Introduction
Hip fracture is common among the elderly population, with approximately 1.6 million cases a year worldwide.Reference Johnell and Kanis1 Despite the introduction of primary prevention programmes in Western countries, the incidence of hip fractures continues to increase because of the ageing of the population.Reference Pasco, Sanders, Hoekstra, Henry, Nicholson and Kotowicz2 In this respect, some estimates indicate that the incidence will have doubled by the year 2040.Reference Zuckerman3
Hip fracture is associated with important morbidity and mortality.Reference Roche, Wenn and Sahota4 The in-hospital mortality rate varies between 2.3 and13.9%,Reference Rapp, Becker, Lamb, Icks and Klenk5–Reference Abrahamsen, van Staa, Ariely, Olson and Cooper7 and reaches 12–36% six months after hospital discharge.Reference Abrahamsen, van Staa, Ariely, Olson and Cooper7–Reference Bentler, Liu, Obrizan and Cook9 Mortality due to hip fracture represents 1.5% of all deaths in the world population over 50 years of age.Reference Kanis, Oden and Johnell10
The increased mortality rate is due not only to hip fracture but also to the age of the population – with over 75% of all cases occurring in people over 70 years of age.Reference Johnell and Kanis1 These patients moreover have important associated co-morbidities.Reference Tarazona-Santabalbina, Belenguer-Varea and Rovira-Daudi8, Reference Nikkel, Fox and Black11 It is this added disease burden, rather than fracture alone, which produces an increased risk of mortality during hospital admission and in the 12 months after hospital discharge. This situation is partly due to the increased risk of in-hospital complications associated with patients of this kind. Roche et al.Reference Roche, Wenn and Sahota4 found the risk of post-operative complications in older patients to be proportional to the number of co-morbid conditions present at the time of admission.
One of the most frequent complications is the development of delirium, characterized by alterations in consciousness, orientation and perception. These circumstances in turn result in temporary cognitive impairment. The incidence of peri-operative delirium varies greatly from 5 to 61%, depending on the literature source.Reference Edelstein, Aharonoff, Karp, Capla, Zuckerman and Koval12, Reference Gustafson, Berggren and Brännström13
On the other hand, from the clinical perspective, and despite the adoption of rehabilitation measures, hip fracture patients suffer considerable functional impairment compared with the pre-fracture situation, with a decrease in the number of individuals who are able to walk without help.Reference Alarcón and González-Montalvo14, Reference Magaziner, Hawkes and Hebel15 In this respect, approximately one-half of all patients recover their independence for activities of daily living, while only one-third recover the same independence in instrumental activities as before fracture. The consequences are poorer quality of life and increased health and social care costs.
Cognitive impairment increases in incidence after the sixth decade of life, with prevalence of dementia increasing from 2.4% at 65 years of age to 34.5% by the age of 85 years.Reference Pressley, Trott and Tang16 An important factor underlying this high prevalence is the increased life expectancy of cognitively impaired patients.Reference Sloane, Zimmerman and Suchindran17 Apart from the problems inherent to cognitive impairment and its associated health and social care costs, it is also implicated in the aetiology of bone fractures in older patients,Reference Buchner and Larson18 with the risk of low-impact falls and of subsequent injuries in such patients increasing 5- and 2.2-fold, respectively, compared with individuals without cognitive impairment.Reference Tinetti, Speechley and Ginter19 The standard of hospital care of geriatric patients with hip fracture has been described as an independent risk factor for poorer functional outcomes,Reference Matsueda and Ishii20 increased immobilization and prolonged or permanent confinement to bed, an increased risk of being unable to walk again with or without help,Reference Parker and Palmer21 an increased risk of institutionalization,Reference Schaller, Sidelnikov and Theiler22 and greater mortality both in hospital and in the 12 months after discharge.Reference Bentler, Liu, Obrizan and Cook9, Reference Soderqvist, Miedel and Ponzer23, Reference Givens, Sanft and Marcantonio24
Cognitive impairment is characterized by different degrees of severity, which in turn are correlated to the vital and functional prognosis.Reference Tarazona-Santabalbina, Belenguer-Varea and Rovira Daudi25 Thus increased severity of cognitive impairment has been associated with increased mortalityReference Bentler, Liu, Obrizan and Cook9, Reference Kyo, Takaoka and Ono26 and poorer functional recovery.Reference Soderqvist, Miedel and Ponzer23, Reference Takayama, Iki and Kusaka27
