une fracture de la hanche (colonne vertébrale, poignet, épaule, du genou). Pour obtenir ces résultants nous avons développé des modèles statistiques bases sur : courbes de survie de Kaplan-Meier, table de mortalité, la régression de Cox à risques proportionnels, régression logistique binaire.
Limitées: Même si l’intervalle du temps dans cet étude est relativement long (6 années), le nombre des sujets qui on pu être poursuivis est assez bas parce que quelques sujets ont manqué leurs contrôles, et autres n’ont pu être contactés, et ainsi quelque indicateurs ont été exclus de l’étude.
Résultats: La cohorte contient 300 sujets de plus de 60 ans avec une fracture de fragilité de la hanche hospitalisés et traités chirurgicalement aux L’hôpital d’urgence Saint Pantelimon section d’Orthopédie pendant 2002-2008. Les critères d’exclusion utilisés ont été : les sujets traités pour ostéoporose, les sujets avec un index cortico-diaphysaire plus de 0,5, les sujets avec des graves maladies neuropsychologiques, les sujets qui sont décédés pendant l’hospitalisation et aussi les patients avec un T-score > -2,5 (méthode DXA) ou les patients avec aucun test diagnostique pour l’ostéoporose. Pendant l’étude nous avons utilisé des questionnaires face to face and le taux de réponse a été 75% (le groupe final ayant 300 sujets). Les sujets dans nôtre étude ont été contrôlés postopératoire à un mois, 3 et 6 mois, 12 mois, après annuel. Les informations nécessaires pour l’analyse de la survie ont été obtenues par téléphone par la famille du sujet.
Conclusions: L’incidence cumulative de fractures ultérieures était 7,4% à 1 année, 16% à 2 ans à partir de la première fracture de la hanche, la valeur médiane de la refracture étaient 49 mois (CI 95% 32,2-65,8). Les hommes ont un risque plus grand d’une nouvelle fracture, même si il n’y a pas une différence significative en ce qui concerne le risque de fracture par gendre. Le risque d’une nouvelle fracture est plus grand dans les sujets plus jeunes (60-69ans) que les patients de plus de 80 ans. Le traitement avec alendronate et vitamine D et suppléments de Ca diminue le risque de fracture ultérieure avec 7%. Les fractures de la hanche sont associées avec une grande mortalité. L’incidence cumulative de la mortalité est de 2% à 1 mois, 5,4% à 3 mois, 17,5% à 6 mois, 27,4% à 1 année et 2 années après la fracture ; le temps médian de survie est de 35 mois. Les sujets avec des fractures récurrentes ont un risque de mortalité plus grand que ceux qui n’ont pas une nouvelle fracture. L’incidence cumulative de morts aux patients qui ont réfracturé est de 31% à 1 année. La prévention de la première fracture de la hanche est idéale, mais même après la première fracture la sensibilisation du médecin devait être augmentée pour prévenir la fracture dans l’avenir par le changement des risques, qui nécessite une gestion optimale des autres maladies diagnostiqués comme la prévention de la tombée.
Mots clés: fracture de la hanche, mortalité, risque de fracture, ostéoporose, traitement avec alendronate, chute accidentelle.
Osteoporosis, a metabolic disease characterized by bone mass reduction and modifications of internal bone structure, leads to increased bone fragility and to an increased fracture risk. When the proper conditions appear, for example, a low energy trauma – usually a fall, the bone resistance is over-passed by the force applied to the bone, and so the fragility fracture appears, the major complication of osteoporosis.
Osteoporotic fractures create an enormous burden on healthcare budgets. In Europe, direct medical costs for osteoporotic fractures are estimated at more than 36 billion Euros annually. In Romania, the evaluation of the economic burden of fragility fractures is complicated due to the lack of a national fracture registry and standardized tracking systems. There is a national evaluation of fractures made by The National Endoprosthetic Registry, but we don’t know if this evolution is real or complete, this registry recording data from the endoprosthetic specialized centers, data from little centers could be missing.
In Romania, according to the last report to the IOF (International Osteoporosis Foundation) for 2007/2008, from the national population of 21.673.328, at the end of 2006, the population over 50 years was 6.678.533 (30,8%), 3.701.511 (55,4%) women and 2.977.022 (44,6%) men. According to the DRG (Diagnosis Related Group, www.drg.ro) at the end of 2006 were recorded 125.931 hip fractures.