In view of the complexity of the hospital management of older patients with hip fracture, specific care units jointly supervised by geriatricians and traumatologists were first introduced in the late 1950s in Hastings and Stoke-on-Trent, in the United Kingdom.Reference Devas28, Reference Clark and Wainwright29 The initiative with the greatest impact was the so-called Hastings model, characterized by collaboration between the traumatology department undertaking surgery in one hospital with a geriatric care unit in another nearby centre. The most complex cases, from the clinical or functional perspective, were referred one week after surgery to the geriatric unit for a specialized rehabilitation programme lasting approximately 5 weeks.Reference Devas and MSJ30
Orthogeriatric units have progressed considerably since these first initiatives and the scientific literature clearly shows earlier geriatric intervention to be associated with improved outcomes.Reference Alarcón, González and Bárcena31 In addition, orthogeriatric units contribute to a reduced number of complications, an increase in the number of operated patients and shorter pre-surgery waiting times, shorter mean stay, improved functional performance at hospital discharge and over the long term, and improved survival rates.Reference González, Alarcón and Pallardo32 In view of the good results obtained, it is not surprising that the collaborative model has been widely adopted. Indeed, at present, 85% of all hospital centres in the United Kingdom have protocols for collaboration between geriatricians and traumatologists.Reference Wajeman, Sheard and Jenner33
A systematic review and first meta-analysis of the efficacy and efficiency of hip fracture management programmes was published in the year 2000,Reference Cameron, Crotty and Currie34, Reference Gillespie35 demonstrating increased functional recovery, a shortening of mean stay, fewer complications and a decrease in economic costs, derived from the fact that a larger number of patients were able to return home after hospital admission.
The prevalence of prior cognitive impairment in hospitalized patients is high, reaching 43% in one study.Reference Ringdal, Ringdal, Juliebø, Wyller, Hjermstad and Loge36 As mentioned above, these patients suffer an increased incidence of falls. A study involving patients admitted to six Spanish hospitals due to hip fracture found that those with a prior diagnosis of dementia had suffered a greater number of recurrent falls before the hip fracture leading to admission.Reference Formiga, Lopez-Soto and Duaso37
Following admission to hospital, patients with dementia and hip fracture present with a high incidence of delirium. Risk factors include older age, severity of functional and cognitive impairment before fracture, pre-existing cerebrovascular or degenerative disease, social isolation and communication difficulties.Reference Juliebø, Bjøro, Krogseth, Skovlund, Ranhoff and Wyller38 Of these factors, the presence of cognitive impairment (defined by a Mini-Mental State Examination (MMSE) score of ≤ 23) was found to be the variable most closely correlated to the incidence of peri-operative delirium.Reference Freter, George, Dunbar, Morrison, Macknight and Rockwood39 Furthermore, delirium is associated with increased severity of pre-existing co-morbidityReference Furlaneto and Garcez-Leme40 and increased mortality risk.Reference Freter, George, Dunbar, Morrison, Macknight and Rockwood39–Reference Chrispal, Mathews and Surekha41
Despite the high prevalence of cognitive impairment before admission and the high incidence of delirium, both conditions remain largely underdiagnosed. A cognitive assessment study of hip fracture patients carried out by nursing professionals underscored the need to periodically apply validated cognitive function assessment tools, due to the frequency of undiagnosis.Reference Söderqvist, Strömberg, Ponzer and Tidermark42 Cameron et al.Reference Cameron, Crotty and Currie34 compared the use of the MMSE for the detection of delirium versus the Confusion Assessment Method (CAM) (regarded as the DSM-IV gold standard) and discarded the use of the former instrument with a cut-off of ≥23, due to the large proportion of false-positive results.
There is a clear need for strategies to prevent or reduce the impact of delirium in these patients. Frequent and thorough geriatric assessment of patients with mental state alterations or cognitive problems is very important. The avoidance of triggering elements (e.g. a drop in blood pressure, malnutrition, peri-operative infections) and active treatment of the underlying factors, are priority concerns.Reference White, Khan and Smitham43 One of the most effective strategies comprises adequate post-operative analgesia, careful selection of the administered drugs and patient management by a multidisciplinary team. In this respect, an interventional study has shown post-operative care in a geriatric unit to reduce the incidence of delirium relative to the number of cases in a traumatology unit (54.9 versus 75.3%, P = 0.003). Furthermore, in the incident cases, geriatric care also shortened the duration of the symptoms.Reference Lundström, Olofsson and Stenvall44 In addition, geriatric care has also been found to reduce the number of complications such as pressure ulcers, nutritional disorders, urinary tract infections, sleep and waking disturbances and accidental falls.
A recent systematic review and meta-analysis of hip fractures in patients with cognitive impairmentReference Smith, Hameed, Cross, Sahota and Fox45 concluded that the NEECHAM Confusion Scale should be considered for the screening of patients for delirium. Furthermore, such screening would be most efficient if clinicians were to take into account the known risk factors for delirium.
However, whilst interdisciplinary geriatric assessment was found to reduce the incidence of delirium compared with the rate in a group not subjected to geriatric intervention (37.2 versus 53.2%, P = 0.04), it did not reduce the duration or severity of the symptoms once delirium was diagnosed.Reference Deschodt, Braes and Flamaing46
The data provided by some of the cited publicationsReference Cameron, Crotty and Currie34, Reference Ringdal, Ringdal, Juliebø, Wyller, Hjermstad and Loge36, Reference Furlaneto and Garcez-Leme40 advise training of nursing professionals in units that attend hip fracture patients in the use of tools for detecting delirium and for assessing the cognitive state.
Functional recovery
Loss of function associated with dementia in hip fracture patients can be recovered through specific post-surgery rehabilitation programmes. Morghen et al.Reference Morghen, Gentile, Ricci, Guerini, Bellelli and Trabucchi47 recorded less recovery of independent walking function with increasing severity of cognitive impairment. Specifically, at hospital discharge, 29.6% of the patients with moderate or severe cognitive impairment, 51.9% of those with mild cognitive impairment and 78.6% of those without cognitive impairment recovered the ability to walk. One year after discharge, these percentages were found to be 57.1, 57.7 and 78.9%, respectively. The capacity to recover independent walking function in turn was correlated to an increased functional level before fracture (as estimated by the Barthel index), independent of the severity of cognitive impairment.
Savino et al.Reference Savino, Martini and Lauretani48 related handgrip strength at the time of hospital admission directly to cognitive function (as assessed by the Short Portable Mental Status Questionnaire, SPMSQ), activities of daily living and vitamin D levels, and inversely to age, depressive symptoms (as scored by the Geriatric Depression Scale), co-morbidity (evaluated with the Charlson Index) and the need for a caregiver. Patients with greatest handgrip strength (highest quartile) were more likely to recover the ability to walk than those in the lowest quartile, after adjusting for age, gender, co-morbidity, vitamin D levels, neuropsychological condition, functionality, surgery delay and rehabilitation time.
Our group has also published results relating to functional recovery at hospital discharge and one and six months after discharge.Reference Tarazona-Santabalbina, Belenguer-Varea and Rovira Daudi25 The percentage of patients with severe cognitive impairment who were able to walk a distance of at least 5 metres without help was 25% at discharge, 37% one month after discharge, and 41% after six months. The respective percentages were 44, 58 and 57% in the case of patients with moderate cognitive impairment, and 68, 82 and 84% in the case of those without cognitive impairment. At the time of hospital discharge, severe dementia, the Barthel score and age were significantly correlated with less walking capacity, while six months after discharge moderate dementia, severe dementia and the Barthel score were significantly correlated with poorer mobility.
One of the reasons for this decrease in functional recovery is the fear of further falls. The final multivariate model of a clinical study on the prevention and treatment of depression after hip fractureReference Oude Voshaar, Banerjee and Horan49 found cognitive status (as scored by the MMSE) and the fear of falls (evaluated by the Modified Falls Efficacy Scale), assessed 6 weeks after surgery, to be predictors of functional recovery.
The functional deterioration associated with hip fracture in patients with dementia increases the probability of institutionalization. One studyReference Harboun, Dorenlot and Cohen50 found that the presence of fracture in the three years before the diagnosis of dementia, the existence of acute congestive heart failure episodes and a body weight loss of over 5% over one year, were independently associated with the institutionalization of patients with dementia.
A percentage of patients classified as being cognitively intact (on the basis of MMSE score) may in fact be cognitively impaired. Hershkovitz et al.Reference Hershkovitz, Jacubovski, Alima Bot, Oshry and Brill51 performed the clock drawing test in patients with hip fracture and MMSE scores of > 23 points and divided the subjects into two groups according to whether the test was correctly executed or not. Those patients who failed to correctly perform the clock drawing test had lower scores in both the cognitive and motor dimensions of the Functional Independent Measure (FIM), as well as a longer hospital stay.