Healthcare costs differ greatly, as do standards of care. However, because hip fractures, unlike vertebral or wrist fractures, almost always result in hospitalization and require surgery in the vast majority of cases, their incidence and related costs are easier to monitor. For example, the average length of hospital stay following a hip fracture is 8 days in Slovakia, 13.9 days in France, and 26 days in the United Kingdom with costs per day of 38 Euros, 720 Euros and 426 Euros respectively. In Romania the cost for a hip fracture in 2006 was 3765.53 ROL (~1200 Euros) according to the DRG.
In Romania osteoporosis is not considered a major healthcare problem, lots of patients with osteoporosis are clueless concerning their condition, despite the fact that they have risk factors who could lead to an early diagnosis of osteoporosis before the appearance of a fragility fracture. Further, in patients who already suffered a fragility fracture, the diagnosis of osteoporosis is frequently neglected, this leading to an increased risk of future fractures. There is also a flawed training of healthcare professionals whom are supposed to diagnose and treat osteoporosis and its complications – fragility fractures - and especially of general practitioners who are supposed to select the patients that should go to a specialist for osteoporosis.
In this context, we decided to evaluate the risk of subsequent hip fractures. This study evaluates the incidence of subsequent fractures, the risk of a subsequent fracture, the mortality risk, and describes the modifications of internal bone structure depending on the compliance with osteoporosis treatment. We consider a subsequent fracture any type of fracture produced after the hip fracture, no matter if is a hip fracture or not (spine, wrist, shoulder, knee).
To obtain these results we developed statistical models based on: survival analysis by Kaplan-Meier, life table, COX proportional hazard regression, regression binary logistic. The survival differences were based on Kaplan-Meier survival curves and for the meaning of the differences the log-rank test was used. The censored events include the individuals who survived till the end of the study and those who were lost during the follow-up, the final event being death. The refracture risk was assessed using the same method: the patients that didn’t suffer a new fracture during the study were censored, and the date when the fracture appeared was the final event. The refracture incidence after a hip fracture was reported in percentages for age groups, gender, types of fracture surgically treated. The level of significance used was p<0.05, and the data was analyzed using SPSS version 13.0 Microsoft Windows.
From clinical and from public health perspective, hip fractures are, by far, the most important type of osteoporosis-related fractures (Melton L.J. III, 2000). Hip fractures disproportionately affect elderly population and, in the majority of patients, result in impairment of mobility (Hannan E.L. et al., 2001), functional decline (Kane R.L. et al., 1998), loss of quality of life (Randell A.G. et al.., 2000), increased health resource utilization (Brainsky A. et al., 1997) and even in death in a significant proportion of male and female patients (Hannan E.L. et al., 2001; Bass E. et al., 2007). These outcomes may be worsened by recurrent or subsequent fractures after hip fracture repair. In fact, hip fracture survivors have a tremendous high risk for subsequent skeletal fractures with and standardized incidence ratio ranking from 2.8 to 8.1 in men and from 2.1 to 3.9 in women (van Staa T.P., 2002; Colon-Emeric C. et al., 2003). Recently, the cumulative incidence of a second hip fracture in a Finnish cohort has been set in 5.08% at 1 year and 8.11% at 2 years after the first fracture (Lönnroos E., 2007); a similar study from the Framingham Study cohort has shown a 14.8% re-fracture rate during a median of 4.2 years of follow-up, the 1-year mortality being significantly increased after the second hip fracture (24.1% vs 15.9%) (Berry S.D. et al., 2007). This high rate of subsequent fractures has been related to the high rate of bone loss—5% to 7% during the first year after the fracture (Karlsson M. et al., 1996)—nevertheless, the influence of osteoporosis undertreatment in patients with hip fracture is a fact that cannot be neglected (Kiebzak G.M. et al., 2002; Hooven F. et al., 2005; Davis J.C. et al.,2006), especially when enough time between the first and second hip fractures for interventions aimed to reduce second hip fractures has elapsed (Berry S.D. et al., 2007).
Patients with hip fracture are at high risk for new vertebral and non-vertebral fractures shortly after the incident hip fracture [van Staa T.P., 2002; Colon-Emeric C. et al., 2003; Lönnroos E., 2007; Berry S.D. et al., 2007] and represent an important population to target for secondary prevention especially when an adequate time between the first and second hip fractures for interventions has elapsed (Berry S.D. et al., 2007)
We developed a retrospective cohort study with a prospective tracking component. The retrospective component of the study analyses the clinical and therapeutically characteristics of the patients over 60 years with fragility hip fractures, admitted and surgically treated at Saint Pantelimon Emergency Hospital / Orthopedics Department during 2002-2008. The exclusion standards used were: the patients treated for osteoporosis (we included the patients diagnosed with osteoporosis, but whom discontinued treatment), patients with cortical-shaft index over 0.5, patients with severe neuropsychical diseases whom couldn’t answer the questions. Further, the patients deceased during hospitalization were excluded, and also the patients with a T-score >-2.5 (DXA method) or the patients with no diagnosis test for osteoporosis and those with more than 30% non-answers at the questions from the questionnaire.