A systematic review carried out in 2009 on the impact of cognitive impairment upon the rehabilitation of geriatric patients admitted to hospital due to hip fracture concluded that there is evidence of the functional benefits of intensive rehabilitation therapy in these subjects after hip fracture repair surgery.Reference Muir and Yohannes52 Such benefits may prove similar to those observed in patients without cognitive impairment. However, there is still not enough information to recommend specific physiotherapeutic interventions in these patients with a view to improving functional outcomes.
Mortality
Cognitive impairment increases the risk of death during hospital admission due to hip fracture and subsequent follow-up. Hershkovitz et al.Reference Hershkovitz, Polatov, Beloosesky and Brill53 analysed the factors associated with mortality over the 24 months after hospital discharge among patients admitted due to hip fracture. The authors found that 26.4% of the patients with dementia had died at the end of the follow-up period, versus only 6.5% of those without cognitive impairment (P < 0.001). Blood albumin levels of < 3.6 g/dl, a FIM score of < 40 points and a patient age > 84 years were also related to increased mortality. In the multivariate analysis, however, only dementia and age were identified as independent mortality risk factors.
Bellelli et al.Reference Bellelli, Frisoni, Turco and Trabucchi54 found patients with dementia and depression (diagnosed by the 15-item Geriatric Depression Scale) to suffer greater mortality than depressed patients without dementia, or patients with dementia but no depression.
In a previous studyReference Tarazona-Santabalbina, Belenguer-Varea and Rovira Daudi25 we recorded increased mortality to be associated with increasing severity of dementia. Patients without cognitive impairment had mortality rates of 4, 7, 15 and 21%, respectively at the four study time points (upon admission and one, six and 12 months after hospital discharge), while those with mild dementia had mortality rates of 5, 8, 18 and 32%, those with moderate dementia had rates of 8, 13, 26 and 46 %, and those with severe dementia had rated of 10, 15, 35 and 53%. The survival analysis after 12 months showed the severe dementia group to present the highest mortality risk (hazard ratio (HR) 1.64; 95% confidence interval (95% CI) 1.02–2.62; P = 0.041) (Figure 1). Patient age, Barthel index score, male gender and the Charlson score were also significantly associated with death.
Survival after hip fracture according to the severity of cognitive impairment Reproduced from: Tarazona-Santabalbina FJ, Belenguer-Varea A, Rovira-Daudi E et al. Severity of cognitive impairment as a prognostic factor for mortality and functional recovery of geriatric patients with hip fracture. Geriatr Gerontol Int (in press).
Evaluations of orthogeriatric units
The heterogeneity of the studies that have attempted to evaluate the efficacy of orthogeriatric units in caring for hip fracture patients with cognitive impairment limits the conclusions that can be drawn. They differ significantly in terms of the characteristics of the interventions, delivering multidisciplinary and/or multicomponent care, conducted in geriatric or rehabilitation areas, involving different intensities of physiotherapy and/or occupational therapy and focusing on management of multiple morbidities, or delirium, or multiple drug therapies, etc. The management of patients in the control groups also differs, ‘usual care’ being the term used in some orthopaedic units, while other studies describe rehabilitation programmes that only differ from those applied to the experimental group in specific aspects related to dementia care. Some studies include institutionalized patients and not all publications stratify the subjects according to the degree of cognitive impairment – a fact that complicates comparisons between the different studies. Nevertheless, there does appear to be general agreement in the literature on the benefits of multidisciplinary and multicomponent care specifically targeted to hip fracture patients with cognitive impairment. These benefits comprise better outcomes in terms of functional recovery, walking and falls, and with less need for institutionalization.Reference Huusko, Karppi, Avikainen, Kautiainen and Sulkava55–Reference McGilton, Davis and Naglie61 Whilst the benefits are lost after four months according to some studies,Reference Huusko, Karppi, Avikainen, Kautiainen and Sulkava55 others describe persisting rehabilitation benefits at 12 months,Reference Stenvall, Berggren, Lundström, Gustafson and Olofsson56 even in patients with severe cognitive impairment.Reference Morghen, Gentile, Ricci, Guerini, Bellelli and Trabucchi47 In one study in which the intervention was continued for three months after discharge, benefits in terms of walking capacity and physical function were observed over two years.Reference Shyu, Tsai and Chen62
A recent systematic review concluded that the benefits of intervention are clear in patients with mild to moderate cognitive impairment, but was unable to draw firm conclusions relating to patients with moderate to severe cognitive impairment, or institutionalized subjects.Reference Allen, Koziak, Buddingh, Liang, Buckingham and Beaupre63