During the study we used face to face questionnaires for the patients selected based on the standards presented, and the response rate was 75% (the final group having 300 patients). The patients kept in the study were followed postoperative at 1 month, 3 and 6 months, 12 months, then yearly. Data required for the survival analysis were obtained over the phone from the patient’s family.
The methodology proposed is quantitative, investigation based on face to face questionnaire and medical papers analysis. The analysis of medical charts and other relevant medical documents (medical tests, X-rays) provided significant data for the study. These quantitative data were enriched with qualitative data obtained based on patient observation protocols during hospitalization, as well as in follow-ups.
In Table 1 we showed the main characteristics of the studied group:
Table 1. Socio-demographic and medical characteristics of the studied group
Age (years) No. (%)
60-69 53 (18)
70-79 127 (42)
80-89 94 (31)
90+ 26 (9)
Mean / Median 78 (77)
Minimum – Maximum / Std. Deviation 60 – 96 (9)
Women 248 (83)
Fracture type and treatment option
Pertrochanteric fractures 140 (47)
DHS 140 (100)
Femoral neck fractures 120 (40)
Internal fixation 40 (33)
Unipolar hemiarthroplasty 40 (33)
Bipolar hemiarthroplasty 2 (1)
Total hip arthroplasty 38 (33)
Subtrochanteric fractures 40 (13)
DHS 20 (50)
Gamma nail 20 (50)
Cardiovascular diseases 213 (71)
Digestive affections 126 (42)
Rheumatic affections 124 (41)
Neuropsychical affections 110 (37)
Endocrine and metabolic diseases 68 (23)
Anemia 83 (28)
Pulmonary affections 40 (13)
Neoplasia 8 (3)
Treatment (mean) 1,27
Minimum – Maximum / Std. Deviation 0-5 (1)
Medical postfracture complications
Cardiac complications 23 (8)
Pulmonary complications 12 (4)
Independence in activities of daily living (prefracture)
Independents 139 (46)
Need some help 111 (37)
Dependents 50 (17)
Exercise (+ 30 min / day) 170 (61)
Rarely 52 (18)
Several times 91 (32)
Many times 70 (25)
Frequently 69 (25)
House factors risk score
0 52 (17)
1 133 (44)
2 86 (29)
3 25 (8)
4 4 (1)
Compliance with alendronate, Ca and vitamin D treatment 57 (35)
Self-evaluation of health status
Good 37 (13)
Medium 85 (31)
Bad 154 (56)
Pain 87 (53)
Concerning comorbidities, in the neurological affections category we included: Parkinson’s disease, paralyses, dementia (Alzheimer’s disease, cerebral atherosclerosis, confusive state), stroke, depressions and epilepsy. The rheumatic affections category include rheumatoid arthritis and arthrosis. Hyperthyroidism, hyperparathyroidism, Cushing’s syndrome, hypothyroidism, diabetes mellitus, endocrine causes of nulliparity are included in the category of endocrine diseases. In the digestive affections category we included hepatic diseases (chronic hepatitis, cirrhosis), intestinal affections that generates malabsorbtion, gastric affections (gastric and duodenal ulcer, gastroesophageal reflux disease). Cardiovascular diseases are represented by arterial hypertension, chronic heart failure, chronic ischemic cardiomyopathy, cardiac arrhythmia. Pulmonary diseases are represented by chronic obstructive bronchopneumopathy, bronchic astma, pulmonary tuberculosis, and neoplasia category include prostate cancer, cervical cancer, breast cancer, lung cancer and leukemias.
In the independents category we included those patients who were able to perform alone basic activities of daily living (dressing, bathing, feeding, going to the toilet), instrumental activities such as shopping, cooking, cleaning, using public transportation, housework. The patients who are able to perform activities of daily living, but need assistance to ambulate in the community, were included in the middle category (need some help). As dependents we classified those who can’t perform activities of daily living.
In the analysis of living conditions, concerning the refracture risk we regarded the following characteristics: poor illumination of the house, high aprons between rooms, slippery floor and abrupt staircase without rail.
To determine the frequency of the falls we asked the patients to quantify de number of falls suffered in the last 12 months before the hip fracture on a scale with 4 degrees : rarely or none, several times, many times, frequently. From the clinical point of view we were interested in sorting the frequency of the falls between „a fall” (rarely) and „recurrent falls” (several times, many times, frequently).
By the compliance with osteoporosis treatment we understood the persistence for one year of alendronate, Calcium supplements and vitamin D treatment. This indicator was measured 1 year after the hip fracture. Also after 1 year the patients were asked if they experienced acute pain in the postfracture period.
Our study has some methodological limitations. Even if the period of time in this study is relatively long (6 years), the number of patients that could be followed for a long period is quite small because some missed their follow-ups, and others couldn’t be reached and so some indicators were excluded from the study. Therefore, one year after the fracture just a little over half patients remained (165), and this fact affects the refracture prognosis. Further, the results obtained must be interpreted with prudence because not all of the statistical indicators are significant for the level of significance p<0.05, and the data can’t be extrapolated to the national population over 60 in Romania because the sample analyzed by us isn’t national representative. Another fact that we must consider is the fact that we don’t have a control group, and the situation of the patients with osteoporosis and fragility fractures as complications can’t be compared with that of other patients who have the same characteristics, but don’t suffer from this disease.
6 of the 41 patients in our study (14%) have suffered a new fracture, 16 (39%) have suffered a subsequent fracture within 12 months from the first fracture, the number of fractures gradually decreasing in time. The cumulative incidence of subsequent fractures was 7,4% at one year, 16 % at 2 years from the first hip fracture, the median value of refracture being 49 months (CI 95% 32,2-65,8). The risk of a subsequent fracture is elevated even 3 years after the first fracture, the patients who didn’t suffered a fracture within the first 36 months have a 7,8% chance of suffering a fracture within 37 months (see chart 1).
Chart 1. Kaplan-Meier survival curve for recurrent fractures
There are no statistically significant difference (Log Rank λ²=0,349, df=1, p=.554) between women and men concerning the follow-up time until a new fracture. However, men apparently have a shorter refracture time than women, the median time of refracture being 47 months for men and 49 months for women. The cumulative incidence of refracture in men is 7% at one year, 10% at 2 years and 22 % at 3 years from the first fracture. In women, the cumulative incidence of refracture is 6,8% at one year, 14,7% at 2 years and 30,7 at 3 years (see chart 2). The survival charts showed that the cumulative incidence of refracture at 6 years is lower in men (22,4%) than women (68%). Analyzing the refracture risk in genders, we notice that men have a greater risk of a new fracture, even though there are no statically significant difference (in men OR=1,19, CI 95% 0,51-2,7).
Chart 2. Kaplan-Meier survival curve for recurrent fractures by gender
There is a significant difference of the refracture time in age groups (Log Rank, =11,06, df=2, p<.005). The patients over 80 years have the shortest refracture time (median time 18 months), the other age groups (60-69 years, 70-79 years) having a longer refracture time (49 months).
Analyzing the cumulative incidence of refracture in age groups, we notice that 2% of the 60-69 years group refracture at one year, 4,2% at 2 years and 9,6% at 3 years from the first fracture. The cumulative incidence of refracture in the next age group (70-79 years) is higher, approximately 4,7% at one year, 17,5% at 2 years. For the age group over 80 years the cumulative incidence of refracture is clearly higher than the other age groups, approximately 17,7% at one year (see chart 3).
Chart 3. Kaplan-Meier survival curve for recurrent fractures by age
The risk of new fracture is higher in younger patients, the 60-69 years age group have a 1,6 (CI 95%, 0,61-4,18) greater chance of refracture than the patients over 80 years, the 70-79 years group having a 1,9 (CI 95% 0,8-3,8) greater chance of refracture. If we consider gender in our analysis, we can say that men over 80 years have a greater risk of refracture than the other age groups. In women the situation is inverted, the women in the 60-69 years age group (OR=3,1, CI 95% 1,2-8,8) and 70-79 years (OR=1,6,CI 95% 0,7-3,7) have a greater refracture risk than those of 80 years or more.
The refracture time has no significant differences with the surgical treatment of the fractures (Log Rank, λ²=7,29, df=4, p>.005). It seems that the femoral neck fractures treated with internal fixation have the longest refracture time (median time 42 months), the greatest hazard (2,4%) being in the first 6-12 months, and after 30 months is stable around 3,3%. The cumulative incidence for the femoral neck fractures treated with internal fixation is 3% at one year and reaches 7% at 2 years from the first fracture (see chart 4).
Chart 4. Kaplan-Meier survival curve – refracture for types of fracture surgically treated
The patients with femoral neck fracture treated with unipolar hemiarthroplasty have the shortest refracture time (median time 36 months), the greatest risk of 30% being in the first 6 months and the risk is kept elevated for 18 months (20%). The cumulative incidence for the femoral neck fractures treated with unipolar hemiarthroplasty is 6% at one year and 20% at 2 years from the first fracture.