Although some authors have reported significantly fewer deaths following interdisciplinary inpatient care,Reference Naglie, Tansey and Kirkland58 the literature offers conflicting findings. In a recent article involving 45,602 elderly patients with hip fracture, Seitz et al.Reference Seitz, Gill and Gruneir64 showed the presence of dementia (in 23.9% of the total) to be related to a significant increase in mortality. However, a diagnosis of dementia was not necessary for this to occur, since mild to moderate cognitive impairment was also correlated with increased mortality, along with male gender and vitamin D deficiency.Reference Schaller, Sidelnikov and Theiler65
Few clinical trials have explored ways to increase survival among hip fracture patients with cognitive impairment. A sub-analysis of the Horizon study has demonstrated an increase in survival after 12 months in hip fracture patients without cognitive impairment treated with zoledronic acid. However, no such effect was noted in patients with cognitive impairment.Reference Prieto-Alhambra, Judge, Arden, Cooper, Lyles and Javaid66 The Oslo Orthogeriatrics Trial recruited 332 hip fracture patients in the period between 17 September 2009 and 5 January 2012, and was designed to offer an orthogeriatric assessment of hip fracture patients, with the inclusion of mortality after four and 12 months as study end-points. The future publication of the results of this trial may help clarify the possible efficacy of orthogeriatric assessment upon the survival of patients with cognitive impairment.Reference Wyller, Watne and Torbergsen67 The CODE study likewise has been designed to investigate the relationship between depression, delirium and dementia and the outcomes of hip fracture management.
Gaps in knowledge
A systematic review has underscored the fact that patients with dementia are largely excluded from studies of hip fracture patient management, since they are unable to correctly answer quality of life questionnaires.Reference Hebert-Davies, Laflamme and Rouleau69 Indeed, two of the most important multicentre trials (FAITH and HEALTH) excluded 22.2 and 36% of the subjects, respectively, due to a diagnosis of dementia. The systematic review of intervention programmes found that, of 17 selected publications, only two compared patient groups with dementia versus groups without dementia. Both of these studies reported significant increases in mortality and complications among the patients with dementia versus the patients without cognitive impairment.
It has been commented above that functional recovery is poorer with increasing severity of dementia. In this respect, a recent study has found that failure to recover was three times more likely in patients with dementia than in individuals without dementia.Reference Bentler, Liu and Obrizan70 The functional condition of the patient before fracture is more important than cognitive level as a predictor of functional recovery. In fact, cognitive impairment does not affect function provided the patients are referred to a rehabilitation unit. Such benefit derived from specific functional rehabilitation programmes in patients with cognitive impairment has been documented in the literature,Reference Al-Ani, Flodin and Söderqvist57 though recovery is found to be slower than in patients without cognitive impairment.Reference Wyller, Watne and Torbergsen67 The recent systematic reviewReference Hershkovitz, Kalandariov and Hermush71 underscores the high probability of functional improvement in patients with cognitive impairment after participating in acute or sub-acute hospital rehabilitation programmes. Such programmes may take longer to implement than in patients without cognitive impairment, but are nevertheless able to maintain the benefits in terms of recovery for up to 12 months. However, no cost-effectiveness studies of specific functional recovery programmes in hip fracture patients with dementia have been published to date.
There is still an important gap in knowledge regarding the types of interventions capable of offering better outcomes in terms of functional recovery and survival in hip fracture patients with dementia.
The future
The lack of results from randomized clinical trials regarding the advantages of a specific interdisciplinary intervention based on the principles of orthogeriatric care should serve to encourage such studies. These in turn should be designed to demonstrate possible reduction in the incidence of falls among patients with cognitive impairment who are institutionalized or live in the community, as well as the attainment of better functional recovery with lower mortality.
Although the benefits of systematic rehabilitation in hip fracture patients with dementia have been clearly established, further studies are needed to determine which orthogeriatric model affords the best results and whether such a model should be specifically designed for these patients or not. Future research should not be limited to confirming immediate treatment outcomes, but should also be able to identify and implement interventions and care models capable of improving patient survival and autonomy over the longer term.
Conflicts of interest
None.