The patients with femoral neck fractures treated with internal fixation have the lowest risk of refracture (OR=0,67 CI 95% 0,2-1,9). Related to this type of fracture, we notice that all the other types of fracture surgically treated have a greater refracture risk. The femoral neck fractures treated with unipolar hemiarthroplasty have a 1,6 (CI 95% 0,4-6,1) times greater risk than those treated with internal fixation. Concerning the subtrochanteric fractures, the refracture risk is 1,9 (CI 95% 0,5-7,1) times greater than femoral neck fractures treated with internal fixation. The refracture risk for those patients who suffered an intertrochanteric or a femoral neck fracture treated with total hip replacement or bipolar hemiarthroplasty is 1,3 (CI 95% 0,4-4,1) and 1,6 times greater (CI 95% 0,4-6,1) than the femoral neck fractures treated with internal fixation.
In the assessment of the subsequent fracture risk we considered the following factors : age, exercise, associated affections, T-score, functional dependence, recurrent falls, medical postoperative complications, self-evaluation of the health status, pain, risk factors in the house, osteoporosis treatment (see table 2). The types of fracture surgically treated were excluded because they are affected by the multicollinearity. From the table we notice that the gender and T-score variables are significant (p<0.05), and that osteoporosis treatment, even if it’s not significant, it has a high level of significance (p=0.07). The other variables must be interpreted with prudence because they don’t have statistical satisfactory significances (p>0.05).
Table 2. Prediction model of recurrent fractures - COX proportional hazard regression
95,0% CI for Exp(B)
Gender(female) -1,698 ,006 ,183 ,054 ,616
Age ,060 ,166 1,062 ,975 1,155
T-score -1,484 ,002 ,227 ,087 ,590
Exercise (<30 min./day) -,454 ,432 ,635 ,205 1,970
House risk factors ,246 ,284 1,279 ,815 2,008
Recurrent falls prefracture ,748 ,271 2,112 ,557 8,007
Cardiovascular diseases ,711 ,178 2,035 ,724 5,717
Rheumatoid arthritis ,360 ,768 1,434 ,130 15,763
Stroke ,508 ,443 1,662 ,454 6,089
Medical complications ,946 ,298 2,574 ,435 15,249
Dependent prefracture ,332 ,662 1,393 ,315 6,159
Alendronate, Ca and vit. D treatment -1,523 ,073 ,218 ,041 1,153
Pain ,262 ,559 1,300 ,539 3,135
Self-evaluation of health status (bad) ,121 ,793 1,129 ,458 2,785
As we already showed, men have a higher new fracture risk than women. Overseeing the other variables of the model, an increase of age by 1 year increases the new fracture hazard by 6,2% (CI 95% 2,5%-15,5%).
In the group studied, the T-score is the most important factor in predicting a subsequent fracture, as the value of the T-score decreases, the chances of a new fracture increase. Keeping in check the other variables, an increase by 1 of the T-score decreases hazard with approximately 77%. In other words, those with a low T-score, have a shorter time till a subsequent fracture.
Cardiovascular diseases, rheumatoid arthritis, stroke are the associated diseases who decrease the time period before a subsequent fracture. The hazard ratio is also elevated (HR=2,6, CI 95% 0,44-15,2) in the patients with medical postfracture complications.
The lack of exercise reduces the refracture risk with approximately 37%. This fact might be due to the decrease of exposure to falls which is a risk factor for refracture. The patients with multiple falls in the last 12 months before the hip fracture have a shorter time till refracture in comparison with those who fell rarely. Also, inappropriate fitted houses for elderly (slippery floor, inappropriate illumination, high aprons) increase de probability of refracture by 28%.
Optimism, self-evaluation of health status, confidence and patient’s motivation in overcoming the difficult situation, in association with the emotional support of the family decrease the risk of a new fracture. The patients who experienced postfracture pain are more likely to refracture in a shorter time than those who didn’t.
The alendronate, Calcium supplements and vitamin D treatment represents a refracture protection factor. If all the variables in the COX proportional hazard regression are kept constant, osteoporosis treatment increases the time before a new fragility fracture with approximately 78%.
This results obtained by us are highlighted by other studies who show that the relative risk of hip fracture rises with age, with the risk of a second fracture being up to 6 times greater after the first hip fracture, and the risk of non-hip fractures also being greater (9–50 times) than would be anticipated (Colon-Emeric C.S. şi colab., 2000; Klotzbuecher C.M. et al. 2000), so the elderly previous ‘fracturers’ potentially benefit the most from treatment. DXA has also been shown to be a predictor of survival that is better than blood pressure or cholesterol measurements (Johansson C. et al., 1998), and yet the test is still underutilized. The apparent ineffectiveness of clinical pathways in reducing mortality or discharge destination (March L.M. et al., 2000) has led some people to believe that treating elderly people with osteoporotic hip fractures is futile.
The alendronate and vitamin D and Calcium supplements treatment diminishes the risk of a subsequent fracture with approximately 7%. The patients with one year alendronate treatment have suffered refracture later than those with no osteoporosis treatment (see chart 5). The median refracture time of the patients with good treatment compliance is significantly greater (48 months) than those patients whom discontinued treatment (38 months). The cumulative incidence of refracture in patients with no osteoporosis treatment is 10,6% at one year and 17,2% at 2 years after the first hip fracture. In patients with good treatment compliance, the cumulative incidence of refracture was 10% at 2 years. Long-time effects of osteoporosis treatment couldn’t be studied because enough patients were lost from tracking, but it can be said that the alendronate, Calcium supplements and vitamin D one year treatment immediately after a hip fracture has positive effects in preventing a subsequent fracture.
Chart 5. Kaplan-Meier survival curve for recurrent fractures by compliance with osteoporotic treatment
As we mentioned already, the T-score is the most important factor in predicting a subsequent fracture, that’s why we are interested in observing the effects of alendronate, Ca supplements and vitamin D treatment on bone mass. In our case, the comparative analysis of the mean values for T-score at one year after the fracture in patients treated with alendronate, Calcium supplements and vitamin D, and in patients whom discontinued treatment or only took supplements, we found that the best T-score is recorded in patients with good compliance in alendronate treatment (average T = - 2,63). The lowest T-score is observed in those patients whom didn’t take any treatment (average T = -3,68).
The alendronate effects on bone mass are undeniable, for 88% of the patients who persisted closely with this treatment for 12 months T-score was improved, for 8% the T-score didn’t change and in 2% of the cases the T-score worsened (association statistically significant between alendronate treatment and T-score evolution, p<.000). Alendronate combined with Calcium supplements and vitamin D determine an average T-score with 1,05 higher than those patients whit no osteoporosis treatment, and with an average T-score higher with 0,44 than the patients treated with just Calcium supplements and vitamin D (see table 3).
Calcium and vitamin D treatment has also positive effects on bone mass, even though weaker than the combination with alendronate, an augmentation of the T-score appearing in only 65% of the patients whit this treatment (association statistically significant, p<.000). Relative to the patients with no osteoporosis treatment, the average T-score is higher with 0,60 in patients treated with Calcium supplements and vitamin D (see table 3).
Table 3 T-score at 1 year postfracture - Tukey HSD
treatment (J) Osteoporosis
treatment Mean Difference (I-J) 95% Confidence Interval
No treatment Alendronate Ca and vit. D -1.046 (*) -1.24 -.85
Ca and vit. D -.603 (*) -.85 -.34
Alendronate,Ca and vit. D No treatment 1.046 (*) .85 1.24
Ca and vit. D .443 (*) .17 .70
Ca and vit. D No treatment .603 (*) .34 .85
Alendronate Ca and vit. D -.443 (*) -.70 -.17
* The mean difference is significant at the .05 level.
Hip fractures are associated with high mortality. At one year after the first fracture in our study left 248 patients (83%), the rest being lost during the follow-up. In the first year after the hip fracture 75 patients died (25%), 17 men and 58 women. In the first 6 months the mortality is the highest, 31 deaths occur in this time, and the percentage on death during hospitalization being 2%. The median survival time is 35 months (CI 95% 28,6-41,4), which means that half of the studied group survives 2,7 years after the hip fracture. The cumulative incidence of mortality is 2% at one month, 5,4% at 3 months, 17,5% at 6 months, 27,4% at one year and 36,8% at 2 year after the fracture. The highest mortality risk (3,1%) is between 6-12 months postfracture, but is maintained elevated (2,8) even after 36 months postfracture (see chart 6).
Chart 6. Kaplan-Meier survival curve for hip fractures
The patients who suffered a subsequent fracture have a shorter survival time than those who didn’t (Log Rank, λ²=4,5, df=1, p=0,034). Half of the patients whom suffered a new fracture survived approximately 32 months, and those who didn’t suffer a new fracture have an average survival time of 45 months. The cumulative incidence of deaths in patients whom suffered refracture was 31 % at 1 year. Analyzing the age group and the subsequent refracture, we observe that the highest mortality risk (9%) for the younger (age groups 60-69 and 70-79 years) is in the interval 3-4 years, the patients over 80 years have a mortality risk of 17% within the first 2 years from de first fracture.
It’s difficult to assign this rise of the refracture’s and hip fracture’s mortality because this patients are in the high-risk mortality group for other causes. Most of the patients treated for hip fracture had a poor health status, suffering from other diseases, with limited mobility and low functional independence. All this features are difficult to control and some of them don’t offer a high accuracy of the measuring attached to the limits of this study.
In conclusion, the incidence of the osteoporotic subsequent hip fractures is higher immediately after the fracture. The cumulative incidence of subsequent fractures is 7,4% at one year, 16% at 2 years from the first hip fracture, and half of the patients refracture before 49 months after the first hip fracture. In women, the cumulative incidence of refracture is 6,8% at one year, 14,7% at 2 years and 30,7 at 3 years (see chart 2). The survival charts showed that the cumulative incidence of refracture at 6 years is lower in men (22,4%) than women (68%). Analyzing the refracture risk in genders, there are no statiscally significant differences. Instead there is a significant difference of refracture time in age groups. The patients over 80 years have the shortest refracture time (median time 18 months), for the other age groups (60-69 years, 70-79 years) median time is elevated, approximately 49 months.
The risk of new fracture is higher in younger patients, the 60-69 years age group have a 1,6 (CI 95%, 0,61-4,18) greater chance of refracture than the patients over 80 years, the 70-79 years group having a 1,9 (CI 95% 0,8-3,8) greater chance of refracture than the patients over 80. Even though we can’t sustain the assumption that younger patients have a greater refracture risk than the elderly (p>0.05), despite the fact that the refracture risk rises with age, we think that this hypothesis should be considered in future research.
The refracture time has no significant differences with the type of fracture surgically treated. It seems that the femoral neck fractures treated with internal fixation have the lowest refracture rate in comparison with the other fracture types surgically treated.
Compliance with osteoporosis treatment diminishes the risk of a subsequent fracture with apx. 7%. The patients with one year alendronate treatment have suffered refracture later than those with no osteoporosis treatment. Alendronate combined with Calcium supplements and vitamin D determine after one year an average T-score with 1,05 higher than those patients whit no osteoporosis treatment.
Hip fractures are associated with high mortality. In the first year after the hip fracture 75 patients died (25%), 17 men and 58 women. The cumulative incidence of mortality is 2% at one month, 5,4% at 3 months, 17,5% at 6 months, 27,4% at one year and 36,8% at 2 year after the fracture. The highest mortality risk (3,1%) is between 6-12 months, but the risk is elevated (2,8%) even after 36 months postfracture. The median survival time is 35 months (CI 95% 28,6-41,4), which means that half of the studied group survives 2,7 years after the hip fracture. The patients with recurrent fractures have a greater mortality risk than those who didn’t suffer a new fracture. The cumulative incidence of deaths in patients who refractured is 31% at one year.
This being said, prevention of the first hip fracture is ideal, but at least after the first fracture the awareness of the physicians should be heightened to prevent future fracture by modifying the risks, which must include optimal management of other medical diagnoses as well as fall prevention (Slemenda C., 1997) and the use of hip protector pads.
Michel et al. (2000) have recently identified typical patient profiles in an attempt to elect those patients who will benefit the most from specific management strategies. In addition, ore research to quantify fracture risk, identify the best treatment for specific subgroups (men, ethnic minorities) and provider education is needed. But this education must be seen to translate into changes in practice to truly affect the management of this devastating disease
Angelescu N. (2001) Tratat de patologie chirurgicală Volumul II , Editura Medicală, Bucureşti,
Antonescu D.M, (2006) Patologia aparatului locomotor- Volumul I, Editura Medicală, Bucureşti
Bass E., French D.D., Bradham D.D., Rubenstein L.Z. (2007) Risk adjusted mortality rates of elderly veterans with hip fractures. Ann. Epidemiol. 17:514–519
Berry S.D., Samelson E.J., Hannan M.T., McLean R.R., Lu M., Cupples L.A., Shaffer M.L., Beiser A.L., Kelly-Hayes M., Kiel D.P. (2007) Second hip fracture in older men and women. The Framingham Study. Arch. Intern. Med. 167:1971–1976
Brainsky A., Glick H., Lydick E., Epstein R., Fox K.M., Hawkes W., Kashner T.M., Zimmerman S.I., Magaziner J. (1997) The economic cost of hip fractures in community-dwelling older adults: a prospective study. J. Am. Geriatr. Soc. 45:281–287
Budică C., Kinetoterapia în Afecţiunile Ortopedico-Traumatice Editura Fundaţiei România De Mâine, Bucureşti, 2005
Colon-Emeric C., Kuchibhatla M., Pieper C., Hawkes W., Fredman L., Magaziner J., Zimmerman S., Lyles K.W. (2003) The contribution of hip fracture to risk of subsequent fractures: data from two longitudinal studies. Osteoporos. Int. 14:879–883
Colon-Emeric C.S., Sloane R., Hawkes W.G., et al. (2000) The risk of subsequent fractures in community-dwelling men and male veterans with hip fracture. Am. J. Med. 109:324–326
Davis J.C., Ashe M.C., Guy P., Khan K.M. (2006) Undertreatment after hip fracture: a retrospective study of osteoporosis overlooked. J. Am. Geriatr. Soc. 54:1019–1020
Fundaţia Internaţională de Osteoporoză, Staying power: Closing the adherence gap in osteoporosis, 2008 http://www.iofbonehealth.org/publications/the-adherence-gap.html, 15 dec. 2008.
Hannan E.L., Magaziner J., Wang J.J., Eastwood E.A., Silberzweig S.B., Gilbert M., Morrison R.S., McLaughlin M.A., Orosz G.M., Siu A.L. (2001) Mortality and locomotion 6 months after hospitalization for hip fracture: risk factors and risk-adjusted hospital outcomes. JAMA 285:2736–2742
Hooven F., Gehlbach S.H., Pekow P., Bertone E., Benjamin E. (2005) Follow-up treatment for osteoporosis after fracture. Osteoporos. Int. 16:296–301
Johansson C., Black D., Johnell O. et al. (1998) Bone mineral density as a predictor of survival. Calcif Tissue Int 63:190–196
Kane R.L., Chen Q., Finch M., Blewett ., Burns R., Moskowitz M.(1998) Functional outcomes of posthospital care for stroke and hip fracture patients under medicare. J. Am. Geriatr. Soc. 46:1525–1533
Karlsson M., Nilsson J.A., Sernbo I., Redlund-Johnell I., Johnell O., Obrant K.J. (1996) Changes of bone mineral mass and soft tissue composition after hip fracture. Bone 18:19–22
Kiebzak G.M., Beinart G.A., Perser K., Ambrose C.G., Siff S.J., Heggeness M.H. (2002) Undertreatment of osteoporosis in men with hip fracture. Arch. Intern. Med. 162:2217–2222
Klotzbuecher C.M., Ross P.D., Landsman P.B., et al. (2000) Patients with prior fractures have an increased risk of future fractures: a summary of the literature and statistical synthesis. J. Bone Miner. Res. 15:721–739
Lönnroos E., Kautiainen H., Karppi P., Hartikainen S., Kiviranta I., Sulkava R. (2007) Incidence of second hip fractures. A population based study. Osteoporos. Int. 18:1279–1285
March L.M., Cameron I.D., Cumming R.G. et al. (2000) Mortality and morbidity after hip fracture: can evidence based clinical pathways make a difference? J. Rheumatol. 27:2227–2231
Melton L.J. III (2000) Who has osteoporosis? A conflict between clinical and public health perspectives. J. Bone Miner. Res. 15:2309–2314
Michel J.P., Hoffmeyer P., Klopfenstein C. et al. (2000) Prognosis of functional recovery 1 year after hip fracture: typical patient profiles through cluster analysis. J. Gerontol. 55A:M508–515
Panait Gh.(2002), Ortopedie-Traumatologie Practică, Editura Publistar, Bucureşti, ,
Panait Gh., Budică C., Panait A., Necula R.(2009) Mineralizarea biologică şi adaptativă biomecanică a oaselor. Osteoporoza Editura Publistar, Bucureşti, 2009
Randell A.G., Nguyen T.V., Bhalerao N., Silverman S.L., Sambrook P.N., Eisman J.A. (2000) Deterioration in quality of life following hip fracture: a prospective study. Osteoporos Int. 11:460–466
Slemenda C. (1997) Prevention of hip fractures: risk factor modification. Am. J. Med. 103:S65–73
Terry Canale, S., CAMPBELL’S OPERATIVE ORTHOPAEDICS 9th ed., Mosby-Year Book, Inc., 1998
Van Staa T.P., Leufkens H.G., Cooper C. (2002) Does a fracture at one site predict later fractures at other sites? A British cohort study. Osteoporos. Int. 13:624–629
Wheeless ' Textbook of Orthopaedics, Data Trace Publishing Company,2005
Wiss Donald A. (2006) Master Techniques in Orthopaedic Surgery: Fractures, 2nd Edition, Lippincott Williams & Wilkins